NeurIPS2026 Notes TODO¶
Total: 955 | Completed: 955 | Pending: 0
- 3d point splatting for mmwave radar novel view synthesis | arXiv: 2609.11894
- 3d-vital visibility-aware identity training from artificial liftings of 3d recon | arXiv: 2610.16692
- a benchmark for omni-modal reasoning in long videos | arXiv: 2512.16978
- a closer look at dynamic scene graph generation in the era of multimodal large l | arXiv: 2503.15846
- a free lunch in llm compression revisiting retraining after pruning | arXiv: 2510.14444
- a latent control model for realistic rodent motion | arXiv: 2610.27279
- a principled optimal transport framework for frame selection in long video under | arXiv: 2610.27692
- a steerable deep network for model-free diffusion mri registration | arXiv: 2501.04794
- a stratified multi-rater evaluation of llm-based virtual standardized patients w | arXiv: 2610.29312
- a systematic analysis of out-of-distribution detection under representation and | arXiv: 2511.11934
- a theoretical analysis of test-driven code generation | arXiv: 2602.06098
- a theoretical analysis of why masked diffusion models mitigate the reversal curs | arXiv: 2602.02133
- a theory of time-sensitive language generation sparse hallucination beats mode c | arXiv: 2605.11302
- a unified framework for image-to-3d part generation via variable granularity | arXiv: 2610.31671
- a unified uncertainty representation for graph neural networks via doubly-spectr | arXiv: Safety
- absolute state-wise constrained policy optimization high-probability state-wise | arXiv: 2410.01212
- accelerating rectified flow models via trajectory-aware caching | arXiv: 2605.16789
- accuracy vs accuracy computational tradeoffs between classification rates and ut | arXiv: 2505.16494
- acdc on a budget - alternating sparse phases | arXiv: 2610.27758
- action chunking proximal policy optimization with feedback correction | arXiv: Robotics
- actionable hallucination detection translating latent uncertainty into agentic c | arXiv: 2608.10430
- active learning of conditional generative models via the transport neural tangen | arXiv: 2610.35652
- adam under generalized smoothness with second-moment-type stochastic gradients | arXiv: 2609.37787
- adaptive compression and targeted perturbation a unified framework for generaliz | arXiv: 2610.16049
- adaptive inference for functionals of m-estimands | arXiv: 2609.39274
- adaptive mass-segmented kv compression for long-context reasoning | arXiv: 2605.23200
- advancing entropy-level credit assignment in rlvr via proximal entropy policy op | arXiv: 2609.39402
- advjudge-zero binary decision flips in llm-as-a-judge via adversarial control to | arXiv: 2512.17375
- aegis generative gradient masking for privacy-preserving medical federated learn | arXiv: 2609.38339
- aeon unifying video and 3d world models | arXiv: 2610.28631
- agenthop a diagnostic benchmark for agentic multi-hop scientific question answer | arXiv: 2609.34428
- align invert solving inverse problems with diffusion and flow-based models via r | arXiv: 2511.16870
- aligning llms with biomedical knowledge using balanced fine-tuning | arXiv: 2511.21075
- all roads lead to rome flow-driven multi-anchor exploration for open-environment | arXiv: 2609.36889
- alphapareto formulaic alpha discovery with llm-guided multi-objective reinforcem | arXiv: 2609.34188
- amortized linear-time exact shapley value for product-kernel methods | arXiv: 2505.16516
- amortized optimal transport from sliced potentials | arXiv: 2604.15114
- an exact information theory of generalization phase transitions in bayesian diff | arXiv: 2607.08041
- analyzing and guiding zero-shot posterior sampling in diffusion models | arXiv: 2602.07715
- anchor3dgs feed-forward 3d gaussian splatting with compact anchor-based represen | arXiv: 2610.21634
- anchoring adversarial trajectories to data manifolds a bilevel transfer optimiza | arXiv: 2609.38991
- anyedit a unified framework for speech and singing voice editing with real-world | arXiv: 2610.15665
- anyhand a large-scale synthetic dataset for rgb-d hand pose estimation | arXiv: 2603.25726
- anymo geometry-aware setup-agnostic modeling of human motion in the wild | arXiv: 2605.22715
- appsolver adaptive patch partitioning for point-wise ship flow prediction on uns | arXiv: 2608.29355
- are text-to-image models inductivist turkeys a counterfactual benchmark for caus | arXiv: 2606.24548
- are we really tilting the mechanics of reward guidance in flow and diffusion mod | arXiv: 2606.02884
- asirf an agentic framework for context-dependent sensitive information redaction | arXiv: Safety
- ask answer and detect role-playing llms for personality detection with question- | arXiv: 2512.08814
- ask answer and detect role-playing llms for personality detection with question- | arXiv: 2512.08814
- assessing the robustness of heterogeneous treatment effects in survival analysis | arXiv: 2510.13397
- asymmetric hierarchical anchoring for robust audiovisual cross-modal generalizat | arXiv: 2602.03570
- asymmetric invertible threat learning reversible privacy defense for face recogn | arXiv: 2605.01217
- asymptotically exact negative guidance of diffusion models via positive-unlabele | arXiv: 2610.19802
- at fulltilt real-time open-set 3d macromolecule detection directly from tilted 2 | arXiv: 2604.10766
- atom-level protein representation learning improves protein structure prediction | arXiv: 2605.22133
- atomic trajectory modeling with state space models for biomolecular dynamics | arXiv: 2603.17633
- atommof all-atom flow matching for mof-adsorbate structure prediction | arXiv: 2602.07351
- attention-discounted adaptive sampler for masked diffusion language models | arXiv: 2606.10829
- audible world models spatially aware sound generation for 3d worlds | arXiv: 2609.38444
- audio-visual flamingo open audio-visual intelligence for long and complex videos | arXiv: 2607.16107
- audiocalm continuous autoregressive language modeling for universal audio genera | arXiv: 2606.23080
- auditing conformal prediction for language models under inference-time model shi | arXiv: 2610.28937
- authority bias in language models source deference and user agreement are not in | arXiv: LLM Alignment
- autoexpert automating 3d lidar annotation from expert-crafted guidelines | arXiv: 2506.02914
- autoregressive appearance prediction for 3d gaussian avatars | arXiv: 2604.00928
- b-calm bias-limited bayesian borrowing for rct-anchored treatment effects under | arXiv: 2607.04036
- ba-t an iterative transformer for two-view bundle adjustment | arXiv: 2606.03287
- balancing frequencies and pixels in flow matching | arXiv: 2609.02748
- Baton: Explicit Semantic Blueprints for Joint Video-Audio Generation | arXiv: 2605.25195
- bayesian decision making observing an expert | arXiv: 2510.08113
- bayesian optimization with fisher information geometry gradient bounds and trust | arXiv: 2609.31107
- behavioral foundation models for quality diversity | arXiv: 2609.35615
- behavioral geometric supervision aligns video foundation models with human socia | arXiv: 2510.01502
- benchmarking compositional generalisation for machine learning interatomic poten | arXiv: 2605.08988
- benchmarking multimodal mathematical reasoning with explicit visual dependency
- benchmarking optimizers for large language model pretraining | arXiv: 2509.01440
- benchmarks are not atomic composition-aware llm evaluation using benchhub | arXiv: 2506.00482
- bernoulli flow models self-consistent generative modeling for binary data | arXiv: 2610.11362
- beyond drug discovery the nanotechnology molecular optimization nmo benchmark | arXiv: Computational Biology
- beyond empirical support structured outlier generation via sinkhorn optimal tran | arXiv: 2609.31470
- beyond interpretability when why and how sparse autoencoders enable label-free v | arXiv: 2506.01247
- beyond maximum likelihood variational inequality estimation for generalized line | arXiv: 2511.03087
- beyond missing rates rethinking incomplete multi-view clustering with protocol d | arXiv: 2606.04857
- beyond normal references discriminative few-shot anomaly detection | arXiv: 2605.23231
- beyond prediction steering vlm agents with retrospective world modeling | arXiv: 2609.39101
- beyond selection token parameterization for extreme visual token compression | arXiv: Multimodal Efficiency
- beyond self-play and scale a behavior benchmark for generalization in autonomous | arXiv: 2605.10034
- beyond spatial-domain supervision a relation constrained space for multi-modal i | arXiv: 2609.38968
- bfm2 brain foundation model via flow matching with splitunet | arXiv: 2606.20812
- bidirectional information flow bif - a sample efficient hierarchical gaussian pr | arXiv: 2505.11294
- bidirectional sparse attention for faster video diffusion training | arXiv: 2509.01085
- bimogen bidirectional motion-text generation via unified masked discrete diffusi | arXiv: 2609.35407
- binding multiple modalities via multimodal wasserstein barycenter | arXiv: Multimodal VLM
- bioblobs unsupervised discovery of functional substructures for protein function | arXiv: 2510.01632
- bird-rl scaling agentic reinforcement learning over stateful data-centric enviro | arXiv: 2610.29426
- birth-death structural learning for 3d gaussian splatting | arXiv: 2610.34344
- bitdance scaling autoregressive generative models with binary tokens | arXiv: 2602.14041
- bitmtp when multi-token prediction meets low-bit large language models | arXiv: 2610.28500
- blarm animating 3d objects from video via blending latent rigid motion primitive | arXiv: 2608.31113
- block sparse flash attention | arXiv: 2512.07011
- boosting text-to-image diffusion models via core token attention-based seed sele | arXiv: 2605.19532
- born-preconditioned learning for iterative neural pde solvers | arXiv: 2603.18527
- bound-conditioned latent inference for progressive image compression | arXiv: 2610.30966
- brainpro process-conditioned pretraining with topographic retrieval for large ee | arXiv: 2509.22050
- branching flows discrete continuous and manifold flow matching with splits and d | arXiv: 2511.09465
- breaking adversarial transferability in fine-tuned speech recognition | arXiv: 2610.09109
- breaking the uniformity trap scaling video diffusion model via splitmoe | arXiv: 2609.38140
- bridge graphical models coupling projection and current-preserving dynamics for | arXiv: 2608.19144
- budgeted multi-source counterfactual annotation for off-policy evaluation | arXiv: 2610.10974
- building transformation layers for riemannian neural networks | arXiv: 2609.35436
- busterx mllm-powered ai-generated video forgery detection and explanation | arXiv: 2505.12620
- c-lora continual low-rank adaptation for pre-trained visual models | arXiv: 2502.17920
- C3P: Contrastive promoter-protein pretraining yields representations capturing bacterial gene regulation | arXiv: 2605.25242
- cairn cross-room 3d scene understanding with topology-aware large multimodal mod | arXiv: 2607.06534
- calibrating generative models to feature distributions with mmd finetuning | arXiv: 2606.19496
- can circuit alignment predict ood generalization | arXiv: 2609.31996
- can large language models develop gambling addiction | arXiv: 2509.22818
- can linguistic reasoning vectors enhance multimodal reasoning ability | arXiv: 2609.31140
- can pixels alone reveal image origin minimax limits and learnable interfaces for | arXiv: 2609.30997
- canvasmar improving masked autoregressive video prediction with canvas | arXiv: 2510.13669
- cass contribution-aware structured sparsity for model merging | arXiv: 2609.34184
- catch your breath adaptive computation for self-paced sequence production | arXiv: 2510.13879
- causal abstractions categorically unified | arXiv: 2510.05033
- causal discovery under hard selection bias a new robust score-matching approach | arXiv: 2610.18075
- causal discovery via transformed low-rank quantile surfaces | arXiv: 2609.16931
- causal inference for sequential settings under interference and latent confoundi | arXiv: 2607.14940
- causal pieces analysing and improving spiking neural networks piece by piece | arXiv: 2504.14015
- Causal Representation Learning for Generalisable Recommendation | arXiv: 2605.27043
- causalbind causal modeling and learning for protein-molecule virtual screening | arXiv: 2610.07340
- causaldrivebench evaluating causal reasoning in vision-language-action models fo | arXiv: 2609.32157
- causalmix data mixture as causal inference for language model training | arXiv: 2607.01104
- cd-rcm generalizable continuous-depth novel view synthesis for reflectance confo | arXiv: 2606.12635
- cellmsa context modeling for single-cell representation learning | arXiv: 2609.38908
- chaining 2-fwl gnns for combinatorial graph alignment | arXiv: 2510.03086
- character mixing for video generation | arXiv: 2510.05093
- chasing label shifters a change-aware framework for dynamic graph node classific | arXiv: 2610.17582
- cheap and powerful tests for supervised subspaces per-component inference for pl | arXiv: 2609.36307
- chem-perturbridge a harmonized compendium of small molecule perturbation transcr | arXiv: 2605.31522
- chi-bench can ai agents automate end-to-end long-horizon policy-rich healthcare | arXiv: 2605.16679
- circuitseer mining high-quality data by probing mathematical reasoning circuits | arXiv: 2510.18470
- clarification as supervision reinforcement learning for vision-language interfac | arXiv: 2509.26594
- code2world a gui world model via renderable code generation | arXiv: 2602.09856
- codecsplat ultra-compact latent coding for feed-forward 3d gaussian splatting | arXiv: 2605.25563
- codemimicry exploiting safety generalization lag in large language models via st | arXiv: 2609.39902
- codescaler scaling code llm training and test-time inference via reward models | arXiv: 2602.17684
- coherence mechanisms for provable self-improvement | arXiv: 2511.08440
- colorconceptbench a benchmark for probabilistic color-concept understanding in t | arXiv: 2601.16836
- colvr enhancing exploratory latent visual reasoning via contrastive optimization | arXiv: 2605.08802
- communication-efficient llm adaptation over decentralized gpu meshes | arXiv: 2609.14339
- commutator memory sparse path-local reading and steering in language models | arXiv: 2609.34348
- compactsplat spatially adaptive gaussian distribution for feedforward scene reco | arXiv: 2610.13177
- compatible likelihoods for flow matching on manifolds | arXiv: 2610.31981
- complete or sparse a tale of two identifiabilities | arXiv: 2610.20851
- complexity-guided regularization for generalizable human gaussian splatting | arXiv: 2610.35152
- concepts in motion temporal concept bottleneck model for interpretable video cla | arXiv: 2509.20899
- concise and logically consistent conformal sets for neuro-symbolic concept-based | arXiv: 2605.18202
- concorde geometry-aware link prediction via decoupled energy minimization | arXiv: 2610.36204
- concurrence of symmetry breaking and nonlocality phase transitions in diffusion | arXiv: 2605.04830
- condensevla learnable history condensation for efficient multi-frame vla | arXiv: 2610.25927
- conditional counterfactual mean embeddings doubly robust estimation and learning | arXiv: 2602.04736
- conformal selective acting anytime-valid risk control for rlvr-trained llms | arXiv: 2605.20270
- conservation laws for diffusion models | arXiv: 2607.10067
- constrained code generation with discrete diffusion | arXiv: 2605.16829
- context value informed in-context reinforcement learning | arXiv: 2509.23209
- contextual flow matching adaptive step selection in flow models for efficient vi | arXiv: 2610.03202
- continual learning in modern hopfield networks with an application to diffusion | arXiv: 2605.27975
- continually evolving skill knowledge in vision language action model | arXiv: 2511.18085
- contractbench can llm agents preserve observation contracts | arXiv: 2605.17281
- contrastive representation shaping for llm unlearning | arXiv: 2601.22028
- control-augmented autoregressive diffusion for data assimilation | arXiv: 2510.06637
- controlflow3d distilling multi-view knowledge into latent flow matching for poin | arXiv: 2610.21190
- controllable dynamic 3d shape generation via 3d trajectories and text | arXiv: 2606.05162
- controllable generative sandbox for causal inference | arXiv: 2603.03587
- convergence guarantees for federated sarsa with local training and heterogeneous | arXiv: 2512.17688
- coreset-induced conditional velocity flow matching | arXiv: 2605.12951
- corrective diffusion language models | arXiv: 2512.15596
- correlating cross-iteration noise for dp-sgd using model curvature | arXiv: 2510.05416
- corvus context optimization and reduction via underlying synchronization for llm | arXiv: 2607.22711
- cost-aware best-llm identification using dueling feedback | arXiv: 2609.30360
- Counterfactual Maps: What They Are and How to Find Them | arXiv: 2602.09128
- counterfactual rollout replay forkable environments as free process rewards for | arXiv: 2609.33875
- coupled integral pinn for discontinuity | arXiv: 2411.11276
- covariate-adjusted deep causal learning for heterogeneous panel data models | arXiv: 2505.20536
- CoWorld-VLA: Thinking in a Multi-Expert World Model for Autonomous Driving | arXiv: 2605.10426
- craft counterfactual-to-interactive reinforcement fine-tuning for driving polici | arXiv: 2605.04470
- crafterdojo a suite of foundation models for building open-ended embodied agents | arXiv: 2508.13530
- crisp fixing flying pixels in latent lidar generation via diffusion decoding | arXiv: 2610.11376
- cs-dialogue a 104-hour dataset of spontaneous mandarin-english code-switching di | arXiv: 2502.18913
- cs-dialogue 基于104小时中英自发对话的语码转换语音识别数据集 | arXiv: 2502.18913
- csflow aligning flow matching with human contrast sensitivity | arXiv: 2606.08833
- cte-bench counterfactual trace evaluation for stateful software simulators | arXiv: Code Intelligence
- cumulative-goodness free-riding in forward-forward networks real repairable but | arXiv: 2605.06240
- cyclic denoising reveals ultrastable memories in diffusion models | arXiv: 2606.24000
- cylindrical geodesic flow matching for quasiperiodic physiological signal transf | arXiv: 2610.08510
- d-gap improving out-of-domain robustness via dataset-agnostic and gradient-guide | arXiv: 2511.11286
- d-opsd on-policy self-distillation for continuously tuning step-distilled diffus | arXiv: 2605.05204
- d2d detector-to-differentiable critic for improved numeracy in text-to-image gen | arXiv: 2510.19278
- data auctions for retrieval augmented generation | arXiv: 2508.16007
- data-driven covariate selection for nonparametric and cycle-agnostic causal effe | arXiv: 2605.06385
- data-driven soft labeling scales dna read classification to whole-body cell-type | arXiv: 2607.04987
- dd-ranking rethinking the evaluation of dataset distillation | arXiv: 2505.13300
- debate a large-scale benchmark for evaluating opinion dynamics in role-playing l | arXiv: 2510.25110
- debiased counterfactual generation via flow matching from observations | arXiv: 2605.07665
- deceptive grounding entity attribution failure in clinical retrieval-augmented g | arXiv: 2607.09349
- decoupled descent exact test error tracking via approximate message passing | arXiv: 2604.27883
- decoupling action from egocentric observation for world simulation | arXiv: 2610.27946
- decoupling conversational dynamics in full-duplex spoken models through reinforc | arXiv: 2607.07148
- deep double q-learning | arXiv: 2507.00275
- deep ensembles for epistemic uncertainty a frequentist perspective | arXiv: 2510.22063
- deep generative models for phylogenetic inference with complex evolutionary proc | arXiv: 2610.19053
- deep minds and shallow probes | arXiv: 2605.11448
- deeparrhythmia segment-contextualized ecg arrhythmia classification via selectiv | arXiv: Medical Imaging
- degradequery counterfactual tuple pretraining for context-aware protac degradati | arXiv: 2608.10595
- delta-adapter scalable exemplar-based image editing with single-pair supervision | arXiv: 2605.07940
- depth2pose a pose-based benchmark for monocular depth estimation without ground- | arXiv: 2605.19797
- depthmaster unified monocular depth estimation for perspective and panoramic ima | arXiv: 2606.12368
- designing reinforcement learning for diffusion models a unified path-space view | arXiv: 2608.14430
- diagnosing and mitigating modality interference in multimodal large language mod | arXiv: 2505.19616
- diagonalizing the softmax hadamard initialization for tractable cross-entropy dy | arXiv: 2512.04006
- didedirect injection with color-texture decoupling for 3d stylization | arXiv: 2610.02044
- diff-ca separating common and salient factors with diffusion models | arXiv: 2606.06120
- differentiable nonlinear model predictive control | arXiv: 2505.01353
- diffpts rethinking diffusion elbo for probabilistic time series forecasting | arXiv: 2609.32363
- diffratio training one-step diffusion models without teacher supervision | arXiv: 2502.08005
- diffusion guidance is a controllable policy improvement operator | arXiv: 2505.23458
- diffusion llms are natural adversaries for any llm | arXiv: 2511.00203
- diffusion masked pretraining for dynamic point cloud | arXiv: 2605.03639
- diffusion models generalization can be characterized by inductive biases toward | arXiv: 2602.06021
- diffusion models observe only gradients a geometric perspective on score matchin | arXiv: 2606.06179
- diffusion models without classifier-free guidance | arXiv: 2502.12154
- digs distribution-informed gaussian splatting for training-free open-vocabulary | arXiv: 2610.16177
- dinocomplete 3d shape completion with distilled semantic priors and state space | arXiv: 2605.26949
- diphon diffusion on graphons for scalable graph generation | arXiv: 2607.07232
- direct product flow matching decoupling radial and angular dynamics for few-shot | arXiv: 2605.05054
- directuv image-conditioned uv texture generation with surface-aware positional e | arXiv: 2609.34651
- disambiguating 2d-3d correspondences in gaussian splatting-based feature fields | arXiv: 2605.07351
- discovering phase space structure in learned hamiltonian systems | arXiv: 2509.23003
- discrete diffusion models exploit asymmetry to solve lookahead planning tasks | arXiv: 2602.19980
- discrete flow matching convergence guarantees under minimal assumptions | arXiv: 2605.08882
- disentangled representation learning via flow matching | arXiv: 2602.05214
- disentangled sparse representations for concept-separated diffusion unlearning | arXiv: 2605.12122
- disentangling dual image references in frequency aware diffusion models for pers | arXiv: 2610.07684
- disrfm polar riemannian flow matching for structure-preserving graph domain adap | arXiv: 2602.00656
- disrfm polar riemannian flow matching for structure-preserving graph domain adap | arXiv: 2602.00656
- distilling multi-teacher scoring principles for unsupervised time series anomaly | arXiv: 2610.29063
- distilling what matters confidence-aware selective distillation for large langua | arXiv: Model Compression
- distinguishing performance from competence in evaluations of humanlike abstract | arXiv: 2510.02125
- do coding agents deceive us detecting and preventing cheating via capped evaluat | arXiv: 2606.07379
- do image editing models understand lighting | arXiv: 2606.26738
- docptbench benchmarking end-to-end photographed document parsing and translation | arXiv: 2511.18434
- dont pay attention plant it pretraining attention via learning-to-rank | arXiv: 2410.23066
- doob-based generative policy optimization | arXiv: 2610.28990
- dream-cubed controllable generative modeling in minecraft by training on billion | arXiv: 2604.22847
- drift flow matching | arXiv: 2605.17244
- drivefuture future-aware latent world models for autonomous driving | arXiv: 2605.09701
- drivehierarchy a benchmark for diagnosing vlm driving capabilities from open-loo | arXiv: 2609.31814
- DriveMind: Mind-Evolving Belief Tracking for Closed-Loop Autonomous Driving | arXiv: 2610.32069
- drivespatial a benchmark for spatiotemporal intelligence in vlms for autonomous | arXiv: 2605.23176
- driving video retrieval for complex queries with structured grounding | arXiv: 2606.09109
- dt-pbo an interpretable tree-based surrogate model for preferential bayesian opt | arXiv: 2512.14263
- dual-pathway circuits of object hallucination in vision-language models | arXiv: 2605.13156
- duet unified dual-space emotion control for diffusion and flow-matching driven t | arXiv: 2606.00066
- dycorm dynamic criterion-aware reward modeling for text-to-image generation | arXiv: 2605.25876
- dynamic model merging made slim | arXiv: 2605.18904
- dynamic regret in online convex optimization with indicator switching costs | arXiv: Optimization
- dynamics of the transformer residual stream coupling spectral geometry to networ | arXiv: 2605.14258
- dynt2i-eval a dynamic evaluation framework for text-to-image models | arXiv: 2605.06170
- dysurface consistent 4d surface reconstruction via bridging explicit gaussians a | arXiv: 2605.10360
- déjà view looping transformers for multi-view 3d reconstruction | arXiv: 2605.30215
- e-geo a testbed for generative engine optimization in e-commerce | arXiv: 2511.20867
- edge of stochastic stability revisiting the edge of stability for sgd | arXiv: 2412.20553
- edit the bits diff the codes bitwise residual editing for visual autoregressive | arXiv: 2606.13558
- edit-r2 context-aware reinforcement learning for multi-turn image editing | arXiv: 2606.05950
- effectiveness of curriculum learning depends on reward sparsity and competing op | arXiv: 2610.31884
- efficient accurate and stable gradients for neural odes | arXiv: 2410.11648
- efficient adjoint matching for fine-tuning diffusion models | arXiv: 2605.11480
- efficient bias mitigation in t2i diffusion models using concept graphs | arXiv: 2607.03397
- efficient dataset distillation for pre-trained self-supervised models via statis | arXiv: Self-Supervised Learning
- efficient dynamic algorithms for graph neural networks with non-linear propagati | arXiv: 2609.32929
- efficient image synthesis with sphere latent encoder | arXiv: 2605.15592
- efficient sam3 adaptation for multi-class semantic segmentation via dense compet | arXiv: 2610.27960
- efficient streaming audio-visual target speaker extraction for real-world acoust | arXiv: 2610.32749
- efficient transferable optimal transport via min-sliced transport plans | arXiv: 2511.19741
- ehrnote-chatqa a benchmark for evidence-grounded multi-turn clinical question an | arXiv: 2606.15735
- eliciting secret knowledge from language models | arXiv: 2510.01070
- elma benchmarking anaphoric compositions for long-term text-to-motion generation | arXiv: 2610.13549
- emerge a benchmark for updating knowledge graphs with emerging textual knowledge | arXiv: 2507.03617
- emergence of distortions in high-dimensional guided diffusion models | arXiv: 2602.00716
- enhancing novel view synthesis via geometry grounded set diffusion | arXiv: 2601.07540
- enhancing speech large language models through reinforced behavior alignment | arXiv: 2509.03526
- ensemble modeling for time series forecasting an adaptive robust optimization ap | arXiv: 2304.04308
- entropy distribution as a fingerprint for hallucinations in generative models | arXiv: 2605.28264
- epic efficient predicate-guided inference-time control for compositional text-to | arXiv: 2605.11722
- equilibrium matching generative modeling with implicit energy-based models | arXiv: 2510.02300
- equivariant spherical transformer for efficient molecular modeling | arXiv: 2505.23086
- ergodic risk measures towards a risk-aware foundation for continual reinforcemen | arXiv: 2510.02945
- ergodic trajectory design by learned pushforward maps provable coverage via cond | arXiv: 2605.13063
- estimating and orthogonalizing unknown pre-training gradients for continual fine | arXiv: 2609.30935
- estimating model-level membership inference vulnerability without reference mode | arXiv: 2510.19773
- estimation of the label-noise transition matrix with performance guarantees via | arXiv: 2609.39829
- even sharper bounds for transductive learning and its applications | arXiv: 2609.28459
- evolutionary foraging in grids intermittent search dynamics emerge in finite dep | arXiv: 2609.39239
- exact flow linear attention exact solution from continuous-time dynamics | arXiv: 2512.12602
- execution guided line-by-line code generation | arXiv: 2506.10948
- exemplar2vqa a scalable exemplar-driven visual question answering generation fra | arXiv: 2609.37655
- expanding the role of diffusion models for robust classifier training | arXiv: 2602.19931
- exploring mllm-diffusion information transfer with metacanvas | arXiv: 2512.11464
- extending roc analysis to uncertainty-aware risk prediction with an interval-bas | arXiv: 2602.04775
- externalized cpdag summaries improve llm causal deduction | arXiv: 2609.31071
- extracting search trees from llm reasoning traces reveals myopic planning | arXiv: 2605.06840
- f2g-pose geometry-aware foundation feature lifting for direct rgb-d category-lev | arXiv: 2610.31623
- fa-lam focus-aware large avatar model for one-shot 4d animatable gaussian head | arXiv: 2607.20922
- facephys state of the heart learning | arXiv: 2512.06275
- factorizedhmr a hybrid framework for video human mesh recovery | arXiv: 2605.14854
- fashionchameleon towards real-time and interactive human-garment video customiza | arXiv: 2605.15824
- fast 4d mesh generation by spatio-temporal attention chains | arXiv: 2605.19786
- fast organic crystal structure prediction with unit cell flow matching | arXiv: 2606.03199
- feedback forensics a toolkit to measure ai personality | arXiv: 2509.26305
- feedback world model enables precise guidance of diffusion policy | arXiv: 2605.15705
- feedforward novel view synthesis for heterogeneous cameras | arXiv: 2610.03522
- film-cam keyed feature modulation for conditional-access watermarking | arXiv: 2610.30347
- finevision open data is all you need | arXiv: 2510.17269
- finite-sample performance of gradient descent in logistic regression with gaussi | arXiv: 2606.21683
- fire fine-grained multimodal reasoning for enhanced image generation | arXiv: 2604.13491
- fixed-point masked generative modeling | arXiv: 2605.31215
- fixed-size active statistical inference | arXiv: 2610.27389
- flame flow enhanced legendre memory models for general time series forecasting | arXiv: 2512.14253
- flare full-modality long-video audiovisual retrieval benchmark with user-simulat | arXiv: 2605.10228
- flashbob io-efficient exact backward-over-backward for softmax attention | arXiv: 2609.24089
- flashmol high-quality molecule generation in as few as four steps | arXiv: 2605.07020
- flow matching for count data | arXiv: 2605.07746
- flow matching for offline reinforcement learning with discrete actions | arXiv: 2602.06138
- flow matching policy optimization with mirror descent and entropy constraints | arXiv: 2603.17685
- Flow-DPPO: Divergence Proximal Policy Optimization for Flow Matching Models | arXiv: 2606.11025
- flowette flow matching with graphette priors for graph generation | arXiv: 2602.23566
- flux geometry-aware longitudinal flow matching with mixture of experts | arXiv: 2605.08648
- fluxlite inference-time proposal control for discrete diffusion models | arXiv: 2609.35947
- flyaoc evaluating agentic ontology curation of drosophila scientific knowledge b | arXiv: 2602.09163
- fmmi flow matching mutual information estimation | arXiv: 2511.08552
- focus benchmarking retinal model generalization from foundation vision encoders | arXiv: 2609.33158
- focusable monocular depth estimation | arXiv: 2605.11756
- focusing influence mechanism for multi-agent reinforcement learning | arXiv: 2506.19417
- follow the mean reference-guided flow matching | arXiv: 2605.10302
- fomemo towards foundation models for expensive multi-objective optimization | arXiv: 2509.03244
- fractional state space transition for long sequence modeling | arXiv: 2609.36314
- fred external data influence estimation via domain knowledge graph grounding | arXiv: 2607.21615
- free draft-and-verification toward lossless parallel decoding for diffusion larg | arXiv: 2510.00294
- frequencyaware flow matching for continuous and consistent robotic action genera | arXiv: 2606.20135
- From Isolated feature to Orbits: Discovering Music Concepts via Multi-SAE Alignment | arXiv: 2610.01864
- from next-token to next-block a principled adaptation path for diffusion llms | arXiv: 2512.06776
- from talking words to sharing thoughts scalable multi-llm aggregation via struct | arXiv: 2606.00405
- from topology to retrieval decoding embedding spaces with unified signatures | arXiv: 2511.22150
- from weak data to strong policy q-targets enable provable in-context reinforceme | arXiv: 2609.30391
- frontier coding agents use metaprogramming to adapt to unfamiliar programming la | arXiv: 2606.10933
- fruc feedforward dynamic scene reconstruction from uncalibrated collaborative dr | arXiv: 2605.29997
- function graph transformers universally approximate operators between function s | arXiv: 2605.17968
- g2tr generation-guided visual token reduction for separate-encoder unified multi | arXiv: 2605.12309
- gate-based feature importance in simulation-based inference for eeg a cautionary | arXiv: 2610.30364
- gaussian splatting-based volumetric video compression with sparse 4d anchors | arXiv: 2609.33969
- gazeflow from human gaze behavior to generative egocentric gaze prediction | arXiv: Human Understanding
- gen-searcher reinforcing agentic search for image generation | arXiv: 2603.28767
- gencope syn2real generalized category-level object pose estimation for robotic p | arXiv: 2610.01758
- genenv difficulty-aligned co-evolution between llm agents and environment simula | arXiv: 2512.19682
- generalized bayes for causal inference | arXiv: 2603.03035
- generalized wasserstein flow matching transport plans everywhere all at once | arXiv: 2605.08424
- generating in the limit with infinitely many hallucinations | arXiv: 2606.28354
- generation navigator a state-aware agentic framework for image generation | arXiv: 2605.17969
- generative conformal prediction with optimized coverage allocation | arXiv: 2410.13735
- generative control as optimization time unconditional flow matching for adaptive | arXiv: 2603.17834
- generative modeling via drifting | arXiv: 2602.04770
- generative scenario rollouts for end-to-end autonomous driving | arXiv: 2601.11475
- genevolve self-evolving image generation agents via tool-orchestrated visual exp | arXiv: 2605.21605
- genrecon bridging generative priors for multi-view 3d scene reconstruction | arXiv: 2605.23888
- genscale a benchmark for relative object scale in image generation and editing | arXiv: 2609.00525
- geofidelity-bench evaluating block-conditioned geographic fidelity in text-to-im | arXiv: 2606.23669
- geometric analysis of neural regression collapse via intrinsic dimension | arXiv: 2510.01105
- geometric inductive biases for semi-supervised equalization the constellation-aw | arXiv: Signal & Communication
- geometry-centered 3d latent world models for growing surfaces | arXiv: 2506.03173
- geotransolver learning physics on irregular domains using multi-scale geometry a | arXiv: 2512.20399
- geowind2plan mission-time 3d urban wind prediction for energy-efficient uav plan | arXiv: Robotics
- gfnx fast and scalable library for generative flow networks in jax | arXiv: 2511.16592
- ghost geometry-hierarchical online streaming token eviction for efficient 3d rec | arXiv: 2605.15852
- glare generating listening heads with appropriate reactions | arXiv: 2609.40317
- global convergence of four-layer matrix factorization under random initializatio | arXiv: 2511.09925
- glut 3d gaussian lookup table for continuous color transformation | arXiv: 2605.19889
- goal-conditioned supervised learning for multi-objective recommendation | arXiv: 2412.08911
- gold panning strategic context shuffling for needle-in-haystack reasoning | arXiv: 2510.09770
- gradient routing localizes and removes unintended behaviors in rl | arXiv: 2610.17062
- graph distance based on cause-effect estimands with latents | arXiv: 2510.25037
- graph energy matching transport-aligned energy-based modeling for graph generati | arXiv: 2603.23398
- graph-regularized sparse autoencoders for llm safety steering | arXiv: 2512.06655
- graphipbench how hard is it to steal a graph neural network and can we stop it | arXiv: 2605.12827
- grasp guided residual adapters with sample-wise partitioning | arXiv: 2512.01675
- greedy alignment principle for optimizer selection | arXiv: 2512.06370
- group distributionally robust optimization with flexible sample queries | arXiv: 2505.15212
- gt-harmbench benchmarking ai safety risks through the lens of game theory | arXiv: 2602.12316
- guaranteed noisy cp tensor recovery via riemannian optimization on the segre man | arXiv: 2510.00569
- guiding visual autoregressive models through spectrum weakening | arXiv: 2511.22991
- gvcc zero-shot video compression via codebook-driven stochastic rectified flow | arXiv: 2603.26571
- h2g hierarchy-aware hyperbolic grouping for 3d scenes | arXiv: 2605.11967
- hadamard representation scaffolding performance across model-free rl
- hallucination in world models is predictable and preventable | arXiv: 2606.27326
- halluworld a controlled benchmark for hallucination via reference world models | arXiv: 2605.19341
- ham-world soft-hamiltonian world models with selective memory for planning | arXiv: Reinforcement Learning
- handedit a unified benchmark for egocentric human-to-robot dexterous hand image | arXiv: 2608.12122
- handwritten text recognition lives in the high-pixel variance subspace | arXiv: 2609.35473
- Hearing is Believing? Evaluating and Analyzing Audio Language Model Sycophancy with SYAUDIO | arXiv: 2601.23149
- heavy-tailed flow matching via random clocks | arXiv: 2607.13841
- hermes a multi-scale spatial-temporal hypergraph network for stock time series f | arXiv: 2509.23668
- hero a heterogeneity-aware benchmark library for federated continual learning | arXiv: 2607.08784
- hero improving the reliability and sensitivity of generative model evaluation us | arXiv: 2606.29784
- hierarchical semantic tree anchoring for clip-based class-incremental learning | arXiv: 2511.15633
- higraph a large-scale hierarchical graph dataset for malware analysis | arXiv: 2509.02113
- hilbertgen-3dhilbert--multifractal conditioning for topology-aware 3d generation | arXiv: 2610.22720
- hilora adaptive hierarchical lora routing for training-free domain generalizatio | arXiv: 2510.12266
- hopse scalable higher-order positional and structural encoder for combinatorial | arXiv: 2505.15405
- horizon-stream long-horizon attention for streaming 3d reconstruction | arXiv: 2605.23889
- hots homophily-aware temperature scaling for graph neural network calibration | arXiv: 2609.32426
- how data scales in agentic reinforcement learning laws and synthesis strategies | arXiv: 2610.23946
- how deep is your network deep vs shallow learning of transfer operators | arXiv: 2509.19930
- how diffusion models memorize | arXiv: 2509.25705
- how do language models compose functions | arXiv: 2510.01685
- how does rl post-training induce skill composition a case study on countdown | arXiv: 2512.01775
- how fine-tuning objectives shape layer-wise information in llm hallucination det | arXiv: 2610.27711
- how much must a private mempool hide exact leakage thresholds for sandwich attac | arXiv: 2609.31379
- how neural reward models learn features for policy optimization a single-index a | arXiv: 2605.24749
- how to make the most of your masked language model for protein engineering | arXiv: 2603.10302
- human cognitive benchmarks reveal foundational visual gaps in mllms | arXiv: 2502.16435
- humanscore benchmarking human motions in generated videos | arXiv: 2604.20157
- hypertransport amortized conditioning of t2i generative models | arXiv: 2605.08254
- i have a stream making self-supervised learning work on continuous video | arXiv: 2609.40333v1
- i-deq a stable inertial deep equilibrium model for image restoration | arXiv: 2605.19705v2
- icats fast video generation via interaction-aware sparse attention and timestep- | arXiv: 2610.11302
- importance-aware obs pruning for diffusion models | arXiv: 2607.20048v3
- impossibility of distribution-free predictive inference for individual treatment | arXiv: 2605.05051
- improved baselines with representation autoencoders | arXiv: 2605.18324
- improving flexible image tokenizers for autoregressive image generation | arXiv: 2601.01535
- improving function space flow matching with kernel optimal transport | arXiv: 2609.38049
- in-context multi-operator learning with deeposets | arXiv: 2512.16074
- information bottleneck-guided adaptive hypergraph transformer for brain disease | arXiv: 2609.37220v1
- information templates a new paradigm for intelligent active feature acquisition | arXiv: 2508.18380
- inline critic steers image editing | arXiv: 2605.12724
- inpainting physics self-supervised learning for context-driven fluid simulation | arXiv: 2605.08832
- inspo unlocking intrinsic self-reflection for llm preference optimization | arXiv: 2512.23126
- instruct-particulate scaling feed-forward 3d object articulation with kinematic | arXiv: 2606.14699
- intact interval-based task activation consolidation for continual learning | arXiv: 2511.17439
- integrating background knowledge for scalable causal discovery | arXiv: 2607.10456
- intelligence per watt measuring intelligence efficiency of local ai | arXiv: 2511.07885
- Intend, Reflect, Refine: An Adaptive Multimodal Reflection Framework for Autonomous Driving | arXiv: 2606.22913
- interpretable but fragile robustness of concept bottlenecks under geometric-sema | arXiv: 2609.38625v1
- interpreting latent protein language model features with geometric annotations | arXiv: 2608.26419
- intrinsically interpretable attention via sparse post-training | arXiv: 2512.05865
- invaria learning scale and density invariance in point clouds via next-resolutio | arXiv: 2605.15923
- its all training a fully synthetic single-stage recipe for llms | arXiv: 2609.37891v1
- jacobian descent for multi-objective optimization | arXiv: 2406.16232
- jointly reinforcing diversity and quality in language model generations | arXiv: 2509.02534
- just ramp-up debiasing regression-based estimator for ab tests under network int | arXiv: 2410.12740
- justitia 任务并行大模型智能体的公平高效调度与选择性饱和服务 | arXiv: 2510.17015
- k12-kgraph a curriculum-aligned knowledge graph for benchmarking and training ed | arXiv: 2605.09635
- kg-guard graph-based hallucination detection for knowledge base question answeri | arXiv: 2606.00328
- Kinetic-Optimal Scheduling with Moment Correction for Metric-Induced Discrete Flow Matching in Zero-Shot Text-to-Speech | arXiv: 2605.09386
- knowledge-level consistency reinforcement learning dual-fact alignment for long- | arXiv: 2509.23765
- koalags complexity-aware continuous level of detail for gaussian splatting | arXiv: 2610.28615
- lafp preserving latent action structure in latent policy learning via flow match | arXiv: 2606.10517
- language-conditioned world modeling for visual navigation | arXiv: 2603.26741v2
- latent riemannian flow matching for geometry-grounded 3d foundation models | arXiv: 2607.19120
- latent video prediction for world modeling an evaluation uncovering intriguing f | arXiv: 2605.15618v2
- leaf language-eeg aligned foundation model for brain-computer interfaces | arXiv: 2509.24302
- leaner transformers can easily learn to cluster | arXiv: 2610.09760
- learning chance-constrained mdps with bellman distributional certificates | arXiv: Reinforcement Learning
- learning discrete riemannian metrics for physical fields with cochain-frame equi | arXiv: 2608.14556
- learning domain trajectories with flow matching for gradual domain adaptation | arXiv: 2610.30685
- learning implicit bias in generative spaces for accelerating protein dynamics em | arXiv: 2606.01833
- learning large-scale competitive team behaviors with mean-field interactions | arXiv: 2504.21164
- learning survival models with right-censored reporting delays | arXiv: 2510.04421
- learning to sample from diffusion models via inverse reinforcement learning | arXiv: 2602.08689
- learning unbiased permutations via flow matching | arXiv: 2605.16755
- learning where and what to restore for composite image restoration | arXiv: Image Restoration
- learning where it matters geometric anchoring for robust preference alignment | arXiv: LLM Alignment
- lemon-zest evolution-informed tokenization for efficient protein language modeli | arXiv: Computational Biology
- less decoder is more encoder geometric representation learning from novel view s | arXiv: 2610.03717
- less supervision better generalization weakly supervised fake region localizatio | arXiv: Safety
- libevobench probing temporal knowledge stratification in code generation models | arXiv: 2606.25402
- lifi lidar generation from multi-view images via geometric and semantic collabor | arXiv: 2610.19786
- lifting biomolecular data acquisition | arXiv: 2512.15984
- lime 基于解耦xor注意力与全局链接嵌入的测试期高效扩展推荐模型 | arXiv: 2510.18239
- linguistic trajectory encoding for efficient long-horizon spatial memory in embo | arXiv: Robotics
- listening to the wise few query-key alignment unlocks latent correct answers in | arXiv: Interpretability
- listening to the wise few querykey alignment unlocks latent correct answers in l | arXiv: 2410.02343
- ll-bench rethinking low-level vision evaluation in the era of large-scale genera | arXiv: 2606.02535
- llm alignment--utility asymmetry under semantic-preserving transformations | arXiv: 2609.32717
- llm judge validation under sparse overlap from inference to design | arXiv: 2609.31857v2
- llm-auction generative auction towards llm-native advertising | arXiv: 2512.10551
- llm-based multi-agent blackboard system for information discovery in data scienc | arXiv: 2510.01285
- load balancing mixture of experts with similarity preserving routers | arXiv: 2506.14038
- log-averaged mirror prox for fast large-scale optimal transport in linear space | arXiv: 2511.11359
- log-likelihood simpsons paradox and the detection of machine-generated text | arXiv: 2605.06294
- logictree-rag logic tree-guided retrieval-augmented generation for long-form pat | arXiv: Information Retrieval & RAG
- logit-gap steering a forward-pass diagnostic for alignment robustness | arXiv: 2506.24056
- logstop temporal scores over prediction sequences for matching and retrieval | arXiv: 2510.06512
- longscape advancing long-horizon embodied world models with context-aware moe | arXiv: 2509.21790
- look before you reason implicit visual thinking for efficient multimodal reasoni | arXiv: 2610.27341
- loop-free inverse reinforcement learning via sequential value recovery with q-sc | arXiv: Reinforcement Learning
- loraspace a pool-wide shared substrate for static and dynamic multi-lora composi | arXiv: 2610.27444
- loratorio an intrinsic approach to lora skill composition
- lost in translation found in embeddings sign language translation and alignment | arXiv: 2512.08040
- lotion smoothing the optimization landscape for quantized training | arXiv: 2510.08757
- low-dimensional adaptation of rectified flow a diffusion and stochastic localiza | arXiv: 2601.15500
- lr-v2x loss-resilient collaborative perception under low-bandwidth communication | arXiv: 2610.11264
- m-plicits neural implicit surfaces via nested multiscale residuals | arXiv: 2609.28684
- machine unlearning in low-dimensional feature subspace | arXiv: 2601.22456
- maeb massive audio embedding benchmark | arXiv: 2602.16008
- majority bit-aware watermarking for large language models | arXiv: 2508.03829
- mamba can learn low-dimensional targets in-context via test-time feature learnin | arXiv: 2510.12026
- managing self-learning experts under per-round budget constraints | arXiv: 2510.22654
- manitaskgen a comprehensive task generator for benchmarking and improving vision | arXiv: 2505.20726
- MAPLE: Latent Multi-Agent Play for End-to-End Autonomous Driving | arXiv: 2605.14201
- mappo maximum a posteriori preference optimization with prior knowledge | arXiv: 2507.21183
- markovian dynamics enforcer feasibility preserving correction on learned dynamic | arXiv: 2609.39888v1
- marktune improving the quality-detectability trade-off in model-embedded llm wat | arXiv: 2512.04044
- marrying optimal transport and odes for unified continuous-time 4d reconstructio | arXiv: 2608.09613
- masksense confronting the visual exploration trap in masked image generation | arXiv: 2610.17009
- mast label-efficient robust and generalizable sound detection for biodiversity m | arXiv: 2609.15221
- matching-based few-shot semantic segmentation models are interpretable by design | arXiv: 2511.18163
- maxitive donsker-varadhan formulation for possibilistic variational inference | arXiv: 2511.21223
- MCSplat: Multi-View Photometric and Geometric Consistent Feed-Forward Gaussian Splatting for Driving Scenes | arXiv: 2610.25345
- measuring what matterssynthetic benchmarks for concept bottleneck models | arXiv: 2606.04326
- mechanism-aware ensemble conditioning for data-limited emulation of extreme even | arXiv: 2609.30746v1
- mechanistic interpretability with sparse autoencoder neural operators | arXiv: 2509.03738
- medevoeval evaluating continual evolution of doctor agents through simulated cli | arXiv: 2606.28900
- medkit evaluating knowledge integration and generalization in large language mod | arXiv: 2609.38543v1
- medmemorybench benchmarking agent memory in personalized healthcare | arXiv: 2605.11814
- megastyle constructing diverse and scalable style dataset via consistent text-to | arXiv: 2604.08364
- mesh bdf barycentric dominance field for 3d native mesh generation | arXiv: 2606.31777
- meshfim local low-poly mesh editing via fill-in-the-middle autoregressive genera | arXiv: 2605.08744
- meta inverse prompting for video generative models | arXiv: 2610.24965
- metacluster enabling deep compression of kolmogorov-arnold network | arXiv: 2510.19105
- metis multi-source egocentric training for integrated dexterous vision-language- | arXiv: 2511.17366
- metric depth estimation from arbitrarily degraded low-resolution depth prompts | arXiv: 2602.09510
- metropolis-adjusted diffusion models | arXiv: 2605.09654
- midpoint generative models | arXiv: 2605.29920
- mind monge inception distance for generative models evaluation | arXiv: 2605.06797
- mini amusement parks maps a testbed for modelling business decisions | arXiv: 2511.15830
- mini-batch kernel k-means | arXiv: 2410.05902
- minimizing modality gap from the input side your speech llm can be a prosody-awa | arXiv: 2605.05927
- mirror descent-ascent for mean-field min-max problems | arXiv: 2402.08106
- misspath-fm flow matching with structured priors for partially observed time ser | arXiv: 2610.17715
- mitigating factual hallucination in large reasoning models via mixed-mode advant | arXiv: 2607.05861
- mitigating object hallucination in large vision-language models via false discov | arXiv: 2609.38979
- mito a millimeter-wave dataset and simulator for non-line-of-sight perception | arXiv: 2502.10259
- mllmeraser achieving test-time unlearning in multimodal large language models th | arXiv: 2510.04217
- mmlongcite a benchmark for evaluating faithfulness of long-context vision-langua | arXiv: 2510.13276
- mobayes a modular bayesian framework for separating reasoning from language in c | arXiv: 2604.20022
- mobilewan closing the quality gap for mobile video diffusion | arXiv: 2607.06173
- moca mixture-of-components attention for scalable compositional 3d generation | arXiv: 2512.07628
- modeling quantum neural network gradient with reinforcement learning | arXiv: 2609.31066v1
- modeling whole-slide images as dynamic tumor microenvironment fields | arXiv: Medical Imaging
- molhit advancing molecular-graph generation with hierarchical discrete diffusion | arXiv: 2602.17602
- momha multi-objective optimization of llm harnesses over accuracy safety and tok | arXiv: LLM Agent
- more than meets the eye uncovering the reasoning-planning disconnect in training | arXiv: 2510.04532
- mosaic-bench measuring compositional vulnerability induction in coding agents | arXiv: 2605.03952
- Motion Cues from Image-based Point Tracking for LiDAR Scene Flow Estimation | arXiv: 2605.16922
- motion forcing a decoupled framework for robust video generation in motion dynam | arXiv: 2603.10408v2
- motionhalluc diagnosing kinematic hallucinations in fine-grained motion reasonin | arXiv: 2606.23061
- mulclip a multi-level alignment framework for enhancing fine-grained long-contex | arXiv: 2512.07128
- multabench benchmarking multimodal tabular learning with text and image | arXiv: 2605.10616
- multi-part object representations via graph structures and co-part discovery | arXiv: 2512.18192
- multidimensional observer model and perceptual dimensions of human image quality | arXiv: 2609.38487
- multimodal llms outperform pathology foundation models in cross-domain histologi | arXiv: 2609.32876
- multiscale supervised unbalanced optimal transport flow matching | arXiv: 2605.16529
- multivariate time series forecasting needs cross variable loss | arXiv: 2608.05742
- murphy feedback-aware grpo with retrospective credit assignment for multi-turn c | arXiv: 2511.07833
- mutable transcripts mitigating context pollution through editable conversation s | arXiv: 2609.31354
- nads navigator-guided data selection for mitigating catastrophic forgetting in f | arXiv: 2610.25751
- namvis next-scale autoregressive multi-view image synthesis | arXiv: 2610.04722
- native audio-visual alignment for generation | arXiv: 2605.30073
- natural language actor-critic scalable off-policy learning in language space | arXiv: 2512.04601
- nearly optimal robust covariance and scatter matrix estimation beyond gaussians | arXiv: 2502.06564
- neat neighborhood-guided efficient autoregressive set transformer for 3d molecul | arXiv: 2512.05844
- nemotron-cascade scaling cascaded reinforcement learning for general-purpose rea | arXiv: 2512.13607
- neoworld-pro programming interactive scenes from monocular images for embodied s | arXiv: 2608.24212
- nestedvla learning to consolidate and generate skills for vision-language-action | arXiv: 2610.23271
- network of theseus like the ship | arXiv: 2512.04198
- neural bayesian filtering | arXiv: 2510.03614
- neural chameleons language models can learn to hide their thoughts from unseen a | arXiv: 2512.11949
- neural harmonic measure operator | arXiv: 2609.35752
- neural proposals symbolic guarantees neuro-symbolic graph generative modeling | arXiv: 2602.16954
- neural structural reasoner a brain-inspired architecture for reasoning over stru | arXiv: 2609.36620
- neurosymbolic object-centric learning with distant supervision | arXiv: 2506.16129
- next-latent prediction transformers learn compact world models | arXiv: 2511.05963
- no pose no problem in 4d feed-forward dynamic gaussians from unposed multi-view | arXiv: 2605.22190
- non-linear pricing restores tractability for a data seller | arXiv: 2609.36589
- nonconvex decentralized stochastic bilevel optimization under heavy-tailed noise | arXiv: 2509.15543
- nonparametric in-context learning under growing geometric complexity minimax opt | arXiv: 2609.31458
- normlift from lifted features to semantic reliability in 3d gaussian splatting | arXiv: 2609.18898
- not all tasks quantize equally fisher-guided quantization for visual geometry tr | arXiv: 2605.15828
- notvla semantics-preserving robot adaptation via narrative action interfaces | arXiv: 2510.03895
- np-lora null space projection for subject-style lora fusion | arXiv: 2511.11051
- nrf-gs neural residual fields for expressive and compact gaussian splatting | arXiv: 2609.37115
- numuon nuclear-norm-constrained muon for compressible llm training | arXiv: 2603.03597
- nuReasoning: A Reasoning-Centric Dataset and Benchmark for Long-Tail Autonomous Driving | arXiv: 2605.31572
- oasis observation-action space alignment via se3 trajectory prediction for robot | arXiv: 2605.25829
- offline constrained reinforcement learning under partial data coverage | arXiv: 2505.17506
- offloading score measuring ai reliance through counterfactual workflows | arXiv: 2605.29392
- ogbench a framework for evaluating graph neural networks on omics data | arXiv: 2605.15511
- omnicapbench a deep-structured evaluation framework for fine-grained audio-visua | arXiv: 2610.12458
- omnisimulator aligning small language models for authentic heterogeneous behavio | arXiv: 2610.29472
- OmniTraffic: A Controllable Generation Pipeline and Benchmark for Spatio-Temporal Traffic Reasoning | arXiv: 2606.15749
- on the depth of monotone relu neural networks and icnns | arXiv: 2505.06169
- on the faithfulness of visual thinking measurement and enhancement | arXiv: 2510.23482
- on the last-iterate convergence ofclipped gradient methods | arXiv: 2610.21026
- on the limits of latent reuse in diffusion models | arXiv: 2605.13448
- On the Robustness of Watermarking for Autoregressive Image Generation | arXiv: 2604.11720
- on the selectivity of generative models in structure-based drug design | arXiv: 2610.24124
- one algorithm two goals dual scoring for parameter and data selection in llm fin | arXiv: 2605.06166
- one step is enough multi-agent reinforcement learning based on one-step policy o | arXiv: 2507.15351
- one threshold does not fit all languages language-conditional deferral for relia | arXiv: 2609.37861
- one view is enough in-the-wild monocular pretraining for novel view generation | arXiv: 2603.23488
- onecanvas 3d scene understanding via panoramic reprojection | arXiv: 2606.19253
- online budget allocation with censored semi-bandit feedback | arXiv: 2508.05844
- online learning via learned latent bayesian tracking | arXiv: 2609.31559
- open vocabulary domain unlearning | arXiv: 2609.31356
- openview empowering mllms with out-of-view vqa | arXiv: 2512.18563
- openwhistle a large-scale longitudinal dataset and benchmark of bottlenose dolph | arXiv: 2609.34839
- optimal convergence analysis of ddpm for general distributions | arXiv: 2510.27562
- optimal in-context adaptivity and distributional robustness of transformers | arXiv: 2510.23254
- orca hunting compositional failures in text-to-image diffusion | arXiv: 2610.09841
- ordinary least squares as an attention mechanism | arXiv: 2504.09663
- otrope optimal transport-based robust off-policy evaluation for large language m | arXiv: 2609.36264
- outbidding and outbluffing elite humans mastering liars poker via self-play and | arXiv: 2511.03724
- Overcoming State Inertia in Full-Duplex Spoken Language Models via Activation Steering | arXiv: 2606.11386
- overlay dense-overlap layout-to-image generation dataset | arXiv: 2610.09071
- p3 joint program-and-proof planning for verified code generation | arXiv: 2608.09277
- p3-vlm a point-based alternative for grounded 3d vision-language models | arXiv: 2610.28018
- pace geometry-aware bridge transport for single-cell trajectory inference | arXiv: 2605.18587
- pair-ci calibrated conditional independence testing for causal discovery with in | arXiv: 2605.04838
- paleoinspired vision from exploring colour vision evolution to inspiring camera | arXiv: 2412.19439
- panoptic scene program diffusion transformer | arXiv: 2609.31780
- parallel rollout approximation for pixel-space autoregressive image generation | arXiv: 2606.27978
- parameter symmetries determine representational geometry in overparameterized no | arXiv: 2609.39078
- parameterized stripe attention for efficient video generation | arXiv: 2609.37001
- passing an endless journey through reconstructed spacetime with ai-generated sou | arXiv: 2609.27489
- patch learnable tile-level hybrid sparsity for llms | arXiv: 2509.23410
- pathways of visual information flow in vision-language models | arXiv: 2607.03358
- penex 基于 adaboost 思想的神经网络正则化 penex adaboost-inspired neural network regularizati | arXiv: 2510.02107
- pepspecbench a unified evaluation benchmark for peptide tandem mass spectrometry | arXiv: 2605.01945
- perceive interact reason building tool-augmented visual agents for spatial reaso | arXiv: 2606.12830
- Perceive-then-Plan: Layout-as-Policy for Monocular 3D Scene Layout Estimation | arXiv: 2605.25326
- performance-driven policy optimization for speculative decoding with adaptive wi | arXiv: 2605.14978
- permutation-invariant spectral learning via dyson diffusion | arXiv: 2510.08535
- persona vectors monitoring and controlling character traits in language models | arXiv: 2507.21509
- personamanifold revealing and exploiting curved geometry in llm persona represen | arXiv: 2609.34571
- pg-lrf physiology-guided latent rectified flow for electro-hemodynamic ppg-to-ec | arXiv: 2605.12541
- phaedra learning high-fidelity discrete tokenization for the physical science | arXiv: Physics & Scientific Computing
- phoebi an open-world benchmark for bacterial identification in phase-contrast mi | arXiv: 2606.22890
- phymo learning physical dynamics with accurate and continuous motion from multi- | arXiv: 2610.29107
- phyprobe rethinking physical consistency evaluation in generated videos | arXiv: 2609.38377
- physical ai smart spaces a large-scale benchmark for multi-camera 3d perception | arXiv: 2610.02580
- pisa piecewise sparse attention is wiser for efficient diffusion transformers | arXiv: 2602.01077
- pisco precise video instance insertion with sparse control | arXiv: 2602.08277
- PitchBench: Measuring Pitch Hearing in Audio-Language Models | arXiv: 2605.26176
- pixeldit2 representation-grounded pixel diffusion transformers | arXiv: 2609.24919
- planing a loosely coupled triangle-gaussian framework for streaming 3d reconstru | arXiv: 2601.22046
- playing markov games without observing payoffs | arXiv: 2509.00179
- pocketve stable and property-guided structure-based drug design with variance-ex | arXiv: Computational Biology
- point cloud sequence encoding for material-conditioned graph network simulators | arXiv: 2605.20978
- pointwise lipschitz continuous graph algorithms | arXiv: 2405.08938
- poisoning attacks on llms require a near-constant number of poison samples | arXiv: 2510.07192
- polytopobench a benchmark for complex vector polygon generation from remote sens | arXiv: 2609.32856
- portool importance-aware policy optimization with rewarded tree for multi-tool-i | arXiv: 2510.26020
- posebridge bridging the skeletonization gap for zero-shot skeleton-based action | arXiv: Human Understanding
- position adopt constraints over fixed penalties in deep learning | arXiv: 2505.20628
- position lets strengthen verifiability if we cant enforce reproducibility | arXiv: 2609.35854
- posterior inference in latent space for scalable constrained black-box optimizat | arXiv: 2507.00480
- ppo in the fisher-rao geometry | arXiv: 2506.03757
- pr-smoother simulator-preserving non-gaussian smoothing for data assimilation | arXiv: 2609.26890
- practical adversarial attacks on stochastic bandits via fake data injection | arXiv: 2505.21938
- precomem predictive cognitive memory for self-evolving long-term dialogue agents | arXiv: 2610.29467
- predictivegenerative drift decomposition for speech enhancement and separation | arXiv: 2605.06189
- preference-guided adaptation for open-vocabulary semantic segmentation via promp | arXiv: 2609.34528
- preserving deg rankings for gene discovery in histology-based spatial gene expre | arXiv: 2609.33928
- pretraining curricula enable selective fine-tuning | arXiv: 2607.04846
- primal-dual representation learning for low-rank constrained mdps | arXiv: 2610.30756
- prism rethinking atmospheric scattering reconstruction as a unified understandin | arXiv: 2604.07048
- pro enabling precise and robust text watermark for open-source llms | arXiv: 2510.23891
- probing persona-dependent preferencesin language models | arXiv: 2605.13339
- proclip progressive vision-language alignment via llm-based embedder | arXiv: 2510.18795
- procobject-10k benchmarking object-centric procedural understanding in instructi | arXiv: 2512.03479
- procompnav proactive instance navigation with comparative judgment for ambiguous | arXiv: 2605.06223
- progressive risk estimation for accident anticipation | arXiv: 2609.32811
- prompts to proxies emulating human preferences via a compact llm ensemble | arXiv: 2509.11311
- property-guided llm program synthesis for planning | arXiv: 2605.16142
- pulse identifying demonstration-utility features with sparse autoencoders | arXiv: 2609.32469
- pure exploration beyond reward feedback the role of post-action context | arXiv: 2502.03061
- pushing biomolecular utility-diversity frontiers with supergroup relative policy | arXiv: 2605.08659
- pygmalion effect in vision image-to-clay translation for reflective geometry rec | arXiv: 2511.21098
- quanvi score-based variational inference via quantum maximally mixed states | arXiv: 2609.39164
- querying counterfactuals on tissue graphs with supervised disentanglement | arXiv: 2606.08493
- qutcc quantile uncertainty training and conformal calibration for imaging invers | arXiv: 2507.14760
- r3 3d reconstruction via relative regression | arXiv: 2605.26519
- rail rethinking auditory intelligence in large audio-language models with a chc- | arXiv: 2606.11260
- ram-h1200 a unified evaluation and dataset on hand radiographs for rheumatoid ar | arXiv: 2605.05616
- rank-constrained adaptation for reliable real-world performance | arXiv: 2602.06924
- rank-transformed dissimilarity profiles for high-dimensional classification | arXiv: 2306.15199
- rankalign vision-language representation alignment via rank transformation | arXiv: 2610.31145
- raw-routed mixture of adapters a causal intervention for routing collapse in tim | arXiv: 2609.39445
- razer pushing the limits of nvfp4 quantization with redundant zero remapping | arXiv: 2501.04052
- re-examining low rank adaptation for private llm fine-tuning | arXiv: 2510.01137
- real in real out what if we only use real data for scene text editing | arXiv: 2610.27592
- reasoning-trace collapse evaluating the loss of explicit reasoning during fine-t | arXiv: Reasoning
- rebalancing reference frame dominance to improve motion in image-to-video models | arXiv: 2605.19398
- recognize -- open-set comic character re-identification | arXiv: 2609.34032
- reconstructing the vocal tract with differentiable acoustic simulation | arXiv: 2609.36737
- redefining instance matching a unified framework for part-aware matching in pano | arXiv: 2605.31094
- reducing credit assignment variance via counterfactual reasoning paths | arXiv: 2605.16302
- reefnet a large-scale dataset and benchmark for fine-grained coral reef recognit | arXiv: 2510.16822
- refineany3d depth refinement as semantic alignment for monocular 3d detection | arXiv: 2608.09147
- reformulating neural operators in d1 dimensions for embedding evolution | arXiv: 2505.11766
- ReGDiff: Guided Diffusion in Regulated Latent Space for Exploring Metamaterial Voxel Geometry | arXiv: 2609.34231
- RegimeVGGT: Layer-Wise Spatially Preserving Redundancy Removal for Visual Geometry Grounded Transformer | arXiv: 2606.18439
- register anything model for generalizable and robust point cloud registration | arXiv: 2610.28821
- relation-aware graph foundation model | arXiv: 2505.12027
- relationvggt visual geometry transformers for 3d spatial relation segmentation | arXiv: 2610.00970
- remember your trace memory-guided long-horizon agentic framework for consistent | arXiv: 2605.14563
- reorggs equivalent distribution reorganization for 3d gaussian splatting | arXiv: 2605.08739
- rep2text decoding full text from a single llm token representation | arXiv: 2511.06571
- replay-buffer engineering for noise-aware quantum circuit optimization | arXiv: Reinforcement Learning
- resbev making bev perception more robust | arXiv: 2603.09529
- rescue re-translation with sentence commitment for unsegmented long-form simulta | arXiv: 2610.03022
- residual calibration via local feature-space refinement | arXiv: 2610.30652
- Resolving Representation Ambiguity in Feedforward Novel View Synthesis Transformer via Semantic-Spatial Decoupling | arXiv: 2605.18599
- rest-rl reinforcing llm reasoning through unified self-training and value-guided | arXiv: 2508.19576
- rethinking cross-layer information routing in diffusion transformers | arXiv: 2605.20708
- rethinking how to remember beyond atomic facts in lifelong llm agent memory | arXiv: 2605.19952
- rethinking infrared small target detection a foundation-driven efficient paradig | arXiv: 2512.05511
- rethinking layer-wise model merging through chain of merges | arXiv: 2508.21421
- rethinking llm fine-tuning via weight space reparameterization preserving safety | arXiv: 2610.26443
- rethinking personalized generation test-time alignment via factorized ranking mo | arXiv: 2609.35695
- rethinking sae evaluation an atomic interpretable unit framework reveals hidden | arXiv: 2610.30474
- rethinking the state update gate for long-sequence recurrent 3d reconstruction | arXiv: 3D Reconstruction & Rendering
- rethinking training targets architectures and data quality for universal speech | arXiv: 2603.02641
- revisiting diffusion fine-tuning for unsupervised domain adaptation | arXiv: 2609.33716
- revitalizing medical time series with vision-informed retrieval a vision-languag | arXiv: 2609.34652
- reviving stale updates data-free knowledge distillation for asynchronous federat | arXiv: 2511.00655
- rifle removal of image flicker-banding via latent diffusion enhancement | arXiv: 2509.24644
- rippleplm structural and property decoupling for protein mutation effect generat | arXiv: 2610.01891
- robust and efficient finetuning of vision foundation models via implicit ensembl | arXiv: 2510.06982
- robust inference-time steering of protein diffusion models via embedding optimiz | arXiv: 2602.05285
- robust residual correction via selective deployment for time series forecasting | arXiv: 2512.22428
- robustness of internal signals for hallucination detection in vision-language mo | arXiv: 2610.27041
- rpc-gs gaussian splatting with native rpc rendering for satellite imagery | arXiv: 2606.06690
- rubato transcribing piano music with timestamps | arXiv: 2605.24291
- rvloss runoff vote loss for self-supervised lidar scene flow estimation | arXiv: 2608.18864
- rxgs receiver-generalizable 3d gaussian splatting for radio-frequency data synth | arXiv: 2605.24290
- s2d sparse-to-dense keymask distillation for unsupervised video instance segment | arXiv: 2512.14440
- saasbench exploring the boundaries of coding agents in long-horizon enterprise s | arXiv: 2605.17526
- safe score matching diffusion policies with hamilton-jacobi reachability for onl | arXiv: 2609.33337
- sagas semantic-aware graph-assisted stitching for offline temporal logic plannin | arXiv: 2512.00775
- sam 3d animal promptable animal 3d reconstruction from images in the wild | arXiv: 2605.07604
- sample-efficient optimization over generative priors via coarse learnability | arXiv: 2503.06917
- sandboxed coding agents are competitive omni-modal task solvers | arXiv: 2606.00579
- satnav a scalable benchmark for long-horizon uav vision-language navigation from | arXiv: 2609.31507
- scalable maximum entropy reinforcement learning for diffusion policies via adjoi | arXiv: 2606.22630
- scale improving guidance-free visual generation via self-contrastive alignment f | arXiv: 2610.23263
- scenescaffold active scene-state construction for unified 3d scene understanding | arXiv: 2609.33518
- schemapose rgb-based category-level pose estimation with parametric category sch | arXiv: 2610.18374
- scir a controllable benchmark for scientific reasoning in llms | arXiv: 2606.13020
- scope-rl stable and quantitative control of policy entropy in rl post-training | arXiv: 2510.08141
- scope-rl 基于温度自适应正样本的策略熵稳定与定量控制 | arXiv: 2510.08141
- seed self-speculative decoding via implicit encoder-decoder | arXiv: 2609.36590
- seeing speech learning visible articulatory dynamics for speech-driven 3d facial | arXiv: Human Understanding
- seg3dparts segmentation-grounded controllable part-level 3d generation | arXiv: 3D Generation & Editing
- selection not fusion radar-modulated state space models for radar-camera depth e | arXiv: 2605.11840
- self-improvement imitation with biologically guided search for protein design un | arXiv: 2605.26690
- semantic motion anchors bridging motion and meaning in co-speech gestures | arXiv: 2605.30608
- semmsa latent semantic-aided robust multimodal sentiment analysis with incomplet | arXiv: Multimodal VLM
- sense semantic neural speech synthesis from brain dynamics via spatial graph enc | arXiv: 2609.37601
- sentinel a multi-level formal framework for safety evaluation of foundation mode | arXiv: 2510.12985
- sequential solution concepts in cooperative games with generalized characteristi | arXiv: 2510.11255
- shared modular recurrence in contextual mdps for universal morphology control | arXiv: 2506.08630
- sharpness of minima in deep matrix factorization | arXiv: 2509.25783
- sheafstain sheaf-theoretic schrödinger bridge for spatially and biologically coh | arXiv: 2606.11846
- short-context dominance how much local context natural language actually needs | arXiv: 2512.08082
- simple extensions of single-objective acquisition functions and hedge strategies | arXiv: 2609.31940
- simple yet effective semi-supervised knowledge distillation from vision-language | arXiv: 2505.07675
- simpleevol an agent-loop framework for llm-driven automated heuristic design wit | arXiv: 2609.37172
- simplihumon simplifying human motion prediction | arXiv: 2603.04399
- slaying a diverse and community-validated dataset of queer slang | arXiv: 2509.17449
- slimoo interpretable multi-objective evolutionary search for llm depth pruning | arXiv: 2610.27812
- slopcodebench benchmarking how coding agents degrade over long-horizon iterative | arXiv: 2603.24755
- smile bridging continuous optimization and discrete symbolic recovery | arXiv: 2609.04639
- snapaudit active auditing of differentially private in-context learning via snap | arXiv: 2511.13502
- SOAR: Regression-based LiDAR Relocalization for UAVs | arXiv: 2602.13267
- social choice foundations for simulation-augmented generation | arXiv: 2609.38287
- soft geometric inductive bias for object centric dynamics | arXiv: 2512.15493
- soft geometric inductive bias for object centric dynamics | arXiv: 2512.15493
- solving every step is not enough milestone oracles reveal a composition gap in l | arXiv: 2609.32235
- sopo socratic guided policy optimization for span-level hallucination detection | arXiv: 2610.32212
- spanuq span-level uncertainty quantification for large language model generation | arXiv: 2607.05721
- spdebench an extensive benchmark for learning stochastic pdes | arXiv: 2505.18511
- spectral feedback for test-time alignment of protein diffusion models | arXiv: 2609.30456
- spectral graph sparsification preserves representation geometry in graph neural | arXiv: 2605.01136
- spectral reversal counteracting singular value bias for graph prompting | arXiv: 2609.32143
- spectral-sphere-constrained hyper-connections | arXiv: LLM Pretraining
- spherical interpolation for backward-compatible multimodal representations | arXiv: Multimodal VLM
- splatlab autonomous scientific discovery for 3dgs research | arXiv: 2610.30468
- sr-grpo stable rank as an intrinsic geometric reward for large language model al | arXiv: 2512.02807
- st-diffeye diffusion-based continuous gaze generation via joint scanpath-traject | arXiv: 2606.15486
- stabilizing knowledge promoting reasoning dual-token constraints for rlvr | arXiv: 2507.15778
- stabilizing the dynamic low-rank training | arXiv: 2609.32615
- staminabench stress-testing coding agents over 100 interaction turns | arXiv: 2606.19613
- star boosting time series foundation models for anomaly detection through state- | arXiv: 2510.16014
- statistical matching via schrödinger bridge beyond conditional independence | arXiv: 2606.22770
- stealth apart harm together skill cascading attacks on skill-based agent systems | arXiv: 2609.30383
- Steer-to-Detect: Probing Hidden Representations for Detection of LLM-Generated Texts | arXiv: 2605.12890
- steering frozen llms adaptive social alignment via online prompt routing | arXiv: 2603.15647
- stepcad mesh-to-cad code generation via llm policy and geometry-guided search | arXiv: 2610.03799
- stochastic interpolants via conditional dependent coupling | arXiv: 2509.23122
- stochastic matching bandits with rare optimization updates | arXiv: 2509.04194
- strand sequence-conditioned transport for single-cell perturbations | arXiv: 2602.10156
- streamgaze gaze-guided temporal reasoning and proactive understanding in streami | arXiv: 2512.01707
- strengthening llms for tabular prediction with structural priors | arXiv: 2510.17385
- stride automated evaluation of text-to-trajectory alignment across diverse conte | arXiv: 2609.34799
- structure-agnostic causal representation learning | arXiv: 2610.00968
- structure-guided masked autoencoders for ultra-high resolution scientific image | arXiv: 2609.30682
- structured sparse memory for recurrent reasoning | arXiv: 2609.33270
- stylestream 20 fast and controllable streaming voice style conversion | arXiv: 2610.28940
- sublinear time quantum sensitivity sampling | arXiv: 2509.16801
- swe atlas benchmarking coding agents beyond issue resolution | arXiv: 2605.08366
- swe-chain benchmarking coding agents on chained release-level package upgrades | arXiv: 2605.14415
- tac timestamped audio captioning | arXiv: 2602.15766
- tact mitigating overthinking and overacting in coding agents via activation stee | arXiv: 2605.05980
- tactile mnist benchmarking active tactile perception | arXiv: 2506.06361
- taking the road less scheduled with adaptive polyak steps | arXiv: 2511.07767
- tales text adventure learning environment suite | arXiv: 2504.14128
- teaching video generators to remember eliciting dynamic memory for out-of-sight | arXiv: 2605.25333
- televal a benchmark designed for spoken language models in chinese interactive s | arXiv: 2507.18061
- tempopfn synthetic pre-training of linear rnns for zero-shot time series forecas | arXiv: 2510.25502
- temporal alignment guidance on-manifold sampling in diffusion models | arXiv: 2510.11057
- temporal gradient inversion for private trajectory reconstruction in embodied re | arXiv: 2609.30258
- test time training enhances in-context learning of nonlinear functions | arXiv: 2509.25741
- test-time defense against adversarial attacks via stochastic resonance of latent | arXiv: 2510.03224
- test-time scaling of diffusions with flow maps | arXiv: 2511.22688
- tgpo temporal grounded policy optimization for signal temporal logic tasks | arXiv: 2510.00225
- the algorithm is not the behavior learned priors override look-ahead in a chess- | arXiv: 2508.21380
- the alignment illusion in multimodal large language models | arXiv: 2609.30210
- the bayesian origin of the probability weighting function in human representatio | arXiv: 2510.04698
- the commit-abstain circuit why language models hallucinate instead of abstaining | arXiv: 2609.32964
- the complexity kink a prompt-side structural complexity index for code-generatio | arXiv: 2609.19616
- the era of agentic organization learning to organize with language models | arXiv: 2510.26658
- the panel complexity of sortition is 12 angry men enough | arXiv: 2504.20508
- the pokémon theorem and other fairness impossibility results | arXiv: 2605.09221
- the price of locality why forward-forward underperforms backpropagation | arXiv: 2609.33240
- the reasoning boundary paradox how reinforcement learning constrains language mo | arXiv: 2510.02230
- the reflexivity threshold a phase transition for multi-agent performative predic | arXiv: 2610.29577
- The Road Ahead in Autonomous Driving: The KITScenes Multimodal Dataset | arXiv: 2606.02956
- the sample complexity of multiple change point identification under bandit feedb | arXiv: 2605.13252
- the sampling complexity of condorcet winner identification in dueling bandits | arXiv: 2603.15189
- the shape of events edge-based inductive biases via cross-domain distillation | arXiv: 2609.30478
- the spectral amplitude principle for dynamics of quantum neural networks | arXiv: 2409.06682
- the superactivator mechanism transformers concentrate reliable concept signals i | arXiv: 2512.05038
- the two clocks and the innovation windowwhen and how generative models learn rul | arXiv: 2605.10019
- the world is not mono enabling spatial understanding in large audio-language mod | arXiv: 2601.02954
- thousandworlds a benchmark for climate emulation of potentially habitable exopla | arXiv: 2606.18338
- tic-grpo provable and efficient optimization for reinforcement learning from hum | arXiv: 2508.02833
- timees probabilistic and deterministic time series forecasting via evolutionary | arXiv: 2609.32384
- timetok granularity-controllable time-series generation via hierarchical tokeniz | arXiv: 2605.01418
- to trust or not to trust attention-based trust management for llm multi-agent sy | arXiv: 2506.02546
- tokenclip token-wise prompt learning for zero-shot anomaly detection | arXiv: 2510.21171
- tokenizer-generator coupling in medical image generation | arXiv: 2608.07713
- tokens-per-parameter coverage is critical for robust llm scaling law extrapolati | arXiv: 2605.08541
- toolsearcher optimizing tool selection at scale via reinforcement learning | arXiv: 2609.30906
- topocurve geometry-aware topology reasoning via bézier curves in autonomous driv | arXiv: 2610.09118
- toward a unified statistical theory of unsupervised pretraining and supervised n | arXiv: 2607.26346
- toward better geometric representations for molecule generative models | arXiv: 2605.07693
- Towards Effective and Transferable Physical Camouflage against Multi-View BEV-based 3D Perception in Autonomous Driving | arXiv: 2610.32230
- towards generalizable 3d anomaly detection via relational inconsistency modeling | arXiv: 2609.35059
- towards hyperparameter transfer for differentially private optimization | arXiv: 2610.22976
- towards identifiable latent additive noise models | arXiv: 2403.15711
- towards mitigating deceptive safety alignment in large reasoning models | arXiv: 2609.36254
- towards scalable context-aware single-cell spatial transcriptomics prediction fr | arXiv: 2609.36429
- towards unified dynamic face landmark detection | arXiv: 2608.10346
- tpo tri-level distributionally robust learning for ood direct preference optimiz | arXiv: 2610.28229
- training language models to explain their own computations | arXiv: 2511.08579
- training optimal large diffusion language models quokka scaling law
- trajevolve trajectory evolution for reinforcement learning with hindsight credit | arXiv: 2610.28060
- transcoda end-to-end zero-shot optical music recognition via data-centric synthe | arXiv: 2605.10835
- TransmissiveGS: Residual-Guided Disentangled Gaussian Splatting for Transmissive Scene Reconstruction and Rendering | arXiv: 2605.10705
- treewalker partial evaluation for grouped tree-ensemble inference | arXiv: 2610.03939
- trust guided decision transformer | arXiv: 2609.31586
- trustworthy retrosynthesis eliminating hallucinations with a diverse ensemble of | arXiv: 2510.10645
- truthful calibration errors for multi-class prediction | arXiv: 2510.06388
- tt-vidt decoupling the temporal axis for efficient motion-centric video pretrain | arXiv: 2609.33419
- turbovggt fast visual geometry reconstruction with adaptive alternating attentio | arXiv: 2605.14315
- two-fidelity best-action identification for stochastic minimax tree | arXiv: 2606.01708
- u-bench a comprehensive understanding of u-net through 100-variant benchmarking | arXiv: 2510.07041
- uncertainty quantification for computer-use agents a benchmark across vision-lan | arXiv: 2606.25760
- understanding and mitigating under-confidence in gnns from the final layer | arXiv: 2505.11335
- understanding gradient orthogonalization for deep learning via non-euclidean tru | arXiv: 2503.12645
- understanding the interplay between memorization and learning in large language | arXiv: 2507.14777
- unfoldart zero-shot recovery of full articulated 3d objects from text or image | arXiv: 2606.30608
- unicon3r unified contact-aware 4d human-scene reconstruction from monocular vide | arXiv: 2604.19923
- uniflow-audio unified flow matching for audio generation from omni-modalities | arXiv: 2509.24391
- unifying goal-conditioned rl and unsupervised skill learning via control-maximiz | arXiv: 2605.06145
- unigs unified geometry-aware gaussian splatting for multimodal rendering | arXiv: 2510.12174
- unirap towards unified part-level physical affordance reasoning and actionable p | arXiv: 2610.29768
- unisharp universal sharp monocular view synthesis | arXiv: 2606.07514
- universal cross-prompt adversarial attacks on promptable concept segmentation | arXiv: 2609.39265
- universe unified video action models for autonomous driving with flexible mask-m | arXiv: 2607.05133
- unlearning that lasts utility-preserving robust and almost irreversible forgetti | arXiv: 2509.02820
- unlearningsoup is repeated tuning necessary for large language model unlearning | arXiv: 2609.37076
- unlocking the duality between flow and field matching | arXiv: 2602.02261
- unrestrained simplex denoising for discrete data a non-markovian approach applie | arXiv: 2603.28572
- urbex agentic spatial grounding for city-scale 3d scenes | arXiv: 2610.18356
- ursa chemistry-aware benchmark for utilitarian retrosynthesis assessment | arXiv: 2607.04688
- v2m-zero zero-pair time-aligned video-to-music generation | arXiv: 2603.11042
- variable-length generative protein design via generalized poisson flow | arXiv: 2607.09039
- velocity ambiguity profiles time-resolved bayes-risk diagnostics for flow matchi | arXiv: 2610.30107
- veri-sure multi-agent rtl code generation with temporal tracing slicing and form | arXiv: 2601.19747
- verifiers in the loop decoding time verification for code translation | arXiv: 2605.17626
- verifying neural networks with reinforcement learning | arXiv: 2609.34553
- very fast bayesian additive regression trees on gpu | arXiv: 2410.23244
- vi-bench benchmarking prompt inversion from aigc videos | arXiv: 2609.08079
- vico a training strategy towards semantic aware dynamic high-resolution | arXiv: 2510.12793
- vidic video difference captioning | arXiv: 2512.03405
- vitex-bench benchmarking high-fidelity video scene text editing | arXiv: 2609.40356
- vocalcoachbench benchmarking audio-language models on expert feedback for singin | arXiv: 2609.04241
- vocalgrad evaluating acoustic perception in audio language models | arXiv: 2610.23745
- volfill single-view amodal 3d scene reconstruction with volumetric flow matching | arXiv: 2605.31466
- watermarking should be treated as a monitoring primitive | arXiv: 2605.13095
- webspline structure-informed splines for real-time 3d gaussians from monocular v | arXiv: 2606.02096
- weisfeiler and leman follow the arrow of time expressive power of message passin | arXiv: 2505.24438
- what comes next and why interpretable next-event prediction with neuro-symbolic | arXiv: 2610.31114
- what mllms learn about when they learn about multimodal reasoning | arXiv: 2510.01719
- when form changes but logic doesnt building logic-invariant llms through logical | arXiv: 2610.25402
- when language overrules revealing text dominance in multimodal large language mo | arXiv: 2508.10552
- when noise meets long-tail feature-threshold dual calibration for robust pseudo- | arXiv: 2609.33668
- when one leak pays forever context binding and the price of deterring collusion | arXiv: 2609.36667
- when reasoning meets its laws | arXiv: 2512.17901
- where do long captions fail position-aware diagnosis and reinforcement learning | arXiv: 2610.31111
- where root cause analysis fails a retrieval-reranking decomposition | arXiv: 2609.36686
- who says what symbolic trimodal binding mechanisms in audio-visual llms | arXiv: 2609.31193
- why deterministic prm guidance underperforms in discrete diffusion reasoning | arXiv: 2609.35472
- witeness overlap directional provenance inside open-weight model families | arXiv: 2609.31784
- words that make language models perceive | arXiv: 2510.02425
- working with ai measuring the applicability of generative ai to occupations | arXiv: 2507.07935
- worldsr harnessing world knowledge search for grounded image super-resolution | arXiv: 2610.28423
- x-avdd cross-attentive audio-visual dataset distillation | arXiv: 2610.13687
- your benchmark is an empirical measure over difficulty | arXiv: 2610.29927
- z0-inf zeroth order approximation for data influence | arXiv: 2510.11832
- zeroth-order sharpness-aware learning with exponential tilting | arXiv: 2510.16157
- 分层保形分类利用类别层次结构构建紧致且具覆盖保证的预测集 hierarchical conformal classification | arXiv: 2508.13288
- 对齐归纳偏置以实现状态空间模型的数据高效泛化 aligning inductive bias for data-efficient generalization | arXiv: 2509.20789
- 记忆决定学习方向状态空间模型中基于梯度的优化理论 | arXiv: 2510.00563