LLM citations vary run-to-run for the same prompt — so how valid are citation-overlap studies?
lilyraynyc · x · 2026-10-20
AI researcher Lily Ray points out a methodological gap in studies comparing citation overlap across LLMs: citations can differ within each LLM for the same prompt each time a response is generated, and she's never understood how such studies account for that internal variability.
More from Models
- Meta launches Muse Wonder coding models with 1M context, $4/$20 per million tokens — AIatMeta · 2026-10-21
- Anthropic models share near-identical personality profiles, with Haiku 5.5 weakest — ArtificialAnlys · 2026-10-21
- Google models consistently high on Slop, low on Transparency in personality profiling — ArtificialAnlys · 2026-10-21
- AI models cluster into three personality families that cross lab boundaries, finds AA — ArtificialAnlys · 2026-10-21
- AA image editing leaderboard v2.1: Krea 3 enters, GPT Image 2.5 max leads at Elo 1199 — ArtificialAnlys · 2026-10-21
- OpenAI admits Pro 500 allowance bug, compensates user with $1,000 in credits — XFreeze · 2026-10-21