arxiv:2605.07865
Choi
yunhowhour
AI & ML interests
None yet
Recent Activity
authored a paper about 8 hours ago
KL for a KL: On-Policy Distillation with Control Variate Baseline authored a paper about 8 hours ago
Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor's Internal States upvoted a paper 3 days ago
KL for a KL: On-Policy Distillation with Control Variate BaselineOrganizations
None yet
models 8
yunhowhour/Qwen3-4B_CRRL_batch_1024_B200_ds_samplelevelmean_step_110
4B • Updated • 75
yunhowhour/Distill-1.5B_GRESO_batch_512_step_120
2B • Updated • 52
yunhowhour/Qwen3-4B_CRRL_batch_1024_B200_w_o_global_norm_step_90
4B • Updated • 44
yunhowhour/Qwen3-4B_CRRL_batch_1024_B200_w_o_global_norm_step_60
4B • Updated • 158
yunhowhour/CRRL_distill_1.5B_w_o_globalnorm_step_120
2B • Updated • 90
yunhowhour/CRRL_distill_1.5B_GRESO_step_90
2B • Updated • 70
yunhowhour/CRRL_batch_1024_step_50
4B • Updated • 99
yunhowhour/DAPO_batch_1024_step_90
4B • Updated • 102
datasets 0
None public yet