4 papers indexed here
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
On-policy distillation (OPD) supervises student-generated responses using sampled-token student-teacher log-ratios. A positive log-ratio penalizes a token that the student assigns higher probability to than the teacher, but its current value does not distinguish sustained high penalties from abrupt increases along a re...
Tail-Corrected Top-$k$ On-Policy Distillation (TT-OPD) is proposed, which preserves the advantages of TK-OPD, including rich distributional supervision and low computational cost, while providing an unbiased estimator of the gradient of the reverse KL divergence.
Direct Advantage Amplification (DAA), which amplifies the advantages of hard-to-sample correct responses on hard prompts, as obtained by Dynamic Sampling, is proposed, which ensures that, when Dynamic Sampling is used, these hard-to-sample responses can be effectively capitalized on, implying higher training efficiency...
RA-OPD selects more reliable trajectories to improve student model performance without requiring additional computational cost and is evaluated on math and code benchmarks using models from the Qwen3 family and the DeepSeek-R1 family.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
