English(EN) Verdicts Without Annotated Evidence: Rejection Sampling or Label-Only Post-Training for Evidence Recovery?

新方法在无人工标注的情况下恢复大语言模型判决的证据段落

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-07 04:00

研究人员开发了新方法，在只有最终判决可用时，为审查工作流恢复证据段落。在ContractNLI数据集上测试了仅标签后训练和拒绝采样技术。与预训练相比，这两种方法在准确率和跨度F1分数上都有所提高，其中仅标签训练的准确率为0.896，跨度F1为0.564。这些方法旨在无需人工标注段落即可增强证据恢复能力。 AI

影响在无需人工标注的情况下，提高了重建大语言模型生成判决支持证据的能力。

排序理由详细介绍大语言模型工作流中证据恢复新方法的论文。[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: 15 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍大语言模型工作流中证据恢复新方法的论文。[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: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Nishanth Nayakanti, Prasang Gupta, Ashutosh Bilthare, Kevin Paul · 2026-10-07 04:00

无标注证据的判决：拒绝采样还是仅标签的训练后用于证据恢复？

arXiv:2610.06962v1 Announce Type: cross Abstract: In many review workflows the verdict is the only thing retained. The passages behind it are not marked, because that annotation costs far more than recording the decision. We measure how much of that evidence a small language mode…

本文如何被排名

报道来源 [1]

无标注证据的判决：拒绝采样还是仅标签的训练后用于证据恢复？

相关实体

相关话题