Pith. sign in

REVIEW 3 cited by

Investigating Data Contamination in Modern Benchmarks for Large Language Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2311.09783 v2 pith:5FP2D5NO submitted 2023-11-16 cs.CL cs.AI

classification cs.CLcs.AI
keywords benchmarkbenchmarksdataevaluationllmsmodelscontaminationcertain
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Recent observations have underscored a disparity between the inflated benchmark scores and the actual performance of LLMs, raising concerns about potential contamination of evaluation benchmarks. This issue is especially critical for closed-source models and certain open-source models where training data transparency is lacking. In this paper we study data contamination by proposing two methods tailored for both open-source and proprietary LLMs. We first introduce a retrieval-based system to explore potential overlaps between evaluation benchmarks and pretraining corpora. We further present a novel investigation protocol named \textbf{T}estset \textbf{S}lot Guessing (\textit{TS-Guessing}), applicable to both open and proprietary models. This approach entails masking a wrong answer in a multiple-choice question and prompting the model to fill in the gap. Additionally, it involves obscuring an unlikely word in an evaluation example and asking the model to produce it. We find that certain commercial LLMs could surprisingly guess the missing option in various test sets. Specifically, in the TruthfulQA benchmark, we find that LLMs exhibit notable performance improvement when provided with additional metadata in the benchmark. Further, in the MMLU benchmark, ChatGPT and GPT-4 demonstrated an exact match rate of 52\% and 57\%, respectively, in guessing the missing options in benchmark test data. We hope these results underscore the need for more robust evaluation methodologies and benchmarks in the field.

Discussion (0). Sign in to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 9 citations worldwide. Full citation record

  1. Relay-Bench: Evaluating LLMs on Multi-Domain Reasoning Chains

    cs.CL 2026-07 conditional novelty 5.0 of 10

    Relay-Bench, a 30-problem benchmark of chained multi-domain tasks with encoded prompts, resists saturation: the best tested model, GPT-5.5, scores 43.3% Pass@1.

  2. SciDA: Scientific Dynamic Assessor of LLMs

    cs.CL 2025-06 conditional novelty 5.0 of 10

    SciDA is a dynamically initialized, multi-discipline olympiad benchmark that shows LLMs perform substantially worse when problem variables are randomized, which the authors attribute to memorization of fixed numerical...

  3. Federated In-Context Learning: Iterative Refinement for Improved Answer Quality

    cs.LG 2025-06 conditional novelty 5.0 of 10

    Fed-ICL iteratively refines QA answers via federated in-context learning with only label transmission, showing convergence on a linear attention model and gains on MMLU and TruthfulQA.

Pith tools