AI models often give the right answers but point to the wrong sources
Quick Answer
AI models like GPT and Gemini frequently provide correct answers but often cite incorrect sources, a phenomenon termed 'attribution hallucination' by researchers at Peking University.
Quick Take
AI models like GPT and Gemini frequently provide correct answers but often cite incorrect sources, a phenomenon termed 'attribution hallucination' by researchers at Peking University. Their new CiteVQA benchmark systematically tests for this issue, which poses risks in regulated fields such as law and medicine.
Key Points
- Attribution hallucination affects the reliability of AI-generated content.
- CiteVQA is the first benchmark to systematically evaluate this issue.
- Correct answers may still be backed by incorrect citations.
- Risks are particularly significant in fields like law and medicine.
- Peking University researchers highlight the need for better citation accuracy.
Article Excerpt
From source RSS / original summaryLeading AI models like GPT and Gemini routinely cite text passages in document analyses that don't actually support their answers. Even when the answer is right, the cited evidence is often wrong. Researchers at Peking University call this "attribution hallucination," a risk for regulated fields like law and medicine. Their new CiteVQA benchmark is the first to test for it systematically. The article AI models often give the right answers but point to the wrong sources appeared first on The Decoder.
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from The Decoder
See more →
An AI model programmed nonstop for 19 days on a single MirrorCode task that cost $2,600 to run
Epoch AI's MirrorCode benchmark reveals Claude Opus 4.7 as the leader with a 56% solve rate, reconstructing a 16,000-line toolkit in 14 hours. Despite this, all models tested struggle with the most complex tasks, highlighting limitations in current AI capabilities. The single task consumed $2,600 over 19 days, raising questions about cost-effectiveness in AI development.
