DARE-bench: Evaluating Modeling and Instruction Fidelity of LLMs in Data Science
This addresses the need for standardized, process-aware evaluation in data science for LLM developers and researchers, though it is incremental as it builds on existing benchmarking efforts.
The paper tackles the problem of evaluating LLMs in data science tasks by introducing DARE-bench, a benchmark with 6,300 Kaggle-derived tasks and verifiable ground truth, which shows that even capable models like gpt-o4-mini struggle, and fine-tuning with it boosts Qwen3-32B's accuracy by 1.83x and Qwen3-4B's by over 8x.
The fast-growing demands in using Large Language Models (LLMs) to tackle complex multi-step data science tasks create an emergent need for accurate benchmarking. There are two major gaps in existing benchmarks: (i) the lack of standardized, process-aware evaluation that captures instruction adherence and process fidelity, and (ii) the scarcity of accurately labeled training data. To bridge these gaps, we introduce DARE-bench, a benchmark designed for machine learning modeling and data science instruction following. Unlike many existing benchmarks that rely on human- or model-based judges, all tasks in DARE-bench have verifiable ground truth, ensuring objective and reproducible evaluation. To cover a broad range of tasks and support agentic tools, DARE-bench consists of 6,300 Kaggle-derived tasks and provides both large-scale training data and evaluation sets. Extensive evaluations show that even highly capable models such as gpt-o4-mini struggle to achieve good performance, especially in machine learning modeling tasks. Using DARE-bench training tasks for fine-tuning can substantially improve model performance. For example, supervised fine-tuning boosts Qwen3-32B's accuracy by 1.83x and reinforcement learning boosts Qwen3-4B's accuracy by more than 8x. These significant improvements verify the importance of DARE-bench both as an accurate evaluation benchmark and critical training data.