English
Related papers

Related papers: TRACE: Distilling Where It Matters via Token-Route…

200 papers

On-Policy Distillation (OPD) has emerged as a dominant post-training paradigm for large language models, especially for reasoning domains. However, OPD remains unstable in practice due to the high gradient variance of its single-sample…

Machine Learning · Computer Science 2026-05-11 Minjae Oh , Sangjun Song , Gyubin Choi , Yunho Choi , Yohan Jo

Reinforcement learning has emerged as a powerful tool for improving diffusion-based text-to-image models, but existing methods are largely limited to single-task optimization. Extending RL to multiple tasks is challenging: joint…

Machine Learning · Computer Science 2026-05-15 Quanhao Li , Junqiu Yu , Kaixun Jiang , Yujie Wei , Zhen Xing , Pandeng Li , Ruihang Chu , Shiwei Zhang , Yu Liu , Zuxuan Wu

Safe Reinforcement Learning (RL) aims to find a policy that achieves high rewards while satisfying cost constraints. When learning from scratch, safe RL agents tend to be overly conservative, which impedes exploration and restrains the…

Robotics · Computer Science 2023-10-16 Jinning Li , Xinyi Liu , Banghua Zhu , Jiantao Jiao , Masayoshi Tomizuka , Chen Tang , Wei Zhan

Improving large language model (LLM) reasoning requires supervision that is both aligned with the model's own test-time states and informative at the token level. Reinforcement learning with verifiable rewards provides on-policy exploration…

Machine Learning · Computer Science 2026-04-30 Zhiquan Tan , Yinrong Hong

Reinforcement learning (RL) post-training has recently driven major gains in long chain-of-thought reasoning large language models (LLMs), but the high inference cost of such models motivates distillation into smaller students. Most…

Machine Learning · Computer Science 2026-04-13 Zhaoyang Zhang , Shuli Jiang , Yantao Shen , Yuting Zhang , Dhananjay Ram , Shuo Yang , Zhuowen Tu , Wei Xia , Stefano Soatto

On-policy distillation (OPD) and on-policy self-distillation (OPSD) have emerged as promising post-training methods for large language models, offering dense token-level supervision on trajectories sampled from the model's own policy.…

Artificial Intelligence · Computer Science 2026-05-26 Siqi Zhu , Xuyan Ye , Hongyu Lu , Weiye Shi , Ge Liu

Reinforcement learning (RL) has emerged as a promising paradigm for inducing explicit reasoning behaviors in large language and vision-language models. However, reasoning-oriented RL post-training remains fundamentally challenging due to…

Artificial Intelligence · Computer Science 2026-05-13 Fan Yang , Rui Meng , Trudi Di Qi , Ali Ezzati , Yuxin Wen

Adapting large language models (LLMs) to long-context tasks requires post-training methods that remain accurate and coherent over thousands of tokens. Existing approaches are limited in several ways: 1) off-policy methods such as supervised…

Computation and Language · Computer Science 2026-05-13 Miguel Moura Ramos , Duarte M. Alves , André F. T. Martins

Search-augmented reasoning agents interleave internal reasoning with calls to an external retriever, and their performance relies on the quality of each issued query. However, under outcome-reward reinforcement learning, every search…

Artificial Intelligence · Computer Science 2026-05-19 Yufei Ma , Zihan Liang , Ben Chen , Zhipeng Qian , Huangyu Dai , Lingtao Mao , Xuxin Zhang , Chenyi Lei , Wenwu Ou

Existing Flow Matching (FM) text-to-image models suffer from two critical bottlenecks under multi-task alignment: the reward sparsity induced by scalar-valued rewards, and the gradient interference arising from jointly optimizing…

Computer Vision and Pattern Recognition · Computer Science 2026-05-26 Zhen Fang , Wenxuan Huang , Yu Zeng , Yiming Zhao , Shuang Chen , Kaituo Feng , Yunlong Lin , Lin Chen , Zehui Chen , Shaosheng Cao , Feng Zhao

Efficient distillation is a key pathway for converting expensive reasoning capability into deployable efficiency, yet in the frontier regime where the student already has strong reasoning ability, naive continual distillation often yields…

Computation and Language · Computer Science 2026-05-22 Zhanming Shen , Jiaqi Hu , Zeyu Qin , Hao Chen , Wentao Ye , Zenan Huang , Yihong Zhuang , Guoshan Lu , Junlin Zhou , Junbo Zhao

Safety alignment often improves robustness to harmful queries at the cost of reasoning ability, a tradeoff known as the safety tax. A common cause is distributional mismatch: supervised fine-tuning trains the target model on safety…

Machine Learning · Computer Science 2026-05-19 Yu Fu , Longxuan Yu , Haz Sameen Shahgir , Zhipeng Wei , Hui Liu , N. Benjamin Erichson , Yue Dong

Training-time privileged information (PI) can enable language models to succeed on tasks they would otherwise fail, making it a powerful tool for reinforcement learning in hard, long-horizon settings. However, transferring capabilities…

Machine Learning · Computer Science 2026-02-17 Emiliano Penaloza , Dheeraj Vattikonda , Nicolas Gontier , Alexandre Lacoste , Laurent Charlin , Massimo Caccia

Reinforcement learning (RL) has been widely used to train LLM agents for multi-turn interactive tasks, but its sample efficiency is severely limited by sparse rewards and long horizons. On-policy self-distillation (OPSD) alleviates this by…

Machine Learning · Computer Science 2026-04-14 Hao Wang , Guozhi Wang , Han Xiao , Yufeng Zhou , Yue Pan , Jichao Wang , Ke Xu , Yafei Wen , Xiaohu Ruan , Xiaoxin Chen , Honggang Qi

On-policy distillation (OPD) trains student models under their own induced distribution while leveraging supervision from stronger teachers. We identify a failure mode of OPD: as training progresses, on-policy rollouts can undergo abrupt…

Computation and Language · Computer Science 2026-04-10 Feng Luo , Yu-Neng Chuang , Guanchu Wang , Zicheng Xu , Xiaotian Han , Tianyi Zhang , Vladimir Braverman

Diffusion and flow models achieve State-Of-The-Art (SOTA) generative performance, yet many practically important behaviors such as fine-grained prompt fidelity, compositional correctness, and text rendering are weakly specified by score or…

Computer Vision and Pattern Recognition · Computer Science 2026-03-17 Yuanzhi Zhu , Xi Wang , Stéphane Lathuilière , Vicky Kalogeiton

On-policy knowledge distillation has proven effective for language models, yet its application to vision-language models (VLMs) remains underexplored. We observe that standard on-policy distillation can improve a student's output quality…

Computer Vision and Pattern Recognition · Computer Science 2026-05-22 Ruiqi Liu , Xiaolei Lv , Gengsheng Li , Ximo Zhu , Zhiheng Wang , Zhengbo Zhang , Junkai Chen , Zhiheng Li , Bo Li , Jun Gao , Shu Wu

As Large Language Models (LLMs) continue to grow in both capability and cost, transferring frontier capabilities into smaller, deployable students has become a central engineering problem, and knowledge distillation remains the dominant…

Machine Learning · Computer Science 2026-05-19 Mingyang Song , Mao Zheng

Transferring reasoning capabilities from larger language models to smaller ones through supervised fine-tuning often fails counterintuitively, with performance degrading despite access to high-quality teacher demonstrations. We identify…

Computation and Language · Computer Science 2025-09-29 Jaehoon Kim , Kwangwook Seo , Dongha Lee

Injecting new reasoning knowledge into Large Language Models (LLMs) via post-training often induces catastrophic forgetting. Recent studies emphasize the importance of on-policy data but suggest that KL-divergence fails to mitigate…

Computation and Language · Computer Science 2026-05-19 Wenye Lin , Kai Han