SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs Paper • 2608.03573 • Published Aug 6 • 60
Beyond Simply Environment Scaling: Designing Effective Environment Distributions for Multimodal Agent Learning Paper • 2608.03571 • Published Aug 6 • 46
The Physics of Multi-Turn Long-Horizon Planning: From Pre-training to Post-training via Single- and Multi-Teacher On-Policy Agentic Distillation Paper • 2607.24720 • Published Jul 27 • 26
Why Multi-Step Tool-Use Reinforcement Learning Collapses and How Supervisory Signals Fix It Paper • 2606.26027 • Published Jun 24 • 19
Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Application Paper • 2606.12191 • Published Jun 10 • 73