The paper introduces a two‑dimensional construct validity framework for evaluating large language models (LLMs) as judges, defining invariance (S) and sensitivity (R) to construct‑preserving and construct‑changing edits. Experiments across seven judges and four domains reveal high invariance (average S = 0.945) but low sensitivity (average R = 0.319), with sensitivity varying by edit type. Audits of public label sets show that surface‑only predictors can reproduce a substantial portion of labels, underscoring that high agreement does not guarantee construct validity.
By Jianlin Chen, Wenhui Chen, Ziyao Lin, Chi Man Vong
arXiv:2606. 15610v1 Announce Type: cross Abstract: LLM-as-a-judge systems are now routinely used for open-ended model evaluation, where human preference annotation is costly, slow, and difficult to reproduce.
By Hiroyasu Usami, Keisuke Hara, Ayato Tsuboi, Naohiko Matsuda
arXiv:2607. 02104v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used as cheap, scalable judges that compare candidate outputs pairwise.
By Jian Xu, Delu Zeng, John Paisley, Qibin Zhao
The study analyzes 373,019 judgments from LLM‑scored benchmarks, decomposing variance into system, item, judge, and interaction components via generalizability theory. It finds that with a single judge, generalizability converges to a ceiling determined by the system‑by‑judge variance, which is substantially lower in pairwise preference settings, allowing one judge to suffice. The research also reveals significant biases in presentation order and highlights that many published win‑rate claims fall below the measured floor of the benchmarks.
By Atul Anand
arXiv:2606. 13685v1 Announce Type: cross Abstract: LLM-as-a-Judge is now widely used to rank model outputs, train reward models, and populate public leaderboards, but its run-to-run reliability remains under-characterized.
By Abel Yagubyan
The paper demonstrates that reading a large language model (LLM) judge’s verdict from the logits of its first generated token—an approach used in constrained decoding and likelihood‑scoring evaluation—introduces a significant distortion in position bias. Because judges do not always start with a verdict token (12–49% of cases for Qwen3 judges and <3% for Llama‑3.1‑8B and Phi‑3.5‑mini), this readout often returns the first response rather than a true judgment, inflating position bias by up to 42 points while barely affecting judge accuracy. The authors recommend reporting the frequency with which a judge leads with a verdict token to provide a more accurate assessment of position bias.
By Gnaneswar Villuri, Hashmath Shaik, Alex Doboli
The article examines how a difference‑in‑differences (DiD) analysis on a censored rating scale can produce misleading effects. It demonstrates that each DiD component is censored by its own share, causing differential attenuation that can fabricate an interaction effect when the two responses are unequally censored. Using a pre‑registered audit of an LLM judge, the authors show that the reported significant interaction is largely an artifact of this censoring mechanism, with the true preference effect being null.
By Shuyi Fan, Boyuan Deng, Mengyu Xu, Xinhong Xie, Chenyang Li, Hongyang Zhang
arXiv:2605.16023v3 Announce Type: replace
Abstract: LLM-as-a-judge has become the dominant paradigm for grading model outputs at scale, yet the same model assigns systematically different scores when...
By Nils Feldhus, Tanja Baeumel, Elena Golimblevskaia, Qianli Wang, Van Bach Nguyen, Aaron Louis Eidt, Selin Kahvecioglu, Christopher Ebert, Wojciech Samek, Jing Yang, Vera Schmitt, Sebastian M\"oller, Simon Ostermann
arXiv:2608. 06940v1 Announce Type: new Abstract: LLM judge panels are a standard evaluation tool, but prior work reports highly correlated panel errors: nine judges provide roughly the effective information of two independent ones, and aggregation closes only a small fraction of the gap.
By Yang Shu
JudgeProfile is a framework that analyzes the subjectivity of large language model (LLM) judges by separating evaluation into perception—how judges compare responses on attributes such as clarity, correctness, and detail—and prioritization—how much each attribute influences the final decision. Using the curated SubjectiveSet dataset of 50,013 response pairs evaluated by 21 judges across 87 attributes, the study finds that judges often agree on attribute judgments even when their overall choices differ. By estimating and adjusting attribute weights, the authors improve agreement with reference labels from 66.48% to 71.97%, outperforming fine‑tuning and rubric prompting.
By Qi Cao, Kangning Liu, Xuan Kan, Shunwen Tan, Yang Pei, Dake Chen, Yatai Ji, Zixuan Ye, Yuanpeng Tu, Daniel Li, Junbiao Tang, Pengtao Xie, Zihao He
arXiv:2607. 18828v1 Announce Type: new Abstract: Readiness stress-testing of medical AI has focused on closed-ended and multimodal benchmarks.
By Koyar Afrasyab
The article examines how classical test theory reliability statistics misrepresent the performance of large language model (LLM) judges. It shows that internal‑consistency coefficients, the dependability index, and Livingston‑Lewis accuracy each conflate judge error with item design or criterion validity, making it impossible to attribute a single reliability value to the judge alone. The authors argue that such misattribution can influence deployment decisions and documentation.
By Louis Yiven Zhu