arXiv subjects

Sebastian Lobentanzer

Publications and source records attributed to Sebastian Lobentanzer.

11 recordsLinked to original sources

arXiv2026

Traceable Trust for action-ready artificial intelligence in bioscience

Artificial intelligence (AI) is becoming part of the working infrastructure of the biosciences. AI models can predict biomolecular structures, design proteins, rank variants, annotate images, recommend strains and optimise experimental conditions. We argue that the decision to use an AI output to guide laboratory action is a key juncture for trustworthy research and should follow a defined, reviewable process. We propose Traceable Trust as a proportionate assessment-and-design framework for this output-to-action boundary. It asks what evidence supports the output, what capability is being claimed, what agency has been delegated, what threshold authorises action, who can override it and how outcomes inform later decisions. We illustrate the framework through three case studies spanning ecosystem resources, project design and laboratory action. Together, the cases show how trust can be documented where AI outputs begin to shape scientific work.

cs.CY↗

arXiv2026

Croissant Tasks: A Metadata Format for Reproducible Machine Learning Evaluations

Reproducibility is fundamental to the scientific method, yet remains a critical challenge in machine learning. Contributing factors include underspecified execution details and brittle software environments. Human-centric remedies, such as checklists and manual verification, help but require intensive effort and fail to scale. To address this, we introduce Croissant Tasks: a declarative, machine-actionable metadata format that abstracts low-level implementation details into high-level specifications. This format enables conceptual reproducibility: verifying claims via independent, agent-generated implementations rather than brittle source code replication. We contribute: (1) the Croissant Tasks specification, formally decoupling task problem from solution; (2) an automated LLM pipeline that retrofits existing benchmarks into this format; and (3) empirical validation showing autonomous agents can ingest these specifications to generate functional, accurate reproduction pipelines from scratch. We envision this format as a new foundation for automated and conceptual reproducibility in machine learning.

cs.AI↗

arXiv2026

Croissant Baker: Metadata Generation for Discoverable, Governable, and Reusable ML Datasets

Croissant has emerged as the metadata standard for machine learning datasets, providing a structured, JSON-LD-based format that makes dataset discovery, automated ingestion, and reproducible analysis machine-checkable across ML platforms. Adoption has accelerated, and NeurIPS now requires Croissant metadata in every submission to its dataset tracks. Yet in practice Croissant generation usually starts with uploading data to a public platform, a path infeasible for governed and large local repositories that hold much of the high-value data ML increasingly relies on. We release Croissant Baker, a local-first, open-source command-line tool that generates validated Croissant metadata directly from a dataset directory through a modular handler registry. We evaluate Croissant Baker on over 140 datasets, scaling to MIMIC-IV at 886 million rows and 374 Parquet files. On held-out comparisons against producer-authored or standards-derived ground truth, Croissant Baker reaches 97-100% agreement across multiple domains.

cs.LG↗

arXiv2026

Quantifying the Expectation-Realisation Gap for Agentic AI Systems

Agentic AI systems are deployed with expectations of substantial productivity gains, yet rigorous empirical evidence reveals systematic discrepancies between pre-deployment expectations and post-deployment outcomes. We review controlled trials and independent validations across software engineering, clinical documentation, and clinical decision support to quantify this expectation-realisation gap. In software development, experienced developers expected a 24% speedup from AI tools but were slowed by 19% -- a 43 percentage-point calibration error. In clinical documentation, vendor claims of multi-minute time savings contrast with measured reductions of less than one minute per note, and one widely deployed tool showed no statistically significant effect. In clinical decision support, externally validated performance falls substantially below developer-reported metrics. These shortfalls are driven by workflow integration friction, verification burden, measurement construct mismatches, and systematic variation in who benefits and who does not. The evidence motivates structured planning frameworks that require explicit, quantified benefit expectations with human oversight costs factored in.

cs.SE↗

arXiv2026

The Agentic Automation Canvas: a structured framework for agentic AI project design

Agentic AI prototypes are being deployed across domains with increasing speed, yet no methodology for their structured design, governance, and prospective evaluation has been established. Existing AI documentation practices and guidelines -- Model Cards, Datasheets, or NIST AI RMF -- are either retrospective or lack machine-readability and interoperability. We present the Agentic Automation Canvas (AAC), a structured framework for the prospective design of agentic systems and a tool to facilitate communication between their users and developers. The AAC captures six dimensions of an automation project: definition and scope; user expectations with quantified benefit metrics; developer feasibility assessments; governance staging; data access and sensitivity; and outcomes. The framework is implemented as a semantic web-compatible metadata schema with controlled vocabulary and mappings to established ontologies such as Schema$\mathrm{.}$org and W3C DCAT. It is made accessible through a privacy-preserving, fully client-side web application with real-time validation. Completed canvases export as FAIR-compliant RO-Crates, yielding versioned, shareable, and machine-interoperable project contracts between users and developers. We describe the schema design, benefit quantification model, and prospective application to diverse use cases from research, clinical, and institutional settings. The AAC and its web application are available as open-source code and interactive web form at https://aac.slolab.ai$\mathrm{.}$

cs.SE↗

arXiv2025

AI Benchmark Democratization and Carpentry

Benchmarks are a cornerstone of modern machine learning, enabling reproducibility, comparison, and scientific progress. However, AI benchmarks are increasingly complex, requiring dynamic, AI-focused workflows. Rapid evolution in model architectures, scale, datasets, and deployment contexts makes evaluation a moving target. Large language models often memorize static benchmarks, causing a gap between benchmark results and real-world performance. Beyond traditional static benchmarks, continuous adaptive benchmarking frameworks are needed to align scientific assessment with deployment risks. This calls for skills and education in AI Benchmark Carpentry. From our experience with MLCommons, educational initiatives, and programs like the DOE's Trillion Parameter Consortium, key barriers include high resource demands, limited access to specialized hardware, lack of benchmark design expertise, and uncertainty in relating results to application domains. Current benchmarks often emphasize peak performance on top-tier hardware, offering limited guidance for diverse, real-world scenarios. Benchmarking must become dynamic, incorporating evolving models, updated data, and heterogeneous platforms while maintaining transparency, reproducibility, and interpretability. Democratization requires both technical innovation and systematic education across levels, building sustained expertise in benchmark design and use. Benchmarks should support application-relevant comparisons, enabling informed, context-sensitive decisions. Dynamic, inclusive benchmarking will ensure evaluation keeps pace with AI evolution and supports responsible, reproducible, and accessible AI deployment. Community efforts can provide a foundation for AI Benchmark Carpentry.

cs.AI↗

arXiv2024

Validation Requirements for AI-based Intervention-Evaluation in Aging and Longevity Research and Practice

The field of aging and longevity research is overwhelmed by vast amounts of data, calling for the use of Artificial Intelligence (AI), including Large Language Models (LLMs), for the evaluation of geroprotective interventions. Such evaluations should be correct, useful, comprehensive, explainable, and they should consider causality, interdisciplinarity, adherence to standards, longitudinal data and known aging biology. In particular, comprehensive analyses should go beyond comparing data based on canonical biomedical databases, suggesting the use of AI to interpret changes in biomarkers and outcomes. Our requirements motivate the use of LLMs with Knowledge Graphs and dedicated workflows employing, e.g., Retrieval-Augmented Generation. While naive trust in the responses of AI tools can cause harm, adding our requirements to LLM queries can improve response quality, calling for benchmarking efforts and justifying the informed use of LLMs for advice on longevity interventions.

cs.HC↗

arXiv2024

Molecular causality in the advent of foundation models

Correlation is not causation. As simple as this widely agreed-upon statement may seem, scientifically defining causality and using it to drive our modern biomedical research is immensely challenging. In this perspective, we attempt to synergise the partly disparate fields of systems biology, causal reasoning, and machine learning, to inform future approaches in the field of systems biology and molecular networks.

q-bio.MN↗

arXiv2023

Drugst.One -- A plug-and-play solution for online systems medicine and network-based drug repurposing

In recent decades, the development of new drugs has become increasingly expensive and inefficient, and the molecular mechanisms of most pharmaceuticals remain poorly understood. In response, computational systems and network medicine tools have emerged to identify potential drug repurposing candidates. However, these tools often require complex installation and lack intuitive visual network mining capabilities. To tackle these challenges, we introduce Drugst.One, a platform that assists specialized computational medicine tools in becoming user-friendly, web-based utilities for drug repurposing. With just three lines of code, Drugst.One turns any systems biology software into an interactive web tool for modeling and analyzing complex protein-drug-disease networks. Demonstrating its broad adaptability, Drugst.One has been successfully integrated with 21 computational systems medicine tools. Available at https://drugst.one, Drugst.One has significant potential for streamlining the drug discovery process, allowing researchers to focus on essential aspects of pharmaceutical treatment research.

q-bio.QM↗

arXiv2023

A Platform for the Biomedical Application of Large Language Models

Current-generation Large Language Models (LLMs) have stirred enormous interest in recent months, yielding great potential for accessibility and automation, while simultaneously posing significant challenges and risk of misuse. To facilitate interfacing with LLMs in the biomedical space, while at the same time safeguarding their functionalities through sensible constraints, we propose a dedicated, open-source framework: BioChatter. Based on open-source software packages, we synergise the many functionalities that are currently developing around LLMs, such as knowledge integration / retrieval-augmented generation, model chaining, and benchmarking, resulting in an easy-to-use and inclusive framework for application in many use cases of biomedicine. We focus on robust and user-friendly implementation, including ways to deploy privacy-preserving local open-source LLMs. We demonstrate use cases via two multi-purpose web apps (https://chat.biocypher.org), and provide documentation, support, and an open community.

q-bio.QM↗

arXiv2023

Democratising Knowledge Representation with BioCypher

Standardising the representation of biomedical knowledge among all researchers is an insurmountable task, hindering the effectiveness of many computational methods. To facilitate harmonisation and interoperability despite this fundamental challenge, we propose to standardise the framework of knowledge graph creation instead. We implement this standardisation in BioCypher, a FAIR (findable, accessible, interoperable, reusable) framework to transparently build biomedical knowledge graphs while preserving provenances of the source data. Mapping the knowledge onto biomedical ontologies helps to balance the needs for harmonisation, human and machine readability, and ease of use and accessibility to non-specialist researchers. We demonstrate the usefulness of this framework on a variety of use cases, from maintenance of task-specific knowledge stores, to interoperability between biomedical domains, to on-demand building of task-specific knowledge graphs for federated learning. BioCypher (https://biocypher.org) frees up valuable developer time; we encourage further development and usage by the community.

q-bio.MN↗