cs.CLMay 25, 2026

A Two-Phase Stability Study of LLM Judges and Bar Council Examiners on Thai Bar-Exam Free-Form Essays

Authors: Pawitsapak Akarajaradwong, Wuttikrai Lertprasertphakorn, Chompakorn Chaksangchaichot, Sarana Nutanong

Organizations: VISAI AI · Independent Researcher · Vidyasirimedhi Institute of Science and Technology

Abstract

Free-form legal essay evaluation in NLP treats expert inter-rater stability as a single ceiling number, and treats LLM-judge agreement with that ceiling as evidence of judge stability. We test both assumptions on the Thai bar examination through an identical-inputs protocol: three Bar Council-trained examiners (A, B, C) and a 26-LLM judge panel score the same 15 cross-graded answers from the same four inputs (question, official Bar Council grading regulation, gold answer, candidate answer). The headline finding is asymmetric. On 10 of 15 cells where the rubric prescribes both axes, all 29 raters converge in a tight band: panel agreement is universal. On the remaining 5 cells where the rubric does not prescribe how to grade a correct final answer that omits a decisive statutory citation, the human panel splits between two coherent readings (B/C majority at the upper rubric band, score 6-8; A minority at the lower band, score 1-2). The LLM judge population does not split symmetrically: 22 of 26 LLMs score in or near B/C's contested band, 3 sit in the regulation-silent middle gap, and only 1 (GPT-5.4 Nano) approaches A's band without consistently scoring within it. Zero LLMs in our 26-judge panel reproduce the minority human reading on the contested cells. The B/C-direction cluster spans every model size, vendor, and price tier we tested. An instrumented three-LLM anchor sub-panel (Claude 4.6 Opus, Gemini 3.1 Pro, GPT-5.4 Pro) carries determinism probes, input ablations, and bootstrap CIs, and reaches anchor panel α=0.77α= 0.77 on the 15 cells against human-panel α=0.36α= 0.36. The high LLM-panel αα reflects systematic convergence on the majority reading rather than balanced reproduction of both readings; a benchmark that selects its LLM judge by maximising agreement with a human reference panel will inherit this asymmetry by construction.

Explore similar work

CardsList
Jun 2, 2026cs.CL

The Geometry of LLM-as-Judge: Why Inter-LLM Consensus Is Not Human Alignment

LLM judges now score most open-ended NLP output, and their mutual agreement is routinely read as evidence that the scores can be trusted. That reading is unsafe: judges may agree because they capture quality, or because they share the same blind spots, and agreement statistics alone cannot tell these apart. We develop a geometric test that can. Treating each judge's scores as a vector, we measure spread, effective rank, the angle to human scores, and the judge-judge, judge-human, and human-human agreement triple for 42 judges on two community-built Indic benchmarks covering four domains and eight languages. Every comparison is reference-matched: a judge and a held-out rater are scored against the same two-rater mean, since an averaged reference otherwise flatters judges by several degrees. On subjective rubrics, judges agree with one another as much as humans do yet reach only 58-66% of human agreement and often concentrate on an axis humans do not weight. On the one rubric with a verifiable answer, most of that gap closes. Ensembles converge on the judges' shared axis rather than the human one, and training widens scores without rotating them. Inter-judge agreement is evidence of human alignment only after this check passes.
Sourabrata Mukherjee, Hamna Hamna, Kalika Bali +1
Llm-As-A-JudgeInter-Rater VariabilityLarge Language Model AlignmentRubricsHumanityConsensus
AskPDF
May 28, 2026cs.CL

Nine Judges, Two Effective Votes: Correlated Errors Undermine LLM Evaluation Panels

LLM-as-a-judge panels aggregate votes from multiple models, with the expectation that diverse models yield more reliable evaluations. We develop a framework to measure the true informational value of such panels and quantify how far their reliability falls short of the independent-voting ideal. Testing a panel of 9 frontier LLMs from 7 model families on three natural language inference datasets (each with 100 human annotations per item), we find that the 9 judges effectively provide only about 2 independent votes' worth of information. Roughly three-quarters of the panel's nominal independence is lost because the models make the same mistakes on the same items. The consequences are stark: the panel's actual accuracy falls 8-22 percentage points short of what independent voting would achieve, and the best single judge matches or outperforms the full panel across all conditions. Neither adding more judges nor using smarter aggregation algorithms helps -- established methods close at most 11% of this gap, even with access to the correct answers. We quantify these findings using the Kish effective sample size (n_eff) and a Condorcet null model, and show the deficit is robust across prompt variants, temperatures, chain-of-thought reasoning, and a pairwise preference task (RewardBench). The bottleneck is correlated judges, not the aggregation algorithm, implying that scaling up panels cannot substitute for genuinely independent evaluation.
Guneet Kohli
Llm-As-A-JudgeJudgesLarge Language Model EvaluationNatural Language Datasets
AskPDF
May 25, 2026cs.CL

Agreement Metrics for LLM-as-Judge Evaluation: What to Report and Why

Validating an LLM judge against human annotations usually means reporting several agreement statistics: accuracy, precision, recall, F1F_1, Cohen's κκ, and one or more rank correlations. A survey of 24 recent LLM-as-judge papers finds metric choice entangled with the judgment scale, tie handling, invalid outputs, and abstention handling, and those choices rarely stated. For binary criteria -- the common case in rubric-based evaluation, where each criterion is graded MET or UNMET -- most of the reported numbers are redundant: Pearson's rr, Spearman's ρρ, Kendall's τbτ_b, the phi coefficient φφ, and the Matthews Correlation Coefficient all reduce to a single number on non-degenerate binary data, so reporting several of them only creates an illusion of corroborating evidence. Cohen's κκ is the one agreement coefficient that adds information: it shares φφ's numerator but normalizes differently, and the gap between them measures how far the judge's positive-label rate has drifted from the human's. We then trace what changes when a judge may abstain with a CANNOT_ASSESS verdict: the three common ways of handling abstentions are not interchangeable preprocessing choices but answer different questions, and they break the binary equivalences. The same equivalences reappear, up to a negligible finite-sample correction, for multi-judge ensembles scored with Fleiss' κκ or Krippendorff's αα. We close with a reporting checklist that names the judgment scale, the abstention and tie handling mode, coverage, the confusion matrix, and the aggregation level alongside any scalar agreement coefficient.
Delip Rao, Chris Callison-Burch
Large Language Model JudgesInter-Rater VariabilityAbstention
AskPDF