lhl616/Qwen3-4B-Base-grpo-nstd-dapo
4B
•
Updated
Haolin Liu
lhl616
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
19 days ago
RelayLLM: Efficient Reasoning via Collaborative Decoding
upvoted
a
paper
20 days ago
Benchmark^2: Systematic Evaluation of LLM Benchmarks
upvoted
a
paper
about 1 month ago
Can LLMs Guide Their Own Exploration? Gradient-Guided Reinforcement Learning for LLM Reasoning
Organizations
None yet
lhl616/Qwen3-4B-Base-grpo-dapo-new
4B
•
Updated
lhl616/Qwen3-4B-Base-grpo-dapo
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-grpo-step-128-8
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-grpo-nstd-step-128-8
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-two_0.0-0.8-start-dapo
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-standard-0.5-0.8-start-new
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-ratio
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-nstd-dense-0.5-0.8-step_reward
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-mixed
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-std-two-relu-0.5-0.8-start-dapo
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-std-two-relu-0.5-0.8-normal-dapo
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-std-two-0.5-0.8-start-dapo
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-std-0.5-0.8-start-real
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-std-0.5-0.8-relu-start-real
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-passk
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.9-0.8-start-new
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.9-0.8-relu-start-real
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.7-0.8-start-new-new
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.7-0.8-start-new
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.7-0.8-relu-start-real
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.7-0.8
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.5-0.8-start-static-prefix
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.5-0.8-start-prm0.9-real
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.5-0.8-start-prm0.3-real
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.5-0.8-start-piecewise-100-200-300-0.5-61729f41
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.5-0.8-start-linear_increase-0.1-10-0.02-real
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.5-0.8-start-linear_decay-0.7-10-0.02-real
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.5-0.8-start-dynamic-prefix
4B
•
Updated
lhl616/Qwen3-4B-Base-axon-error-aware-128-8-dense-nstd-0.5-0.8-start-dapo
4B
•
Updated