Automatically Finding Reward Model Biases
Fuente:
arXiv
Saved in:
| Main Authors: | Wang, Atticus, Arcuschin, Iván, Conmy, Arthur |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Similar Items
Base Models Know How to Reason, Thinking Models Learn When
by: Venhoff, Constantin, et al.
Published: (2025)
by: Venhoff, Constantin, et al.
Published: (2025)
Understanding Reasoning in Thinking Language Models via Steering Vectors
by: Venhoff, Constantin, et al.
Published: (2025)
by: Venhoff, Constantin, et al.
Published: (2025)
Chain-of-Thought Reasoning In The Wild Is Not Always Faithful
by: Arcuschin, Iván, et al.
Published: (2025)
by: Arcuschin, Iván, et al.
Published: (2025)
Biases in the Blind Spot: Detecting What LLMs Fail to Mention
by: Arcuschin, Iván, et al.
Published: (2026)
by: Arcuschin, Iván, et al.
Published: (2026)
Inference-Time Toxicity Mitigation in Protein Language Models
by: Burda, Manuel Fernández, et al.
Published: (2026)
by: Burda, Manuel Fernández, et al.
Published: (2026)
Applying sparse autoencoders to unlearn knowledge in language models
by: Farrell, Eoin, et al.
Published: (2024)
by: Farrell, Eoin, et al.
Published: (2024)
Improving Steering Vectors by Targeting Sparse Autoencoder Features
by: Chalnev, Sviatoslav, et al.
Published: (2024)
by: Chalnev, Sviatoslav, et al.
Published: (2024)
Thought Anchors: Which LLM Reasoning Steps Matter?
by: Bogdan, Paul C., et al.
Published: (2025)
by: Bogdan, Paul C., et al.
Published: (2025)
Combining Causal Models for More Accurate Abstractions of Neural Networks
by: Pîslar, Theodora-Mara, et al.
Published: (2025)
by: Pîslar, Theodora-Mara, et al.
Published: (2025)
Measuring Chain-of-Thought Monitorability Through Faithfulness and Verbosity
by: Meek, Austin, et al.
Published: (2025)
by: Meek, Austin, et al.
Published: (2025)
What Are Step-Level Reward Models Rewarding? Counterintuitive Findings from MCTS-Boosted Mathematical Reasoning
by: Ma, Yiran, et al.
Published: (2024)
by: Ma, Yiran, et al.
Published: (2024)
Scaling sparse feature circuit finding for in-context learning
by: Kharlapenko, Dmitrii, et al.
Published: (2025)
by: Kharlapenko, Dmitrii, et al.
Published: (2025)
Automatic Reward Shaping from Multi-Objective Human Heuristics
by: Xie, Yuqing, et al.
Published: (2025)
by: Xie, Yuqing, et al.
Published: (2025)
Automatic Reward Shaping from Confounded Offline Data
by: Li, Mingxuan, et al.
Published: (2025)
by: Li, Mingxuan, et al.
Published: (2025)
Reward Models Inherit Value Biases from Pretraining
by: Christian, Brian, et al.
Published: (2026)
by: Christian, Brian, et al.
Published: (2026)
Automatically Finding Rule-Based Neurons in OthelloGPT
by: Singh, Aditya, et al.
Published: (2025)
by: Singh, Aditya, et al.
Published: (2025)
How do LLMs Compute Verbal Confidence
by: Kumaran, Dharshan, et al.
Published: (2026)
by: Kumaran, Dharshan, et al.
Published: (2026)
Improving Dictionary Learning with Gated Sparse Autoencoders
by: Rajamanoharan, Senthooran, et al.
Published: (2024)
by: Rajamanoharan, Senthooran, et al.
Published: (2024)
Building Production-Ready Probes For Gemini
by: Kramár, János, et al.
Published: (2026)
by: Kramár, János, et al.
Published: (2026)
Evaluating Open-Source Sparse Autoencoders on Disentangling Factual Knowledge in GPT-2 Small
by: Chaudhary, Maheep, et al.
Published: (2024)
by: Chaudhary, Maheep, et al.
Published: (2024)
Confounding Robust Continuous Control via Automatic Reward Shaping
by: Juliani, Mateo, et al.
Published: (2026)
by: Juliani, Mateo, et al.
Published: (2026)
Reinforcement Learning with Stochastic Reward Machines
by: Corazza, Jan, et al.
Published: (2025)
by: Corazza, Jan, et al.
Published: (2025)
How Causal Abstraction Underpins Computational Explanation
by: Geiger, Atticus, et al.
Published: (2025)
by: Geiger, Atticus, et al.
Published: (2025)
How Do Transformers Learn Variable Binding in Symbolic Programs?
by: Wu, Yiwei, et al.
Published: (2025)
by: Wu, Yiwei, et al.
Published: (2025)
Beyond Reward Hacking: Causal Rewards for Large Language Model Alignment
by: Wang, Chaoqi, et al.
Published: (2025)
by: Wang, Chaoqi, et al.
Published: (2025)
SemiReward: A General Reward Model for Semi-supervised Learning
by: Li, Siyuan, et al.
Published: (2023)
by: Li, Siyuan, et al.
Published: (2023)
Intra-Trajectory Consistency for Reward Modeling
by: Zhou, Chaoyang, et al.
Published: (2025)
by: Zhou, Chaoyang, et al.
Published: (2025)
Adversarial Training for Process Reward Models
by: Juneja, Gurusha, et al.
Published: (2025)
by: Juneja, Gurusha, et al.
Published: (2025)
PRDP: Proximal Reward Difference Prediction for Large-Scale Reward Finetuning of Diffusion Models
by: Deng, Fei, et al.
Published: (2024)
by: Deng, Fei, et al.
Published: (2024)
Learning Causally Invariant Reward Functions from Diverse Demonstrations
by: Ovinnikov, Ivan, et al.
Published: (2024)
by: Ovinnikov, Ivan, et al.
Published: (2024)
SPARE: Single-Pass Annotation with Reference-Guided Evaluation for Automatic Process Supervision and Reward Modelling
by: Rizvi, Md Imbesat Hassan, et al.
Published: (2025)
by: Rizvi, Md Imbesat Hassan, et al.
Published: (2025)
AgentRewardBench: Evaluating Automatic Evaluations of Web Agent Trajectories
by: Lù, Xing Han, et al.
Published: (2025)
by: Lù, Xing Han, et al.
Published: (2025)
WARP: On the Benefits of Weight Averaged Rewarded Policies
by: Ramé, Alexandre, et al.
Published: (2024)
by: Ramé, Alexandre, et al.
Published: (2024)
Optimal Transport for LLM Reward Modeling from Noisy Preference
by: Pan, Licheng, et al.
Published: (2026)
by: Pan, Licheng, et al.
Published: (2026)
Finding Culture-Sensitive Neurons in Vision-Language Models
by: Zhao, Xiutian, et al.
Published: (2025)
by: Zhao, Xiutian, et al.
Published: (2025)
RewardAnything: Generalizable Principle-Following Reward Models
by: Yu, Zhuohao, et al.
Published: (2025)
by: Yu, Zhuohao, et al.
Published: (2025)
Adversarial Reward Auditing for Active Detection and Mitigation of Reward Hacking
by: Beigi, Mohammad, et al.
Published: (2026)
by: Beigi, Mohammad, et al.
Published: (2026)
CHARM: Calibrating Reward Models With Chatbot Arena Scores
by: Zhu, Xiao, et al.
Published: (2025)
by: Zhu, Xiao, et al.
Published: (2025)
Beyond Correctness: Confidence-Aware Reward Modeling for Enhancing Large Language Model Reasoning
by: He, Qianxi, et al.
Published: (2025)
by: He, Qianxi, et al.
Published: (2025)
InfoRM: Mitigating Reward Hacking in RLHF via Information-Theoretic Reward Modeling
by: Miao, Yuchun, et al.
Published: (2024)
by: Miao, Yuchun, et al.
Published: (2024)
Similar Items
-
Base Models Know How to Reason, Thinking Models Learn When
by: Venhoff, Constantin, et al.
Published: (2025) -
Understanding Reasoning in Thinking Language Models via Steering Vectors
by: Venhoff, Constantin, et al.
Published: (2025) -
Chain-of-Thought Reasoning In The Wild Is Not Always Faithful
by: Arcuschin, Iván, et al.
Published: (2025) -
Biases in the Blind Spot: Detecting What LLMs Fail to Mention
by: Arcuschin, Iván, et al.
Published: (2026) -
Inference-Time Toxicity Mitigation in Protein Language Models
by: Burda, Manuel Fernández, et al.
Published: (2026)