#on-policy

5 articles

01 ·Inference & Serving·★ MEMBER·PAPER·7 min read Paper Explained: When EOS Tokens Disagree — Why On-Policy Distillation Students Forget How to Stop In on-policy distillation, student responses can grow until they hit the generation budget. This paper (arXiv:2609.20511) traces a major cause to student and teacher marking 'the end' with different special tokens. We walk through it from scratch, using only the paper itself. 02 ·Audio & Speech·★ MEMBER·PAPER·10 min read Paper Walkthrough: X-AuT — Pull Audio-Encoder Layers Out Two at a Time, Then Backfill from a Bigger Teacher Delete whole layers from a speech LLM's audio encoder and the decoder starts guessing 'end of sentence' too early, dropping characters. X-AuT picks which layers to remove with short trial runs, then backfills in three stages: distillation from a larger teacher, supervision on the student's own decoding context, and LoRA finetuning. Qwen3-ASR-0.6B goes from 18 to 16 layers with macro-average error falling 5.61% → 5.27%, and 14 layers land at 5.75% — single-run numbers, read here through the equations and the recipe. 03 ·Distillation & Compression·★ MEMBER·PAPER·9 min read On-Policy Distillation — Learning From What the Student Actually Writes Classic distillation has the student copy sentences the teacher wrote. On-policy distillation has the student write, then lets the teacher mark it up. The difference is one symbol in the loss — and that symbol removes exposure bias, turns distillation into a form of RL, and opens the door to self-distillation methods like u-OPSD and AgentOPSD. 04 ·FREE·PAPER·12 min read Paper Review: S²VOPD — Don't Strengthen the Teacher, Blur the Student's View A walkthrough of S²VOPD, which improves vision-language models without labels, rewards, or a stronger teacher. The trick is inverting where the asymmetry comes from: degrade the student's view instead of privileging the teacher's. 05 ·Inference & Serving·FREE·PAPER·8 min read Paper Walkthrough: DAPD — Breaking the Teacher's "Cheat-Sheet Illusion" in Distillation with Dual Anchors When you distill from a teacher that can see the reference solution, the student learns to act as if invisible answers were still there — the privilege illusion. DAPD (arXiv:2608.01735) traces this failure to information asymmetry and fixes it with two levels of anchoring. A walkthrough grounded strictly in the paper itself.