English
Related papers

Related papers: Scaling Reward Modeling without Human Supervision

200 papers

Process reward models (PRMs) that provide dense, step-level feedback have shown promise for reinforcement learning, yet their adoption remains limited by the need for expensive step-level annotations or ground truth references. We propose…

Machine Learning · Computer Science 2025-12-04 Salman Rahman , Sruthi Gorantla , Arpit Gupta , Swastik Roy , Nanyun Peng , Yang Liu

Large language models (LLMs) excel at reasoning, yet post-training remains critical for aligning their behavior with task goals. Existing reinforcement learning (RL) methods often depend on costly human annotations or external reward…

Computation and Language · Computer Science 2025-06-12 Pengyi Li , Matvey Skripkin , Alexander Zubrey , Andrey Kuznetsov , Ivan Oseledets

Step-level reward models (SRMs) can significantly enhance mathematical reasoning performance through process supervision or step-level preference alignment based on reinforcement learning. The performance of SRMs is pivotal, as they serve…

Artificial Intelligence · Computer Science 2025-03-11 Yiran Ma , Zui Chen , Tianqiao Liu , Mi Tian , Zhuo Liu , Zitao Liu , Weiqi Luo

Reinforcement learning (RL) has recently proven great success in various domains. Yet, the design of the reward function requires detailed domain expertise and tedious fine-tuning to ensure that agents are able to learn the desired…

Robotics · Computer Science 2023-03-06 Murad Dawood , Nils Dengler , Jorge de Heuvel , Maren Bennewitz

Recent advancements in improving the reasoning capabilities of Large Language Models have underscored the efficacy of Process Reward Models (PRMs) in addressing intermediate errors through structured feedback mechanisms. This study analyzes…

Computation and Language · Computer Science 2025-06-03 Zhengyu Chen , Yudong Wang , Teng Xiao , Ruochen Zhou , Xuesheng Yang , Wei Wang , Zhifang Sui , Jingang Wang

Reinforcement learning provides a general framework for learning robotic skills while minimizing engineering effort. However, most reinforcement learning algorithms assume that a well-designed reward function is provided, and learn a single…

Robotics · Computer Science 2020-04-28 Archit Sharma , Michael Ahn , Sergey Levine , Vikash Kumar , Karol Hausman , Shixiang Gu

Reward-based fine-tuning steers a pretrained diffusion or flow-based generative model toward higher-reward samples while remaining close to the pretrained model. Although existing methods are derived from different perspectives, we show…

Machine Learning · Computer Science 2026-05-08 Jeongjae Lee , Jinho Chang , Jeongsol Kim , Jong Chul Ye

In offline reinforcement learning (RL), learning from fixed datasets presents a promising solution for domains where real-time interaction with the environment is expensive or risky. However, designing dense reward signals for offline…

Machine Learning · Computer Science 2025-04-15 Younghwan Lee , Tung M. Luu , Donghoon Lee , Chang D. Yoo

Reward Models (RMs) are crucial for online alignment of language models (LMs) with human preferences. However, RM-based preference-tuning is vulnerable to reward hacking, whereby LM policies learn undesirable behaviors from flawed RMs. By…

Computation and Language · Computer Science 2026-03-05 Daniel Fein , Max Lamparth , Violet Xiang , Mykel J. Kochenderfer , Nick Haber

Deep Reinforcement Learning (RL) has emerged as a powerful paradigm to solve a range of complex yet specific control tasks. Yet training generalist agents that can quickly adapt to new tasks remains an outstanding challenge. Recent advances…

Machine Learning · Computer Science 2021-10-29 Michael Laskin , Denis Yarats , Hao Liu , Kimin Lee , Albert Zhan , Kevin Lu , Catherine Cang , Lerrel Pinto , Pieter Abbeel

Reward shaping is an effective technique for incorporating domain knowledge into reinforcement learning (RL). Existing approaches such as potential-based reward shaping normally make full use of a given shaping reward function. However,…

Machine Learning · Computer Science 2020-11-06 Yujing Hu , Weixun Wang , Hangtian Jia , Yixiang Wang , Yingfeng Chen , Jianye Hao , Feng Wu , Changjie Fan

Reward shaping has been applied widely to accelerate Reinforcement Learning (RL) agents' training. However, a principled way of designing effective reward shaping functions, especially for complex continuous control problems, remains…

Machine Learning · Computer Science 2026-02-12 Mateo Juliani , Mingxuan Li , Elias Bareinboim

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

Model-free reinforcement learning (RL) is a powerful approach for learning control policies directly from high-dimensional state and observation. However, it tends to be data-inefficient, which is especially costly in robotic learning…

Robotics · Computer Science 2020-10-14 Xubo Lyu , Mo Chen

Reward models have become a staple in modern NLP, serving as not only a scalable text evaluator, but also an indispensable component in many alignment recipes and inference-time algorithms. However, while recent reward models increase…

Computation and Language · Computer Science 2025-09-22 Zhaofeng Wu , Michihiro Yasunaga , Andrew Cohen , Yoon Kim , Asli Celikyilmaz , Marjan Ghazvininejad

Reward models (RMs) play a crucial role in Reinforcement Learning from Human Feedback by serving as proxies for human preferences in aligning large language models. However, they suffer from various biases which could lead to reward…

Artificial Intelligence · Computer Science 2026-03-18 Xiao Zhu , Chenmien Tan , Pinzhen Chen , Rico Sennrich , Huiming Wang , Yanlin Zhang , Hanxu Hu

In reinforcement learning (RL), sparse rewards are a natural way to specify the task to be learned. However, most RL algorithms struggle to learn in this setting since the learning signal is mostly zeros. In contrast, humans are good at…

Learning from human feedback is a popular approach to train robots to adapt to user preferences and improve safety. Existing approaches typically consider a single querying (interaction) format when seeking human feedback and do not…

Robotics · Computer Science 2026-01-16 Yashwanthi Anand , Nnamdi Nwagwu , Kevin Sabbe , Naomi T. Fitter , Sandhya Saisubramanian

Teaching robots novel skills with demonstrations via human-in-the-loop data collection techniques like kinesthetic teaching or teleoperation puts a heavy burden on human supervisors. In contrast to this paradigm, it is often significantly…

Robotics · Computer Science 2024-04-24 Daniel Yang , Davin Tjia , Jacob Berg , Dima Damen , Pulkit Agrawal , Abhishek Gupta

Semi-supervised learning (SSL) has witnessed great progress with various improvements in the self-training framework with pseudo labeling. The main challenge is how to distinguish high-quality pseudo labels against the confirmation bias.…

Machine Learning · Computer Science 2024-02-21 Siyuan Li , Weiyang Jin , Zedong Wang , Fang Wu , Zicheng Liu , Cheng Tan , Stan Z. Li