一篇新的研究论文识别出直接决策模型中存在的“序数尺度利用偏差”，即JEV和KEV等模型未能充分利用提供的用于分类和评估的序数尺度。尽管准确率很高，这些模型通常会压缩其预测，只使用可用尺度的一小部分。训练后修改，如BA-LoRA，表明这种偏差是学习到的并且可以缓解，从而显著提高尺度利用率。 AI
影响 突出了AI模型在分类任务可靠性方面的一个潜在局限性，并指出了模型训练和评估的改进方向。
排序理由 该集群包含一篇发表在arXiv上的研究论文，详细介绍了关于AI模型行为的新发现。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2609.38827v1 Announce Type: new Abstract: Direct-decision models turn text into low-latency structured labels and scores, making them attractive for classification and automatic evaluation. Yet reliability requires more than accuracy: a model must also use the ordinal decis…
Direct-decision models turn text into low-latency structured labels and scores, making them attractive for classification and automatic evaluation. Yet reliability requires more than accuracy: a model must also use the ordinal decision scale supplied by the user faithfully. We an…
Direct-decision models turn text into low-latency structured labels and scores, making them attractive for classification and automatic evaluation. Yet reliability requires more than accuracy: a model must also use the ordinal decision scale supplied by the user faithfully. We an…