paper-with-me

홈 › Papers

Overestimation in LLM Evaluation: A Controlled Large-Scale Study on Data Contamination's Impact on Machine Translation

2025-01-30 · Muhammed Yusuf Kocyigit, Eleftheria Briakou, Daniel Deutsch, Jiaming Luo, Colin Cherry, Markus Freitag

Data contamination -- the accidental consumption of evaluation examples within the pre-training data -- can undermine the validity of evaluation benchmarks. In this paper, we present a rigorous analysis of the effects of contamination on language models at 1B and 8B scales on the machine translation task. Starting from a carefully decontaminated train-test split, we systematically introduce contamination at various stages, scales, and data formats to isolate its effect and measure its impact on performance metrics. Our experiments reveal that contamination with both source and target substantially inflates BLEU scores, and this inflation is 2.5 times larger (up to 30 BLEU points) for 8B compared to 1B models. In contrast, source-only and target-only contamination generally produce smaller, less consistent over-estimations. Finally, we study how the temporal distribution and frequency of contaminated samples influence performance over-estimation across languages with varying degrees of data resources.

📄 PDF Abstract BibTeX arXiv:2501.18771

Code (0)

등록된 구현이 없습니다.

Tasks

Machine Translation

Similar Papers 제목 키워드 기반

Data Quality Matters For Adversarial Training: An Empirical Study

2021-02-15 · chengyu dong, Liyuan Liu, Jingbo Shang

Multiple intriguing problems are hovering in adversarial training, including robust overfitting, robustness overestimation, and robustness-accuracy trade-off. These problems pose great challenges to both reliable evaluat…

Adversarial Robustness

Adversarial Robustness Overestimation and Instability in TRADES

2024-10-10 · Jonathan Weiping Li, Ren-Wei Liang, Cheng-Han Yeh, Cheng-Chang Tsai 외

This paper examines the phenomenon of probabilistic robustness overestimation in TRADES, a prominent adversarial training method. Our study reveals that TRADES sometimes yields disproportionately high PGD validation accu…

Adversarial Robustness

Evaluation Pitfalls and Challenges in Multimedia Event Extraction

2026-06-25 · Philipp Seeberger, Steffen Freisinger, Tobias Bocklet, Korbinian Riedhammer arxiv

Multimedia event extraction aims to jointly identify events and their arguments across multiple modalities, such as text and images, to support more comprehensive event understanding. While recent work reports steady and…

Event Extraction

Rethinking the effects of data contamination in Code Intelligence

2025-06-03 · Zhen Yang, Hongyi Lin, Yifan He, Jie Xu 외

In recent years, code intelligence has gained increasing importance in the field of automated software engineering. Meanwhile, the widespread adoption of Pretrained Language Models (PLMs) and Large Language Models (LLMs)…

Code GenerationCode SummarizationCode Translation

RobustBench: a standardized adversarial robustness benchmark

2020-10-19 · Francesco Croce, Maksym Andriushchenko, Vikash Sehwag, Edoardo Debenedetti 외

As a research community, we are still lacking a systematic understanding of the progress on adversarial robustness which often makes it hard to identify the most promising ideas in training robust models. A key challenge…

Adversarial RobustnessBenchmarkingFairnessimage-classification+2