arxiv:2610.20572
Kaixiang Zhao
kzhao5
AI & ML interests
None yet
Recent Activity
authored a paper 12 days ago
Recovering Reliable Supervision via Critique-Guided On-Policy Distillation authored a paper 12 days ago
An RL View of OPD: Least Square Policy Distillation for Sample-Efficient LLM Reasoning upvoted a paper 12 days ago
Recovering Reliable Supervision via Critique-Guided On-Policy DistillationOrganizations
None yet
models 8
kzhao5/opd-baselines-pt06b
Updated
kzhao5/mea-soda-defenses
Updated
kzhao5/mea-qedks-defenses
Updated
kzhao5/mea-seqkd-defenses
Updated
kzhao5/opd-rethink-k2-horizon-7b-semantic-class-step200
9B • Updated • 13
kzhao5/opd-rethink-gemma3-4b-semantic-class-step200
5B • Updated • 7
kzhao5/opd-rethink-llama3.2-3b-semantic-class-step200
4B • Updated • 9
kzhao5/opd-rethink-qwen3-1.7b-semantic-class-step200
2B • Updated • 12