AI Research · research

Few-Shot Degradation Is Not What It Seems: Behavioral Evidence, Representation Analysis, and a Rando...

论文摘要：Few-shot prompting sometimes degrades language models instead of helping them, but why this happens is unknown. We evaluate 12 open-weight models on two Ukrainian tasks news c...

首次发现 2026-09-16T22:05:30Z · 最近更新 2026-09-17T02:52:59Z · 置信度 high · 重要性 7.5

为什么重要

该论文来自一手研究源并通过 3qk7 Research Radar 的新鲜度、来源与研究价值筛选；后续代码、模型权重、复现和相关产品事件可继续关联到该记录。

检测到的变化

研究论文进入 3qk7 高价值 Research Radar

一手证据

arXiv cs.CL · 2026-09-17T02:52:59Z