4 articles — 🔴 1 critical , 🟡 2 important , 🟢 1 interesting
Process Reward Agents enable small frozen models (0.5B-8B) to significantly improve medical reasoning without any training — Qwen3-4B achieves a new state-of-the-art of 80.8% on MedQA.
Sequence-Level PPO reformulates LLM reasoning as a contextual bandit problem, achieving the performance of expensive group methods like GRPO with dramatically fewer resources — without multi-sampling.
A new benchmark reveals a universal judgment deficiency in AI agents — when specifications are incomplete, no frontier model achieves more than a fraction of its full performance. Researchers show this skill can be trained with RL.
A new benchmark for customer services reveals two phenomena: 'Execution Gap' (models correctly classify intents but don't perform the correct actions) and 'Empathy Resilience' (models remain polite while making logical errors).