# Who can we trust? LLM-as-a-jury for Comparative Assessment

arXiv: https://arxiv.org/abs/2602.16610 · HTML: https://mlsanity.com/paper/2602.16610/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 53 references that are in the map.
- First posted on arXiv 18 Feb 2026.

## Builds on (most cited first, 30 of 34)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference](https://mlsanity.com/paper/2403.04132/) (2024)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [Can Large Language Models Be an Alternative to Human Evaluations?](https://mlsanity.com/paper/2305.01937/) (2023)
- [SummEval: Re-evaluating Summarization Evaluation](https://mlsanity.com/paper/2007.12626/) (2020)
- [DeepSeek LLM: Scaling Open-Source Language Models with Longtermism](https://mlsanity.com/paper/2401.02954/) (2024)
- [From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipeline](https://mlsanity.com/paper/2406.11939/) (2024)
- [Is ChatGPT a Good NLG Evaluator? A Preliminary Study](https://mlsanity.com/paper/2303.04048/) (2023)
- [Large Language Models Are State-of-the-Art Evaluators of Translation Quality](https://mlsanity.com/paper/2302.14520/) (2023)
- [MoverScore: Text Generation Evaluating with Contextualized Embeddings and Earth Mover Distance](https://mlsanity.com/paper/1909.02622/) (2019)
- [GPTScore: Evaluate as You Desire](https://mlsanity.com/paper/2302.04166/) (2023)
- [Is ChatGPT Good at Search? Investigating Large Language Models as Re-Ranking Agents](https://mlsanity.com/paper/2304.09542/) (2023)
- [Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models](https://mlsanity.com/paper/2404.18796/) (2024)
- [MT-Bench-101: A Fine-Grained Benchmark for Evaluating Large Language Models in Multi-Turn Dialogues](https://mlsanity.com/paper/2402.14762/) (2024)
- [QuestEval: Summarization Asks for Fact-based Evaluation](https://mlsanity.com/paper/2103.12693/) (2021)
- [USR: An Unsupervised and Reference Free Evaluation Metric for Dialog Generation](https://mlsanity.com/paper/2005.00456/) (2020)
- [Of Human Criteria and Automatic Metrics: A Benchmark of the Evaluation of Story Generation](https://mlsanity.com/paper/2208.11646/) (2022)
- [LLM Comparative Assessment: Zero-shot NLG Evaluation through Pairwise Comparisons using Large Language Models](https://mlsanity.com/paper/2307.07889/) (2023)
- [Leveraging Large Language Models for NLG Evaluation: Advances and Challenges](https://mlsanity.com/paper/2401.07103/) (2024)
- [Reference-Guided Verdict: LLMs-as-Judges in Automatic Evaluation of Free-Form QA](https://mlsanity.com/paper/2408.09235/) (2024)
- [Efficient LLM Comparative Assessment: a Product of Experts Framework for Pairwise Comparisons](https://mlsanity.com/paper/2405.05894/) (2024)
- [TrustJudge: Inconsistencies of LLM-as-a-Judge and How to Alleviate Them](https://mlsanity.com/paper/2509.21117/) (2025)
- [CrossCheckGPT: Universal Hallucination Ranking for Multimodal Foundation Models](https://mlsanity.com/paper/2405.13684/) (2024)
- [Language Model Council: Democratically Benchmarking Foundation Models on Highly Subjective Tasks](https://mlsanity.com/paper/2406.08598/) (2024)
- [LLM-RankFusion: Mitigating Intrinsic Inconsistency in LLM-based Ranking](https://mlsanity.com/paper/2406.00231/) (2024)
- [Aligning with Logic: Measuring, Evaluating and Improving Logical Preference Consistency in Large Language Models](https://mlsanity.com/paper/2410.02205/) (2024)

## Built on by (most cited first, 7 of 7)

- [Calibrate, Don't Curate: Label-Efficient Estimation from Noisy LLM Judges](https://mlsanity.com/paper/2605.09702/) (2026)
- [A Finite-Calibration Regime Map for LLM Judge Panels](https://mlsanity.com/paper/2606.01034/) (2026)
- [CoEval: Ranking Language Models for Custom Tasks Without Labeled Data or Trustworthy Benchmarks](https://mlsanity.com/paper/2606.03650/) (2026)
- [Can LLMs Rank? A Tale of Triads and Triage](https://mlsanity.com/paper/2606.30412/) (2026)
- [How Do Large Language Models Judge Social Attraction? Evidence from Theory-Grounded Persona Ratings Across Multiple LLMs and Humans](https://mlsanity.com/paper/2608.09717/) (2026)
- [Multi-Expert Conformal Risk Control for Pairwise LLM Judging in Open-Ended Dialogue](https://mlsanity.com/paper/2608.26529/) (2026)
- [Multiple latent orderings better predict language model preferences](https://mlsanity.com/paper/2609.22170/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are increasingly applied as automatic evaluators for natural language generation assessment often using pairwise comparative judgements. Existing approaches typically rely on single judges or aggregate multiple judges assuming equal reliability. In practice, LLM judges vary substantially in performance across tasks and evaluation aspects, and their judgment probabilities may be biased and inconsistent. Furthermore, human-labelled supervision for judge calibration may be unavailable. We first empirically demonstrate that inconsistencies in LLM comparison probabilities exist and show that it limits the effectiveness of direct probability-based ranking. To address this, we study the LLM-asa-jury setting and propose BT-sigma, a judge-aware extension of the Bradley-Terry model that introduces a discriminator parameter for each judge to jointly infer item rankings and judge reliability from pairwise comparisons alone. Experiments on benchmark NLG evaluation datasets show that BT-sigma consistently outperforms averaging-based aggregation methods, and that the learned discriminators strongly correlate with independent measures of the cycle consistency of LLM judgments. Further analysis reveals that BT-sigma can be interpreted as an unsupervised calibration mechanism that improves aggregation by modelling judge reliability.