PulseAugur
中
实时 03:53:11
English(EN) Return-to-Go Is More Than a Number: Q-Guided Alignment for Return-Conditioned Supervised Learning

新的 Q-Guided Alignment 框架增强了序列模型的可控性

作者 PulseAugur 编辑部 · [1 个来源] ·

研究人员推出了一种名为 Q-ALIGN DT 的新框架，旨在通过将面向返回（RTG）的输入与学习策略的实际性能进行对齐来增强条件序列模型（CSM）。该方法利用 Q 函数来指导 CSM，确保更高的 RTG 对应于具有更高预期回报的轨迹。在 D4RL 基准测试上的实验表明，Q-ALIGN DT 提供了卓越的可控性和性能，有效地学习了能够泛化到速度跟踪等复杂任务的结构化策略，而之前的方​​法在此类任务上表现不佳。 AI

影响 该框架有望在序列决策任务中实现更具可控性和更高性能的 AI 代理。

排序理由 这是一篇详细介绍新框架和实验结果的研究论文。[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 →

新的 Q-Guided Alignment 框架增强了序列模型的可控性

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
这是一篇详细介绍新框架和实验结果的研究论文。[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
105 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]

  1. arXiv cs.AI TIER_1 English(EN) · Yuxiao Yang, Weitong Zhang ·

    Return-to-Go 不仅仅是一个数字：面向回溯条件监督学习的 Q 引导对齐

    arXiv:2605.29028v1 Announce Type: cross Abstract: Conditioned Sequence Models (CSMs) learn policies by treating return-to-go (RTG) as a control signal. However, existing CSMs often treat the RTGs as simple numerical inputs rather than aligning them with the performance of their p…

相关话题