Metraipolis › archive › dc_07c97dfec662

SciGame: Let Agents Play the Scientific Discovery Games in the Real and Counterfactual World — arXiv:2610.37919

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.37919
License (as archived)
CC0-1.0
Archived
2026-10-31T06:08:40.666Z
SHA-256 of the archived text
c75a713b2db53cb53d611b814ebb717627abddc9ffb4f8917076ce868ea18bcf
Stable URL of this page
https://metraipolis.com/documento/dc_07c97dfec662

How to cite

SciGame: Let Agents Play the Scientific Discovery Games in the Real and Counterfactual World — arXiv:2610.37919 — arXiv (en), https://arxiv.org/abs/2610.37919 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_07c97dfec662 (Metraipolis archive dc_07c97dfec662, sha256 c75a713b2db53cb53d611b814ebb717627abddc9ffb4f8917076ce868ea18bcf)
BibTeX
@misc{dc07c97dfec662,
  title = {SciGame: Let Agents Play the Scientific Discovery Games in the Real and Counterfactual World — arXiv:2610.37919},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.37919},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_07c97dfec662; sha256 c75a713b2db53cb53d611b814ebb717627abddc9ffb4f8917076ce868ea18bcf},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.37919. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language model agents are increasingly used for scientific discovery. However, existing benchmarks mix law inference with programming, score data fitting instead of prediction on unseen inputs, and do not separate empirical evidence from pretrained priors. We introduce SciGame, a controlled benchmark that converts scientific literature into structured scientific cards. Each card yields a pair of Original and dimensionally consistent Counterfactual environments with the same task interface. SciGame contains a Historical collection of canonical laws and a live Temporal collection curated from papers published after model knowledge cutoffs. It evaluates agents in candidate-guided Identification and open-ended Discovery settings. Each setting runs under Prior-only, Fixed, and Adaptive protocols. Experiments on five frontier models yield three findings. First, familiar priors hurt performance once the governing relation is altered. Second, adaptive experimentation helps the strongest models but harms weaker ones under limited budgets. Third, good explanations can hide quantitative failures. Recovering the exact symbolic form remains the main bottleneck, even when agents interpret evidence correctly. Current agents can investigate systems actively, but inducing exact laws in unfamiliar settings remains an open challenge.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.37919 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-31T06:07:08.440Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).