Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2505.11855 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915291609432064 |
|---|---|
| author | Son, Guijin Hong, Jiwoo Fan, Honglu Nam, Heejeong Ko, Hyunwoo Lim, Seungwon Song, Jinyeop Choi, Jinha Paulo, Gonçalo Yu, Youngjae Biderman, Stella |
| author_facet | Son, Guijin Hong, Jiwoo Fan, Honglu Nam, Heejeong Ko, Hyunwoo Lim, Seungwon Song, Jinyeop Choi, Jinha Paulo, Gonçalo Yu, Youngjae Biderman, Stella |
| contents | Recent advances in large language models (LLMs) have fueled the vision of automated scientific discovery, often called AI Co-Scientists. To date, prior work casts these systems as generative co-authors responsible for crafting hypotheses, synthesizing code, or drafting manuscripts. In this work, we explore a complementary application: using LLMs as verifiers to automate the \textbf{academic verification of scientific manuscripts}. To that end, we introduce SPOT, a dataset of 83 published papers paired with 91 errors significant enough to prompt errata or retraction, cross-validated with actual authors and human annotators. Evaluating state-of-the-art LLMs on SPOT, we find that none surpasses 21.1\% recall or 6.1\% precision (o3 achieves the best scores, with all others near zero). Furthermore, confidence estimates are uniformly low, and across eight independent runs, models rarely rediscover the same errors, undermining their reliability. Finally, qualitative analysis with domain experts reveals that even the strongest models make mistakes resembling student-level misconceptions derived from misunderstandings. These findings highlight the substantial gap between current LLM capabilities and the requirements for dependable AI-assisted academic verification. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_11855 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | When AI Co-Scientists Fail: SPOT-a Benchmark for Automated Verification of Scientific Research Son, Guijin Hong, Jiwoo Fan, Honglu Nam, Heejeong Ko, Hyunwoo Lim, Seungwon Song, Jinyeop Choi, Jinha Paulo, Gonçalo Yu, Youngjae Biderman, Stella Computation and Language Recent advances in large language models (LLMs) have fueled the vision of automated scientific discovery, often called AI Co-Scientists. To date, prior work casts these systems as generative co-authors responsible for crafting hypotheses, synthesizing code, or drafting manuscripts. In this work, we explore a complementary application: using LLMs as verifiers to automate the \textbf{academic verification of scientific manuscripts}. To that end, we introduce SPOT, a dataset of 83 published papers paired with 91 errors significant enough to prompt errata or retraction, cross-validated with actual authors and human annotators. Evaluating state-of-the-art LLMs on SPOT, we find that none surpasses 21.1\% recall or 6.1\% precision (o3 achieves the best scores, with all others near zero). Furthermore, confidence estimates are uniformly low, and across eight independent runs, models rarely rediscover the same errors, undermining their reliability. Finally, qualitative analysis with domain experts reveals that even the strongest models make mistakes resembling student-level misconceptions derived from misunderstandings. These findings highlight the substantial gap between current LLM capabilities and the requirements for dependable AI-assisted academic verification. |
| title | When AI Co-Scientists Fail: SPOT-a Benchmark for Automated Verification of Scientific Research |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2505.11855 |