arxiv:2507.08794
Haolin Liu
lhl616
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
about 8 hours ago
Training Data Efficiency in Multimodal Process Reward Models
upvoted
a
paper
1 day ago
Parallel-Probe: Towards Efficient Parallel Thinking via 2D Probing
upvoted
a
paper
27 days ago
RelayLLM: Efficient Reasoning via Collaborative Decoding
Organizations
None yet
models
117
lhl616/Qwen3-8B-axon-error-aware-128-8-ratio
8B
•
Updated
lhl616/Qwen3-8B-axon-error-aware-128-8-mixed
8B
•
Updated
lhl616/Qwen3-8B-Base-axon-ppo
8B
•
Updated
lhl616/Qwen3-8B-Base-axon-grpo-step-128-8
8B
•
Updated
lhl616/Qwen3-8B-Base-axon-error-aware-128-8-ratio-new
8B
•
Updated
lhl616/Qwen3-8B-Base-axon-error-aware-128-8-dense-passk
8B
•
Updated
lhl616/Qwen3-8B-Base-axon-error-aware-128-8-dense-nstd-0.5-0.8-step-2
8B
•
Updated
lhl616/Qwen3-8B-Base-axon-error-aware-128-8-dense-nstd-0.5-0.8-start-relu
8B
•
Updated
lhl616/Qwen3-8B-Base-axon-error-aware-128-8-dense-nstd-0.5-0.8-start
8B
•
Updated
lhl616/Qwen3-8B-Base-axon-error-aware-128-8-dense-nstd-0.5-0.8-relu-normal-real
8B
•
Updated
datasets
0
None public yet