SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.04197 · 2024

DICE: Detecting In-distribution Contamination in LLM's Fine-tuning Phase for Math Reasoning

arXiv · PDF · Open in the Atlas

Code that ran

4 of 5 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 3 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; 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/2406.04197.json · badge: /api/ran/2406.04197.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
copy not recorded community
not in the paper's declared repositories
4 ran · 1 no recorded run
FunctionStatusWhere it lives
generation_prompt_template Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from thu-keg/dice
pointer only · get_code("925b9c7288304f71")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

prompt_template Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from thu-keg/dice
pointer only · get_code("f94987f7381d494f")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

read_json Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from agromanou/crab
pointer only; the stored copy get_code serves, from agromanou/crab, is pointer only (licence: NONE) · get_code("f8c65c92ec213846")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

test_DINM Ran
ran, honoured the contract we drafted
this paper's copy was not recorded; identical code first harvested from thu-keg/dice
pointer only · get_code("47eafd807076448c")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

predict Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from thu-keg/dice
pointer only · get_code("8f12c20fe27a3248")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 1 later) and 0 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.

FunctionAlso present inEarliest-datedOthers named
read_json 7 other papers
4 dated earlier, 3 later
2310.15239 (2023-10)
CROW: Benchmarking Commonsense Reasoning in Real-World Tasks
2311.04284 (2023-11)
CRAB: Assessing the Strength of Causal Relationships Between Real-World Events
2402.02823 (2024-02)
Evading Data Contamination Detection for Language Models is (too) Easy
and 4 more not named here
generation_prompt_template 1 other paper
1 dated earlier
2402.02823 (2024-02)
Evading Data Contamination Detection for Language Models is (too) Easy
predict 1 other paper
1 dated earlier
2402.02823 (2024-02)
Evading Data Contamination Detection for Language Models is (too) Easy
prompt_template 1 other paper
1 dated earlier
2402.02823 (2024-02)
Evading Data Contamination Detection for Language Models is (too) Easy
test_DINM 1 other paper
1 dated earlier
2402.02823 (2024-02)
Evading Data Contamination Detection for Language Models is (too) Easy

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

The advancement of large language models (LLMs) relies on evaluation using public benchmarks, but data contamination can lead to overestimated performance. Previous researches focus on detecting contamination by determining whether the model has seen the exact same data during training. Besides, prior work has already shown that even training on data similar to benchmark data inflates performance, namely \emph{In-distribution contamination}. In this work, we argue that in-distribution contamination can lead to the performance drop on OOD benchmarks. To effectively detect in-distribution contamination, we propose DICE, a novel method that leverages the internal states of LLMs to locate-then-detect the contamination. DICE first identifies the most sensitive layer to contamination, then trains a classifier based on the internal states of that layer. Experiments reveal DICE's high accuracy in detecting in-distribution contamination across various LLMs and math reasoning datasets. We also show the generalization capability of the trained DICE detector, which is able to detect contamination across multiple benchmarks with similar distributions. Additionally, we find that DICE's predictions correlate with the performance of LLMs fine-tuned by either us or other organizations, achieving a coefficient of determination ($R^2$) between 0.61 and 0.75. The code and data are available at https://github.com/THU-KEG/DICE.

For agents

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

get_harvested_code_for_paper("2406.04197")
get_code_for_paper("2406.04197")
have("2406.04197")

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

curl https://syntology.ai/api/ran/2406.04197.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.04197.svg)](https://syntology.ai/paper/2406.04197)

Connect an agent — have() is free.