Showing 1–50 of 51 results for author: Moor, M

AgentGUI: An Interface for Observing and Steering Long-Running AI Agents

Authors: Xuan Zhao, Jiwoong Sohn, Qinyue Zheng, Michael Moor

Abstract: AI agents are increasingly adept at tackling complex, long-running tasks. With the rapid surge of autonomous capabilities, human oversight is systematically lagging behind due to limited human-centered interfacing. Aiming to address this, we introduce AgentGUI, a user-friendly, locally hosted GUI for seamlessly observing and steering AI agents amid multiple concurrent, long-running sessions. Agent… ▽ More AI agents are increasingly adept at tackling complex, long-running tasks. With the rapid surge of autonomous capabilities, human oversight is systematically lagging behind due to limited human-centered interfacing. Aiming to address this, we introduce AgentGUI, a user-friendly, locally hosted GUI for seamlessly observing and steering AI agents amid multiple concurrent, long-running sessions. AgentGUI features 1) rich agent trajectory visualizations, 2) effective manual and automated steering, and 3) integration with and coordination between open-source and frontier agent frameworks. A controlled user study demonstrates statistically significant reduction in the time it takes to identify key elements from agent traces (38% faster, p = 0.023). In a preliminary experiment, AgentGUI's automated drift prevention feature raises the task completion rate of small local agents by as high as 34pp across a 0.8B--9B model ladder (N=50 runs per model). AgentGUI is publicly available through its project website (https://agent-gui-project.github.io) and open-source repository (https://github.com/eth-medical-ai-lab/agent-gui), along with a demo video (https://youtube.com/watch?v=GSDyxN1gTF0). △ Less

Submitted 4 August, 2026; v1 submitted 28 July, 2026; originally announced July 2026.

20 Second Parity Lifetime in an InAs--Pb Tetron Device

Authors: Morteza Aghaee, Zulfi Alam, Mariusz Andrzejczuk, Andrey Antipov, Theodora Asimakidis, Mikhail Astafev, Lukas Avilovas, Ahmad Azizimanesh, Amin Barzegar, Bela Bauer, Jonathan Becker, Umesh Kumar Bhaskar, Andrea G. Boa, Srini Boddapati, Nichlaus Bohac, Jouri Bommer, Jan Borovsky, Léo Bourdet, Samuel Boutin, Srivatsa Chakravarthi, Benjamin J. Chapman, Nikolaos Chatzaras, Tzu-Chiao Chien, Jason Cho, Patrick T. Codd , et al. (140 additional authors not shown)

Abstract: A central promise of topological quantum computing is that increasing the excitation gap improves device performance significantly. Here, we experimentally validate this principle in an InAs--Pb tetron device via interferometric single-shot parity measurements. By replacing aluminum with the higher-gap superconductor lead in our superconductor-semiconductor hybrid devices, we have improved the rob… ▽ More A central promise of topological quantum computing is that increasing the excitation gap improves device performance significantly. Here, we experimentally validate this principle in an InAs--Pb tetron device via interferometric single-shot parity measurements. By replacing aluminum with the higher-gap superconductor lead in our superconductor-semiconductor hybrid devices, we have improved the robustness of our topological phase. In addition, to enable fast and precise bring-up at scale, we have developed an rf measurement technique that resolves low-energy wire-end states and directly measures their energy splitting with $μ\text{eV}$ precision. We employ this technique to bring up a device in a multi-tetron array and perform parity measurements of one of the tetron's hybrid nanowires (NWs). By controllably switching the wire parity, we observe $h/2e$-periodic bimodal shifts in the quantum capacitance of a quantum dot coupled to the hybrid nanowire in an interference loop. Further time-resolved measurements reveal a characteristic parity switching time of $\sim 20$ s with some instances reaching minute-scale. Such extremely long parity lifetimes are orders of magnitude longer than typical qubit operation times, which are on the order of $μ\text{s}$. Finally, we discuss potential implications for the fidelity of Pauli measurements. △ Less

Submitted 2 June, 2026; v1 submitted 2 June, 2026; originally announced June 2026.

Agentic-J: An AI Agent for Biological Microscopy Image Analysis

Authors: Lukas Johanns, Marilin Moor, Davide Panzeri, Yu Zhou, Xinyi Chen, Nora F. K. Pauly, Zixuan Pan, Matthias Gunzer, Andreas Müller, Yiyu Shi, Hedi Peterson, Jianxu Chen

Abstract: Biological image analysis increasingly demands integration across heterogeneous tools, programming environments, and domain knowledge that few researchers can command simultaneously. We present Agentic-J, a containerised, multi-agent AI assistant, primarily for ImageJ/Fiji that enables biologists to specify analysis tasks in natural language, from nuclei segmentation and cell tracking to multi-con… ▽ More Biological image analysis increasingly demands integration across heterogeneous tools, programming environments, and domain knowledge that few researchers can command simultaneously. We present Agentic-J, a containerised, multi-agent AI assistant, primarily for ImageJ/Fiji that enables biologists to specify analysis tasks in natural language, from nuclei segmentation and cell tracking to multi-condition quantification. The agent generates executable scripts organised into a documented project structure, so every analysis decision is traceable and the workflow can be reproduced or shared. The specialised sub-agents handle plugin management, code generation, debugging, quality assurance, and statistical reporting. In this paper we introduce the system's design, demonstrate real biological microscopy image analysis workflows, and detailed the technical implementation. △ Less

Submitted 1 June, 2026; originally announced June 2026.

Comments: Presented at Cell Biology at Scale 2026 (Poster). The Agentic-J project is available at https://mmv-lab.github.io/Agentic-J/

Universal Boosts, Specific Suppressors: Sparse Autoencoder Steering of Medical Vision-Language Models

Authors: Farhad Nooralahzadeh, Benjamin Gundersen, Nicolas Deperrois, Hidetoshi Matsuom, Mizuho Nishio, Thomas Frauenfelder, Ahmed Allam, Christian Blüthgen, Michael Moor, Michael Krauthammer

Abstract: Medical vision-language models (VLMs) often hallucinate findings when generating chest X-ray reports: they fabricate findings that are not present in the image, miss important ones, or locate them incorrectly. We mitigate this without weight updates by decoding-time residual steering on a per-token sparse autoencoder (SAE) basis: Top-$K$ SAEs on late layers, causal steering against clinical errors… ▽ More Medical vision-language models (VLMs) often hallucinate findings when generating chest X-ray reports: they fabricate findings that are not present in the image, miss important ones, or locate them incorrectly. We mitigate this without weight updates by decoding-time residual steering on a per-token sparse autoencoder (SAE) basis: Top-$K$ SAEs on late layers, causal steering against clinical errors, then combined suppress/boost intervention at inference time. On the MIMIC-CXR test split, our inference-only method improves the quality of generated reports for three radiology VLMs (RadVLM, LLaVA-Rad, and CheXOne), with relative improvements of +5.4%, +7.2%, and +17.0% in the clinical composite metric, and statistically significant GREEN gains on all backbones. A cross-model feature alignment shows that the quality-promoting (boost) directions overlap strongly across architectures, whereas hallucination-linked (suppress) directions are model-specific. Therefore, transferable steering must treat suppression per-backbone, rather than sharing a universal suppress list. The same recipe transfers zero-shot to IU-Xray (Green $+7.7\%$ rel.) without retraining, confirming that the identified features are properties of the model, not of the training corpus. We release causal feature sets and an interactive feature dashboard: https://cxr-sparse-feature-dashboard.netlify.app/. △ Less

Submitted 24 May, 2026; originally announced May 2026.

RadAgent: A tool-using AI agent for stepwise interpretation of chest computed tomography

Authors: Mélanie Roschewitz, Kenneth Styppa, Yitian Tao, Jiwoong Sohn, Jean-Benoit Delbrouck, Benjamin Gundersen, Nicolas Deperrois, Christian Bluethgen, Julia E. Vogt, Bjoern Menze, Farhad Nooralahzadeh, Michael Krauthammer, Michael Moor

Abstract: Vision-language models (VLM) have markedly advanced AI-driven interpretation and reporting of complex medical imaging, such as computed tomography (CT). Yet, existing methods largely relegate clinicians to passive observers of final outputs, offering no interpretable reasoning trace for them to inspect, validate, or refine. To address this, we introduce RadAgent, a tool-using AI agent that generat… ▽ More Vision-language models (VLM) have markedly advanced AI-driven interpretation and reporting of complex medical imaging, such as computed tomography (CT). Yet, existing methods largely relegate clinicians to passive observers of final outputs, offering no interpretable reasoning trace for them to inspect, validate, or refine. To address this, we introduce RadAgent, a tool-using AI agent that generates CT reports through a stepwise and interpretable process. Each resulting report is accompanied by a fully inspectable trace of intermediate decisions and tool interactions, allowing clinicians to examine how the reported findings are derived. In our experiments, we observe that RadAgent improves chest CT report generation over its 3D VLM counterpart, CT-Chat, across three dimensions. Clinical accuracy improves by 5.8 points (35.4% relative) in macro-F1 and 5.1 points (18.6% relative) in micro-F1. Robustness under adversarial conditions improves by 24.7 points (41.9% relative). Furthermore, RadAgent achieves 37.0% in faithfulness, a new capability entirely absent in its 3D VLM counterpart. By structuring the interpretation of chest CT as an explicit, tool-augmented and iterative reasoning trace, RadAgent brings us closer toward transparent and reliable AI for radiology. △ Less

Submitted 1 June, 2026; v1 submitted 16 April, 2026; originally announced April 2026.

Process Reward Agents for Steering Knowledge-Intensive Reasoning

Authors: Jiwoong Sohn, Tomasz Sternal, Kenneth Styppa, Torsten Hoefler, Michael Moor

Abstract: Reasoning in knowledge-intensive domains remains challenging as intermediate steps are often not locally verifiable: unlike math or code, evaluating step correctness may require synthesizing clues across large external knowledge sources. As a result, subtle errors can propagate through reasoning traces, potentially never to be detected. Prior work has proposed process reward models (PRMs), includi… ▽ More Reasoning in knowledge-intensive domains remains challenging as intermediate steps are often not locally verifiable: unlike math or code, evaluating step correctness may require synthesizing clues across large external knowledge sources. As a result, subtle errors can propagate through reasoning traces, potentially never to be detected. Prior work has proposed process reward models (PRMs), including retrieval-augmented variants, but these methods operate post hoc, scoring completed trajectories, which prevents their integration into dynamic inference procedures. Here, we introduce Process Reward Agents (PRA), an inference-time method for providing domain-grounded, online, step-wise rewards to a frozen policy. In contrast to prior retrieval-augmented PRMs, PRA enables search-based decoding to rank and prune candidate trajectories at every generation step. Experiments on multiple medical reasoning benchmarks demonstrate that PRA consistently outperforms strong baselines, achieving 81.9% accuracy on MedQA with Qwen3-4B, a new state of the art at the 4B scale. Importantly, PRA generalizes to unseen frozen policy models ranging from 0.5B to 8B parameters, improving their accuracy by up to 25.7% without any policy model updates. More broadly, PRA suggests a paradigm in which frozen reasoners are decoupled from domain-specific reward modules, allowing the deployment of new backbones in complex domains without retraining. △ Less

Submitted 1 June, 2026; v1 submitted 10 April, 2026; originally announced April 2026.

Comments: Accepted to ICML 2026

Fauna Sprout: A lightweight, approachable, developer-ready humanoid robot

Authors: Fauna Robotics, :, Diego Aldarondo, Ana Pervan, Daniel Corbalan, Dave Petrillo, Bolun Dai, Aadhithya Iyer, Nina Mortensen, Erik Pearson, Sridhar Pandian Arunachalam, Emma Reznick, David Weis, Jacob Davison, Samuel Patterson, Tess Carella, Michael Suguitan, David Ye, Oswaldo Ferro, Nilesh Suriyarachchi, Spencer Ling, Erik Su, Daniel Giebisch, Peter Traver, Sam Fonseca , et al. (26 additional authors not shown)

Abstract: Recent advances in learned control, large-scale simulation, and generative models have accelerated progress toward general-purpose robotic controllers, yet the field still lacks platforms suitable for safe, expressive, long-term deployment in human environments. Most existing humanoids are either closed industrial systems or academic prototypes that are difficult to deploy and operate around peopl… ▽ More Recent advances in learned control, large-scale simulation, and generative models have accelerated progress toward general-purpose robotic controllers, yet the field still lacks platforms suitable for safe, expressive, long-term deployment in human environments. Most existing humanoids are either closed industrial systems or academic prototypes that are difficult to deploy and operate around people, limiting progress in robotics. We introduce Sprout, a developer platform designed to address these limitations through an emphasis on safety, expressivity, and developer accessibility. Sprout adopts a lightweight form factor with compliant control, limited joint torques, and soft exteriors to support safe operation in shared human spaces. The platform integrates whole-body control, manipulation with integrated grippers, and virtual-reality-based teleoperation within a unified hardware-software stack. An expressive head further enables social interaction -- a domain that remains underexplored on most utilitarian humanoids. By lowering physical and technical barriers to deployment, Sprout expands access to capable humanoid platforms and provides a practical basis for developing embodied intelligence in real human environments. △ Less

Submitted 26 January, 2026; originally announced January 2026.

Meta-RL Induces Exploration in Language Agents

Authors: Yulun Jiang, Liangze Jiang, Damien Teney, Michael Moor, Maria Brbic

Abstract: Reinforcement learning (RL) has enabled the training of large language model (LLM) agents to interact with the environment and to solve multi-turn long-horizon tasks. However, the RL-trained agents often struggle in tasks that require active exploration and fail to efficiently adapt from trial-and-error experiences. In this paper, we present LaMer, a general Meta-RL framework that enables LLM agen… ▽ More Reinforcement learning (RL) has enabled the training of large language model (LLM) agents to interact with the environment and to solve multi-turn long-horizon tasks. However, the RL-trained agents often struggle in tasks that require active exploration and fail to efficiently adapt from trial-and-error experiences. In this paper, we present LaMer, a general Meta-RL framework that enables LLM agents to actively explore and learn from the environment feedback at test time. LaMer consists of two key components: (i) a cross-episode training framework to encourage exploration and long-term rewards optimization; and (ii) in-context policy adaptation via reflection, allowing the agent to adapt their policy from task feedback signal without gradient update. Experiments across diverse environments show that LaMer significantly improves performance over RL baselines, with 11%, 14%, and 19% performance gains on Sokoban, MineSweeper and Webshop, respectively. Moreover, LaMer also demonstrates better generalization to more challenging or previously unseen tasks compared to the RL-trained agents. Overall, our results demonstrate that Meta-RL provides a principled approach to induce exploration in language agents, enabling more robust adaptation to novel environments through learned exploration strategies. △ Less

Submitted 8 March, 2026; v1 submitted 18 December, 2025; originally announced December 2025.

Comments: ICLR 2026

Enhancing Radiology Report Generation and Visual Grounding using Reinforcement Learning

Authors: Benjamin Gundersen, Nicolas Deperrois, Samuel Ruiperez-Campillo, Thomas M. Sutter, Julia E. Vogt, Michael Moor, Farhad Nooralahzadeh, Michael Krauthammer

Abstract: Recent advances in vision-language models (VLMs) have improved Chest X-ray (CXR) interpretation in multiple aspects. However, many medical VLMs rely solely on supervised fine-tuning (SFT), which optimizes next-token prediction without evaluating answer quality. In contrast, reinforcement learning (RL) can incorporate task-specific feedback, and its combination with explicit intermediate reasoning… ▽ More Recent advances in vision-language models (VLMs) have improved Chest X-ray (CXR) interpretation in multiple aspects. However, many medical VLMs rely solely on supervised fine-tuning (SFT), which optimizes next-token prediction without evaluating answer quality. In contrast, reinforcement learning (RL) can incorporate task-specific feedback, and its combination with explicit intermediate reasoning ("thinking") has demonstrated substantial gains on verifiable math and coding tasks. To investigate the effects of RL and thinking in a CXR VLM, we perform large-scale SFT on CXR data to build an updated RadVLM based on Qwen3-VL, followed by a cold-start SFT stage that equips the model with basic thinking ability. We then apply Group Relative Policy Optimization (GRPO) with clinically grounded, task-specific rewards for report generation and visual grounding, and run matched RL experiments on both domain-specific and general-domain Qwen3-VL variants, with and without thinking. Across these settings, we find that while strong SFT remains crucial for high base performance, RL provides additional gains on both tasks, whereas explicit thinking does not appear to further improve results. Under a unified evaluation pipeline, the RL-optimized RadVLM models outperform their baseline counterparts and reach state-of-the-art performance on both report generation and grounding, highlighting clinically aligned RL as a powerful complement to SFT for medical VLMs. △ Less

Submitted 11 December, 2025; originally announced December 2025.

Comments: 10 pages main text (3 figures, 3 tables), 31 pages in total

MTBBench: A Multimodal Sequential Clinical Decision-Making Benchmark in Oncology

Authors: Kiril Vasilev, Alexandre Misrahi, Eeshaan Jain, Phil F Cheng, Petros Liakopoulos, Olivier Michielin, Michael Moor, Charlotte Bunne

Abstract: Multimodal Large Language Models (LLMs) hold promise for biomedical reasoning, but current benchmarks fail to capture the complexity of real-world clinical workflows. Existing evaluations primarily assess unimodal, decontextualized question-answering, overlooking multi-agent decision-making environments such as Molecular Tumor Boards (MTBs). MTBs bring together diverse experts in oncology, where d… ▽ More Multimodal Large Language Models (LLMs) hold promise for biomedical reasoning, but current benchmarks fail to capture the complexity of real-world clinical workflows. Existing evaluations primarily assess unimodal, decontextualized question-answering, overlooking multi-agent decision-making environments such as Molecular Tumor Boards (MTBs). MTBs bring together diverse experts in oncology, where diagnostic and prognostic tasks require integrating heterogeneous data and evolving insights over time. Current benchmarks lack this longitudinal and multimodal complexity. We introduce MTBBench, an agentic benchmark simulating MTB-style decision-making through clinically challenging, multimodal, and longitudinal oncology questions. Ground truth annotations are validated by clinicians via a co-developed app, ensuring clinical relevance. We benchmark multiple open and closed-source LLMs and show that, even at scale, they lack reliability -- frequently hallucinating, struggling with reasoning from time-resolved data, and failing to reconcile conflicting evidence or different modalities. To address these limitations, MTBBench goes beyond benchmarking by providing an agentic framework with foundation model-based tools that enhance multi-modal and longitudinal reasoning, leading to task-level performance gains of up to 9.0% and 11.2%, respectively. Overall, MTBBench offers a challenging and realistic testbed for advancing multimodal LLM reasoning, reliability, and tool-use with a focus on MTB environments in precision oncology. △ Less

Submitted 25 November, 2025; originally announced November 2025.

Comments: Accepted to NeurIPS 2025

Agentic Systems in Radiology: Design, Applications, Evaluation, and Challenges

Authors: Christian Bluethgen, Dave Van Veen, Daniel Truhn, Jakob Nikolas Kather, Michael Moor, Malgorzata Polacin, Akshay Chaudhari, Thomas Frauenfelder, Curtis P. Langlotz, Michael Krauthammer, Farhad Nooralahzadeh

Abstract: Building agents, systems that perceive and act upon their environment with a degree of autonomy, has long been a focus of AI research. This pursuit has recently become vastly more practical with the emergence of large language models (LLMs) capable of using natural language to integrate information, follow instructions, and perform forms of "reasoning" and planning across a wide range of tasks. Wi… ▽ More Building agents, systems that perceive and act upon their environment with a degree of autonomy, has long been a focus of AI research. This pursuit has recently become vastly more practical with the emergence of large language models (LLMs) capable of using natural language to integrate information, follow instructions, and perform forms of "reasoning" and planning across a wide range of tasks. With its multimodal data streams and orchestrated workflows spanning multiple systems, radiology is uniquely suited to benefit from agents that can adapt to context and automate repetitive yet complex tasks. In radiology, LLMs and their multimodal variants have already demonstrated promising performance for individual tasks such as information extraction and report summarization. However, using LLMs in isolation underutilizes their potential to support complex, multi-step workflows where decisions depend on evolving context from multiple information sources. Equipping LLMs with external tools and feedback mechanisms enables them to drive systems that exhibit a spectrum of autonomy, ranging from semi-automated workflows to more adaptive agents capable of managing complex processes. This review examines the design of such LLM-driven agentic systems, highlights key applications, discusses evaluation methods for planning and tool use, and outlines challenges such as error cascades, tool-use efficiency, and health IT integration. △ Less

Submitted 13 October, 2025; v1 submitted 10 October, 2025; originally announced October 2025.

Distinct Lifetimes for $X$ and $Z$ Loop Measurements in a Majorana Tetron Device

Authors: Morteza Aghaee, Zulfi Alam, Rikke Andersen, Mariusz Andrzejczuk, Andrey Antipov, Mikhail Astafev, Lukas Avilovas, Ahmad Azizimanesh, Eric Banek, Bela Bauer, Jonathan Becker, Umesh Kumar Bhaskar, Andrea G. Boa, Srini Boddapati, Nichlaus Bohac, Jouri D. S. Bommer, Jan Borovsky, Léo Bourdet, Samuel Boutin, Lucas Casparis, Srivatsa Chakravarthi, Hamidreza Chalabi, Benjamin J. Chapman, Nikolaos Chatzaras, Tzu-Chiao Chien , et al. (142 additional authors not shown)

Abstract: We present a hardware realization and measurements of a tetron qubit device in a superconductor-semiconductor heterostructure. The device architecture contains two parallel superconducting nanowires, which support four Majorana zero modes (MZMs) when tuned into the topological phase, and a trivial superconducting backbone. Two distinct readout interferometers are formed by connecting the supercond… ▽ More We present a hardware realization and measurements of a tetron qubit device in a superconductor-semiconductor heterostructure. The device architecture contains two parallel superconducting nanowires, which support four Majorana zero modes (MZMs) when tuned into the topological phase, and a trivial superconducting backbone. Two distinct readout interferometers are formed by connecting the superconducting structure to a series of quantum dots. We perform single-shot interferometric measurements of the fermion parity for the two loops, designed to implement Pauli-$X$ and $Z$ measurements of the tetron. Performing repeated single-shot measurements yields two widely separated time scales $τ_X = 14.5\pm 0.3 \, \mathrm{μs}$ and $τ_Z = 12.4\pm 0.4\, \mathrm{ms}$ for parity switches observed in the $X$ and $Z$ measurement loops, which we attribute to intra-wire parity switches and external quasiparticle poisoning, respectively. We estimate assignment errors of $\mathrm{err}^X_a=16\%$ and $\mathrm{err}^Z_a=0.5\%$ for $X$ and $Z$ measurement-based operations, respectively. △ Less

Submitted 4 September, 2025; v1 submitted 11 July, 2025; originally announced July 2025.

Comments: Extended discussion of switching dynamics and multiple time scales in App. A. Added App. C on alternative scenarios. Corrected Figs. 2(i) & A4. Explained the method for extracting multiple time scales from long time traces in App. A

MARBLE: A Hard Benchmark for Multimodal Spatial Reasoning and Planning

Authors: Yulun Jiang, Yekun Chai, Maria Brbić, Michael Moor

Abstract: The ability to process information from multiple modalities and to reason through it step-by-step remains a critical challenge in advancing artificial intelligence. However, existing reasoning benchmarks focus on text-only reasoning, or employ multimodal questions that can be answered by directly retrieving information from a non-text modality. Thus, complex reasoning remains poorly understood in… ▽ More The ability to process information from multiple modalities and to reason through it step-by-step remains a critical challenge in advancing artificial intelligence. However, existing reasoning benchmarks focus on text-only reasoning, or employ multimodal questions that can be answered by directly retrieving information from a non-text modality. Thus, complex reasoning remains poorly understood in multimodal domains. Here, we present MARBLE, a challenging multimodal reasoning benchmark that is designed to scrutinize multimodal language models (MLLMs) in their ability to carefully reason step-by-step through complex multimodal problems and environments. MARBLE is composed of two highly challenging tasks, M-Portal and M-Cube, that require the crafting and understanding of multistep plans under spatial, visual, and physical constraints. We find that current MLLMs perform poorly on MARBLE -- all the 12 advanced models obtain near-random performance on M-Portal and 0% accuracy on M-Cube. Only in simplified subtasks some models outperform the random baseline, indicating that complex reasoning is still a challenge for existing MLLMs. Moreover, we show that perception remains a bottleneck, where MLLMs occasionally fail to extract information from the visual inputs. By shedding a light on the limitations of MLLMs, we hope that MARBLE will spur the development of the next generation of models with the ability to reason and plan across many, multimodal reasoning steps. △ Less

Submitted 28 June, 2025; originally announced June 2025.

SMMILE: An Expert-Driven Benchmark for Multimodal Medical In-Context Learning

Authors: Melanie Rieff, Maya Varma, Ossian Rabow, Subathra Adithan, Julie Kim, Ken Chang, Hannah Lee, Nidhi Rohatgi, Christian Bluethgen, Mohamed S. Muneer, Jean-Benoit Delbrouck, Michael Moor

Abstract: Multimodal in-context learning (ICL) remains underexplored despite significant potential for domains such as medicine. Clinicians routinely encounter diverse, specialized tasks requiring adaptation from limited examples, such as drawing insights from a few relevant prior cases or considering a constrained set of differential diagnoses. While multimodal large language models (MLLMs) have shown adva… ▽ More Multimodal in-context learning (ICL) remains underexplored despite significant potential for domains such as medicine. Clinicians routinely encounter diverse, specialized tasks requiring adaptation from limited examples, such as drawing insights from a few relevant prior cases or considering a constrained set of differential diagnoses. While multimodal large language models (MLLMs) have shown advances in medical visual question answering (VQA), their ability to learn multimodal tasks from context is largely unknown. We introduce SMMILE, the first expert-driven multimodal ICL benchmark for medical tasks. Eleven medical experts curated problems, each including a multimodal query and multimodal in-context examples as task demonstrations. SMMILE encompasses 111 problems (517 question-image-answer triplets) covering 6 medical specialties and 13 imaging modalities. We further introduce SMMILE++, an augmented variant with 1038 permuted problems. A comprehensive evaluation of 15 MLLMs demonstrates that most models exhibit moderate to poor multimodal ICL ability in medical tasks. In open-ended evaluations, ICL contributes only an 8% average improvement over zero-shot on SMMILE and 9.4% on SMMILE++. We observe a susceptibility for irrelevant in-context examples: even a single noisy or irrelevant example can degrade performance by up to 9.5%. Moreover, we observe that MLLMs are affected by a recency bias, where placing the most relevant example last can lead to substantial performance improvements of up to 71%. Our findings highlight critical limitations and biases in current MLLMs when learning multimodal medical tasks from context. SMMILE is available at https://smmile-benchmark.github.io. △ Less

Submitted 5 February, 2026; v1 submitted 26 June, 2025; originally announced June 2025.

Comments: NeurIPS 2025 (Datasets & Benchmarks Track)

Med-PRM: Medical Reasoning Models with Stepwise, Guideline-verified Process Rewards

Authors: Jaehoon Yun, Jiwoong Sohn, Jungwoo Park, Hyunjae Kim, Xiangru Tang, Yanjun Shao, Yonghoe Koo, Minhyeok Ko, Qingyu Chen, Mark Gerstein, Michael Moor, Jaewoo Kang

Abstract: Large language models have shown promise in clinical decision making, but current approaches struggle to localize and correct errors at specific steps of the reasoning process. This limitation is critical in medicine, where identifying and addressing reasoning errors is essential for accurate diagnosis and effective patient care. We introduce Med-PRM, a process reward modeling framework that lever… ▽ More Large language models have shown promise in clinical decision making, but current approaches struggle to localize and correct errors at specific steps of the reasoning process. This limitation is critical in medicine, where identifying and addressing reasoning errors is essential for accurate diagnosis and effective patient care. We introduce Med-PRM, a process reward modeling framework that leverages retrieval-augmented generation to verify each reasoning step against established medical knowledge bases. By verifying intermediate reasoning steps with evidence retrieved from clinical guidelines and literature, our model can precisely assess the reasoning quality in a fine-grained manner. Evaluations on five medical QA benchmarks and two open-ended diagnostic tasks demonstrate that Med-PRM achieves state-of-the-art performance, with improving the performance of base models by up to 13.50% using Med-PRM. Moreover, we demonstrate the generality of Med-PRM by integrating it in a plug-and-play fashion with strong policy models such as Meerkat, achieving over 80\% accuracy on MedQA for the first time using small-scale models of 8 billion parameters. Our code and data are available at: https://med-prm.github.io/ △ Less

Submitted 22 September, 2025; v1 submitted 13 June, 2025; originally announced June 2025.

Comments: Accepted to EMNLP 2025 (Oral)

MIRIAD: Augmenting LLMs with millions of medical query-response pairs

Authors: Qinyue Zheng, Salman Abdullah, Sam Rawal, Cyril Zakka, Sophie Ostmeier, Maximilian Purk, Eduardo Reis, Eric J. Topol, Jure Leskovec, Michael Moor

Abstract: LLMs are bound to transform healthcare with advanced decision support and flexible chat assistants. However, LLMs are prone to generate inaccurate medical content. To ground LLMs in high-quality medical knowledge, LLMs have been equipped with external knowledge via RAG, where unstructured medical knowledge is split into small text chunks that can be selectively retrieved and integrated into the LL… ▽ More LLMs are bound to transform healthcare with advanced decision support and flexible chat assistants. However, LLMs are prone to generate inaccurate medical content. To ground LLMs in high-quality medical knowledge, LLMs have been equipped with external knowledge via RAG, where unstructured medical knowledge is split into small text chunks that can be selectively retrieved and integrated into the LLMs context. Yet, existing RAG pipelines rely on raw, unstructured medical text, which can be noisy, uncurated and difficult for LLMs to effectively leverage. Systematic approaches to organize medical knowledge to best surface it to LLMs are generally lacking. To address these challenges, we introduce MIRIAD, a large-scale, curated corpus of 5,821,948 medical QA pairs, each rephrased from and grounded in a passage from peer-reviewed medical literature using a semi-automated pipeline combining LLM generation, filtering, grounding, and human annotation. Unlike prior medical corpora, which rely on unstructured text, MIRIAD encapsulates web-scale medical knowledge in an operationalized query-response format, which enables more targeted retrieval. Experiments on challenging medical QA benchmarks show that augmenting LLMs with MIRIAD improves accuracy up to 6.7% compared to unstructured RAG baselines with the same source corpus and with the same amount of retrieved text. Moreover, MIRIAD improved the ability of LLMs to detect medical hallucinations by 22.5 to 37% (increase in F1 score). We further introduce MIRIAD-Atlas, an interactive map of MIRIAD spanning 56 medical disciplines, enabling clinical users to visually explore, search, and refine medical knowledge. MIRIAD promises to unlock a wealth of down-stream applications, including medical information retrievers, enhanced RAG applications, and knowledge-grounded chat interfaces, which ultimately enables more reliable LLM applications in healthcare. △ Less

Submitted 9 June, 2025; v1 submitted 6 June, 2025; originally announced June 2025.

Comments: Preprint

ACM Class: I.2.7

Response to recent comments on Phys. Rev. B 107, 245423 (2023) and Subsection S4.3 of the Supp. Info. for Nature 638, 651-655 (2025)

Authors: Morteza Aghaee, Zulfi Alam, Mariusz Andrzejczuk, Andrey E. Antipov, Mikhail Astafev, Amin Barzegar, Bela Bauer, Jonathan Becker, Umesh Kumar Bhaskar, Alex Bocharov, Srini Boddapati, David Bohn, Jouri Bommer, Leo Bourdet, Samuel Boutin, Benjamin J. Chapman, Sohail Chatoor, Anna Wulff Christensen, Patrick Codd, William S. Cole, Paul Cooper, Fabiano Corsetti, Ajuan Cui, Andreas Ekefjärd, Saeed Fallahi , et al. (105 additional authors not shown)

Abstract: The topological gap protocol (TGP) is a statistical test designed to identify a topological phase with high confidence and without human bias. It is used to determine a promising parameter regime for operating topological qubits. The protocol's key metric is the probability of incorrectly identifying a trivial region as topological, referred to as the false discovery rate (FDR). Two recent manuscr… ▽ More The topological gap protocol (TGP) is a statistical test designed to identify a topological phase with high confidence and without human bias. It is used to determine a promising parameter regime for operating topological qubits. The protocol's key metric is the probability of incorrectly identifying a trivial region as topological, referred to as the false discovery rate (FDR). Two recent manuscripts [arXiv:2502.19560, arXiv:2503.08944] engage with the topological gap protocol and its use in Phys. Rev. B 107, 245423 (2023) and Subsection S4.3 of the Supplementary Information for Nature 638, 651-655 (2025), although they do not explicitly dispute the main results of either one. We demonstrate that the objections in arXiv:2502.19560 and arXiv:2503.08944 are unfounded, and we uphold the conclusions of Phys. Rev. B 107, 245423 (2023) and Nature 638, 651-655 (2025). Specifically, we show that no flaws have been identified in our estimate of the false discovery rate (FDR). We provide a point-by-point rebuttal of the comments in arXiv:2502.19560 and arXiv:2503.08944. △ Less

Submitted 17 April, 2025; originally announced April 2025.

Comments: Response to arXiv:2502.19560 and arXiv:2503.08944. 11 pages, 5 figures, 2 tables, code for reproduction

AgentRxiv: Towards Collaborative Autonomous Research

Authors: Samuel Schmidgall, Michael Moor

Abstract: Progress in scientific discovery is rarely the result of a single "Eureka" moment, but is rather the product of hundreds of scientists incrementally working together toward a common goal. While existing agent workflows are capable of producing research autonomously, they do so in isolation, without the ability to continuously improve upon prior research results. To address these challenges, we int… ▽ More Progress in scientific discovery is rarely the result of a single "Eureka" moment, but is rather the product of hundreds of scientists incrementally working together toward a common goal. While existing agent workflows are capable of producing research autonomously, they do so in isolation, without the ability to continuously improve upon prior research results. To address these challenges, we introduce AgentRxiv-a framework that lets LLM agent laboratories upload and retrieve reports from a shared preprint server in order to collaborate, share insights, and iteratively build on each other's research. We task agent laboratories to develop new reasoning and prompting techniques and find that agents with access to their prior research achieve higher performance improvements compared to agents operating in isolation (11.4% relative improvement over baseline on MATH-500). We find that the best performing strategy generalizes to benchmarks in other domains (improving on average by 3.3%). Multiple agent laboratories sharing research through AgentRxiv are able to work together towards a common goal, progressing more rapidly than isolated laboratories, achieving higher overall accuracy (13.7% relative improvement over baseline on MATH-500). These findings suggest that autonomous agents may play a role in designing future AI systems alongside humans. We hope that AgentRxiv allows agents to collaborate toward research goals and enables researchers to accelerate discovery. △ Less

Submitted 23 March, 2025; originally announced March 2025.

Roadmap to fault tolerant quantum computation using topological qubit arrays

Authors: David Aasen, Morteza Aghaee, Zulfi Alam, Mariusz Andrzejczuk, Andrey Antipov, Mikhail Astafev, Lukas Avilovas, Amin Barzegar, Bela Bauer, Jonathan Becker, Juan M. Bello-Rivas, Umesh Bhaskar, Alex Bocharov, Srini Boddapati, David Bohn, Jouri Bommer, Parsa Bonderson, Jan Borovsky, Leo Bourdet, Samuel Boutin, Tom Brown, Gary Campbell, Lucas Casparis, Srivatsa Chakravarthi, Rui Chao , et al. (157 additional authors not shown)

Abstract: We describe a concrete device roadmap towards a fault-tolerant quantum computing architecture based on noise-resilient, topologically protected Majorana-based qubits. Our roadmap encompasses four generations of devices: a single-qubit device that enables a measurement-based qubit benchmarking protocol; a two-qubit device that uses measurement-based braiding to perform single-qubit Clifford operati… ▽ More We describe a concrete device roadmap towards a fault-tolerant quantum computing architecture based on noise-resilient, topologically protected Majorana-based qubits. Our roadmap encompasses four generations of devices: a single-qubit device that enables a measurement-based qubit benchmarking protocol; a two-qubit device that uses measurement-based braiding to perform single-qubit Clifford operations; an eight-qubit device that can be used to show an improvement of a two-qubit operation when performed on logical qubits rather than directly on physical qubits; and a topological qubit array supporting lattice surgery demonstrations on two logical qubits. Devices that enable this path require a superconductor-semiconductor heterostructure that supports a topological phase, quantum dots and coupling between those quantum dots that can create the appropriate loops for interferometric measurements, and a microwave readout system that can perform fast, low-error single-shot measurements. We describe the key design components of these qubit devices, along with the associated protocols for demonstrations of single-qubit benchmarking, Clifford gate execution, quantum error detection, and quantum error correction, which differ greatly from those in more conventional qubits. Finally, we comment on implications and advantages of this architecture for utility-scale quantum computation. △ Less

Submitted 18 July, 2025; v1 submitted 17 February, 2025; originally announced February 2025.

Comments: v2: 12+8 pages, 9+5 figures, significant main text and appendix revisions

Agent Laboratory: Using LLM Agents as Research Assistants

Authors: Samuel Schmidgall, Yusheng Su, Ze Wang, Ximeng Sun, Jialian Wu, Xiaodong Yu, Jiang Liu, Michael Moor, Zicheng Liu, Emad Barsoum

Abstract: Historically, scientific discovery has been a lengthy and costly process, demanding substantial time and resources from initial conception to final results. To accelerate scientific discovery, reduce research costs, and improve research quality, we introduce Agent Laboratory, an autonomous LLM-based framework capable of completing the entire research process. This framework accepts a human-provide… ▽ More Historically, scientific discovery has been a lengthy and costly process, demanding substantial time and resources from initial conception to final results. To accelerate scientific discovery, reduce research costs, and improve research quality, we introduce Agent Laboratory, an autonomous LLM-based framework capable of completing the entire research process. This framework accepts a human-provided research idea and progresses through three stages--literature review, experimentation, and report writing to produce comprehensive research outputs, including a code repository and a research report, while enabling users to provide feedback and guidance at each stage. We deploy Agent Laboratory with various state-of-the-art LLMs and invite multiple researchers to assess its quality by participating in a survey, providing human feedback to guide the research process, and then evaluate the final paper. We found that: (1) Agent Laboratory driven by o1-preview generates the best research outcomes; (2) The generated machine learning code is able to achieve state-of-the-art performance compared to existing methods; (3) Human involvement, providing feedback at each stage, significantly improves the overall quality of research; (4) Agent Laboratory significantly reduces research expenses, achieving an 84% decrease compared to previous autonomous research methods. We hope Agent Laboratory enables researchers to allocate more effort toward creative ideation rather than low-level coding and writing, ultimately accelerating scientific discovery. △ Less

Submitted 17 June, 2025; v1 submitted 7 January, 2025; originally announced January 2025.

Reverse Image Retrieval Cues Parametric Memory in Multimodal LLMs

Authors: Jialiang Xu, Michael Moor, Jure Leskovec

Abstract: Despite impressive advances in recent multimodal large language models (MLLMs), state-of-the-art models such as from the GPT-4 suite still struggle with knowledge-intensive tasks. To address this, we consider Reverse Image Retrieval (RIR) augmented generation, a simple yet effective strategy to augment MLLMs with web-scale reverse image search results. RIR robustly improves knowledge-intensive vis… ▽ More Despite impressive advances in recent multimodal large language models (MLLMs), state-of-the-art models such as from the GPT-4 suite still struggle with knowledge-intensive tasks. To address this, we consider Reverse Image Retrieval (RIR) augmented generation, a simple yet effective strategy to augment MLLMs with web-scale reverse image search results. RIR robustly improves knowledge-intensive visual question answering (VQA) of GPT-4V by 37-43%, GPT-4 Turbo by 25-27%, and GPT-4o by 18-20% in terms of open-ended VQA evaluation metrics. To our surprise, we discover that RIR helps the model to better access its own world knowledge. Concretely, our experiments suggest that RIR augmentation helps by providing further visual and textual cues without necessarily containing the direct answer to a query. In addition, we elucidate cases in which RIR can hurt performance and conduct a human evaluation. Finally, we find that the overall advantage of using RIR makes it difficult for an agent that can choose to use RIR to perform better than an approach where RIR is the default setting. △ Less

Submitted 29 May, 2024; originally announced May 2024.

AgentClinic: a multimodal agent benchmark to evaluate AI in simulated clinical environments

Authors: Samuel Schmidgall, Rojin Ziaei, Carl Harris, Eduardo Reis, Jeffrey Jopling, Michael Moor

Abstract: Evaluating large language models (LLM) in clinical scenarios is crucial to assessing their potential clinical utility. Existing benchmarks rely heavily on static question-answering, which does not accurately depict the complex, sequential nature of clinical decision-making. Here, we introduce AgentClinic, a multimodal agent benchmark for evaluating LLMs in simulated clinical environments that incl… ▽ More Evaluating large language models (LLM) in clinical scenarios is crucial to assessing their potential clinical utility. Existing benchmarks rely heavily on static question-answering, which does not accurately depict the complex, sequential nature of clinical decision-making. Here, we introduce AgentClinic, a multimodal agent benchmark for evaluating LLMs in simulated clinical environments that include patient interactions, multimodal data collection under incomplete information, and the usage of various tools, resulting in an in-depth evaluation across nine medical specialties and seven languages. We find that solving MedQA problems in the sequential decision-making format of AgentClinic is considerably more challenging, resulting in diagnostic accuracies that can drop to below a tenth of the original accuracy. Overall, we observe that agents sourced from Claude-3.5 outperform other LLM backbones in most settings. Nevertheless, we see stark differences in the LLMs' ability to make use of tools, such as experiential learning, adaptive retrieval, and reflection cycles. Strikingly, Llama-3 shows up to 92% relative improvements with the notebook tool that allows for writing and editing notes that persist across cases. To further scrutinize our clinical simulations, we leverage real-world electronic health records, perform a clinical reader study, perturb agents with biases, and explore novel patient-centric metrics that this interactive environment firstly enables. △ Less

Submitted 24 May, 2025; v1 submitted 13 May, 2024; originally announced May 2024.

Almanac Copilot: Towards Autonomous Electronic Health Record Navigation

Authors: Cyril Zakka, Joseph Cho, Gracia Fahed, Rohan Shad, Michael Moor, Robyn Fong, Dhamanpreet Kaur, Vishnu Ravi, Oliver Aalami, Roxana Daneshjou, Akshay Chaudhari, William Hiesinger

Abstract: Clinicians spend large amounts of time on clinical documentation, and inefficiencies impact quality of care and increase clinician burnout. Despite the promise of electronic medical records (EMR), the transition from paper-based records has been negatively associated with clinician wellness, in part due to poor user experience, increased burden of documentation, and alert fatigue. In this study, w… ▽ More Clinicians spend large amounts of time on clinical documentation, and inefficiencies impact quality of care and increase clinician burnout. Despite the promise of electronic medical records (EMR), the transition from paper-based records has been negatively associated with clinician wellness, in part due to poor user experience, increased burden of documentation, and alert fatigue. In this study, we present Almanac Copilot, an autonomous agent capable of assisting clinicians with EMR-specific tasks such as information retrieval and order placement. On EHR-QA, a synthetic evaluation dataset of 300 common EHR queries based on real patient data, Almanac Copilot obtains a successful task completion rate of 74% (n = 221 tasks) with a mean score of 2.45 over 3 (95% CI:2.34-2.56). By automating routine tasks and streamlining the documentation process, our findings highlight the significant potential of autonomous agents to mitigate the cognitive load imposed on clinicians by current EMR systems. △ Less

Submitted 14 May, 2024; v1 submitted 30 April, 2024; originally announced May 2024.

Interferometric Single-Shot Parity Measurement in an InAs-Al Hybrid Device

Authors: Morteza Aghaee, Alejandro Alcaraz Ramirez, Zulfi Alam, Rizwan Ali, Mariusz Andrzejczuk, Andrey Antipov, Mikhail Astafev, Amin Barzegar, Bela Bauer, Jonathan Becker, Umesh Kumar Bhaskar, Alex Bocharov, Srini Boddapati, David Bohn, Jouri Bommer, Leo Bourdet, Arnaud Bousquet, Samuel Boutin, Lucas Casparis, Benjamin James Chapman, Sohail Chatoor, Anna Wulff Christensen, Cassandra Chua, Patrick Codd, William Cole , et al. (137 additional authors not shown)

Abstract: The fusion of non-Abelian anyons or topological defects is a fundamental operation in measurement-only topological quantum computation. In topological superconductors, this operation amounts to a determination of the shared fermion parity of Majorana zero modes. As a step towards this, we implement a single-shot interferometric measurement of fermion parity in indium arsenide-aluminum heterostruct… ▽ More The fusion of non-Abelian anyons or topological defects is a fundamental operation in measurement-only topological quantum computation. In topological superconductors, this operation amounts to a determination of the shared fermion parity of Majorana zero modes. As a step towards this, we implement a single-shot interferometric measurement of fermion parity in indium arsenide-aluminum heterostructures with a gate-defined nanowire. The interferometer is formed by tunnel-coupling the proximitized nanowire to quantum dots. The nanowire causes a state-dependent shift of these quantum dots' quantum capacitance of up to 1 fF. Our quantum capacitance measurements show flux h/2e-periodic bimodality with a signal-to-noise ratio of 1 in 3.7 $μ$s at optimal flux values. From the time traces of the quantum capacitance measurements, we extract a dwell time in the two associated states that is longer than 1 ms at in-plane magnetic fields of approximately 2 T. These results are consistent with a measurement of the fermion parity encoded in a pair of Majorana zero modes that are separated by approximately 3 $μ$m and subjected to a low rate of poisoning by non-equilibrium quasiparticles. The large capacitance shift and long poisoning time enable a parity measurement error probability of 1%. △ Less

Submitted 2 April, 2024; v1 submitted 17 January, 2024; originally announced January 2024.

Comments: Added data on a second measurement of device A and a measurement of device B, expanded discussion of a trivial scenario. Refs added, author list updated

Journal ref: Nature 638, 651 (2025)

Style-Aware Radiology Report Generation with RadGraph and Few-Shot Prompting

Authors: Benjamin Yan, Ruochen Liu, David E. Kuo, Subathra Adithan, Eduardo Pontes Reis, Stephen Kwak, Vasantha Kumar Venugopal, Chloe P. O'Connell, Agustina Saenz, Pranav Rajpurkar, Michael Moor

Abstract: Automatically generated reports from medical images promise to improve the workflow of radiologists. Existing methods consider an image-to-report modeling task by directly generating a fully-fledged report from an image. However, this conflates the content of the report (e.g., findings and their attributes) with its style (e.g., format and choice of words), which can lead to clinically inaccurate… ▽ More Automatically generated reports from medical images promise to improve the workflow of radiologists. Existing methods consider an image-to-report modeling task by directly generating a fully-fledged report from an image. However, this conflates the content of the report (e.g., findings and their attributes) with its style (e.g., format and choice of words), which can lead to clinically inaccurate reports. To address this, we propose a two-step approach for radiology report generation. First, we extract the content from an image; then, we verbalize the extracted content into a report that matches the style of a specific radiologist. For this, we leverage RadGraph -- a graph representation of reports -- together with large language models (LLMs). In our quantitative evaluations, we find that our approach leads to beneficial performance. Our human evaluation with clinical raters highlights that the AI-generated reports are indistinguishably tailored to the style of individual radiologist despite leveraging only a few examples as context. △ Less

Submitted 31 October, 2023; v1 submitted 26 October, 2023; originally announced October 2023.

Comments: Accepted to Findings of EMNLP 2023

Med-Flamingo: a Multimodal Medical Few-shot Learner

Authors: Michael Moor, Qian Huang, Shirley Wu, Michihiro Yasunaga, Cyril Zakka, Yash Dalmia, Eduardo Pontes Reis, Pranav Rajpurkar, Jure Leskovec

Abstract: Medicine, by its nature, is a multifaceted domain that requires the synthesis of information across various modalities. Medical generative vision-language models (VLMs) make a first step in this direction and promise many exciting clinical applications. However, existing models typically have to be fine-tuned on sizeable down-stream datasets, which poses a significant limitation as in many medical… ▽ More Medicine, by its nature, is a multifaceted domain that requires the synthesis of information across various modalities. Medical generative vision-language models (VLMs) make a first step in this direction and promise many exciting clinical applications. However, existing models typically have to be fine-tuned on sizeable down-stream datasets, which poses a significant limitation as in many medical applications data is scarce, necessitating models that are capable of learning from few examples in real-time. Here we propose Med-Flamingo, a multimodal few-shot learner adapted to the medical domain. Based on OpenFlamingo-9B, we continue pre-training on paired and interleaved medical image-text data from publications and textbooks. Med-Flamingo unlocks few-shot generative medical visual question answering (VQA) abilities, which we evaluate on several datasets including a novel challenging open-ended VQA dataset of visual USMLE-style problems. Furthermore, we conduct the first human evaluation for generative medical VQA where physicians review the problems and blinded generations in an interactive app. Med-Flamingo improves performance in generative medical VQA by up to 20\% in clinician's rating and firstly enables multimodal medical few-shot adaptations, such as rationale generation. We release our model, code, and evaluation app under https://github.com/snap-stanford/med-flamingo. △ Less

Submitted 27 July, 2023; originally announced July 2023.

Comments: Preprint

Almanac: Retrieval-Augmented Language Models for Clinical Medicine

Authors: Cyril Zakka, Akash Chaurasia, Rohan Shad, Alex R. Dalal, Jennifer L. Kim, Michael Moor, Kevin Alexander, Euan Ashley, Jack Boyd, Kathleen Boyd, Karen Hirsch, Curt Langlotz, Joanna Nelson, William Hiesinger

Abstract: Large-language models have recently demonstrated impressive zero-shot capabilities in a variety of natural language tasks such as summarization, dialogue generation, and question-answering. Despite many promising applications in clinical medicine, adoption of these models in real-world settings has been largely limited by their tendency to generate incorrect and sometimes even toxic statements. In… ▽ More Large-language models have recently demonstrated impressive zero-shot capabilities in a variety of natural language tasks such as summarization, dialogue generation, and question-answering. Despite many promising applications in clinical medicine, adoption of these models in real-world settings has been largely limited by their tendency to generate incorrect and sometimes even toxic statements. In this study, we develop Almanac, a large language model framework augmented with retrieval capabilities for medical guideline and treatment recommendations. Performance on a novel dataset of clinical scenarios (n = 130) evaluated by a panel of 5 board-certified and resident physicians demonstrates significant increases in factuality (mean of 18% at p-value < 0.05) across all specialties, with improvements in completeness and safety. Our results demonstrate the potential for large language models to be effective tools in the clinical decision-making process, while also emphasizing the importance of careful testing and deployment to mitigate their shortcomings. △ Less

Submitted 31 May, 2023; v1 submitted 28 February, 2023; originally announced March 2023.

Zero-shot causal learning

Authors: Hamed Nilforoshan, Michael Moor, Yusuf Roohani, Yining Chen, Anja Šurina, Michihiro Yasunaga, Sara Oblak, Jure Leskovec

Abstract: Predicting how different interventions will causally affect a specific individual is important in a variety of domains such as personalized medicine, public policy, and online marketing. There are a large number of methods to predict the effect of an existing intervention based on historical data from individuals who received it. However, in many settings it is important to predict the effects of… ▽ More Predicting how different interventions will causally affect a specific individual is important in a variety of domains such as personalized medicine, public policy, and online marketing. There are a large number of methods to predict the effect of an existing intervention based on historical data from individuals who received it. However, in many settings it is important to predict the effects of novel interventions (e.g., a newly invented drug), which these methods do not address. Here, we consider zero-shot causal learning: predicting the personalized effects of a novel intervention. We propose CaML, a causal meta-learning framework which formulates the personalized prediction of each intervention's effect as a task. CaML trains a single meta-model across thousands of tasks, each constructed by sampling an intervention, its recipients, and its nonrecipients. By leveraging both intervention information (e.g., a drug's attributes) and individual features~(e.g., a patient's history), CaML is able to predict the personalized effects of novel interventions that do not exist at the time of training. Experimental results on real world datasets in large-scale medical claims and cell-line perturbations demonstrate the effectiveness of our approach. Most strikingly, \method's zero-shot predictions outperform even strong baselines trained directly on data from the test interventions. △ Less

Submitted 22 February, 2024; v1 submitted 28 January, 2023; originally announced January 2023.

InAs-Al Hybrid Devices Passing the Topological Gap Protocol

Authors: Morteza Aghaee, Arun Akkala, Zulfi Alam, Rizwan Ali, Alejandro Alcaraz Ramirez, Mariusz Andrzejczuk, Andrey E Antipov, Pavel Aseev, Mikhail Astafev, Bela Bauer, Jonathan Becker, Srini Boddapati, Frenk Boekhout, Jouri Bommer, Esben Bork Hansen, Tom Bosma, Leo Bourdet, Samuel Boutin, Philippe Caroff, Lucas Casparis, Maja Cassidy, Anna Wulf Christensen, Noah Clay, William S Cole, Fabiano Corsetti , et al. (102 additional authors not shown)

Abstract: We present measurements and simulations of semiconductor-superconductor heterostructure devices that are consistent with the observation of topological superconductivity and Majorana zero modes. The devices are fabricated from high-mobility two-dimensional electron gases in which quasi-one-dimensional wires are defined by electrostatic gates. These devices enable measurements of local and non-loca… ▽ More We present measurements and simulations of semiconductor-superconductor heterostructure devices that are consistent with the observation of topological superconductivity and Majorana zero modes. The devices are fabricated from high-mobility two-dimensional electron gases in which quasi-one-dimensional wires are defined by electrostatic gates. These devices enable measurements of local and non-local transport properties and have been optimized via extensive simulations to ensure robustness against non-uniformity and disorder. Our main result is that several devices, fabricated according to the design's engineering specifications, have passed the topological gap protocol defined in Pikulin et al. [arXiv:2103.12217]. This protocol is a stringent test composed of a sequence of three-terminal local and non-local transport measurements performed while varying the magnetic field, semiconductor electron density, and junction transparencies. Passing the protocol indicates a high probability of detection of a topological phase hosting Majorana zero modes as determined by large-scale disorder simulations. Our experimental results are consistent with a quantum phase transition into a topological superconducting phase that extends over several hundred millitesla in magnetic field and several millivolts in gate voltage, corresponding to approximately one hundred micro-electron-volts in Zeeman energy and chemical potential in the semiconducting wire. These regions feature a closing and re-opening of the bulk gap, with simultaneous zero-bias conductance peaks at both ends of the devices that withstand changes in the junction transparencies. The extracted maximum topological gaps in our devices are 20-60 $μ$eV. This demonstration is a prerequisite for experiments involving fusion and braiding of Majorana zero modes. △ Less

Submitted 8 March, 2024; v1 submitted 6 July, 2022; originally announced July 2022.

Comments: Final version

Journal ref: Phys. Rev. B 107, 245423 (2023)

Predicting sepsis in multi-site, multi-national intensive care cohorts using deep learning

Authors: Michael Moor, Nicolas Bennet, Drago Plecko, Max Horn, Bastian Rieck, Nicolai Meinshausen, Peter Bühlmann, Karsten Borgwardt

Abstract: Despite decades of clinical research, sepsis remains a global public health crisis with high mortality, and morbidity. Currently, when sepsis is detected and the underlying pathogen is identified, organ damage may have already progressed to irreversible stages. Effective sepsis management is therefore highly time-sensitive. By systematically analysing trends in the plethora of clinical data availa… ▽ More Despite decades of clinical research, sepsis remains a global public health crisis with high mortality, and morbidity. Currently, when sepsis is detected and the underlying pathogen is identified, organ damage may have already progressed to irreversible stages. Effective sepsis management is therefore highly time-sensitive. By systematically analysing trends in the plethora of clinical data available in the intensive care unit (ICU), an early prediction of sepsis could lead to earlier pathogen identification, resistance testing, and effective antibiotic and supportive treatment, and thereby become a life-saving measure. Here, we developed and validated a machine learning (ML) system for the prediction of sepsis in the ICU. Our analysis represents the largest multi-national, multi-centre in-ICU study for sepsis prediction using ML to date. Our dataset contains $156,309$ unique ICU admissions, which represent a refined and harmonised subset of five large ICU databases originating from three countries. Using the international consensus definition Sepsis-3, we derived hourly-resolved sepsis label annotations, amounting to $26,734$ ($17.1\%$) septic stays. We compared our approach, a deep self-attention model, to several clinical baselines as well as ML baselines and performed an extensive internal and external validation within and across databases. On average, our model was able to predict sepsis with an AUROC of $0.847 \pm 0.050$ (internal out-of sample validation) and $0.761 \pm 0.052$ (external validation). For a harmonised prevalence of $17\%$, at $80\%$ recall our model detects septic patients with $39\%$ precision 3.7 hours in advance. △ Less

Submitted 12 July, 2021; originally announced July 2021.

In-plane selective area InSb-Al nanowire quantum networks

Authors: Roy L. M. Op het Veld, Di Xu, Vanessa Schaller, Marcel A. Verheijen, Stan M. E. Peters, Jason Jung, Chuyao Tong, Qingzhen Wang, Michiel W. A. de Moor, Bart Hesselmann, Kiefer Vermeulen, Jouri D. S. Bommer, Joon Sue Lee, Andrey Sarikov, Mihir Pendharkar, Anna Marzegalli, Sebastian Koelling, Leo P. Kouwenhoven, Leo Miglio, Chris J. Palmstrøm, Hao Zhang, Erik P. A. M. Bakkers

Abstract: Strong spin-orbit semiconductor nanowires coupled to a superconductor are predicted to host Majorana zero modes. Exchange (braiding) operations of Majorana modes form the logical gates of a topological quantum computer and require a network of nanowires. Here, we develop an in-plane selective-area growth technique for InSb-Al semiconductor-superconductor nanowire networks with excellent quantum tr… ▽ More Strong spin-orbit semiconductor nanowires coupled to a superconductor are predicted to host Majorana zero modes. Exchange (braiding) operations of Majorana modes form the logical gates of a topological quantum computer and require a network of nanowires. Here, we develop an in-plane selective-area growth technique for InSb-Al semiconductor-superconductor nanowire networks with excellent quantum transport properties. Defect-free transport channels in InSb nanowire networks are realized on insulating, but heavily mismatched InP substrates by 1) full relaxation of the lattice mismatch at the nanowire/substrate interface on a (111)B substrate orientation, 2) nucleation of a complete network from a single nucleation site, which is accomplished by optimizing the surface diffusion length of the adatoms. Essential quantum transport phenomena for topological quantum computing are demonstrated in these structures including phase-coherent transport up to 10 $μ$m and a hard superconducting gap accompanied by 2$e$-periodic Coulomb oscillations with an Al-based Cooper pair island integrated in the nanowire network. △ Less

Submitted 11 March, 2021; originally announced March 2021.

Comments: Data repository is available at https://doi.org/10.5281/zenodo.4589484 . Author version of the text before peer review, while see DOI for the published version

Journal ref: Commun. Phys. 3, 59 (2020)

Topological Graph Neural Networks

Authors: Max Horn, Edward De Brouwer, Michael Moor, Yves Moreau, Bastian Rieck, Karsten Borgwardt

Abstract: Graph neural networks (GNNs) are a powerful architecture for tackling graph learning tasks, yet have been shown to be oblivious to eminent substructures such as cycles. We present TOGL, a novel layer that incorporates global topological information of a graph using persistent homology. TOGL can be easily integrated into any type of GNN and is strictly more expressive (in terms the Weisfeiler--Lehm… ▽ More Graph neural networks (GNNs) are a powerful architecture for tackling graph learning tasks, yet have been shown to be oblivious to eminent substructures such as cycles. We present TOGL, a novel layer that incorporates global topological information of a graph using persistent homology. TOGL can be easily integrated into any type of GNN and is strictly more expressive (in terms the Weisfeiler--Lehman graph isomorphism test) than message-passing GNNs. Augmenting GNNs with TOGL leads to improved predictive performance for graph and node classification tasks, both on synthetic data sets, which can be classified by humans using their topology but not by ordinary GNNs, and on real-world data. △ Less

Submitted 17 March, 2022; v1 submitted 15 February, 2021; originally announced February 2021.

Journal ref: Tenth International Conference on Learning Representations (ICLR), 2022

Large zero-bias peaks in InSb-Al hybrid semiconductor-superconductor nanowire devices

Authors: Hao Zhang, Michiel W. A. de Moor, Jouri D. S. Bommer, Di Xu, Guanzhong Wang, Nick van Loo, Chun-Xiao Liu, Sasa Gazibegovic, John A. Logan, Diana Car, Roy L. M. Op het Veld, Petrus J. van Veldhoven, Sebastian Koelling, Marcel A. Verheijen, Mihir Pendharkar, Daniel J. Pennachio, Borzoyeh Shojaei, Joon Sue Lee, Chris J. Palmstrøm, Erik P. A. M. Bakkers, S. Das Sarma, Leo P. Kouwenhoven

Abstract: We report electron transport studies on InSb-Al hybrid semiconductor-superconductor nanowire devices. Tunnelling spectroscopy is used to measure the evolution of subgap states while varying magnetic field and voltages applied to various nearby gates. At magnetic fields between 0.7 and 0.9 T, the differential conductance contains large zero bias peaks (ZBPs) whose height reaches values on the order… ▽ More We report electron transport studies on InSb-Al hybrid semiconductor-superconductor nanowire devices. Tunnelling spectroscopy is used to measure the evolution of subgap states while varying magnetic field and voltages applied to various nearby gates. At magnetic fields between 0.7 and 0.9 T, the differential conductance contains large zero bias peaks (ZBPs) whose height reaches values on the order 2e2/h. We investigate these ZBPs for large ranges of gate voltages in different devices. We discuss possible interpretations in terms of disorder-induced subgap states, Andreev bound states and Majorana zero modes. △ Less

Submitted 27 January, 2021; originally announced January 2021.

Comments: This manuscript replaces "Quantized Majorana conductance" Nature 556, 74 (2018). Technical errors in Nature 556, 74 (2018) are corrected and the original claims now have a wider interpretation. A Retraction Note (in preparation) on Nature 556, 74 (2018) will include a detailed description of errors and the corrected data analyses

Learning Individualized Treatment Rules with Estimated Translated Inverse Propensity Score

Authors: Zhiliang Wu, Yinchong Yang, Yunpu Ma, Yushan Liu, Rui Zhao, Michael Moor, Volker Tresp

Abstract: Randomized controlled trials typically analyze the effectiveness of treatments with the goal of making treatment recommendations for patient subgroups. With the advance of electronic health records, a great variety of data has been collected in clinical practice, enabling the evaluation of treatments and treatment policies based on observational data. In this paper, we focus on learning individual… ▽ More Randomized controlled trials typically analyze the effectiveness of treatments with the goal of making treatment recommendations for patient subgroups. With the advance of electronic health records, a great variety of data has been collected in clinical practice, enabling the evaluation of treatments and treatment policies based on observational data. In this paper, we focus on learning individualized treatment rules (ITRs) to derive a treatment policy that is expected to generate a better outcome for an individual patient. In our framework, we cast ITRs learning as a contextual bandit problem and minimize the expected risk of the treatment policy. We conduct experiments with the proposed framework both in a simulation study and based on a real-world dataset. In the latter case, we apply our proposed method to learn the optimal ITRs for the administration of intravenous (IV) fluids and vasopressors (VP). Based on various offline evaluation methods, we could show that the policy derived in our framework demonstrates better performance compared to both the physicians and other baselines, including a simple treatment prediction approach. As a long-term goal, our derived policy might eventually lead to better clinical guidelines for the administration of IV and VP. △ Less

Submitted 2 July, 2020; originally announced July 2020.

Journal ref: 2020 IEEE International Conference on Healthcare Informatics (ICHI)

Path Imputation Strategies for Signature Models of Irregular Time Series

Authors: Michael Moor, Max Horn, Christian Bock, Karsten Borgwardt, Bastian Rieck

Abstract: The signature transform is a 'universal nonlinearity' on the space of continuous vector-valued paths, and has received attention for use in machine learning on time series. However, real-world temporal data is typically observed at discrete points in time, and must first be transformed into a continuous path before signature techniques can be applied. We make this step explicit by characterising i… ▽ More The signature transform is a 'universal nonlinearity' on the space of continuous vector-valued paths, and has received attention for use in machine learning on time series. However, real-world temporal data is typically observed at discrete points in time, and must first be transformed into a continuous path before signature techniques can be applied. We make this step explicit by characterising it as an imputation problem, and empirically assess the impact of various imputation strategies when applying signature-based neural nets to irregular time series data. For one of these strategies, Gaussian process (GP) adapters, we propose an extension~(GP-PoM) that makes uncertainty information directly available to the subsequent classifier while at the same time preventing costly Monte-Carlo (MC) sampling. In our experiments, we find that the choice of imputation drastically affects shallow signature models, whereas deeper architectures are more robust. Next, we observe that uncertainty-aware predictions (based on GP-PoM or indicator imputations) are beneficial for predictive performance, even compared to the uncertainty-aware training of conventional GP adapters. In conclusion, we have demonstrated that the path construction is indeed crucial for signature models and that our proposed strategy leads to competitive performance in general, while improving robustness of signature models in particular. △ Less

Submitted 6 June, 2020; v1 submitted 25 May, 2020; originally announced May 2020.

From Andreev to Majorana bound states in hybrid superconductor-semiconductor nanowires

Authors: Elsa Prada, Pablo San-Jose, Michiel W. A. de Moor, Attila Geresdi, Eduardo J. H. Lee, Jelena Klinovaja, Daniel Loss, Jesper Nygård, Ramón Aguado, Leo P. Kouwenhoven

Abstract: Electronic excitations above the ground state must overcome an energy gap in superconductors with spatially-homogeneous s-wave pairing. In contrast, inhomogeneous superconductors such as those with magnetic impurities or weak links, or heterojunctions containing normal metals or quantum dots, can host subgap electronic excitations that are generically known as Andreev bound states (ABSs). With the… ▽ More Electronic excitations above the ground state must overcome an energy gap in superconductors with spatially-homogeneous s-wave pairing. In contrast, inhomogeneous superconductors such as those with magnetic impurities or weak links, or heterojunctions containing normal metals or quantum dots, can host subgap electronic excitations that are generically known as Andreev bound states (ABSs). With the advent of topological superconductivity, a new kind of ABS with exotic qualities, known as Majorana bound state (MBS), has been discovered. We review the main properties of ABSs and MBSs, and the state-of-the-art techniques for their detection. We focus on hybrid superconductor-semiconductor nanowires, possibly coupled to quantum dots, as one of the most flexible and promising experimental platforms. We discuss how the combined effect of spin-orbit coupling and Zeeman field in these wires triggers the transition from ABSs into MBSs. We show theoretical progress beyond minimal models in understanding experiments, including the possibility of different types of robust zero modes that may emerge without a band-topological transition. We examine the role of spatial non-locality, a special property of MBS wavefunctions that, together with non-Abelian braiding, is the key to realizing topological quantum computation. △ Less

Submitted 4 October, 2020; v1 submitted 11 November, 2019; originally announced November 2019.

Comments: Review. 23 pages, 8 figures, 1 table. Shareable published version by Springer Nature at https://rdcu.be/b7DWT (free to read but not to download)

Journal ref: Nat. Rev. Phys. 2, 575 (2020)

Set Functions for Time Series

Authors: Max Horn, Michael Moor, Christian Bock, Bastian Rieck, Karsten Borgwardt

Abstract: Despite the eminent successes of deep neural networks, many architectures are often hard to transfer to irregularly-sampled and asynchronous time series that commonly occur in real-world datasets, especially in healthcare applications. This paper proposes a novel approach for classifying irregularly-sampled time series with unaligned measurements, focusing on high scalability and data efficiency.… ▽ More Despite the eminent successes of deep neural networks, many architectures are often hard to transfer to irregularly-sampled and asynchronous time series that commonly occur in real-world datasets, especially in healthcare applications. This paper proposes a novel approach for classifying irregularly-sampled time series with unaligned measurements, focusing on high scalability and data efficiency. Our method SeFT (Set Functions for Time Series) is based on recent advances in differentiable set function learning, extremely parallelizable with a beneficial memory footprint, thus scaling well to large datasets of long time series and online monitoring scenarios. Furthermore, our approach permits quantifying per-observation contributions to the classification outcome. We extensively compare our method with existing algorithms on multiple healthcare time series datasets and demonstrate that it performs competitively whilst significantly reducing runtime. △ Less

Submitted 14 September, 2020; v1 submitted 26 September, 2019; originally announced September 2019.

Comments: Accepted at the International Conference on Machine Learning (ICML) 2020

Forecasting U.S. Textile Comparative Advantage Using Autoregressive Integrated Moving Average Models and Time Series Outlier Analysis

Authors: Zahra Saki, Lori Rothenberg, Marguerite Moor, Ivan Kandilov, A. Blanton Godfrey

Abstract: To establish an updated understanding of the U.S. textile and apparel (TAP) industrys competitive position within the global textile environment, trade data from UN-COMTRADE (1996-2016) was used to calculate the Normalized Revealed Comparative Advantage (NRCA) index for 169 TAP categories at the four-digit Harmonized Schedule (HS) code level. Univariate time series using Autoregressive Integrated… ▽ More To establish an updated understanding of the U.S. textile and apparel (TAP) industrys competitive position within the global textile environment, trade data from UN-COMTRADE (1996-2016) was used to calculate the Normalized Revealed Comparative Advantage (NRCA) index for 169 TAP categories at the four-digit Harmonized Schedule (HS) code level. Univariate time series using Autoregressive Integrated Moving Average (ARIMA) models forecast short-term future performance of Revealed categories with export advantage. Accompanying outlier analysis examined permanent level shifts that might convey important information about policy changes, influential drivers and random events. △ Less

Submitted 13 August, 2019; originally announced August 2019.

Comments: 11 pages, 1Figure and 9 tables

Journal ref: 2018 Joint Statistical Meeting, 1996-2006

Topological Autoencoders

Authors: Michael Moor, Max Horn, Bastian Rieck, Karsten Borgwardt

Abstract: We propose a novel approach for preserving topological structures of the input space in latent representations of autoencoders. Using persistent homology, a technique from topological data analysis, we calculate topological signatures of both the input and latent space to derive a topological loss term. Under weak theoretical assumptions, we construct this loss in a differentiable manner, such tha… ▽ More We propose a novel approach for preserving topological structures of the input space in latent representations of autoencoders. Using persistent homology, a technique from topological data analysis, we calculate topological signatures of both the input and latent space to derive a topological loss term. Under weak theoretical assumptions, we construct this loss in a differentiable manner, such that the encoding learns to retain multi-scale connectivity information. We show that our approach is theoretically well-founded and that it exhibits favourable latent representations on a synthetic manifold as well as on real-world image data sets, while preserving low reconstruction errors. △ Less

Submitted 31 May, 2021; v1 submitted 3 June, 2019; originally announced June 2019.

Comments: Accepted at the International Conference on Machine Learning (ICML) 2020; camera-ready version

Disc Tearing and Bardeen-Petterson Alignment in GRMHD Simulations of Highly Tilted Thin Accretion Discs

Authors: M. Liska, C. Hesp, A. Tchekhovskoy, A. Ingram, M. van der Klis, S. B. Markoff, M. Van Moer

Abstract: Luminous active galactic nuclei (AGN) and X-Ray binaries (XRBs) often contain geometrically thin, radiatively cooled accretion discs. According to theory, these are -- in many cases -- initially highly misaligned with the black hole equator. In this work, we present the first general relativistic magnetohydrodynamic simulations of very thin (h/r~0.015-0.05) accretion discs around rapidly spinning… ▽ More Luminous active galactic nuclei (AGN) and X-Ray binaries (XRBs) often contain geometrically thin, radiatively cooled accretion discs. According to theory, these are -- in many cases -- initially highly misaligned with the black hole equator. In this work, we present the first general relativistic magnetohydrodynamic simulations of very thin (h/r~0.015-0.05) accretion discs around rapidly spinning (a~0.9) black holes and tilted by 45-65 degrees. We show that the inner regions of the discs with h/r<0.03 align with the black hole equator, though out to smaller radii than predicted by analytic work. The inner aligned and outer misaligned disc regions are separated by a sharp break in tilt angle accompanied by a sharp drop in density. We find that frame-dragging by the spinning black hole overpowers the disc viscosity, which is self-consistently produced by magnetized turbulence, tearing the disc apart and forming a rapidly precessing inner sub-disc surrounded by a slowly precessing outer sub-disc. We find that the system produces a pair of relativistic jets for all initial tilt values. At small distances the black hole launched jets precess rapidly together with the inner sub-disc, whereas at large distances they partially align with the outer sub-disc and precess more slowly. If the tearing radius can be modeled accurately in future work, emission model independent measurements of black hole spin based on precession-driven quasi-periodic oscillations may become possible. △ Less

Submitted 10 August, 2022; v1 submitted 17 April, 2019; originally announced April 2019.

Comments: 8 pages, 5 figures, accompanying animations included in YouTube playlist: https://www.youtube.com/playlist?list=PLDO1oeU33GwlaPSME1TdCto1Y3P6yG91L

Machine learning for early prediction of circulatory failure in the intensive care unit

Authors: Stephanie L. Hyland, Martin Faltys, Matthias Hüser, Xinrui Lyu, Thomas Gumbsch, Cristóbal Esteban, Christian Bock, Max Horn, Michael Moor, Bastian Rieck, Marc Zimmermann, Dean Bodenham, Karsten Borgwardt, Gunnar Rätsch, Tobias M. Merz

Abstract: Intensive care clinicians are presented with large quantities of patient information and measurements from a multitude of monitoring systems. The limited ability of humans to process such complex information hinders physicians to readily recognize and act on early signs of patient deterioration. We used machine learning to develop an early warning system for circulatory failure based on a high-res… ▽ More Intensive care clinicians are presented with large quantities of patient information and measurements from a multitude of monitoring systems. The limited ability of humans to process such complex information hinders physicians to readily recognize and act on early signs of patient deterioration. We used machine learning to develop an early warning system for circulatory failure based on a high-resolution ICU database with 240 patient years of data. This automatic system predicts 90.0% of circulatory failure events (prevalence 3.1%), with 81.8% identified more than two hours in advance, resulting in an area under the receiver operating characteristic curve of 94.0% and area under the precision-recall curve of 63.0%. The model was externally validated in a large independent patient cohort. △ Less

Submitted 19 April, 2019; v1 submitted 16 April, 2019; originally announced April 2019.

Comments: 5 main figures, 1 main table, 13 supplementary figures, 5 supplementary tables; 250ppi images

Early Recognition of Sepsis with Gaussian Process Temporal Convolutional Networks and Dynamic Time Warping

Authors: Michael Moor, Max Horn, Bastian Rieck, Damian Roqueiro, Karsten Borgwardt

Abstract: Sepsis is a life-threatening host response to infection associated with high mortality, morbidity, and health costs. Its management is highly time-sensitive since each hour of delayed treatment increases mortality due to irreversible organ damage. Meanwhile, despite decades of clinical research, robust biomarkers for sepsis are missing. Therefore, detecting sepsis early by utilizing the affluence… ▽ More Sepsis is a life-threatening host response to infection associated with high mortality, morbidity, and health costs. Its management is highly time-sensitive since each hour of delayed treatment increases mortality due to irreversible organ damage. Meanwhile, despite decades of clinical research, robust biomarkers for sepsis are missing. Therefore, detecting sepsis early by utilizing the affluence of high-resolution intensive care records has become a challenging machine learning problem. Recent advances in deep learning and data mining promise to deliver a powerful set of tools to efficiently address this task. This empirical study proposes two novel approaches for the early detection of sepsis: a deep learning model and a lazy learner based on time series distances. Our deep learning model employs a temporal convolutional network that is embedded in a Multi-task Gaussian Process Adapter framework, making it directly applicable to irregularly-spaced time series data. Our lazy learner, by contrast, is an ensemble approach that employs dynamic time warping. We frame the timely detection of sepsis as a supervised time series classification task. For this, we derive the most recent sepsis definition in an hourly resolution to provide the first fully accessible early sepsis detection environment. Seven hours before sepsis onset, our methods improve area under the precision--recall curve from 0.25 to 0.35/0.40 over the state of the art. This demonstrates that they are well-suited for detecting sepsis in the crucial earlier stages when management is most effective. △ Less

Submitted 15 October, 2020; v1 submitted 5 February, 2019; originally announced February 2019.

Comments: Accepted at the Machine Learning for Healthcare 2019 Conference (MLHC). Camera-ready version

Neural Persistence: A Complexity Measure for Deep Neural Networks Using Algebraic Topology

Authors: Bastian Rieck, Matteo Togninalli, Christian Bock, Michael Moor, Max Horn, Thomas Gumbsch, Karsten Borgwardt

Abstract: While many approaches to make neural networks more fathomable have been proposed, they are restricted to interrogating the network with input data. Measures for characterizing and monitoring structural properties, however, have not been developed. In this work, we propose neural persistence, a complexity measure for neural network architectures based on topological data analysis on weighted strati… ▽ More While many approaches to make neural networks more fathomable have been proposed, they are restricted to interrogating the network with input data. Measures for characterizing and monitoring structural properties, however, have not been developed. In this work, we propose neural persistence, a complexity measure for neural network architectures based on topological data analysis on weighted stratified graphs. To demonstrate the usefulness of our approach, we show that neural persistence reflects best practices developed in the deep learning community such as dropout and batch normalization. Moreover, we derive a neural persistence-based stopping criterion that shortens the training process while achieving comparable accuracies as early stopping based on validation loss. △ Less

Submitted 27 September, 2019; v1 submitted 23 December, 2018; originally announced December 2018.

Comments: Published as a conference paper at ICLR 2019

Electric field tunable superconductor-semiconductor coupling in Majorana nanowires

Authors: Michiel W. A. de Moor, Jouri D. S. Bommer, Di Xu, Georg W. Winkler, Andrey E. Antipov, Arno Bargerbos, Guanzhong Wang, Nick van Loo, Roy L. M. Op het Veld, Sasa Gazibegovic, Diana Car, John A. Logan, Mihir Pendharkar, Joon Sue Lee, Erik P. A. M. Bakkers, Chris J. Palmstrøm, Roman M. Lutchyn, Leo P. Kouwenhoven, Hao Zhang

Abstract: We study the effect of external electric fields on superconductor-semiconductor coupling by measuring the electron transport in InSb semiconductor nanowires coupled to an epitaxially grown Al superconductor. We find that the gate voltage induced electric fields can greatly modify the coupling strength, which has consequences for the proximity induced superconducting gap, effective g-factor, and sp… ▽ More We study the effect of external electric fields on superconductor-semiconductor coupling by measuring the electron transport in InSb semiconductor nanowires coupled to an epitaxially grown Al superconductor. We find that the gate voltage induced electric fields can greatly modify the coupling strength, which has consequences for the proximity induced superconducting gap, effective g-factor, and spin-orbit coupling, which all play a key role in understanding Majorana physics. We further show that level repulsion due to spin-orbit coupling in a finite size system can lead to seemingly stable zero bias conductance peaks, which mimic the behavior of Majorana zero modes. Our results improve the understanding of realistic Majorana nanowire systems. △ Less

Submitted 4 June, 2018; originally announced June 2018.

Comments: 10 pages, 5 figures, supplemental information as ancillary file

Journal ref: 2018 New J. Phys. 20 103049

Quantized Majorana conductance

Authors: Hao Zhang, Chun-Xiao Liu, Sasa Gazibegovic, Di Xu, John A. Logan, Guanzhong Wang, Nick van Loo, Jouri D. S. Bommer, Michiel W. A. de Moor, Diana Car, Roy L. M. Op het Veld, Petrus J. van Veldhoven, Sebastian Koelling, Marcel A. Verheijen, Mihir Pendharkar, Daniel J. Pennachio, Borzoyeh Shojaei, Joon Sue Lee, Chris J. Palmstrom, Erik P. A. M. Bakkers, S. Das Sarma, Leo P. Kouwenhoven

Abstract: Majorana zero-modes hold great promise for topological quantum computing. Tunnelling spectroscopy in electrical transport is the primary tool to identify the presence of Majorana zero-modes, for instance as a zero-bias peak (ZBP) in differential-conductance. The Majorana ZBP-height is predicted to be quantized at the universal conductance value of 2e2/h at zero temperature. Interestingly, this qua… ▽ More Majorana zero-modes hold great promise for topological quantum computing. Tunnelling spectroscopy in electrical transport is the primary tool to identify the presence of Majorana zero-modes, for instance as a zero-bias peak (ZBP) in differential-conductance. The Majorana ZBP-height is predicted to be quantized at the universal conductance value of 2e2/h at zero temperature. Interestingly, this quantization is a direct consequence of the famous Majorana symmetry, 'particle equals antiparticle'. The Majorana symmetry protects the quantization against disorder, interactions, and variations in the tunnel coupling. Previous experiments, however, have shown ZBPs much smaller than 2e2/h, with a recent observation of a peak-height close to 2e2/h. Here, we report a quantized conductance plateau at 2e2/h in the zero-bias conductance measured in InSb semiconductor nanowires covered with an Al superconducting shell. Our ZBP-height remains constant despite changing parameters such as the magnetic field and tunnel coupling, i.e. a quantized conductance plateau. We distinguish this quantized Majorana peak from possible non-Majorana origins, by investigating its robustness on electric and magnetic fields as well as its temperature dependence. The observation of a quantized conductance plateau strongly supports the existence of non-Abelian Majorana zero-modes in the system, consequently paving the way for future braiding experiments. △ Less

Submitted 29 October, 2017; originally announced October 2017.

Comments: 5 figures

Journal ref: Nature (2018)

Ballistic superconductivity in semiconductor nanowires

Authors: Hao Zhang, Önder Gül, Sonia Conesa-Boj, Michał P. Nowak, Michael Wimmer, Kun Zuo, Vincent Mourik, Folkert K. de Vries, Jasper van Veen, Michiel W. A. de Moor, Jouri D. S. Bommer, David J. van Woerkom, Diana Car, Sébastien R. Plissard, Erik P. A. M. Bakkers, Marina Quintero-Pérez, Maja C. Cassidy, Sebastian Koelling, Srijit Goswami, Kenji Watanabe, Takashi Taniguchi, Leo P. Kouwenhoven

Abstract: Semiconductor nanowires have opened new research avenues in quantum transport owing to their confined geometry and electrostatic tunability. They have offered an exceptional testbed for superconductivity, leading to the realization of hybrid systems combining the macroscopic quantum properties of superconductors with the possibility to control charges down to a single electron. These advances brou… ▽ More Semiconductor nanowires have opened new research avenues in quantum transport owing to their confined geometry and electrostatic tunability. They have offered an exceptional testbed for superconductivity, leading to the realization of hybrid systems combining the macroscopic quantum properties of superconductors with the possibility to control charges down to a single electron. These advances brought semiconductor nanowires to the forefront of efforts to realize topological superconductivity and Majorana modes. A prime challenge to benefit from the topological properties of Majoranas is to reduce the disorder in hybrid nanowire devices. Here, we show ballistic superconductivity in InSb semiconductor nanowires. Our structural and chemical analyses demonstrate a high-quality interface between the nanowire and a NbTiN superconductor which enables ballistic transport. This is manifested by a quantized conductance for normal carriers, a strongly enhanced conductance for Andreev-reflecting carriers, and an induced hard gap with a significantly reduced density of states. These results pave the way for disorder-free Majorana devices. △ Less

Submitted 10 July, 2017; originally announced July 2017.

Comments: This submission contains the first part of arXiv:1603.04069. The second part of arXiv:1603.04069 is included in a separate paper

Journal ref: Nature Communications 8, 16025 (2017)

Epitaxy of Advanced Nanowire Quantum Devices

Authors: Sasa Gazibegovic, Diana Car, Hao Zhang, Stijn C. Balk, John A. Logan, Michiel W. A. de Moor, Maja C. Cassidy, Rudi Schmits, Di Xu, Guanzhong Wang, Peter Krogstrup, Roy L. M. Op het Veld, Jie Shen, Daniël Bouman, Borzoyeh Shojaei, Daniel Pennachio, Joon Sue Lee, Petrus J. van Veldhoven, Sebastian Koelling, Marcel A. Verheijen, Leo P. Kouwenhoven, Chris J. Palmstrøm, Erik P. A. M. Bakkers

Abstract: Semiconductor nanowires provide an ideal platform for various low-dimensional quantum devices. In particular, topological phases of matter hosting non-Abelian quasi-particles can emerge when a semiconductor nanowire with strong spin-orbit coupling is brought in contact with a superconductor. To fully exploit the potential of non-Abelian anyons for topological quantum computing, they need to be exc… ▽ More Semiconductor nanowires provide an ideal platform for various low-dimensional quantum devices. In particular, topological phases of matter hosting non-Abelian quasi-particles can emerge when a semiconductor nanowire with strong spin-orbit coupling is brought in contact with a superconductor. To fully exploit the potential of non-Abelian anyons for topological quantum computing, they need to be exchanged in a well-controlled braiding operation. Essential hardware for braiding is a network of single-crystalline nanowires coupled to superconducting islands. Here, we demonstrate a technique for generic bottom-up synthesis of complex quantum devices with a special focus on nanowire networks having a predefined number of superconducting islands. Structural analysis confirms the high crystalline quality of the nanowire junctions, as well as an epitaxial superconductor-semiconductor interface. Quantum transport measurements of nanowire "hashtags" reveal Aharonov-Bohm and weak-antilocalization effects, indicating a phase coherent system with strong spin-orbit coupling. In addition, a proximity-induced hard superconducting gap is demonstrated in these hybrid superconductor-semiconductor nanowires, highlighting the successful materials development necessary for a first braiding experiment. Our approach opens new avenues for the realization of epitaxial 3-dimensional quantum device architectures. △ Less

Submitted 10 December, 2021; v1 submitted 3 May, 2017; originally announced May 2017.

Comments: data of the paper can be found at DOI: 10.5281/zenodo.4572619 or link: https://zenodo.org/record/5025868#.YbMBfi-iFHg

Journal ref: Nature 548, 434-438 (2017)

InSb Nanowires with Built-In GaxIn1-xSb Tunnel Barriers for Majorana Devices

Authors: Diana Car, Sonia Conesa-Boj, Hao Zhang, Roy L. M. Op het Veld, Michiel W. A. de Moor, Elham M. T. Fadaly, Önder Gül, Sebastian Kölling, Sebastien R. Plissard, Vigdis Toresen, Michael T. Wimmer, Kenji Watanabe, Takashi Taniguchi, Leo P. Kouwenhoven, Erik P. A. M. Bakkers

Abstract: Majorana zero modes (MZMs), prime candidates for topological quantum bits, are detected as zero bias conductance peaks (ZBPs) in tunneling spectroscopy measurements. Implementation of a narrow and high tunnel barrier in the next generation of Majorana devices can help to achieve the theoretically predicted quantized height of the ZBP. We propose a material-oriented approach to engineer a sharp and… ▽ More Majorana zero modes (MZMs), prime candidates for topological quantum bits, are detected as zero bias conductance peaks (ZBPs) in tunneling spectroscopy measurements. Implementation of a narrow and high tunnel barrier in the next generation of Majorana devices can help to achieve the theoretically predicted quantized height of the ZBP. We propose a material-oriented approach to engineer a sharp and narrow tunnel barrier by synthesizing a thin axial segment of GaxIn1-xSb within an InSb nanowire. By varying the precursor molar fraction and the growth time, we accurately control the composition and the length of the barriers. The height and the width of the GaxIn1-xSb tunnel barrier are extracted from the Wentzel-Kramers-Brillouin (WKB)-fits to the experimental I-V traces. △ Less

Submitted 12 July, 2021; v1 submitted 17 November, 2016; originally announced November 2016.

Comments: See doi: 10.5281/zenodo.5064400 for source data. No changes in this version compared to the previous version

Journal ref: Nano Lett.17, 721 (2017)

Ballistic Majorana nanowire devices

Authors: Önder Gül, Hao Zhang, Jouri D. S. Bommer, Michiel W. A. de Moor, Diana Car, Sébastien R. Plissard, Erik P. A. M. Bakkers, Attila Geresdi, Kenji Watanabe, Takashi Taniguchi, Leo P. Kouwenhoven

Abstract: Majorana modes are zero-energy excitations of a topological superconductor that exhibit non-Abelian statistics. Following proposals for their detection in a semiconductor nanowire coupled to an s-wave superconductor, several tunneling experiments reported characteristic Majorana signatures. Reducing disorder has been a prime challenge for these experiments because disorder can mimic the zero-energ… ▽ More Majorana modes are zero-energy excitations of a topological superconductor that exhibit non-Abelian statistics. Following proposals for their detection in a semiconductor nanowire coupled to an s-wave superconductor, several tunneling experiments reported characteristic Majorana signatures. Reducing disorder has been a prime challenge for these experiments because disorder can mimic the zero-energy signatures of Majoranas, and renders the topological properties inaccessible. Here, we show characteristic Majorana signatures in InSb nanowire devices exhibiting clear ballistic transport properties. Application of a magnetic field and spatial control of carrier density using local gates generates a zero bias peak that is rigid over a large region in the parameter space of chemical potential, Zeeman energy, and tunnel barrier potential. The reduction of disorder allows us to resolve separate regions in the parameter space with and without a zero bias peak, indicating topologically distinct phases. These observations are consistent with the Majorana theory in a ballistic system, and exclude for the first time the known alternative explanations that invoke disorder or a nonuniform chemical potential. △ Less

Submitted 26 April, 2021; v1 submitted 13 March, 2016; originally announced March 2016.

Comments: See https://doi.org/10.5281/zenodo.4721357 for source data. No changes in this version (v3) compared to the previous version (v2)

Journal ref: Nature Nanotechnology (2018)

Conductance Quantization at zero magnetic field in InSb nanowires

Authors: Jakob Kammhuber, Maja C. Cassidy, Hao Zhang, Önder Gül, Fei Pei, Michiel W. A. de Moor, Bas Nijholt, Kenji Watanabe, Takashi Taniguchi, Diana Car, Sebastien R. Plissard, Erik P. A. M. Bakkers, Leo P. Kouwenhoven

Abstract: Ballistic electron transport is a key requirement for existence of a topological phase transition in proximitized InSb nanowires. However, measurements of quantized conductance as direct evidence of ballistic transport have so far been obscured due to the increased chance of backscattering in one dimensional nanowires. We show that by improving the nanowire-metal interface as well as the dielectri… ▽ More Ballistic electron transport is a key requirement for existence of a topological phase transition in proximitized InSb nanowires. However, measurements of quantized conductance as direct evidence of ballistic transport have so far been obscured due to the increased chance of backscattering in one dimensional nanowires. We show that by improving the nanowire-metal interface as well as the dielectric environment we can consistently achieve conductance quantization at zero magnetic field. Additionally, studying the sub-band evolution in a rotating magnetic field reveals an orbital degeneracy between the second and third sub-bands for perpendicular fields above 1T. △ Less

Submitted 11 March, 2016; originally announced March 2016.

Journal ref: Nano lett. 16(6), 3482-3486 (2016)