按年被提到
20262026 · 2 次2026
大模型与语言模型›幻觉与语言模型局限·arXiv:2602.01031 · cs.AI

HalluHard: A Hard Multi-Turn Hallucination Benchmark

Dongyang Fan, Sebastien Delsad, Nicolas Flammarion, Maksym Andriushchenko · 2026
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2026-05-05 · 最近 2026-06-13
2不同的人
2活跃月份
2讨论串
4被引(OpenAlex)
arXiv 摘要(候选层不写导读)

Large language models (LLMs) still produce plausible-sounding but ungrounded factual claims, a problem that worsens in multi-turn dialogue as context grows and early errors cascade. We introduce $\textbf{HalluHard}$, a challenging multi-turn hallucination benchmark with 950 seed questions spanning four high-stakes domains: legal cases, research questions, medical guidelines, and coding. We operationalize groundedness by requiring inline citations for factual assertions. To support reliable evaluation in open-ended settings, we propose a judging pipeline that iteratively retrieves evidence via web search. It can fetch, filter, and parse full-text sources (including PDFs) to assess whether cited material actually supports the generated content. Across a diverse set of frontier proprietary and open-weight models, hallucinations remain substantial even with web search ($\approx 30\%$ for the strongest configuration, Opus-4.5 with web search), with content-grounding errors persisting at high rates. Finally, we show that hallucination behavior is shaped by model capacity, turn position, effective reasoning, and the type of knowledge required.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

没有评论直接链到它;下面是它被提交的讨论串里回复最多的评论(不计入统计)

讨论串里没有 ≥100 字的评论,只有提交记录。

被提交为 story · 2 次