Papers › Judging the Judges: A Systematic Study of Position Bias in LLM-as-a-Judge

Judging the Judges: A Systematic Study of Position Bias in LLM-as-a-Judge

12 Jun 2024arXiv:2406.07791archive 2025-07-28

Lin Shi, Chiyu Ma, Wenhua Liang, Xingjian Diao, Weicheng Ma, Soroush Vosoughi

LLM-as-a-Judge has emerged as a promising alternative to human evaluators across various tasks, yet inherent biases - particularly position bias, the tendency to favor solutions based on their position within the prompt - compromise its reliability. This exploratory study evaluates position bias in LLM judges across pairwise and list-wise comparison settings, introducing three metrics: repetition stability, position consistency, and preference fairness. Our experiments, involving 15 LLM judges across MTBench and DevBench with 22 tasks and approximately 40 solution-generating models, result in over 150,000 evaluation instances. We identify Judge-Level, Candidate-Level, and Task-Level factors contributing to bias. The findings confirm that position bias is not due to random chance and varies significantly across judges and tasks. While position bias is weakly influenced by the length of prompt components, it is strongly affected by the quality gap between solutions. Our agreement and disagreement analysis among judges further provides insights into the distribution of judging difficulty across the dataset, and highlights the potential for dataset modifications.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 16 code samples harvested from 1 repository linked to this paper: 16 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 3 honoured, 1 violated, 7 with no contract checked; 5 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 5 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 16 samples from 1 repository, 16 ran (11 with no instrument failure; 5 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

Slimshilin/Position-Bias-Analyzer-Demo officialmentioned in papermentioned on GitHubSyntology: 16 samples, 16 ran (5 instrument-failed) · 0 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

16 samples harvested: 16 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 3 honoured, 1 violated, 7 with no contract checked; 5 where Syntology's instrument failed) · 0 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · honoured contract
1ran · violated contract
5ran · our draft was wrong
7ran

Licence: 16 of the 16 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from Slimshilin/Position-Bias-Analyzer-Demo. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

Slimshilin/Position-Bias-Analyzer-Demo official repository 16 samples, 16 ran (5 instrument-failed) · 0 unverified
Claude_context_window subeval/chat_api/claude.py ran · honoured contract fingerprinted no licence file found · pointer only · b3186cda66ba698e · report fixture: signature model:str/0/str · returned int · level recorded 2026-09-24
GPT_context_window in 5 papers subeval/chat_api/gpt.py ran · honoured contract fingerprinted no licence file found · pointer only · acd86dcb3b21fb02 · report fixture: signature model:str/0/str · returned int · level recorded 2026-09-24
Gemini_context_window subeval/chat_api/gemini.py ran · honoured contract fingerprinted no licence file found · pointer only · 4edd4f61d223fe52 · report fixture: signature model:str/0/str · returned int · level recorded 2026-09-24
aggregate_judge_data position_bias_analyzer/calculate_consistency_and_preference/consistency_preference_util.py ran no licence file found · pointer only · 6a696ae76c55b5e2 · report fixture: signature directory_path:str/0/str, nopt:int/0/int · returned DataFrame · level recorded 2026-09-24
augment_length_stats position_bias_analyzer/position_bias_vs_factors/position_bias_vs_lengths/calculate_length_stats.py ran no licence file found · pointer only · bffbd3fb4e1ab9a6 · report fixture: signature previous_results_df:DataFrame/0/?, length_results_df:DataFrame/0/? · returned DataFrame · level recorded 2026-09-24
augment_results_with_overall_winrates position_bias_analyzer/position_bias_vs_factors/position_bias_vs_winrates/calculate_overall_winrate.py ran no licence file found · pointer only · ff69251cef4fa95a · report fixture: signature previous_results_df:DataFrame/0/?, overall_winrate_results_df:DataFrame/0/? · returned DataFrame · level recorded 2026-09-24
calculate_length_stats position_bias_analyzer/position_bias_vs_factors/position_bias_vs_lengths/calculate_length_stats.py ran no licence file found · pointer only · 390a970b1fdce16c · report fixture: signature data:DataFrame/0/?, benchmark:str/0/str, reference_model:str/0/str · returned DataFrame · level recorded 2026-09-24
calculate_overall_winrate position_bias_analyzer/position_bias_vs_factors/position_bias_vs_winrates/calculate_overall_winrate.py ran no licence file found · pointer only · 1ec9bc4d7069a870 · report fixture: signature data:DataFrame/0/?, reference_model:str/0/str · returned DataFrame · level recorded 2026-09-24
calculate_positional_consistency_and_preference_score position_bias_analyzer/calculate_consistency_and_preference/consistency_preference_calculation.py ran no licence file found · pointer only · 2a367c5cad3f8101 · report fixture: signature data:DataFrame/0/?, reference_model:str/0/str · returned tuple · level recorded 2026-09-24
cn_string subeval/smp.py ran · violated contract no licence file found · pointer only · e8277b4e02292cbe · report fixture: signature s:str/0/str · returned bool · level recorded 2026-09-24
extract_answer in 112 papers position_bias_analyzer/calculate_consistency_and_preference/consistency_preference_util.py ran · our draft was wrong no licence file found · pointer only · 726da9cc7c6c575a · report fixture: signature data:DataFrame/0/?, benchmark:str/0/str · returned DataFrame · level recorded 2026-09-24
match_answer in 6 papers position_bias_analyzer/calculate_consistency_and_preference/consistency_preference_util.py ran · our draft was wrong no licence file found · pointer only · ecf4b23852cb6c61 · report fixture: signature s:str/0/str, benchmark:str/0/str · returned str · level recorded 2026-09-24
match_answer in 6 papers subeval/subjective/analyze_util.py ran · our draft was wrong no licence file found · pointer only · 1f3d4cff9e4b57fd · report fixture: signature s:str/0/str, benchmark:str/0/str · returned str · level recorded 2026-09-24
proc_task subeval/subjective/analyze_util.py ran · our draft was wrong no licence file found · pointer only · 9f7a492d259f95bd · report fixture: signature tasks:list/1/str · returned tuple · level recorded 2026-09-24
splitlen subeval/smp.py ran fingerprinted no licence file found · pointer only · 046fc285c89d080d · report fixture: signature s:str/0/str, sep:str/0/str · returned int · level recorded 2026-09-24
timestr subeval/smp.py ran · our draft was wrong no licence file found · pointer only · b893e74b04f04876 · report fixture: signature second:bool/0/bool, minute:bool/0/bool · returned str · level recorded 2026-09-24

Tasks

Fairness

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutGPT-4Label SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections