Learning Smooth Reward Models with Temporal Difference for LLM RL and Inference
Dan Zhang
zd21
AI & ML interests
None yet
Recent Activity
published a dataset 2 days ago
zd21/EntiWeave published a dataset 2 days ago
zd21/PathRefiner authored a paper about 1 month ago
ReST-MCTS*: LLM Self-Training via Process Reward Guided Tree SearchOrganizations
None yet
models 18
zd21/qwen2.5-7b-td2
Updated
zd21/qwen2.5-7b-baseline-prm
Updated
zd21/DeepSeek-TD1-PRM
Updated
zd21/GLM-Z1-9B-0414-TDRM
9B • Updated • 1
zd21/GLM4-9B-0414-TDRM
9B • Updated • 3
zd21/Qwen2.5-1.5B-TDRM
Updated
zd21/Qwen2.5-0.5B-TDRM
Updated
zd21/Qwen2.5-Math-7B-TDRM
Updated
zd21/Qwen2.5-Math-1.5B-TDRM
Updated
zd21/DS-R1-Distill-Qwen-7.5B-TDRM
Updated
datasets 29
zd21/EntiWeave
Updated • 4
zd21/PathRefiner
Updated • 4
zd21/DataSciBench
Preview • Updated • 106 • 2
zd21/TDRM-3-step-TD
Viewer • Updated • 1.41M • 10
zd21/TDRM-2-step-TD
Viewer • Updated • 1.41M • 10
zd21/TDRM-1-step-TD
Viewer • Updated • 1.41M • 17
zd21/ReST-MCTS_SciGLM-6B_Self-Rewarding-DPO_2nd
Viewer • Updated • 1 • 7
zd21/ReST-MCTS_SciGLM-6B_ReST-MCTS_Policy_2nd
Viewer • Updated • 40.9k • 9
zd21/ReST-MCTS_SciGLM-6B_ReST-EM-CoT_2nd
Viewer • Updated • 28.9k • 10
zd21/ReST-MCTS_Mistral-MetaMATH-7b-Instruct_Self-Rewarding-DPO_2nd
Viewer • Updated • 1 • 5