English
Related papers

Related papers: Rethinking KL Regularization in RLHF: From Value E…

200 papers

Reinforcement Learning from Human Feedback (RLHF) or Verifiable Rewards (RLVR) are two key steps in the post-training of modern Language Models (LMs). A common problem is reward hacking, where the policy may exploit inaccuracies of the…

Machine Learning · Computer Science 2026-02-23 Johannes Ackermann , Michael Noukhovitch , Takashi Ishida , Masashi Sugiyama

Reinforcement Learning from Human Feedback (RLHF) plays a crucial role in aligning large language models (LLMs) with human values and preferences. However, the quality and stability of the trained reward model largely determine the final…

Machine Learning · Computer Science 2025-12-17 Chunjin Jian , Xinhua Zhu

Alignment and adaptation in large language models heavily rely on reinforcement learning from human feedback (RLHF); yet, theoretical understanding of its generalisability remains premature, especially when the learned reward could shift,…

Machine Learning · Computer Science 2026-02-26 Kenton Tang , Yuzhu Chen , Fengxiang He

Reinforcement Learning from Human Feedback (RLHF) is crucial for aligning Large Language Models (LLMs) with human values. However, RLHF has been continuously challenged by its high complexity in implementation and computation consumption,…

Machine Learning · Computer Science 2026-03-24 Yuhao Du , Zhuo Li , Pengyu Cheng , Zhihong Chen , Yuejiao Xie , Xiang Wan , Anningzhe Gao

Reinforcement Learning from Human Feedback (RLHF) has advanced alignment capabilities significantly but remains hindered by two core challenges: \textbf{reward hacking} and \textbf{stable optimization}. Current solutions independently…

Machine Learning · Computer Science 2026-02-13 Li He , Qiang Qu , He Zhao , Stephen Wan , Dadong Wang , Lina Yao , Tongliang Liu

The effectiveness of reinforcement learning (RL) agents in continuous control robotics tasks is mainly dependent on the design of the underlying reward function, which is highly prone to reward hacking. A misalignment between the reward…

Alignment via reinforcement learning from human feedback (RLHF) has become the dominant paradigm for controlling the quality of outputs from large language models (LLMs). However, existing theories do not provide strong justification for…

Machine Learning · Computer Science 2026-05-19 Jihun Yun , Juno Kim , Jongho Park , Junhyuck Kim , Jongha Jon Ryu , Jaewoong Cho , Kwang-Sung Jun

Reinforcement learning from human feedback (RLHF) is a prevalent approach to align AI systems with human values by learning rewards from human preference data. Due to various reasons, however, such data typically takes the form of rankings…

Machine Learning · Computer Science 2024-06-06 Ilgee Hong , Zichong Li , Alexander Bukharin , Yixiao Li , Haoming Jiang , Tianbao Yang , Tuo Zhao

Reinforcement Learning from Human Feedback (RLHF) and its variants have emerged as the dominant approaches for aligning Large Language Models with human intent. While empirically effective, the theoretical generalization properties of these…

Machine Learning · Computer Science 2026-01-26 Zhaochun Li , Mingyang Yi , Yue Wang , Shisheng Cui , Yong Liu

The reasoning performance of large language models (LLMs) can be substantially improved by training them with reinforcement learning (RL). The RL objective for LLM training involves a regularization term, which is the reverse…

Group Relative Policy Optimization (GRPO) is widely used for critic-free Large Language Model (LLM) post-training, but its KL regularization is usually implemented as a local loss-side token penalty. We show that this misses the…

Machine Learning · Computer Science 2026-05-26 Jiarui Yao , Ruida Wang , Hao Bai , Tong Zhang

This paper studies the alignment process of generative models with Reinforcement Learning from Human Feedback (RLHF). We first identify the primary challenges of existing popular methods like offline PPO and offline DPO as lacking in…

Machine Learning · Computer Science 2024-05-02 Wei Xiong , Hanze Dong , Chenlu Ye , Ziqi Wang , Han Zhong , Heng Ji , Nan Jiang , Tong Zhang

Reinforcement learning from human feedback (RLHF) or verifiable rewards (RLVR), the standard paradigm for aligning LLMs or building recent SOTA reasoning models, is highly sensitive to noise from inconsistent or erroneous rewards. Yet, the…

Machine Learning · Computer Science 2026-05-20 Omar El Mansouri , Fathinah Asma Izzati , Mohamed El Amine Seddik , Salem Lahlou

Standard reinforcement learning from human feedback (RLHF) trains a reward model on pairwise preference data and then uses it for policy optimization. However, while reward models are optimized to capture relative preferences, existing…

Machine Learning · Computer Science 2026-02-05 Kyuseong Choi , Dwaipayan Saha , Woojeong Kim , Anish Agarwal , Raaz Dwivedi

Recent work shows that preference alignment objectives can be interpreted as divergence estimators between aligned (preferred) & unaligned (less-preferred) distributions, yielding a principled recipe for designing alignment losses. However,…

Machine Learning · Computer Science 2026-05-12 Rajdeep Haldar , Lantao Mei , Guang Lin , Yue Xing , Qifan Song

Reinforcement Learning with Verified Reward (RLVR) has emerged as a critical paradigm for advancing the reasoning capabilities of Large Language Models (LLMs). Most existing RLVR methods, such as GRPO and its variants, ensure stable updates…

Machine Learning · Computer Science 2026-02-10 Qingyuan Wu , Yuhui Wang , Simon Sinong Zhan , Yanning Dai , Shilong Deng , Sarra Habchi , Qi Zhu , Matthias Gallé , Chao Huang

Reinforcement learning from human feedback (RLHF) is an essential technique for ensuring that large language models (LLMs) are aligned with human values and preferences during the post-training phase. As an effective RLHF approach, group…

Machine Learning · Computer Science 2025-06-18 Zonglin Yang , Zhexuan Gu , Houduo Qi , Yancheng Yuan

The alignment of large language models (LLMs) with human preferences is commonly achieved through Reinforcement Learning from Human Feedback (RLHF). Direct Preference Optimization (DPO) simplified this paradigm by establishing a direct…

Machine Learning · Computer Science 2025-09-26 Yuandong Tan

While Reinforcement Learning (RL) has been proven essential for tuning large language models (LLMs), it can lead to reward over-optimization (ROO). Existing approaches address ROO by adding KL regularization, requiring computationally…

Computation and Language · Computer Science 2024-05-01 Mathieu Rita , Florian Strub , Rahma Chaabouni , Paul Michel , Emmanuel Dupoux , Olivier Pietquin

Aligning human preference and value is an important requirement for building contemporary foundation models and embodied AI. However, popular approaches such as reinforcement learning with human feedback (RLHF) break down the task into…

Artificial Intelligence · Computer Science 2024-12-03 Chenliang Li , Siliang Zeng , Zeyi Liao , Jiaxiang Li , Dongyeop Kang , Alfredo Garcia , Mingyi Hong
‹ Prev 1 2 3 10 Next ›