1
Quantifying Prior Dominance in RAG Systems
量化RAG系统的先验主导，破解认知盲点，提升知识检索与生成的可信度。
arXiv:2606.23695v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) grounds Large Language Models in external knowledge, yet curren…
2
The Language Blind Spot: How Query Language and Brand Recognition Tier Shape AI-Constructed Brand Reputation Across Twelve European Languages
非英语查询揭示AI品牌声誉评估巨大偏差，十二种欧洲语言实验颠覆英语中心假设
arXiv:2606.23165v1 Announce Type: cross Abstract: Large language models (LLMs) increasingly mediate how people form impressions of organisations, yet …
3
Trust, but Don't Verify: Epistemic Blind Spots in LLM Source Evaluation
揭示大模型在信息验证中的致命盲点：为何AI会轻易信任却放弃核查来源？
arXiv:2606.05403v1 Announce Type: cross Abstract: Language models increasingly act as epistemic proxies, synthesizing evidence from multiple sources t…
4
Biases in the Blind Spot: Detecting What LLMs Fail to Mention
论文揭示了LLM在回答时隐藏的“盲点偏见”，帮你发现模型没说什么，提升AI对齐与安全认知。
arXiv:2602.10117v5 Announce Type: replace-cross Abstract: Large Language Models (LLMs) often provide chain-of-thought (CoT) reasoning traces that appe…
5
Positional Failures in Long-Context LLMs: A Blind Spot in Reasoning Benchmarks
揭示长上下文LLM因位置偏差导致推理失败的盲点，挑战现有基准评估体系。
arXiv:2605.23170v1 Announce Type: cross Abstract: Position-controlled evaluation is standard for retrieval tasks such as Needle-in-a-Haystack and RULE…
6
Which Way Did It Move? Diagnosing and Overcoming Directional Motion Blindness in Video-LLMs
视频大模型竟会“运动盲视”？研究揭示其方向识别缺陷并提出数据与模型改进方案。
arXiv:2605.22823v1 Announce Type: new Abstract: Video Large Language Models (Video-LLMs) have made rapid progress on temporal video understanding, yet…
7
Overcoming Dynamics-Blindness: Training-Free Pace-and-Path Correction for VLA Models
VLA模型存在动力学盲点？这篇论文提出无需训练的步速与路径校正方法，突破单帧观测局限。
arXiv:2605.11459v2 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) models achieve remarkable flexibility and generalization beyond…