arXiv:2606. 04036v1 Announce Type: new Abstract: On-policy self-distillation, where a language model conditions on privileged context to supervise its own generations, is a promising source of dense supervision for sparse-reward reinforcement learning.
arXiv:2607. 18955v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) has substantially improved the reasoning capabilities of large language models on tasks such as mathematical reasoning and code generation.
arXiv:2607. 28076v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) is effective for training large language model agents.
Recent advances in post-training Large Language Models (LLMs) increasingly rely on Reinforcement Learning with Verifiable Rewards (RLVR) or On-Policy Self-Distillation (OPSD). While OPSD provides dense, logit-level supervision, it inherently suffers from exposure bias due to the privileged information of the self-teacher.
Continual post-training enables foundation models to acquire new knowledge while preserving existing capabilities. Recent work suggests that on-policy learning can mitigate forgetting, with on-policy self-distillation emerging as a particularly attractive approach.
arXiv:2607. 28026v1 Announce Type: new Abstract: Recent advances in post-training Large Language Models (LLMs) increasingly rely on Reinforcement Learning with Verifiable Rewards (RLVR) or On-Policy Self-Distillation (OPSD).
arXiv:2601. 19897v2 Announce Type: replace Abstract: Continual learning, enabling models to acquire new skills and knowledge without degrading existing capabilities, remains a fundamental challenge for foundation models.
arXiv:2607. 01763v1 Announce Type: new Abstract: Continual post-training enables foundation models to acquire new knowledge while preserving existing capabilities.
arXiv:2606. 03620v1 Announce Type: cross Abstract: Self-distilled policy optimization (SDPO) has become a popular paradigm for LLM post-training, where a model learns from its own predictions conditioned on privileged information.
RISE (Recursive Improvement via Self-Extrapolating Policy Distillation) is a new method that builds a synthetic teacher from a language model’s own RLVR training trajectory. By extrapolating the displacement between the current checkpoint and a trailing anchor in parameter or logit space, RISE transforms sparse outcome-based updates into dense token-level targets without external models or privileged conditioning. The approach recursively refines the student model, combining RLVR and on‑policy distillation, and demonstrates superior performance across mathematical reasoning, STEM, code generation, and multi‑turn agentic tasks.
arXiv:2609.16937v1 Announce Type: cross Abstract: On-policy distillation (OPD) has emerged as an effective approach for large language model post-training, yet existing objectives face a trade-off be...
arXiv:2606. 13795v1 Announce Type: new Abstract: RL post-training has become increasingly pivotal for improving diffusion policies, but existing diffusion policy-gradient methods are often unstable and cannot achieve reliable policy improvement.
Post-training Large Language Models (LLMs) with Reinforcement Learning (RL) has become an important tool for improving model capabilities, but the LLM action-space structure introduces challenges distinct from classical RL, with implications for inducing exploration. New methods are required that leverage the broad knowledge and flexibility of pre-trained LLMs to deliberately generate diverse experience at training time.