# Are LLMs Ready for Scientific Discovery? A Capability-Oriented Benchmark for AI Scientists

arXiv: https://arxiv.org/abs/2607.11079 · HTML: https://mlsanity.com/paper/2607.11079/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 36 references that are in the map.
- First posted on arXiv 13 Jul 2026.

## Builds on (most cited first, 18 of 18)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [GLM-5: from Vibe Coding to Agentic Engineering](https://mlsanity.com/paper/2602.15763/) (2026)
- [Kimi K2.5: Visual Agentic Intelligence](https://mlsanity.com/paper/2602.02276/) (2026)
- [DS-1000: A Natural and Reliable Benchmark for Data Science Code Generation](https://mlsanity.com/paper/2211.11501/) (2022)
- [ToolQA: A Dataset for LLM Question Answering with External Tools](https://mlsanity.com/paper/2306.13304/) (2023)
- [Data Interpreter: An LLM Agent For Data Science](https://mlsanity.com/paper/2402.18679/) (2024)
- [DSBench: How Far Are Data Science Agents from Becoming Data Science Experts?](https://mlsanity.com/paper/2409.07703/) (2024)
- [DiscoveryBench: Towards Data-Driven Discovery with Large Language Models](https://mlsanity.com/paper/2407.01725/) (2024)
- [BLADE: Benchmarking Language Model Agents for Data-Driven Science](https://mlsanity.com/paper/2408.09667/) (2024)
- [Benchmark Self-Evolving: A Multi-Agent Framework for Dynamic LLM Evaluation](https://mlsanity.com/paper/2402.11443/) (2024)
- [Are LLMs Capable of Data-based Statistical and Causal Reasoning? Benchmarking Advanced Quantitative Reasoning with Data](https://mlsanity.com/paper/2402.17644/) (2024)
- [DataSciBench: An LLM Agent Benchmark for Data Science](https://mlsanity.com/paper/2502.13897/) (2025)
- [DeepAnalyze: Agentic Large Language Models for Autonomous Data Science](https://mlsanity.com/paper/2510.16872/) (2025)
- [KramaBench: A Benchmark for AI Systems on Data-to-Insight Pipelines over Data Lakes](https://mlsanity.com/paper/2506.06541/) (2025)
- [Hypothesis Generation for Materials Discovery and Design Using Goal-Driven and Constraint-Guided LLM Agents](https://mlsanity.com/paper/2501.13299/) (2025)
- [DSCodeBench: A Realistic Benchmark for Data Science Code Generation](https://mlsanity.com/paper/2505.15621/) (2025)
- [SciTS: Scientific Time Series Understanding and Generation with LLMs](https://mlsanity.com/paper/2510.03255/) (2025)
- [AtmosSci-Bench: Evaluating the Recent Advance of Large Language Model for Atmospheric Science](https://mlsanity.com/paper/2502.01159/) (2025)

## Built on by (most cited first, 1 of 1)

- [K-Bench: measuring model performance on real scientific agent requests](https://mlsanity.com/paper/2608.21601/) (2026)

## Abstract (arXiv)

Existing benchmarks for scientific data analysis evaluate LLMs primarily on code execution or workflow completion, overlooking that scientific analysis serves to support distinct types of scientific claims: hypothesis exploration, statistical inference, mechanistic explanation, each with different assumptions and validity criteria. We introduce SDABench, a benchmark that reorganizes evaluation around six capabilities (descriptive, exploratory, inferential, predictive, causal, and mechanistic) across five domains (Biology, Chemistry, Environment, Geography, Physics). SDABench comprises 527 real-data instances (SDA-Real) and 6000 synthetic instances (SDA-Synth), each in both multiple-choice and open-ended formats, constructed through an automated pipeline. Evaluating 15 representative LLMs, we find that models handle descriptive analysis well but degrade sharply on tasks requiring assumption selection, latent-process modeling, or mechanistic reasoning. SDABench further provides a five-stage error analysis framework that locates where LLMs fail: more advanced models more reliably identify the relevant scope and variables, but still struggle to select appropriate analytical procedures, model variable relationships, and draw valid conclusions.