内容库 / 论文

arXiv:2608.14277 · 工程化

SimpleOPD: Simple Tokenizer-Agnostic On-Policy Distillation for Long-Context Reasoning

SimpleOPD：面向长上下文推理的简单、与 Tokenizer 无关的 On-Policy Distillation

SimpleOPD: Simple Tokenizer-Agnostic On-Policy Distillation for Long-Context Reasoning

类型：arxiv
标识：2608.14277
链接：https://arxiv.org/abs/2608.14277
主分类：engineering
形态：method
被引：0
被引来源：Semantic Scholar
S2被引：0
影响力被引：0
TLDR：This work introduces a student reference KL loss and mask the advantages of special termination tokens to mitigate the problem of excessive generation length and frequent truncation, and improves on science benchmarks such as HLE and HiPhO, suggesting that OPD transfers reasoning capabilities that generalize beyond the mathematical training domain.
待LLM分类：否
标题中文：SimpleOPD：面向长上下文推理的简单、与 Tokenizer 无关的 On-Policy Distillation
TLDR中文：引入 student reference KL 损失并 mask 特殊终止 token 的 advantage，以缓解生成过长和频繁截断的问题；在 HLE 和 HiPhO 等科学 benchmark 上取得改进，表明 OPD 传递的推理能力可泛化至数学训练领域之外。
来源文件：
/inbox/tom/_candidates/2026-08-17-agent-rag-longcontext-candidates.json
[S2 enrich]

相关 · 按语义匹配

仓库 nick7nlp/Awesome-LLM-On-Policy-Distillation

笔记 Cool Papers · cs.CL (papers.cool) · RSS 摘要

笔记 flyP 精读与批判 · 2026-07-28 15:50

文档 Tom 评 flyP 的产出：On-Policy Distillation 三大要点解读（2607.13399）

笔记 MMProLong × ContextRL 关联性更新(短评 · flyP)

笔记 ContextRL:Context-Aware 强化学习用于 Agentic 与多模态 LLM(精读 · flyP)

arXiv

2608.14277

TLDR

引入 student reference KL 损失并 mask 特殊终止 token 的 advantage，以缓解生成过长和频繁截断的问题；在 HLE 和 HiPhO 等科学 benchmark 上取得改进，表明 OPD 传递的推理能力可泛化至数学训练领域之外。This work introduces a student reference KL loss and mask the advantages of special termination tokens to mitigate the problem of excessive generation length and frequent truncation, and improves on science benchmarks such as HLE and HiPhO, suggesting that OPD transfers reasoning capabilities that generalize beyond the mathematical training domain.

被引 · S2

0

S2 0

影响力 0

相关解读

arXiv ↗ 中文全文