Deepthika S, Pawni Yadav, B. Sankar · 2026 · arXiv
DOI: 10.48550/arXiv.2610.31579 · open-access PDF
Pesquisa sobre a sífania pergunta se os grandes modelos de linguagem (LLMs) se inclinam para o que seus usuários querem ouvir. Nós perguntamos a pergunta inversa: quando um LLM é confiante, consistente e errado, os usuários se inclinam para ele? Em um experimento dentro de tópicos, 32 participantes aprenderam até cinco tópicos científicos conversando com cinco assistentes construídos no mesmo modelo (GPT-5-mini). Desconhecido para os participantes, o briefing de cada assistente tinha 0, 25, 50, 75 ou 100% de seus fatos substituídos por falsidades plausíveis, e cada assistente falou com total confiança. Um quiz surpresa depois de cada tópico ofereceu o fato correto, t
Machine translation (Meta M2M100); the English original follows.
Research on sycophancy asks whether large language models (LLMs) bend towards what their users want to hear. We ask the reverse question: when an LLM is confident, consistent, and wrong, do users bend towards it? In a within-subjects experiment, 32 participants learned up to five science topics by chatting with five assistants built on the same model (GPT-5-mini). Unknown to participants, each assistant's briefing had 0, 25, 50, 75, or 100% of its facts replaced by plausible falsehoods, and every assistant spoke with full confidence. A surprise quiz after each topic offered the correct fact, t
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.31579, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange