FrontierScience: Evaluating AI's Ability to Perform Expert-Level Scientific Tasks
This addresses the gap in AI evaluation for high-level scientific tasks, providing a more challenging benchmark for researchers and developers, though it is incremental as it builds on existing benchmarking efforts.
The authors tackled the problem of evaluating expert-level scientific reasoning in language models by introducing FrontierScience, a benchmark with Olympiad and Research tracks, resulting in a dataset of several hundred questions including 160 in a gold set, with granular rubric-based evaluation for research tasks.
We introduce FrontierScience, a benchmark evaluating expert-level scientific reasoning in frontier language models. Recent model progress has nearly saturated existing science benchmarks, which often rely on multiple-choice knowledge questions or already published information. FrontierScience addresses this gap through two complementary tracks: (1) Olympiad, consisting of international olympiad problems at the level of IPhO, IChO, and IBO, and (2) Research, consisting of PhD-level, open-ended problems representative of sub-tasks in scientific research. FrontierScience contains several hundred questions (including 160 in the open-sourced gold set) covering subfields across physics, chemistry, and biology, from quantum electrodynamics to synthetic organic chemistry. All Olympiad problems are originally produced by international Olympiad medalists and national team coaches to ensure standards of difficulty, originality, and factuality. All Research problems are research sub-tasks written and verified by PhD scientists (doctoral candidates, postdoctoral researchers, or professors). For Research, we introduce a granular rubric-based evaluation framework to assess model capabilities throughout the process of solving a research task, rather than judging only a standalone final answer.