Anthropic's Claude Sonnet 4.5 is experiencing issues with its citation index, causing untrustworthy hyperlinks in its self-assessments. The model's internal citation numbers are drawing from an accumulated index that spans the entire conversation, rather than just the most recent search results. While the factual content of the citations remains accurate, the associated links may point to unrelated sources, necessitating manual navigation for users seeking verifiable information. AI
IMPACT Highlights potential issues with citation accuracy in current LLMs, impacting their reliability for research.
RANK_REASON User-generated post discussing a perceived issue with a specific model's functionality.
Read on Mastodon — fosstodon.org →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
