arxiv: 2605.04375 · v1 · submitted 2026-05-06 · 📡 eess.SY · cs.AI· cs.SY

Recognition: unknown

Experiment-as-Code Labs: A Declarative Stack for AI-Driven Scientific Discovery

Zhenning Yang , Yuhan Chen , Patrick Tser Jern Kon , Tongyuan Miao , Hongyi Lin , Venkat Viswanathan , Danai Koutra , Ang Chen

Authors on Pith no claims yet

Pith reviewed 2026-05-08 17:25 UTC · model grok-4.3

classification 📡 eess.SY cs.AIcs.SY

keywords experiment-as-codedeclarative configurationsAI for scienceautonomous labsscientific discoverysystems stackprogrammable instruments

0 comments

The pith

Experiments can be encoded as declarative configurations that AI agents generate and systems compile to control any lab instrument.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper advances a new paradigm called Experiment-as-Code Labs to let AI agents move beyond simulated environments and directly operate physical scientific instruments. Experiments are expressed as declarative configurations rather than imperative code; a systems layer then analyzes them for safety and resources before compiling to device APIs. This stack is presented as independent of any particular science, laboratory, or instrument. A sympathetic reader would care because it promises to connect increasingly capable AI agents to the real-world observations that often drive discovery, including course changes prompted by unexpected results during runs.

Core claim

By encoding experiments as declarative configurations that AI agents produce, a systems layer can perform program analysis, safety checks, resource assignment, and job orchestration, after which the configurations compile to device-level APIs for actual execution. This yields a general stack that operates without dependence on specific scientific domains, laboratories, or instruments.

What carries the argument

The declarative configuration that encodes an experiment, compiled by the systems layer to device APIs after analysis and orchestration.

If this is right

AI agents can propose and revise experiments without direct exposure to hardware APIs.
The systems layer automatically enforces safety checks and assigns resources before any physical action occurs.
The same experimental description can be reused across different laboratories and instruments.
Discovery workflows remain portable and independent of particular scientific fields or equipment setups.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Portable declarative descriptions could increase reproducibility by allowing the same experiment to execute identically on equivalent hardware elsewhere.
High-throughput screening campaigns across chemistry, biology, and materials science could share a common orchestration backend.
Real-time adaptation during runs would still require extensions to the declarative model if the paper's weakest assumption does not hold.

Load-bearing premise

Declarative configurations can capture the complexity, adaptability, and real-time decision-making required for physical experiments, including handling unexpected observations during execution.

What would settle it

A physical experiment in which an unforeseen sensor reading requires an immediate instrument adjustment that cannot be pre-expressed in any declarative configuration the agent could have generated beforehand.

read the original abstract

To unleash the full potential of AI for Science, we must untether the agents from a purely digital environment. The agent's ability to control and explore in real-world labs is essential because the physical lab remains foundational to scientific discovery. While some tasks can be performed on a computer (e.g., data analysis, running simulated experiments), Eureka moments could occur at any time while operating lab instruments (e.g., when a scientist notices unexpected clues, intuition may prompt a real-time course change). Although autonomous labs are on the rise, which expose programmable APIs to control scientific instruments via software, bridging the gap between increasingly powerful AI agents and automated lab equipment requires innovation that draws insights from computer systems. We propose a new paradigm called ``Experiment-as-Code (EaC) Labs,'' where a core concept is to encode experiments as declarative configurations that can be compiled down to device-level APIs. AI agents come up with hypotheses and experiments, written as an ensemble of declarative configurations. The systems layer performs program analysis, safety checks, resource assignment, and job orchestration. Finally, programmatic experimentation occurs via actuating the device APIs. This is a general stack that is science-, lab-, and instrument-independent, representing a novel synthesis across the physical, systems, and intelligence layers to unleash the next breakthrough in AI for Science.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper sketches a three-layer declarative stack for AI agents to drive physical labs but stays at the level of an untested architecture with no implementation or validation.

read the letter

The new piece here is the specific synthesis: AI agents generate hypotheses as ensembles of declarative experiment configs, a systems layer runs program analysis, safety checks, and orchestration, and everything compiles down to device APIs. They position this as a general, science- and instrument-independent stack that pulls in ideas from programming languages and operating systems to close the gap between digital agents and real hardware. That framing is clean and directly addresses why current autonomous-lab APIs alone are not enough for serendipitous discovery. The motivation section does a good job noting that Eureka moments often happen during physical operation, not just in simulation or post-hoc analysis. Credit for keeping the proposal high-level and portable rather than tying it to one domain or one robot arm. The soft spot is exactly the one the stress-test note flags. Declarative configs plus post-hoc orchestration sound fine for scripted runs, but the paper gives no mechanism for embedding live sensor-driven branching, conditional re-planning, or keeping the AI agent in the loop when an observation deviates from the pre-declared plan. Physical experiments routinely require exactly those mid-run adjustments, and without them the stack risks collapsing to better automation rather than AI-driven discovery. There are also no examples, pseudocode, error models, or even a toy implementation, so we cannot check whether the compilation step preserves safety or whether the systems layer can actually schedule conflicting instrument access. This is for people working on AI-for-Science architectures who want a high-level organizing idea rather than evaluated results. It could usefully seed discussion in a reading group or workshop, but it does not yet deliver something a practitioner could build from. I would send it to peer review as a vision paper because the gap it names is real and the three-layer breakdown is coherent, though any serious referee will ask for at least a prototype or concrete encoding examples before accepting the central claim.

Referee Report

1 major / 2 minor

Summary. The paper proposes 'Experiment-as-Code (EaC) Labs,' a declarative stack for AI-driven scientific discovery. AI agents formulate hypotheses and experiments as declarative configurations, which are compiled to device APIs. A systems layer handles program analysis, safety checks, resource assignment, and orchestration, enabling programmatic control of lab instruments. The approach is claimed to be general across sciences, labs, and instruments, synthesizing physical, systems, and intelligence layers.

Significance. This conceptual proposal could have substantial impact on AI for Science by providing a framework to integrate AI agents with physical experimentation, potentially enabling more adaptive and autonomous discovery processes. The novelty lies in the declarative approach drawing from computer systems principles. However, without any implementation, experiments, or detailed mechanisms, the significance remains prospective and hinges on whether the architecture can address the complexities of real-world labs.

major comments (1)

Abstract: The claim that the stack enables handling of unexpected observations and real-time course changes during experiments is not supported by the described components. The architecture specifies compilation of declarative configurations to device APIs and post-hoc orchestration, but provides no mechanism for embedding observation-driven control flow, dynamic branching, or maintaining AI agents in the loop during execution. This is load-bearing for distinguishing the proposal from scripted automation.

minor comments (2)

Abstract: The term 'ensemble of declarative configurations' is introduced without a concrete example or definition of its structure, which would aid clarity.
The manuscript would benefit from citing related work on autonomous laboratories and declarative programming in scientific contexts to better position the contribution.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the detailed and constructive review of our manuscript. The major comment raises an important point about the alignment between the abstract claims and the described architecture, which we address below. We will incorporate revisions to strengthen the presentation of the proposal.

read point-by-point responses

Referee: Abstract: The claim that the stack enables handling of unexpected observations and real-time course changes during experiments is not supported by the described components. The architecture specifies compilation of declarative configurations to device APIs and post-hoc orchestration, but provides no mechanism for embedding observation-driven control flow, dynamic branching, or maintaining AI agents in the loop during execution. This is load-bearing for distinguishing the proposal from scripted automation.

Authors: We agree that the referee's assessment is correct: the manuscript's current description of the EaC Labs stack centers on declarative configurations being compiled to device APIs, with the systems layer performing analysis, safety checks, resource assignment, and orchestration, without explicitly specifying mechanisms for observation-driven control flow, dynamic branching during execution, or continuous AI agent involvement in real-time loops. This leaves the abstract's reference to handling unexpected observations and course changes insufficiently supported by the architecture details provided. To resolve this, we will revise the manuscript by expanding the systems layer description to outline how monitoring of device outputs can trigger iterative re-compilation of configurations (potentially involving AI agents for hypothesis updates) or by adjusting the abstract and introduction to more precisely reflect the core declarative and orchestration contributions while noting dynamic adaptation as an enabled extension rather than a fully detailed feature. This revision will better clarify the distinction from scripted automation and align claims with the proposal's scope. revision: yes

Circularity Check

0 steps flagged

No circularity: declarative architecture proposal without derivations or self-referential reductions

full rationale

The paper presents a conceptual proposal for Experiment-as-Code Labs, describing a stack where AI agents generate hypotheses as declarative configurations that compile to device APIs, with a systems layer handling analysis, safety, and orchestration. No equations, fitted parameters, predictions, or derivation chains appear in the abstract or described content. No self-citations are used to justify uniqueness theorems, ansatzes, or load-bearing premises. The central claim is an architectural synthesis across physical, systems, and intelligence layers, which stands as an independent forward-looking description rather than reducing to its own inputs by construction. This is a standard non-circular outcome for a systems/architecture paper.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 1 invented entities

The proposal rests on domain assumptions about AI capability and declarative expressiveness rather than new entities or fitted parameters.

axioms (2)

domain assumption AI agents can reliably generate useful hypotheses and experiments in declarative form
Invoked when stating that agents come up with experiments written as declarative configurations.
domain assumption Declarative configurations can be compiled to device APIs while preserving safety and intent
Central to the systems layer performing program analysis, safety checks, and orchestration.

invented entities (1)

Experiment-as-Code Labs stack no independent evidence
purpose: To provide a general, science-independent layer connecting AI agents to physical lab execution
New architectural concept introduced in the proposal without prior empirical grounding.

pith-pipeline@v0.9.0 · 5562 in / 1358 out tokens · 26889 ms · 2026-05-08T17:25:42.152041+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

209 extracted references · 114 canonical work pages

[1]

2012 , doi =

Hawkes, Nigel , title =. 2012 , doi =. https://www.bmj.com/content/344/bmj.e2555.full.pdf , journal =

2012
[2]

1,500 scientists lift the lid on reproducibility

Baker, Monya , title=. Nature , year=. doi:10.1038/533452a , url=

work page doi:10.1038/533452a
[3]

Wolfram language: Programming language + built-in knowledge , url=
[4]

2025 , publisher =

Yang, Zhenning and Bhatnagar, Archit and Qiu, Yiming and Miao, Tongyuan and Tser Jern Kon, Patrick and Xiao, Yunming and Huang, Yibo and Casado, Martin and Chen, Ang , title =. 2025 , publisher =

2025
[5]

2024 , eprint=

The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery , author=. 2024 , eprint=

2024
[6]

arXiv preprint arXiv:2405.09783 , year=

Llm and simulation as bilevel optimizers: A new paradigm to advance physical scientific discovery , author=. arXiv preprint arXiv:2405.09783 , year=

work page arXiv
[7]

Interesting scientific idea generation using knowledge graphs and llms: Evaluations with 100 research group leaders.arXiv:2405.17044, 2024

Generation and human-expert evaluation of interesting research ideas using knowledge graphs and large language models , author=. arXiv preprint arXiv:2405.17044 , year=

work page arXiv
[8]

, author Chen, S

Scienceagentbench: Toward rigorous assessment of language agents for data-driven scientific discovery , author=. arXiv preprint arXiv:2410.05080 , year=

work page arXiv
[9]

2024 , eprint=

Large Language Models as Biomedical Hypothesis Generators: A Comprehensive Evaluation , author=. 2024 , eprint=

2024
[10]

arXiv preprint arXiv:2507.02554 , year=

AI Research Agents for Machine Learning: Search, Exploration, and Generalization in MLE-bench , author=. arXiv preprint arXiv:2507.02554 , year=

work page arXiv
[11]

2025 , eprint=

The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search , author=. 2025 , eprint=

2025
[12]

2025 , eprint=

AI-Researcher: Autonomous Scientific Innovation , author=. 2025 , eprint=

2025
[13]

2025 , eprint=

When AI Co-Scientists Fail: SPOT-a Benchmark for Automated Verification of Scientific Research , author=. 2025 , eprint=

2025
[14]

2025 , eprint=

Large Language Models Penetration in Scholarly Writing and Peer Review , author=. 2025 , eprint=

2025
[15]

Computer Methods and Programs in Biomedicine Update , volume=

Using artificial intelligence in academic writing and research: An essential productivity tool , author=. Computer Methods and Programs in Biomedicine Update , volume=. 2024 , publisher=

2024
[16]

Advances in Simulation , volume=

Artificial intelligence-assisted academic writing: recommendations for ethical use , author=. Advances in Simulation , volume=. 2025 , publisher=

2025
[17]

BMC Medical Education , volume=

Exploring the potential of artificial intelligence to enhance the writing of english academic papers by non-native english-speaking medical students-the educational application of ChatGPT , author=. BMC Medical Education , volume=. 2024 , publisher=

2024
[18]

2026 , eprint=

A Survey of AI Scientists , author=. 2026 , eprint=

2026
[19]

Nature Communications , volume=

Evaluating large language model agents for automation of atomic force microscopy , author=. Nature Communications , volume=. 2025 , publisher=

2025
[20]

Nature , volume=

Autonomous chemical research with large language models , author=. Nature , volume=. 2023 , publisher=

2023
[21]

2025 , eprint=

BLADE: Benchmarking Language Model Agents for Data-Driven Science , author=. 2025 , eprint=

2025
[22]

2024 , eprint=

DS-Agent: Automated Data Science by Empowering Large Language Models with Case-Based Reasoning , author=. 2024 , eprint=

2024
[23]

Advanced Materials , volume=

SciAgents: automating scientific discovery through bioinspired multi-agent intelligent graph reasoning , author=. Advanced Materials , volume=. 2025 , publisher=

2025
[24]

H yp ER : Literature-grounded Hypothesis Generation and Distillation with Provenance

Vasu, Rosni and Basu, Chandrayee and Dalvi Mishra, Bhavana and Sarasua, Cristina and Clark, Peter and Bernstein, Abraham. H yp ER : Literature-grounded Hypothesis Generation and Distillation with Provenance. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025. doi:10.18653/v1/2025.emnlp-main.1292

work page doi:10.18653/v1/2025.emnlp-main.1292 2025
[25]

Many Heads Are Better Than One: Improved Scientific Idea Generation by A LLM -Based Multi-Agent System

Su, Haoyang and Chen, Renqi and Tang, Shixiang and Yin, Zhenfei and Zheng, Xinzhe and Li, Jinzhe and Qi, Biqing and Wu, Qi and Li, Hui and Ouyang, Wanli and Torr, Philip and Zhou, Bowen and Dong, Nanqing. Many Heads Are Better Than One: Improved Scientific Idea Generation by A LLM -Based Multi-Agent System. Proceedings of the 63rd Annual Meeting of the As...

work page doi:10.18653/v1/2025.acl-long.1368 2025
[26]

Hypothesis Generation with Large Language Models , url=

Zhou, Yangqiaoyu and Liu, Haokun and Srivastava, Tejes and Mei, Hongyuan and Tan, Chenhao , year=. Hypothesis Generation with Large Language Models , url=. doi:10.18653/v1/2024.nlp4science-1.10 , booktitle=

work page doi:10.18653/v1/2024.nlp4science-1.10 2024
[27]

Litllm: A toolkit for literature review with large language models.arXiv preprint arXiv:2402.01788, 2024

Litllm: A toolkit for scientific literature review , author=. arXiv preprint arXiv:2402.01788 , year=

work page arXiv
[28]

Rodriques, and Andrew D

Paperqa: Retrieval-augmented generative agent for scientific research , author=. arXiv preprint arXiv:2312.07559 , year=

work page arXiv
[29]

2025 , eprint=

Agent Laboratory: Using LLM Agents as Research Assistants , author=. 2025 , eprint=

2025
[30]

Advanced Materials Technologies , year =

Scientific Discovery at the Press of a Button: Navigating Automated and Cloud Laboratory Landscapes , author =. Advanced Materials Technologies , year =
[31]

PLoS ONE , year =

Automation in the Life Science Research Laboratory , author =. PLoS ONE , year =
[32]

2024 , url =

Top 10 Lab Automation Software Solutions , author =. 2024 , url =

2024
[33]

2025 , url =

LabLynx Laboratory Software Suites: Transforming Your Lab to Meet Every Need , author =. 2025 , url =

2025
[34]

2026 , url =

Top 15 LIMS Vendors in 2026: Based on Real User Reviews , author =. 2026 , url =

2026
[35]

2026 , note =

Terraform by. 2026 , note =

2026
[36]

2025 , eprint=

Automated Cloud Infrastructure-as-Code Reconciliation with AI Agents , author=. 2025 , eprint=

2025
[37]

2025 , publisher =

Yang, Zhenning and Bhatnagar, Archit and Qiu, Yiming and Miao, Tongyuan and Tser Jern Kon, Patrick and Xiao, Yunming and Huang, Yibo and Casado, Martin and Chen, Ang , title =. 2025 , publisher =. doi:10.1145/3759441.3759443 , journal =

work page doi:10.1145/3759441.3759443 2025
[38]

Automated Bug Discovery in Cloud Infrastructure-as-Code Updates with LLM Agents , year=

Xiang, Yiming and Yang, Zhenning and Peng, Jingjia and Bauer, Hermann and Kon, Patrick Tser Jern and Qiu, Yiming and Chen, Ang , booktitle=. Automated Bug Discovery in Cloud Infrastructure-as-Code Updates with LLM Agents , year=
[39]

Automated Lifting for Cloud Infrastructure-as-Code Programs , year=

Peng, Jingjia and Qiu, Yiming and Kon, Patrick Tser Jern and Zhao, Pinhan and Huang, Yibo and Guo, Zheng and Wang, Xinyu and Chen, Ang , booktitle=. Automated Lifting for Cloud Infrastructure-as-Code Programs , year=
[40]

Artificial intelligence in drug development

Zhang, Kang and Yang, Xin and Wang, Yifei and Yu, Yunfang and Huang, Niu and Li, Gen and Li, Xiaokun and Wu, Joseph C and Yang, Shengyong. Artificial intelligence in drug development. Nat. Med
[41]

No ´e, A

Noé, Frank and Tkatchenko, Alexandre and Müller, Klaus-Robert and Clementi, Cecilia. Machine Learning for Molecular Simulation. Annual Review of Physical Chemistry. 2020. doi:https://doi.org/10.1146/annurev-physchem-042018-052331

work page doi:10.1146/annurev-physchem-042018-052331 2020
[42]

AI4Materials: Transforming the landscape of materials science and enigneering , journal =

Xue Jiang and Dezhen Xue and Yang Bai and William Yi Wang and Jianjun Liu and Mingli Yang and Yanjing Su , keywords =. AI4Materials: Transforming the landscape of materials science and enigneering , journal =. 2025 , issn =. doi:https://doi.org/10.1016/j.revmat.2025.100010 , url =

work page doi:10.1016/j.revmat.2025.100010 2025
[43]

and Seshadri, Ram , title =

Cheetham, Anthony K. and Seshadri, Ram , title =. Chemistry of Materials , volume =. 2024 , doi =

2024
[44]

Scaling deep learning for materials discovery , volume =

Merchant, Amil and Batzner, Simon and Schoenholz, Samuel and Aykol, Muratahan and Cheon, Gowoon and Cubuk, Ekin , year =. Scaling deep learning for materials discovery , volume =. Nature , doi =
[45]

Accurate structure prediction of biomolecular interactions with AlphaFold 3

Abramson, Josh and Adler, Jonas and Dunger, Jack and Evans, Richard and Green, Tim and Pritzel, Alexander and Ronneberger, Olaf and Willmore, Lindsay and Ballard, Andrew J and Bambrick, Joshua and Bodenstein, Sebastian W and Evans, David A and Hung, Chia-Chun and O'Neill, Michael and Reiman, David and Tunyasuvunakool, Kathryn and Wu, Zachary and Z emgulyt...
[46]

Nature Communications , author =

Autonomous platforms for data-driven organic synthesis , volume =. Nature Communications , author =. 2022 , note =. doi:10.1038/s41467-022-28736-4 , abstract =

work page doi:10.1038/s41467-022-28736-4 2022
[47]

Nature Synthesis , author =

Empowering scientists with data-driven automated experimentation , copyright =. Nature Synthesis , author =. 2023 , note =. doi:10.1038/s44160-023-00337-z , abstract =

work page doi:10.1038/s44160-023-00337-z 2023
[48]

doi:10.1038/s44160-022-00231-0

The rise of self-driving labs in chemical and materials sciences , copyright =. Nature Synthesis , author =. 2023 , note =. doi:10.1038/s44160-022-00231-0 , abstract =

work page doi:10.1038/s44160-022-00231-0 2023
[49]

Chemistry of Materials , author =

Self-. Chemistry of Materials , author =. 2023 , note =. doi:10.1021/acs.chemmater.2c03593 , abstract =

work page doi:10.1021/acs.chemmater.2c03593 2023
[50]

Digital Discovery , author =

Driving school for self-driving labs , volume =. Digital Discovery , author =. 2023 , note =. doi:10.1039/D3DD00150D , abstract =

work page doi:10.1039/d3dd00150d 2023
[51]

Nature Communications , author =

Autonomous optimization of non-aqueous. Nature Communications , author =. 2022 , note =. doi:10.1038/s41467-022-32938-1 , abstract =

work page doi:10.1038/s41467-022-32938-1 2022
[52]

Nature Communications , author =

Differentiable modeling and optimization of non-aqueous. Nature Communications , author =. 2024 , note =. doi:10.1038/s41467-024-51653-7 , abstract =

work page doi:10.1038/s41467-024-51653-7 2024
[53]

Nature Materials , author =

Accelerating advanced-materials commercialization , volume =. Nature Materials , author =. 2016 , note =. doi:10.1038/nmat4625 , abstract =

work page doi:10.1038/nmat4625 2016
[54]

doi:10.26434/chemrxiv-2025-6wq3x , abstract =

Chen, Yuhan and Lin, Hongyi and Zhang, Tianyi and Dave, Adarsh and Mitchell, Jared and Whitacre, Jay and Viswanathan, Venkatasubramanian , month = aug, year =. doi:10.26434/chemrxiv-2025-6wq3x , abstract =

work page doi:10.26434/chemrxiv-2025-6wq3x 2025
[55]

Nature Synthesis , author =

Navigating phase diagram complexity to guide robotic inorganic materials synthesis , volume =. Nature Synthesis , author =. 2024 , note =. doi:10.1038/s44160-024-00502-y , abstract =

work page doi:10.1038/s44160-024-00502-y 2024
[56]

Nature , author =

Nanoscale synthesis and affinity ranking , volume =. Nature , author =. 2018 , note =. doi:10.1038/s41586-018-0056-8 , abstract =

work page doi:10.1038/s41586-018-0056-8 2018
[57]

Ament , author M

Autonomous materials synthesis via hierarchical active learning of nonequilibrium phase diagrams , volume =. Science Advances , author =. 2021 , note =. doi:10.1126/sciadv.abg4930 , abstract =

work page doi:10.1126/sciadv.abg4930 2021
[58]

Science Advances , author =

Real-time experiment-theory closed-loop interaction for autonomous materials science , volume =. Science Advances , author =. 2025 , note =. doi:10.1126/sciadv.adu7426 , abstract =

work page doi:10.1126/sciadv.adu7426 2025
[59]

Chem , author =

Synthesis in the. Chem , author =. 2016 , keywords =. doi:10.1016/j.chempr.2016.06.002 , abstract =

work page doi:10.1016/j.chempr.2016.06.002 2016
[60]

doi: 10.1038/s41586-021-03819-2

Highly accurate protein structure prediction with. Nature , author =. 2021 , note =. doi:10.1038/s41586-021-03819-2 , abstract =

work page doi:10.1038/s41586-021-03819-2 2021
[61]

Proteins: Structure, Function, and Bioinformatics , author =

Critical assessment of methods of protein structure prediction (. Proteins: Structure, Function, and Bioinformatics , author =. 2021 , note =. doi:10.1002/prot.26237 , abstract =

work page doi:10.1002/prot.26237 2021
[62]

Kearnes, Michael R

The. Journal of the American Chemical Society , author =. 2021 , note =. doi:10.1021/jacs.1c09820 , abstract =

work page doi:10.1021/jacs.1c09820 2021
[63]

Organic Process Research & Development , author =

Designing. Organic Process Research & Development , author =. 2023 , note =. doi:10.1021/acs.oprd.3c00186 , abstract =

work page doi:10.1021/acs.oprd.3c00186 2023
[64]

Chemical Science , author =

Chemistry informer libraries: a chemoinformatics enabled approach to evaluate and advance synthetic methods , volume =. Chemical Science , author =. 2016 , note =. doi:10.1039/C5SC04751J , abstract =

work page doi:10.1039/c5sc04751j 2016
[65]

Accounts of Chemical Research , author =

Chemistry. Accounts of Chemical Research , author =. 2021 , note =. doi:10.1021/acs.accounts.0c00760 , abstract =

work page doi:10.1021/acs.accounts.0c00760 2021
[66]

Chemical Science , author =

Predicting reaction conditions from limited data through active transfer learning , volume =. Chemical Science , author =. 2022 , note =. doi:10.1039/D1SC06932B , abstract =

work page doi:10.1039/d1sc06932b 2022
[67]

Nature Communications , author =

Rapid planning and analysis of high-throughput experiment arrays for reaction discovery , volume =. Nature Communications , author =. 2023 , note =. doi:10.1038/s41467-023-39531-0 , abstract =

work page doi:10.1038/s41467-023-39531-0 2023
[68]

The International Journal of Robotics Research , author =

Bridging the gap between safety and real-time performance in receding-horizon trajectory design for mobile robots , volume =. The International Journal of Robotics Research , author =. 2020 , note =. doi:10.1177/0278364920943266 , abstract =

work page doi:10.1177/0278364920943266 2020
[69]

2024 , keywords =

IEEE Transactions on Robotics , author =. 2024 , keywords =. doi:10.1109/TRO.2024.3366819 , abstract =

work page doi:10.1109/tro.2024.3366819 2024
[70]

IEEE Transactions on Robotics , author =

Let us. IEEE Transactions on Robotics , author =. 2025 , keywords =. doi:10.1109/TRO.2025.3584559 , abstract =

work page doi:10.1109/tro.2025.3584559 2025
[71]

IEEE Transactions on Robotics , author =

Can. IEEE Transactions on Robotics , author =. 2025 , keywords =. doi:10.1109/TRO.2025.3584557 , abstract =

work page doi:10.1109/tro.2025.3584557 2025
[72]

IEEE Robotics and Automation Letters , author =

Serving. IEEE Robotics and Automation Letters , author =. 2024 , keywords =. doi:10.1109/LRA.2024.3355731 , abstract =

work page doi:10.1109/lra.2024.3355731 2024
[73]

IEEE Robotics and Automation Letters , author =

Bring the. IEEE Robotics and Automation Letters , author =. 2025 , keywords =. doi:10.1109/LRA.2025.3547299 , abstract =

work page doi:10.1109/lra.2025.3547299 2025
[74]

arXiv preprint arXiv:2508.13355 , year=

Mu, Wenhao and Cao, Zhi and Uludag, Mehmed and Rodríguez, Alexander , month = sep, year =. Counterfactual. doi:10.48550/arXiv.2508.13355 , abstract =

work page doi:10.48550/arxiv.2508.13355
[75]

2023 , keywords =

Proceedings of the AAAI Conference on Artificial Intelligence , author =. 2023 , keywords =. doi:10.1609/aaai.v37i12.26690 , abstract =

work page doi:10.1609/aaai.v37i12.26690 2023
[76]

Computers and Geotechnics , author =

Leveraging physics-informed neural networks in geotechnical earthquake engineering:. Computers and Geotechnics , author =. 2025 , keywords =. doi:10.1016/j.compgeo.2025.107137 , abstract =

work page doi:10.1016/j.compgeo.2025.107137 2025
[77]

InPro- ceedings of the International AAAI Conference on Web and Social Media, volume 19, pages 341–371

Chopra, Ayush and Rodríguez, Alexander and Subramanian, Jayakumar and Quera-Bofarull, Arnau and Krishnamurthy, Balaji and Prakash, B. Aditya and Raskar, Ramesh , month = may, year =. Differentiable. doi:10.48550/arXiv.2207.09714 , abstract =

work page doi:10.48550/arxiv.2207.09714
[78]

Vaccine , author =

Using neural networks to calibrate agent based models enables improved regional evidence for vaccine strategy and policy , volume =. Vaccine , author =. 2023 , keywords =. doi:10.1016/j.vaccine.2023.08.060 , abstract =

work page doi:10.1016/j.vaccine.2023.08.060 2023
[79]

Li, Ruipu and Rodríguez, Alexander , month = dec, year =. Neural. doi:10.48550/arXiv.2412.18144 , abstract =

work page doi:10.48550/arxiv.2412.18144
[80]

Adaptive

Li, Ruipu and Menacho, Daniel and Rodríguez, Alexander , month = aug, year =. Adaptive. doi:10.48550/arXiv.2508.13362 , abstract =

work page doi:10.48550/arxiv.2508.13362

Showing first 80 references.