💡 LLM Reasoning

🎞️ ECCV2026 · 1 paper notes

📌 Same area in other venues: 📷 CVPR2026 (16) · 🔬 ICLR2026 (241) · 💬 ACL2026 (82) · 🧪 ICML2026 (78) · 🤖 AAAI2026 (37) · 🧠 NeurIPS2025 (82)

ProAct: Agentic Lookahead in Interactive Environments

ProAct moves lookahead out of inference time and into data construction: it first probes the real environment with MCTS and compresses the resulting futures into observation-analysis-conclusion causal chains for SFT (GLAD), then stabilizes PPO/GRPO training with a parameter-free Monte-Carlo Critic (MC-Critic) that estimates values from cheap random-policy rollouts, letting a 4B model substantially outperform open-source baselines of the same and larger scale on 2048 and Sokoban.