The paper investigates how many human annotators are equivalent to a panel of 32 large‑language‑model (LLM) judges. By comparing the panel’s label distributions to empirical human labels on three ChaosNLI tasks, the authors find two distinct effective panel sizes: distribution‑error matching yields effective sizes of 2.304, 3.750, and 3.445, while spectral matching gives 4.242, 6.459, and 6.499, indicating a 1.72–1.89× gap. The study also explores how spectral diversity, participation ratio, and panel composition affect effective size, and demonstrates that carefully chosen panels can outperform baseline accuracy while improving effective size.
By Chao Li, Yingying Yu, Yunfeng Li
arXiv:2608.22432v1 Announce Type: cross
Abstract: Multilingual LLM judges produce different evaluator-backbone rankings depending on the prompt language: on an eight-language Agent-as-a-Judge benchma...
By Alhasan Mahmood, Samir Abdaljalil, Hasan Kurban
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
arXiv:2608.29517v1 Announce Type: new
Abstract: Large language models (LLMs) are increasingly used as essay graders in learning analytics, evaluated almost exclusively with agreement statistics. Educ...
By Veerendra Kumar Sunkavalli
The paper investigates whether panels of vision‑language models (VLMs) can reliably judge image aesthetics. It shows that a panel of holistic judges rarely outperforms its best member, but when each model scores images on five rubric‑defined dimensions and these dimension scores are fused across model families, the panel consistently beats the best single VLM on two datasets (EVA and PARA). The study demonstrates that the value of a panel depends on the type of input it receives, and that dimension‑based fusion yields measurable gains at the cost of additional labeling and API usage.
By Amit Jadhav, Shaurya Beriwala, Beomjin Kim
arXiv:2606. 03650v1 Announce Type: cross Abstract: Choosing or ranking language models for a specific application is hardest when no task-specific labeled data exists, and standard public benchmarks cannot be trusted, their items having likely leaked into pretraining, so scores reflect memorization rather than fitness.
By Alexander Apartsin, Yehudit Aperstein
The study evaluates how sixteen language models from ten families generate diverse formulations of a psychotherapeutic case, finding an average semantic diversity of 1.69 distinct formulations versus 1.43 for a single-model baseline. It introduces the Vendi Score to quantify diversity and defines a per-model dissent metric to identify the most divergent voice within an ensemble. The analysis shows that model identity significantly influences dissent, but this effect varies across model pairs and panel compositions, indicating that ensemble dispersion is a measurable property rather than an assumed one.
By Mario Vega-Barbas, Lidia Mora-Valenciano, Iv\'an Pau, Fernando Seoane, Farhad Abtahi
Prior work on human label variation (HLV) in natural language inference (NLI) has often relied on re-annotation resources that select items by disagreement level. An earlier study (arXiv:2607.
The paper investigates whether consensus among large language model (LLM) judges truly reflects human alignment. By treating each judge’s scores as vectors, the authors measure spread, effective rank, and angles to human scores across 42 judges on Indic benchmarks, revealing that inter‑judge agreement often mirrors shared blind spots rather than human judgments. They find that while judges agree as much as humans, they only reach 58‑66% of human agreement and frequently focus on axes humans do not weight, indicating that ensemble agreement alone is insufficient evidence of alignment.
By Sourabrata Mukherjee, Hamna Hamna, Kalika Bali, Sunayana Sitaram
The paper reports that in English all‑words word sense disambiguation (WSD), the scarcity of high‑quality labels—not the models—has become the limiting factor. The authors introduce lexEN, a human‑adjudicated correction layer over the Maru2022 ALL_NEW benchmark, and SenseBench, a living leaderboard for LLM WSD evaluation. They show that frontier large language models reach about 95 % accuracy on lexEN‑v1, that relabeling corpora with these models improves downstream systems, and that fine‑grained WordNet senses are often ill‑posed, with coarsening improving both annotator agreement and model performance.
"whyItMatters":"The study highlights that improving label quality and managing annotation costs are now the critical challenges for advancing WSD performance, as model accuracy is already near its theoretical ceiling."
By Vassili Philippov, Amro Salman, Dmitrii Andreev, Penny Hands, Emil Kaiumov, Pavel Katunin, Anton Nikolaev
JudgeSense is a benchmark comprising 880 items from human‑labelled corpora, each presented under two differently worded instructions that ask the same question. The study evaluates 25 judges from six providers across four tasks, measuring how rewording affects agreement with the judge’s own verdicts. Results show that rewording reduces agreement on all tasks, with significant effects on two, and that stability varies across tasks and is not predicted by parameter count.
By Rohith Reddy Bellibatlu, Edward Raff, Wenbin Zhang
The study evaluates multi‑agent debate (MAD) in small language models, testing whether cognitive diversity—via personas, sampling temperature, or model identity—drives performance gains. Across 23 models, five tasks, and over 5,500 runs, MAD consistently outperforms single‑model inference but, when matched for generation budget, it ties or falls behind self‑consistency sampling, with persona prompting actually reducing accuracy. The authors find that MAD’s benefits largely stem from the first answer exchange and that many reported gains are due to ensemble‑sampling effects rather than true diversity, highlighting the need for budget‑matched, contamination‑checked baselines.
whyItMatters":"The findings clarify that MAD’s perceived advantages may be overestimated and that future debate mechanisms must be evaluated against rigorous, budget‑matched baselines to ensure genuine performance improvements."
By Leonardo Ferreira, Gardenia Liu, Kaden Zheng