arXiv:2607. 18966v1 Announce Type: new Abstract: Language models trained with reinforcement learning may learn to optimize the grader's judgment rather than the intended objective.
Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.
arXiv:2608. 13622v1 Announce Type: new Abstract: Open-ended real-world interaction admits multiple valid behaviors: an agent may answer directly, ask for clarification, provide progress updates, or confirm before acting.
arXiv:2606. 26300v1 Announce Type: new Abstract: A classical intuition holds that verifying a solution is easier than producing one.
arXiv:2607. 05394v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) is a powerful recipe for improving language-model reasoning, but it is expensive to repeat on every new strong model because the target model must generate many rollouts during training.
arXiv:2608. 04794v1 Announce Type: new Abstract: Self-distillation (SD) has emerged as a compute-efficient alternative to reinforcement learning with verifiable rewards: a self-teacher, conditioned on privileged information (PI) about the answer such as a reference solution, supplies dense per-token supervision to a student that never sees it.
arXiv:2607. 05904v1 Announce Type: new Abstract: Training a language model against its own reference-free judgments (the premise of self-rewarding, self-play, and LLM-as-a-judge pipelines) assumes a model's verdict on a shown answer tracks correctness.
arXiv:2602. 12124v2 Announce Type: replace Abstract: While most AI alignment research focuses on preventing models from generating explicitly harmful content, a more subtle risk arises from capability-seeking RL training in vulnerable environments.