PulseAugur
中
实时 06:51:34

RISE 方法通过自我外推和策略蒸馏增强语言模型

作者 PulseAugur 编辑部 · [2 个来源] ·

研究人员推出了一种名为 RISE 的新方法，通过自我外推和策略蒸馏来改进语言模型。与依赖外部教师或有限的上下文内学习的先前方法不同，RISE 从模型自身的训练轨迹构建了一个合成教师。该合成教师通过外推模型的进展提供密集的、token 级别的监督。实验表明，在数学推理、STEM 和代码生成等各种任务上，RISE 的表现优于标准的人类反馈强化学习 (RLHF) 和同策略自我蒸馏。 AI

影响 这项新的自我外推技术可能带来更高效、更有效的语言模型训练，从而提高在复杂推理和生成任务上的性能。

排序理由 该集群包含一篇详细介绍改进语言模型新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 Hugging Face Daily Papers 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

RISE 方法通过自我外推和策略蒸馏增强语言模型

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该集群包含一篇详细介绍改进语言模型新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, model release
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
3 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.
Coverage growth since scoring
+1 source(s) since last score
New sources have picked up this story since our last re-score. Score will update on the next scoring pass.

完整方法见我们的编辑标准。

报道来源 [2]

  1. arXiv cs.AI TIER_1 English(EN) · Yang Li, Semih Yavuz, Shafiq Joty ·

    RISE: 通过自外推策略蒸馏的递归改进

    arXiv:2609.05295v1 Announce Type: new Abstract: On-policy distillation (OPD) provides dense, per-token supervision for language model post-training, but its effectiveness is bottlenecked by teacher quality: external teachers suffer from distribution mismatch, while self-distillat…

  2. Hugging Face Daily Papers TIER_1 English(EN) ·

    RISE: 通过自外推策略蒸馏实现递归改进

    RISE improves language model post-training by recursively generating dense token-level supervision from the model's own reinforcement learning trajectory via self-extrapolation, avoiding external teachers.

相关实体

相关话题