SWE-MERA: A Dynamic Benchmark for Agenticly Evaluating Large Language Models on Software Engineering Tasks
This addresses critical limitations in benchmarking LLMs for software engineering, though it is incremental as it builds on prior datasets like SWE-bench.
The authors tackled the problem of data contamination and inadequate test cases in existing software engineering benchmarks for LLMs by introducing SWE-MERA, a dynamic benchmark with automated collection from GitHub, which showed strong discriminative power in evaluating a dozen recent models.
The rapid advancement of Large Language Models (LLMs) in software engineering has revealed critical limitations in existing benchmarks, particularly the widely used SWE-bench dataset. Recent studies have uncovered severe data contamination issues, e.g. SWE-bench reports 32.67% of successful patches involve direct solution leakage and 31.08% pass due to inadequate test cases. We introduce SWE-MERA, a dynamic, continuously updated benchmark designed to address these fundamental challenges through an automated collection of real-world GitHub issues and rigorous quality validation. Our approach implements a reliable pipeline that ensures quality while minimizing contamination risks, resulting in approximately 10,000 potential tasks with 300 samples currently available. Evaluation using the Aider coding agent demonstrates strong discriminative power in state-of-the-art models. We report performance across a dozen recent LLMs evaluated on tasks collected between September 2024 and June 2025.