Qwen3-4B Chemistry checkpoints from the SciKnowEval-style generalization runs: GRPO baseline, RLSD, and SDPO.
Seongryong Jung
SeongryongJung
AI & ML interests
Post-training, Knowledge Distillation, Self-Evolving AI
Recent Activity
updated a model 2 days ago
SeongryongJung/Qwen3-4B-Physics-RLSD published a model 2 days ago
SeongryongJung/Qwen3-4B-Physics-RLSD updated a collection 2 days ago
Qwen3-4B Chemistry RL Fine-tuningOrganizations
None yet
models 23
SeongryongJung/Qwen3-4B-Physics-RLSD
Text Generation • 4B • Updated • 1 • 1
SeongryongJung/Qwen3-4B-Chemistry-SDPO
Text Generation • 4B • Updated • 4 • 1
SeongryongJung/Qwen3-4B-Chemistry-RLSD
Text Generation • 4B • Updated • 19
SeongryongJung/Qwen3-4B-Chemistry-GRPO
Text Generation • 4B • Updated • 20
SeongryongJung/Qwen-8b-base-SDPO
Text Generation • 8B • Updated • 14
SeongryongJung/Qwen-8b-base-RLRT
Text Generation • 8B • Updated • 18
SeongryongJung/Qwen-8b-base-RLSD
Text Generation • 8B • Updated • 24
SeongryongJung/Qwen-4b-base-RLRT
Text Generation • 4B • Updated • 10
SeongryongJung/Qwen-8b-base-GRPO
Text Generation • 8B • Updated • 7
SeongryongJung/Qwen-4b-base-RLSD
Text Generation • 4B • Updated • 10
datasets 6
SeongryongJung/information-asymmetry-qwen3-4b
Updated • 2.08k
SeongryongJung/opsd-plain-4b-rollouts
Viewer • Updated • 748 • 84
SeongryongJung/opsd-plain-8b-rollouts
Viewer • Updated • 768 • 82
SeongryongJung/factory-agent-rollouts
Preview • Updated • 67
SeongryongJung/powerplant-shortqa-rows-4001-5000
Viewer • Updated • 3k • 25
SeongryongJung/medical-o1-reasoning-sft-gpt-4.1-mini-rewrite-hints
Viewer • Updated • 19.7k • 22