English
Related papers

Related papers: OPSD Compresses What RLVR Teaches: A Post-RL Compa…

200 papers

The standard post-training recipe for large multimodal models (LMMs) applies supervised fine-tuning (SFT) on curated demonstrations followed by reinforcement learning with verifiable rewards (RLVR). However, SFT introduces distributional…

Computer Vision and Pattern Recognition · Computer Science 2026-05-05 Sudong Wang , Weiquan Huang , Xiaomin Yu , Zuhao Yang , Hehai Lin , Keming Wu , Chaojun Xiao , Chen Chen , Wenxuan Wang , Beier Zhu , Yunjian Zhang , Chengwei Qin

Standard knowledge distillation for autoregressive models often suffers from distribution mismatch. While on-policy methods mitigate this by leveraging student-generated outputs, they rely on computationally expensive Reinforcement Learning…

Machine Learning · Computer Science 2026-05-08 Miao Rang , Zhenni Bi , Hang Zhou , Kai Han , Xuechun Wang , An Xiao , Xinghao Chen , Yunhe Wang , Hanting Chen

Scaling test-time compute via long Chain-of-Thought unlocks remarkable gains in reasoning capabilities, yet it faces practical limits due to the linear growth of KV cache and quadratic attention complexity. In this paper, we introduce…

Artificial Intelligence · Computer Science 2026-04-10 Zhicheng Yang , Zhijiang Guo , Yinya Huang , Yongxin Wang , Wenlei Shi , Yiwei Wang , Xiaodan Liang , Jing Tang

Training reasoning language models (LMs) with reinforcement learning (RL) for one-hot correctness inherently relies on the LM being able to explore and solve its task with some chance at initialization. Furthermore, a key use case of…

Machine Learning · Computer Science 2025-10-30 Edoardo Cetin , Tianyu Zhao , Yujin Tang

Supervised fine-tuning (SFT) is the predominant method for adapting large language models (LLMs), yet it often struggles with generalization compared to reinforcement learning (RL). In this work, we posit that this performance disparity…

Computation and Language · Computer Science 2026-02-03 Rui Ming , Haoyuan Wu , Shoubo Hu , Zhuolun He , Bei Yu

Current multimodal models often suffer from shallow reasoning, leading to errors caused by incomplete or inconsistent thought processes. To address this limitation, we propose Self-Verification and Self-Rectification (SVSR), a unified…

Artificial Intelligence · Computer Science 2026-05-29 Zhe Qian , Nianbing Su , Zhonghua Wang , Hebei Li , Zhongxing Xu , Yueying Li , Fei Luo , Zhuohan Ouyang , Yanbiao Ma

Most provably-efficient learning algorithms introduce optimism about poorly-understood states and actions to encourage exploration. We study an alternative approach for efficient exploration, posterior sampling for reinforcement learning…

Machine Learning · Statistics 2013-12-30 Ian Osband , Daniel Russo , Benjamin Van Roy

Aligned models can misbehave in several ways: they are often sycophantic, fall victim to jailbreaks, or fail to include appropriate safety warnings. Consistency training is a promising new alignment paradigm to mitigate such failures by…

Machine Learning · Computer Science 2026-05-22 Andy Han , Kristina Fujimoto , Avidan Shah , Kiet Nguyen , Kai Xu , Chen Yueh-Han , Ilia Sucholutsky , Rico Angell

Latent reasoning represents a new development in Transformer language models that has shown potential in compressing reasoning lengths compared to chain-of-thought reasoning. By directly passing the information-rich previous final latent…

Machine Learning · Computer Science 2025-11-27 Alex Ning , Yen-Ling Kuo , Gabe Gomes

Reinforcement learning (RL) has emerged as a promising paradigm for training reasoning-oriented models by leveraging rule-based reward signals. However, RL training typically tends to improve single-sample success rates (i.e., Pass@1) while…

Computation and Language · Computer Science 2026-04-21 Yifu Huo , Chenglong Wang , Ziming Zhu , Shunjie Xing , Peinan Feng , Tongran Liu , Qiaozhi He , Tianhua Zhou , Xiaojia Chang , Jingbo Zhu , Zhengtao Yu , Tong Xiao

Looped language models (LoopLMs) perform iterative latent computation to refine internal representations, offering a promising alternative to explicit chain-of-thought (CoT) reasoning. However, existing reinforcement learning (RL) paradigms…

Computation and Language · Computer Science 2026-03-23 Guo Tang , Shixin Jiang , Heng Chang , Nuo Chen , Yuhan Li , Huiming Fan , Jia Li , Ming Liu , Bing Qin

In sequential decision-making problems, Return-Conditioned Supervised Learning (RCSL) has gained increasing recognition for its simplicity and stability in modern decision-making tasks. Unlike traditional offline reinforcement learning (RL)…

Machine Learning · Computer Science 2025-06-11 Zhishuai Liu , Yu Yang , Ruhan Wang , Pan Xu , Dongruo Zhou

This study investigates the effectiveness of reinforcement learning (RL) fine-tuning techniques on a compact language model (Qwen2.5-0.5B Base) for two challenging tasks: instruction following and mathematical reasoning. We compare…

Computation and Language · Computer Science 2025-07-29 Yifu Han , Geo Zhang

Post-training has become central to turning pretrained large language models (LLMs) into aligned, capable, and deployable systems. Recent progress spans supervised fine-tuning (SFT), preference optimization, reinforcement learning (RL),…

Computation and Language · Computer Science 2026-04-17 Shiwan Zhao , Zhihu Wang , Xuyang Zhao , Jiaming Zhou , Caiyue Xu , Chenfei Liu , Liting Zhang , Yuhang Jia , Yanzhe Zhang , Hualong Yu , Zichen Xu , Qicheng Li , Yong Qin

Off-policy learning ability is an important feature of reinforcement learning (RL) for practical applications. However, even one of the most elementary RL algorithms, temporal-difference (TD) learning, is known to suffer form divergence…

Machine Learning · Computer Science 2025-04-21 Han-Dong Lim , Donghwan Lee

Reinforcement Learning with Verifiable Rewards (RLVR) enhances reasoning of Large Language Models (LLMs) but usually exhibits limited generation diversity due to the over-incentivization of positive rewards. Although methods like Negative…

Machine Learning · Computer Science 2026-05-11 Zihan Lin , Xiaohan Wang , Jie Cao , Jiajun Chai , Li Wang , Xiaodong Lu , Wei Lin , Ran He , Guojun Yin

Deep reinforcement learning has demonstrated remarkable achievements across diverse domains such as video games, robotic control, autonomous driving, and drug discovery. Common methodologies in partially-observable domains largely lean on…

Machine Learning · Computer Science 2024-02-15 Michael Lanier , Ying Xu , Nathan Jacobs , Chongjie Zhang , Yevgeniy Vorobeychik

Foundation models encode rich structural knowledge but often rely on post-training procedures to adapt their reasoning behavior to specific tasks. Popular approaches such as reinforcement learning with verifiable rewards (RLVR) and…

Machine Learning · Computer Science 2026-01-21 Dake Bu , Wei Huang , Andi Han , Atsushi Nitanda , Bo Xue , Qingfu Zhang , Hau-San Wong , Taiji Suzuki

Through reinforcement learning (RL) with outcome correctness rewards, large reasoning models (LRMs) with scaled inference computation have demonstrated substantial success on complex reasoning tasks. However, the one-sided reward, focused…

Computation and Language · Computer Science 2025-11-21 Jiashu Yao , Heyan Huang , Shuang Zeng , Chuwei Luo , WangJie You , Jie Tang , Qingsong Liu , Yuhang Guo , Yangyang Kang

Reinforcement learning using verifiable rewards (RLVR) improves LLM reasoning, but the conditions under which it transfers across domains -- and why it does so -- remain under-explored. We study cross-domain transfer in a 7B model whose SFT…

Machine Learning · Computer Science 2026-05-29 Mayug Maniparambil , Arjun Karuvally , Terrence Sejnowski , Fergal Reid