Probing Preference Representations: A Multi-Dimensional Evaluation and Analysis Method for Reward Models
wangchenglong
wangclnlp
AI & ML interests
None yet
Recent Activity
upvoted a paper about 19 hours ago
Foundations of Large Language Models upvoted a paper 2 months ago
ToFu: A White-Box, Token-Efficient Agent Harness for Researchers upvoted a paper 5 months ago
LLMs Improving LLMs: Agentic Discovery for Test-Time ScalingOrganizations
Self-Training Generative Foundation Reward Models for Reward Reasoning
-
GRAM-R^2: Self-Training Generative Foundation Reward Models for Reward Reasoning
Paper • 2509.02492 • Published • 2 -
wangclnlp/GRAM-RR-LLaMA-3.1-8B-RewardModel
Text Generation • 8B • Updated • 19 • 2 -
wangclnlp/GRAM-RR-LLaMA-3.2-3B-RewardModel
Text Generation • 3B • Updated • 20 -
wangclnlp/GRAM-RR-TrainingData
Updated • 30
Probing Preference Representations: A Multi-Dimensional Evaluation and Analysis Method for Reward Models
Generative Foundation Reward Models for Reward Generalization
Self-Training Generative Foundation Reward Models for Reward Reasoning
-
GRAM-R^2: Self-Training Generative Foundation Reward Models for Reward Reasoning
Paper • 2509.02492 • Published • 2 -
wangclnlp/GRAM-RR-LLaMA-3.1-8B-RewardModel
Text Generation • 8B • Updated • 19 • 2 -
wangclnlp/GRAM-RR-LLaMA-3.2-3B-RewardModel
Text Generation • 3B • Updated • 20 -
wangclnlp/GRAM-RR-TrainingData
Updated • 30
A Robust Visual Reward Model Optimized via Auxiliary Textual Preference Data