🧊 3D Vision

💬 ACL2026 · 1 paper notes

📌 Same area in other venues: 🧪 ICML2026 (27) · 📷 CVPR2026 (230) · 🔬 ICLR2026 (63) · 🤖 AAAI2026 (76) · 🧠 NeurIPS2025 (112) · 📹 ICCV2025 (264)

CodeBind: Decoupled Representation Learning for Multimodal Alignment with Unified Compositional Codebook

CodeBind enhances ImageBind/ViT-Lens style multimodal alignment using shared-specific representation decoupling and a unified compositional VQ codebook. It simultaneously improves cross-modal classification/retrieval across nine modalities while preserving stronger modality-specific fine-grained information.