AntiLeakBench: Preventing Data Contamination by Automatically Constructing Benchmarks with Updated Real-World Knowledge
This addresses the issue of unfair LLM evaluation for researchers and developers by automating benchmark updates, though it is incremental as it builds on existing contamination prevention methods.
The paper tackles the problem of data contamination in LLM evaluation by proposing AntiLeak-Bench, an automated framework that constructs benchmarks with new knowledge to ensure contamination-free testing, reducing maintenance costs and demonstrating effectiveness in experiments.
Data contamination hinders fair LLM evaluation by introducing test data into newer models' training sets. Existing studies solve this challenge by updating benchmarks with newly collected data. However, they fail to guarantee contamination-free evaluation as the newly collected data may contain pre-existing knowledge, and their benchmark updates rely on intensive human labor. To address these issues, we in this paper propose AntiLeak-Bench, an automated anti-leakage benchmarking framework. Instead of simply using newly collected data, we construct samples with explicitly new knowledge absent from LLMs' training sets, which thus ensures strictly contamination-free evaluation. We further design a fully automated workflow to build and update our benchmark without human labor. This significantly reduces the cost of benchmark maintenance to accommodate emerging LLMs. Through extensive experiments, we highlight that data contamination likely exists before LLMs' cutoff time and demonstrate AntiLeak-Bench effectively overcomes this challenge.