Evaluation is proposed that separates actions, valid experiments, campaigns and confirmed claims, framing open challenges as controlled comparisons with measurable failure conditions.
Abstract
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.
Scientific reasoning requires interaction with the physical world: designing experiments, interpreting outcomes, and deciding what experiment should come next. Here we show that general-purpose large language models (LLMs) already possess substantial capabilities for experimental reasoning. We connected LLMs to a liqui...
Realizing Embodied AI for Science requires robots to translate experimental goals into reliable physical procedures, reducing hazardous and repetitive human work while generating evidence for scientific discovery. Although language-based planning and learned manipulation provide complementary capabilities, their integr...
Shi-Tong Liu, Xu Yang, J. Zhang et al.· 0 citations
This work introduces EmbodiedLab, a benchmark in which a robot must act to know, and evaluates six frontier models in their developers' coding agents with native observation and control, with added perception and motion planning, and with manipulation skills distilled from the strongest agents.
Hong-Cheng Gao, Ze Zheng, Alex Zhou et al.· 1 citation
An embodied brain, or System 2, must turn an abstract goal into actions while discovering relevant scene conditions and checking whether executed skills achieved their intended effects. Open-loop evaluations cannot test observations shaped by an agent's decisions, while some closed-loop evaluations provide limited evid...
The results suggest that the frontier of Physical AI depends not only on stronger action models, but also on executable harnesses that integrate perception, task understanding and reasoning, and action execution into a unified, verifiable, and feedback-driven system.
EmbodiedEvo is a self-evolving coding agent that distills physical feedback into reusable code Skills and diagnostic Lessons that improves weaker executors and shows the largest gains on memory-dependent tasks and successful reuse of coordinated procedures.