← Back to all news
arXiv Computation and Language By Zengqing Wu, Chuan Xiao

Language Models Give Group-Linked Fields the Wrong Weight

Read the original on arXiv Computation and Language →

The paper introduces an evidence‑referenced audit for large language models that assesses how much predictive information proxy fields contribute, comparing model reliance to a known outcome rule and a Bayesian linear regression. Applied to a simulated patient‑ranking task, the study finds that models rely on non‑informative fields, grow less reliance as proxies become predictive, are sensitive to display format, and that ranking accuracy does not reliably indicate proxy reliance. The authors argue that audits should report reliance against explicit predictive references alongside accuracy to properly evaluate model behavior.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

Related stories

arXiv Computation and Language
Oct 23

Invisibilization as a Source of Probability Bias in Language Models

The paper investigates how large language models (LLMs) from OpenAI and Anthropic exhibit probability bias when deciding whether to trust information from reporters of different racial descriptions. Using a two‑urn Bayesian game and 6,600 sessions, the authors find that models often refuse to assess a reporter’s credibility—especially for Black‑described reporters—yet when they do assess, Black reporters receive slightly higher trust. The bias is subtle, growing with reasoning effort, and is not detected by standard audits that compare average trust or only assessed cases.

By Christian Arnold, Ernesto Calvo, Guadalupe Gonzalez, Sebasti\'an Vallejo Vera, Tiago Ventura
llmssafety
More like this →
arXiv Computation and Language
Sep 30

Pair Difficulty Matters: Rethinking Pairwise LLM-as-a-Judge Evaluation and Consistency

Large Language Model judges are commonly used to rank texts via pairwise comparison, with reliability traditionally measured by position bias, transitivity, and pairwise agreement. This paper argues that these proxies are misleading because they are dominated by close‑rank‑gap pairs, which contribute little to the overall ranking, while far‑gap pairs carry the true ranking signal. Experiments on simulations and human‑rated corpora show weak correlation between the proxies and actual ranking accuracy, suggesting judges should be evaluated using rank‑gap‑conditional metrics against human rankings.

By Bruno Brocai, Maria Becker
llmsbenchmarkssafety
More like this →
arXiv Computation and Language
Sep 16

Bias Audits Detect Bias but Disagree on Ranking: Evidence from Ten Instruments and Ten Frontier Models

The study evaluates ten bias audit instruments across ten advanced language models on occupational gender, age, and socioeconomic status. While each tool reliably detects bias, their rankings of model performance are essentially random, indicating that different audits measure distinct constructs. The findings show that a single audit can identify bias direction within its own framework, but no audit can consistently rank models against one another.

By William Guey, Pierrick Bougault, Wei Zhang, Vitor D. de Moura, Jos\'e O. Gomes
benchmarkssafety
More like this →
arXiv AI
Aug 17

Whose doctor does the AI recommend? An algorithm audit of reputation and demographic signals in large language model-assisted physician choice

arXiv:2608. 14399v1 Announce Type: cross Abstract: Patients increasingly ask large language model (LLM) assistants which doctor to see, making these systems AI infomediaries: algorithms that intermediate one person's choice among other people and thereby decide, silently and at scale, which physicians become visible.

By Syeda Anshrah Gillani, Mirza Samad Ahmed Baig
llms
More like this →