DeepScientist: Advancing Frontier-Pushing Scientific Findings Progressively
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908569382682624 |
|---|---|
| author | Weng, Yixuan Zhu, Minjun Xie, Qiujie Sun, Qiyao Lin, Zhen Liu, Sifan Zhang, Yue |
| author_facet | Weng, Yixuan Zhu, Minjun Xie, Qiujie Sun, Qiyao Lin, Zhen Liu, Sifan Zhang, Yue |
| contents | While previous AI Scientist systems can generate novel findings, they often lack the focus to produce scientifically valuable contributions that address pressing human-defined challenges. We introduce DeepScientist, a system designed to overcome this by conducting goal-oriented, fully autonomous scientific discovery over month-long timelines. It formalizes discovery as a Bayesian Optimization problem, operationalized through a hierarchical evaluation process consisting of "hypothesize, verify, and analyze". Leveraging a cumulative Findings Memory, this loop intelligently balances the exploration of novel hypotheses with exploitation, selectively promoting the most promising findings to higher-fidelity levels of validation. Consuming over 20,000 GPU hours, the system generated about 5,000 unique scientific ideas and experimentally validated approximately 1100 of them, ultimately surpassing human-designed state-of-the-art (SOTA) methods on three frontier AI tasks by 183.7\%, 1.9\%, and 7.9\%. This work provides the first large-scale evidence of an AI achieving discoveries that progressively surpass human SOTA on scientific tasks, producing valuable findings that genuinely push the frontier of scientific discovery. To facilitate further research into this process, we will open-source all experimental logs and system code at https://github.com/ResearAI/DeepScientist/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_26603 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DeepScientist: Advancing Frontier-Pushing Scientific Findings Progressively Weng, Yixuan Zhu, Minjun Xie, Qiujie Sun, Qiyao Lin, Zhen Liu, Sifan Zhang, Yue Computation and Language Machine Learning While previous AI Scientist systems can generate novel findings, they often lack the focus to produce scientifically valuable contributions that address pressing human-defined challenges. We introduce DeepScientist, a system designed to overcome this by conducting goal-oriented, fully autonomous scientific discovery over month-long timelines. It formalizes discovery as a Bayesian Optimization problem, operationalized through a hierarchical evaluation process consisting of "hypothesize, verify, and analyze". Leveraging a cumulative Findings Memory, this loop intelligently balances the exploration of novel hypotheses with exploitation, selectively promoting the most promising findings to higher-fidelity levels of validation. Consuming over 20,000 GPU hours, the system generated about 5,000 unique scientific ideas and experimentally validated approximately 1100 of them, ultimately surpassing human-designed state-of-the-art (SOTA) methods on three frontier AI tasks by 183.7\%, 1.9\%, and 7.9\%. This work provides the first large-scale evidence of an AI achieving discoveries that progressively surpass human SOTA on scientific tasks, producing valuable findings that genuinely push the frontier of scientific discovery. To facilitate further research into this process, we will open-source all experimental logs and system code at https://github.com/ResearAI/DeepScientist/. |
| title | DeepScientist: Advancing Frontier-Pushing Scientific Findings Progressively |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2509.26603 |