English
Related papers

Related papers: Reward Shaping to Mitigate Reward Hacking in RLHF

200 papers

Despite notable advancements in Multimodal Large Language Models (MLLMs), most state-of-the-art models have not undergone thorough alignment with human preferences. This gap exists because current alignment research has primarily achieved…

Reinforcement learning from human feedback (RLHF) emerges as a promising paradigm for aligning large language models (LLMs). However, a notable challenge in RLHF is overoptimization, where beyond a certain threshold, the pursuit of higher…

Machine Learning · Computer Science 2024-01-02 Yuanzhao Zhai , Han Zhang , Yu Lei , Yue Yu , Kele Xu , Dawei Feng , Bo Ding , Huaimin Wang

Advancements in Natural Language Processing (NLP), have led to the emergence of Large Language Models (LLMs) such as GPT, Llama, Claude, and Gemini, which excel across a range of tasks but require extensive fine-tuning to align their…

Computation and Language · Computer Science 2025-04-01 Angela Lopez-Cardona , Carlos Segura , Alexandros Karatzoglou , Sergi Abadal , Ioannis Arapakis

In many real-world scenarios, reward signal for agents are exceedingly sparse, making it challenging to learn an effective reward function for reward shaping. To address this issue, the proposed approach in this paper performs reward…

Machine Learning · Computer Science 2026-05-18 Wenyun Li , Wenjie Huang , Chen Sun

Reinforcement learning from human feedback (RLHF) with reward models has advanced alignment of generative models to human aesthetic and perceptual preferences. However, jointly optimizing multiple rewards often incurs an alignment tax,…

Computer Vision and Pattern Recognition · Computer Science 2026-03-17 Chieh-Yun Chen , Zhonghao Wang , Qi Chen , Zhifan Ye , Min Shi , Yue Zhao , Yinan Zhao , Hui Qu , Wei-An Lin , Yiru Shen , Ajinkya Kale , Irfan Essa , Humphrey Shi

While Reinforcement Learning from Human Feedback (RLHF) is widely used to align Large Language Models (LLMs) with human preferences, it typically assumes homogeneous preferences across users, overlooking diverse human values and minority…

Computation and Language · Computer Science 2025-10-28 Yijiang River Dong , Tiancheng Hu , Yinhong Liu , Ahmet Üstün , Nigel Collier

A centerpiece of the ever-popular reinforcement learning from human feedback (RLHF) approach to fine-tuning autoregressive language models is the explicit training of a reward model to emulate human feedback, distinct from the language…

Computation and Language · Computer Science 2023-05-22 Wanqiao Xu , Shi Dong , Dilip Arumugam , Benjamin Van Roy

Reward models (RMs) used in reinforcement learning from human feedback (RLHF) are vulnerable to reward hacking: as the policy maximizes a learned proxy reward, true quality plateaus or degrades. We make the assumption that reward hacking is…

Machine Learning · Computer Science 2026-04-06 Shinnosuke Ono , Johannes Ackermann , Soichiro Nishimori , Takashi Ishida , Masashi Sugiyama

Reinforcement Learning from Human Feedback (RLHF) has shown great potential in fine-tuning Large Language Models (LLMs) to align with human preferences. Existing methods perform preference alignment from a fixed dataset, which can be…

Machine Learning · Computer Science 2025-02-10 Chenjia Bai , Yang Zhang , Shuang Qiu , Qiaosheng Zhang , Kang Xu , Xuelong Li

Reinforcement Learning from Human Feedback (RLHF) has enabled significant advancements within language modeling for powerful, instruction-following models. However, the alignment of these models remains a pressing challenge as the policy…

Machine Learning · Computer Science 2024-06-21 Ahmed M. Ahmed , Rafael Rafailov , Stepan Sharkov , Xuechen Li , Sanmi Koyejo

In natural human-to-human conversations, participants often receive feedback signals from one another based on their follow-up reactions. These reactions can include verbal responses, facial expressions, changes in emotional state, and…

Computation and Language · Computer Science 2025-02-25 Chen Zhang , Dading Chong , Feng Jiang , Chengguang Tang , Anningzhe Gao , Guohua Tang , Haizhou Li

Recent advances in machine learning have shown that Reinforcement Learning from Human Feedback (RLHF) can improve machine learning models and align them with human preferences. Although very successful for Large Language Models (LLMs),…

Computer Vision and Pattern Recognition · Computer Science 2023-11-07 Katie Z Luo , Zhenzhen Liu , Xiangyu Chen , Yurong You , Sagie Benaim , Cheng Perng Phoo , Mark Campbell , Wen Sun , Bharath Hariharan , Kilian Q. Weinberger

Reinforcement learning (RL) systems typically optimize scalar reward functions that assume precise and reliable evaluation of outcomes. However, real-world objectives--especially those derived from human preferences--are often uncertain,…

Machine Learning · Computer Science 2026-04-30 Disha Singha

Reward models (RMs) are essential in reinforcement learning from human feedback (RLHF) to align large language models (LLMs) with human values. However, RM training data is commonly recognized as low-quality, containing inductive biases…

Machine Learning · Computer Science 2026-05-20 Zhuo Li , Pengyu Cheng , Zhechao Yu , Feifei Tong , Anningzhe Gao , Tsung-Hui Chang , Xiang Wan , Erchao Zhao , Xiaoxi Jiang , Guanjun Jiang

Learning rewards from preference feedback has become an important tool in the alignment of agentic models. Preference-based feedback, often implemented as a binary comparison between multiple completions, is an established method to acquire…

Machine Learning · Computer Science 2025-03-03 Yannick Metz , András Geiszl , Raphaël Baur , Mennatallah El-Assady

Reinforcement learning, which acquires a policy maximizing long-term rewards, has been actively studied. Unfortunately, this learning type is too slow and difficult to use in practical situations because the state-action space becomes huge…

Machine Learning · Computer Science 2024-10-28 Takato Okudo , Seiji Yamada

Modern language models often rely on Reinforcement Learning from Human Feedback (RLHF) to encourage safe behaviors. However, they remain vulnerable to adversarial attacks due to three key limitations: (1) the inefficiency and high cost of…

Existing alignment methods share a common topology of information flow, where reward information is collected from humans, modeled with preference learning, and used to tune language models. However, this shared topology has not been…

Machine Learning · Computer Science 2025-05-29 Tianyi Qiu , Fanzhi Zeng , Jiaming Ji , Dong Yan , Kaile Wang , Jiayi Zhou , Yang Han , Josef Dai , Xuehai Pan , Yaodong Yang

Uniform-reward reinforcement learning from human feedback (RLHF), which trains a single reward model to represent the preferences of all annotators, fails to capture the diversity of opinions across sub-populations, inadvertently favoring…

Machine Learning · Computer Science 2025-09-05 Arpan Mukherjee , Marcello Bullo , Deniz Gündüz

Faithfulness, expressiveness, and elegance is the constant pursuit in machine translation. However, traditional metrics like \textit{BLEU} do not strictly align with human preference of translation quality. In this paper, we explore…

Computation and Language · Computer Science 2024-02-28 Nuo Xu , Jun Zhao , Can Zu , Sixian Li , Lu Chen , Zhihao Zhang , Rui Zheng , Shihan Dou , Wenjuan Qin , Tao Gui , Qi Zhang , Xuanjing Huang