Driving · 2026-11-04

Embodied AI for Scientific Discovery: From Physical Action to Scientific Evidence

Wenhao Xu, Changwei Wang, Shunpeng Chen, Yukun Song, Sifan Zhou, Kaize Luo, Shibiao Xu, Yuehua Zhao, Rongtao Xu

arXiv:2611.04371PDF

AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.

TL;DR

This review defines “embodied AI for science” as learned decision procedures coupled to physical experiments with partial observability, costly/irreversible actions, and feedback that updates subsequent decisions. It organizes prior work using three closures—operational, epistemic, evidential—within a record-linked evidence cycle (Record–Diagnose–Calibrate–Confirm). It proposes an evaluation and reporting framework that separates actions, valid experiments, campaign progress, and confirmed claims.

Why it matters

The authors argue that throughput/task success alone is insufficient for scientific discovery: systems must be evaluated by how deeply their experimental records support claims. By separating commanded vs realized vs observed conditions and defining operational/epistemic/evidential closures, the review provides a structured way to compare methods, analyze failure modes, and specify what evidence is required for confirmation.

Method

  • Defines embodied AI for science as a decision procedure coupled to a physical interface, with feedback that informs subsequent decisions; the experiment is partially observed (specimen/apparatus/measurement process latent).
  • Organizes literature via three closures: operational closure (intended protocol → realized experiment), epistemic closure (observations → belief/next experiment), and evidential closure (record → claim).
  • Proposes evaluation separating actions, valid experiments, campaigns, and confirmed claims, with reporting standards grounded in a record-linked evidence cycle: Record → Diagnose → Calibrate → Confirm.
Abstract (from arXiv)

Scientific experiments are increasingly selected, executed and interpreted by learned systems that act through physical interfaces. Robot learning builds physical competence, whereas AI for science chooses which experiment to run, each reporting different endpoints. This review treats their interface as one problem, embodied AI for science, where the experiment itself is partially observed: specimen state, apparatus state and measurement processes are latent, actions are costly and irreversible, and missingness depends on execution and outcome. We organize the literature by three closures operating at distinct levels: operational closure (the experiment realized is the one intended), epistemic closure (the result updates beliefs or selects subsequent experiments) and evidential closure (the record supports a claim). We analyze workflows through a record-linked evidence cycle - Record, Diagnose, Calibrate, Confirm - as analytic functions with feedback. A taxonomy crosses these closures with six delegated decisions and five embodiment classes. Our formulation separates commanded, realized and observed conditions and models diagnosis with explicit resource, risk and sample-change costs. We compare experiment-selection methods by terminal decision, review perception, protocol compilation, visuomotor policies, world models, recovery and transfer by evidence emission requirements, and compare chemistry, materials, biology, microscopy and field workflows by what their records establish. An appendix provides an evidence map of 28 workflows (26 physical, two software-boundary), characterizing validation pathways and claim confirmation requirements. We propose evaluation that separates actions, valid experiments, campaigns and confirmed claims, framing open challenges as controlled comparisons with measurable failure conditions.

Related papers

Publication record

Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.

First public release: 2026-11-03 · Indexed by Pub-AI: 2026-11-04 UTC

preprint

Embodied AI for Scientific Discovery: From Physical Action to Scientific Evidence

Wenhao Xu, Changwei Wang, Shunpeng Chen, Yukun Song, Sifan Zhou, Kaize Luo, Shibiao Xu, Yuehua Zhao, Rongtao Xu

Preprint posted: 2026-11-03 · Original record ↗

arxiv
2611.04371
Explicit source versions (1)
  • v1 · 2026-11-03
  1. Preprint available

    Source date: 2026-11-03 · First detected: 2026-11-04 UTC · Historical baseline

    Source evidence (2)
Abstract · arxiv

Scientific experiments are increasingly selected, executed and interpreted by learned systems that act through physical interfaces. Robot learning builds physical competence, whereas AI for science chooses which experiment to run, each reporting different endpoints. This review treats their interface as one problem, embodied AI for science, where the experiment itself is partially observed: specimen state, apparatus state and measurement processes are latent, actions are costly and irreversible, and missingness depends on execution and outcome. We organize the literature by three closures operating at distinct levels: operational closure (the experiment realized is the one intended), epistemic closure (the result updates beliefs or selects subsequent experiments) and evidential closure (the record supports a claim). We analyze workflows through a record-linked evidence cycle - Record, Diagnose, Calibrate, Confirm - as analytic functions with feedback. A taxonomy crosses these closures with six delegated decisions and five embodiment classes. Our formulation separates commanded, realized and observed conditions and models diagnosis with explicit resource, risk and sample-change costs. We compare experiment-selection methods by terminal decision, review perception, protocol compilation, visuomotor policies, world models, recovery and transfer by evidence emission requirements, and compare chemistry, materials, biology, microscopy and field workflows by what their records establish. An appendix provides an evidence map of 28 workflows (26 physical, two software-boundary), characterizing validation pathways and claim confirmation requirements. We propose evaluation that separates actions, valid experiments, campaigns and confirmed claims, framing open challenges as controlled comparisons with measurable failure conditions.