SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2510.09259 · ICLR · 2025

Detecting Data Contamination from Reinforcement Learning Post-training for Large Language Models

Ge Li, Kechi Zhang, Yihong Dong, Yongding Tao, Huanyu Liu, Tian Wang, Xiaolong Hu

arXiv · PDF · Open in the Atlas

Code that ran

2 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 1 failure of our instrument; 1 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2510.09259.json · badge: /api/ran/2510.09259.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
yongding-tao/RL-Data-Contamination community
Papers with Code archive
2 ran · 1 no recorded run
FunctionStatusWhere it lives
SelfCritiqueDetector Ran
ran on a synthesized input
yongding-tao/RL-Data-Contamination/detectors/self_critique.py
pointer only (licence: NONE) · get_code("b984df7960f96771")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yongding-tao/RL-Data-Contamination when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

calculate_penalized_cosine_similarity Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yongding-tao/RL-Data-Contamination/detectors/self_critique.py
pointer only (licence: NONE) · get_code("cd15a5ec9d5a687a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yongding-tao/RL-Data-Contamination when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

BaseDetector Not yet run
unverified (harvested, no recorded run)
yongding-tao/RL-Data-Contamination/detectors/self_critique.py
pointer only (licence: NONE) · get_code("0b0bc9b6266d1e5f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yongding-tao/RL-Data-Contamination when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 3 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Data contamination poses a significant threat to the reliable evaluation of Large Language Models (LLMs). This issue arises when benchmark samples may inadvertently appear in training sets, compromising the validity of reported performance. While detection methods have been developed for the pre-training and Supervised Fine-Tuning stages, a critical research gap exists for the increasingly significant phase of Reinforcement Learning (RL) post-training. As RL posttraining becomes pivotal for advancing LLM reasoning, the absence of specialized contamination detection methods in this paradigm presents a critical vulnerability. To address this, we conduct the first systematic study of data detection within RL post-training scenario and propose Self-Critique. Our method is motivated by a key observation: after RL phase, the output entropy distribution of LLMs tends to collapse into highly specific and sparse modes. Self-Critique probes for the underlying policy collapse, i.e., the model's convergence to a narrow reasoning path, which causes this entropy reduction. To facilitate this research, we also introduce RL-MIA, a benchmark constructed to simulate this specific contamination scenario. Extensive experiments show that Self-Critique significantly outperforms baseline methods across multiple models and contamination tasks, achieving an AUC improvement of up to 30%. Whereas existing methods are close to a random guess for RL-phase contamination, our method makes detection possible. Our benchmark and code are available at https://github.com/yongding-tao/RL-Data-Contamination.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2510.09259")
get_code_for_paper("2510.09259")
have("2510.09259")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2510.09259.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2510.09259.svg)](https://syntology.ai/paper/2510.09259)

Connect an agent — have() is free.