Claims

LLMs reconstruct highly cited researchers' coauthor lists at approximately twice the rate of lower-cited peers, with the disparity growing with model size

Claim · All claims
Evidence · 1 study · associational r? · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims

Reading the evidence codes

q is the design tier, i the impact, n the sample, and the kind with its r the rigour judged within that kind: what each code means.

This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.

q2 i? Highly cited researchers are reconstructed by LLMs at approximately twice the rate of lower-cited peers, and the disparity grows with model size. → Yifan Liu 2026

Evidence

Yifan Liu 2026

Yifan Liu, Jaime Arguello, Orland Hoeber, Chang Liu, Soo Young Rieh, Luanne Sinnamon, et al. (2026). Report on CHIIR 2026 Workshop on Generative AI and Academic Search (GAI&AS). ACM SIGIR Forum. https://arxiv.org/abs/2606.08936 doi:10.48550/arxiv.2606.08936

q2 · i? · associational · r?

A lightning-talk investigation using the Discoverable Name Extraction (DNE) set-based metric evaluated three LLMs (DeepSeek R1, Llama 4 Scout, Mixtral 8×7B) against bibliographic ground truth from OpenAlex and Google Scholar spanning 1,596 seed authors across 10 disciplines and 8 global regions. The report states "highly cited researchers are reconstructed by LLMs at approximately twice the rate of lower-cited peers". No effect size is printed.

"Their central finding is unambiguous: highly cited researchers are reconstructed by LLMs at approximately twice the rate of lower-cited peers, a disparity that grows with model size and reflects the disproportionate repetition of prominent names in citation-heavy training corpora."

Discussion

Page metadata
Field Value
type claim
id llm-coauthor-reconstruction-citation-bias
title LLMs reconstruct highly cited researchers' coauthor lists at approximately twice the rate of lower-cited peers, with the disparity growing with model size
description LLMs reconstruct highly cited researchers' coauthor lists at approximately twice the rate of lower-cited peers, with the disparity growing with model size
status review
evidence_strength moderate
newest_evidence 2026
recheck_by 2029
generated by process:wiki-ingest on 2026-10-10
trust unverified — no one has confirmed this page's content
sources 1 — yifan-liu-2026

Frontmatter is the page's machine-readable record — see the Schema & Guide.