sparse-supervision

Tag

Cards List
#sparse-supervision

1% of Tokens Can Be Enough: On Gradient Estimation in On-Policy Distillation

Hugging Face Daily Papers ↗ · 3d ago Cached

The paper introduces an information-efficiency ratio (IER) for optimizing token selection in sparse on-policy distillation, demonstrating that using only 1% of tokens can achieve performance comparable to full supervision.

0 favorites 0 likes
#sparse-supervision

Rethinking Critic Learning in PPO: Understanding and Mitigating Value Flattening

Hugging Face Daily Papers ↗ · 2026-09-16 Cached

The paper identifies Value Flattening as a failure mode in PPO critic learning for LLMs and introduces SP3O, a sparse supervision method, to mitigate it, showing consistent improvements in experiments.

0 favorites 0 likes
#sparse-supervision

CODEBLOCK: Learning to Supervise Code at the Right Granularity

arXiv cs.LG ↗ · 2026-06-18 Cached

Proposes CodeBlock, a structure-aware sparse supervision framework for supervised fine-tuning of code LLMs. It selects high-quality instruction-response pairs and partitions code responses into syntactically coherent coding items, applying loss only to selected items to achieve stronger pass@1 rates using only 1.9% of supervised response tokens.

0 favorites 0 likes
← Back to home

Submit Feedback