English(EN) Training Language Models to Cooperate with Inference-Time Controllers

新的CALM框架训练LLM以适应各种推理控制器

作者 PulseAugur 编辑部 · [1 个来源] · 2026-07-28 04:00

研究人员开发了CALM，一个旨在提高大型语言模型（LLM）对各种推理时控制器适应性的训练后框架。与之前为单一交互模式进行优化的方法不同，CALM将控制器直接集成到训练循环中。这种多任务强化学习方法通过在控制器诱导的交互协议上进行训练，使LLM能够更好地泛化到各种工作流程，包括思维链（Chain-of-Thought）、自洽性（self-consistency）和验证流程（verification pipelines）。 AI

影响增强了LLM对各种推理和交互协议的适应性，有可能提高在复杂、多步任务中的性能。

排序理由该集群包含一篇详细介绍新语言模型训练框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

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新语言模型训练框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
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: 72 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Moumita Choudhury, Vanshaj Khattar, Jing Liu, Toshiaki Koike-Akino, Ankush Chakrabarty, Shlomo Zilberstein, Ye Wang · 2026-07-28 04:00

训练语言模型与推理时控制器协作

arXiv:2607.23771v1 Announce Type: new Abstract: Large language model (LLM) performance increasingly depends not only on the base model, but also on the inference-time controller used to organize reasoning. Existing post-training methods, however, typically optimize for a single f…

本文如何被排名

报道来源 [1]

训练语言模型与推理时控制器协作

相关实体

相关话题