-
Guiding Posterior Exploration with Optimizer-Derived Geometry
Authors:
Moritz Schlager,
Emanuel Sommer,
Thomas Möllenhoff,
David Rügamer
Abstract:
Sampling-based methods offer a principled approach to uncertainty quantification in Bayesian neural networks. Their practical use, however, is often challenged by the computational cost of exploring high-dimensional and multimodal posterior distributions. To overcome these difficulties, Bayesian Deep Ensembles, i.e., warmstarting the sampling from several optimized solutions, have proven to be an…
▽ More
Sampling-based methods offer a principled approach to uncertainty quantification in Bayesian neural networks. Their practical use, however, is often challenged by the computational cost of exploring high-dimensional and multimodal posterior distributions. To overcome these difficulties, Bayesian Deep Ensembles, i.e., warmstarting the sampling from several optimized solutions, have proven to be an effective strategy. In this paper, we demonstrate that curvature estimates computed during the warmstart as a byproduct in adaptive optimizers such as AdamW can inform the sampling phase at negligible additional cost. Specifically, our proposed preconditioned sampling strategy based on optimizer-derived geometries can substantially reduce or even eliminate the need for a lengthy sampling burn-in phase and leads to greater numerical stability. This approach consistently maintains or improves predictive performance and uncertainty quantification without any additional computational costs. We confirm the consistency of our findings across various datasets and network architectures.
△ Less
Submitted 28 July, 2026;
originally announced July 2026.
-
Patients With Personality: Realistic Patient Simulation through Controlled Diversity and Selective Disclosure
Authors:
Moritz Schlager,
Friederike Jungmann,
Samuel Schmidgall,
Philipp Raffler,
Franziska Hartl,
Eva Wende,
Paula Roßmüller,
Conrad Ketzer,
Avinatan Hassidim,
Dale R. Webster,
Yossi Matias,
Yun Liu,
Daniel Rueckert,
Mike Schaekermann,
Paul Hager
Abstract:
Simulating realistic patient interactions is a key requirement to testing clinical applications of LLMs at scale without time-consuming and expensive user studies. However, existing approaches often lack realism and controllability, often oversharing information unprompted, and failing to capture the wide variability of patient behavior. Here, we introduce PatientsWithPersonality (PWP), a patient…
▽ More
Simulating realistic patient interactions is a key requirement to testing clinical applications of LLMs at scale without time-consuming and expensive user studies. However, existing approaches often lack realism and controllability, often oversharing information unprompted, and failing to capture the wide variability of patient behavior. Here, we introduce PatientsWithPersonality (PWP), a patient simulation framework that generates realistic yet diverse virtual patient responses through explicit personality parametrization over a latent patient state. Grounded in HEXACO, a six-dimensional personality space used to quantify and parameterize human behavioral traits, our approach enables fine-grained control over conversational style, cooperativeness, and information disclosure within a unified framework. In a clinician evaluation, PWP is judged nearly as realistic as recorded human actors and clearly ahead of prior simulators, while being flagged as "too informative" far less often. Conditioning on HEXACO axes yields personas whose configured traits are recoverable by both clinicians and an autorater, span a substantially wider behavioral footprint than the closest baseline, and prevent oversharing. Altogether, our framework paves the way for more accurate and informative LLM benchmarking through our realistic and steerable patient simulator.
△ Less
Submitted 11 August, 2026; v1 submitted 13 May, 2026;
originally announced June 2026.
-
Vendor-agnostic 4D Phase Contrast MRI: a complete open-source pipeline for velocities, displacement, and strain analysis
Authors:
Marta B. Maggioni,
Sabine M. Räuber,
Katarina Puš,
Bostjan Šimunič,
Xeni Deligianni,
Regina M. M. Schlaeger,
Francesco Santini
Abstract:
Phase contrast MRI (PC MRI) enables quantitative assessment of tissue motion and strain. Although it is increasingly used, standardized, vendor-agnostic pipelines for accelerated acquisitions remain scarce. We present a fully open-source 4D flow PC-MRI pipeline integrating a compressed sensing-accelerated sequence implemented in PyPulseq, BART-based reconstruction, and strain analysis. Additionall…
▽ More
Phase contrast MRI (PC MRI) enables quantitative assessment of tissue motion and strain. Although it is increasingly used, standardized, vendor-agnostic pipelines for accelerated acquisitions remain scarce. We present a fully open-source 4D flow PC-MRI pipeline integrating a compressed sensing-accelerated sequence implemented in PyPulseq, BART-based reconstruction, and strain analysis. Additionally, a gradient probing sequence was developed to ensure correct velocity sign assignment across scanner orientations and vendors. The pipeline was validated across two Siemens MRI systems (3T MAGNETOM Prisma and 3T Vida Fit) in two anatomical applications: forearm (Flexor Digitorum Superficialis, n=9) and thigh (Vastus Lateralis, n=10) during Neuromuscular Electrical Stimulation (NMES)-induced contractions. Compressed sensing reduced acquisition times from 35 and 80 minutes to 5 and 11 minutes for the arm and leg acquisitions, respectively. Muscle strain maps and sigmoid-fitted strain curves enabled extraction of peak strain, mean strain, and buildup rate. Strains in the Vastus Lateralis were approximately one order of magnitude higher than in the Flexor Digitorum Superficialis (median peak strain 0.49 vs. 0.063, mean strain 0.31 vs. 0.031). The pipeline demonstrates multi-platform compatibility and provides a reproducible, open framework for quantitative muscle imaging.
△ Less
Submitted 8 June, 2026;
originally announced June 2026.
-
MO-CAPO: Multi-Objective Cost-Aware Prompt Optimization
Authors:
Jan Büssing,
Moritz Schlager,
Timo Heiß,
Tom Zehle,
Matthias Feurer
Abstract:
Large language models (LLMs) achieve strong performance across a wide range of tasks but are highly sensitive to prompt design, motivating the need for automatic prompt optimization. Existing methods predominantly focus on performance alone, ignoring competing objectives such as inference cost or latency. At the same time, existing work on multi-objective prompt optimization relies on off-the-shel…
▽ More
Large language models (LLMs) achieve strong performance across a wide range of tasks but are highly sensitive to prompt design, motivating the need for automatic prompt optimization. Existing methods predominantly focus on performance alone, ignoring competing objectives such as inference cost or latency. At the same time, existing work on multi-objective prompt optimization relies on off-the-shelf NSGA-II, ignoring optimization efficiency. As a remedy, we introduce MO-CAPO, a novel multi-objective prompt optimization algorithm that jointly optimizes performance and inference cost while leveraging budget allocation for cost-efficient optimization. We further propose a deployment-oriented cost objective that captures the full computational profile of LLM inference. We evaluate our approach across four tasks and three LLMs and compare it to an NSGA-II-based multi-objective method and state-of-the-art single-objective prompt optimizers. Results show that MO-CAPO consistently identifies strong, robust, and diverse Pareto front approximations while maintaining cost-efficiency. It outperforms the NSGA-II baseline on 8 out of 12 cases in terms of the noisy R2 metric and achieves competitive performances often already at a considerably lower budget. The discovered solution sets span diverse performance-cost trade-offs that are omitted by single-objective optimizers, yet the top-performance candidates remain competitive with single-objective solutions. Additionally, we conduct the first evaluation of multi-objective machine learning experiments that considers generalization and robustness through noisy R2 and approximation gap, enabling a more realistic assessment of solution quality. MO-CAPO enables practitioners to select from an efficiently discovered set of multiple prompts offering different trade-offs between performance and cost.
△ Less
Submitted 15 May, 2026;
originally announced May 2026.
-
promptolution: A Unified, Modular Framework for Prompt Optimization
Authors:
Tom Zehle,
Timo Heiß,
Moritz Schlager,
Matthias Aßenmacher,
Matthias Feurer
Abstract:
Prompt optimization has become crucial for enhancing the performance of large language models (LLMs) across a broad range of tasks. Although many research papers demonstrate its effectiveness, practical adoption is hindered because existing implementations are often tied to unmaintained, isolated research codebases or require invasive integration into application frameworks. To address this, we in…
▽ More
Prompt optimization has become crucial for enhancing the performance of large language models (LLMs) across a broad range of tasks. Although many research papers demonstrate its effectiveness, practical adoption is hindered because existing implementations are often tied to unmaintained, isolated research codebases or require invasive integration into application frameworks. To address this, we introduce promptolution, a unified, modular open-source framework that provides all components required for prompt optimization within a single extensible system for both practitioners and researchers. It integrates multiple contemporary discrete prompt optimizers, supports systematic and reproducible benchmarking, and returns framework-agnostic prompt strings, enabling seamless integration into existing LLM pipelines while remaining agnostic to the underlying model implementation.
△ Less
Submitted 23 February, 2026; v1 submitted 2 December, 2025;
originally announced December 2025.
-
CAPO: Cost-Aware Prompt Optimization
Authors:
Tom Zehle,
Moritz Schlager,
Timo Heiß,
Matthias Feurer
Abstract:
Large language models (LLMs) have revolutionized natural language processing by solving a wide range of tasks simply guided by a prompt. Yet their performance is highly sensitive to prompt formulation. While automatic prompt optimization addresses this challenge by finding optimal prompts, current methods require a substantial number of LLM calls and input tokens, making prompt optimization expens…
▽ More
Large language models (LLMs) have revolutionized natural language processing by solving a wide range of tasks simply guided by a prompt. Yet their performance is highly sensitive to prompt formulation. While automatic prompt optimization addresses this challenge by finding optimal prompts, current methods require a substantial number of LLM calls and input tokens, making prompt optimization expensive. We introduce CAPO (Cost-Aware Prompt Optimization), an algorithm that enhances prompt optimization efficiency by integrating AutoML techniques. CAPO is an evolutionary approach with LLMs as operators, incorporating racing to save evaluations and multi-objective optimization to balance performance with prompt length. It jointly optimizes instructions and few-shot examples while leveraging task descriptions for improved robustness. Our extensive experiments across diverse datasets and LLMs demonstrate that CAPO outperforms state-of-the-art discrete prompt optimization methods in 11/15 cases with improvements up to 21%p in accuracy. Our algorithm achieves better performances already with smaller budgets, saves evaluations through racing, and decreases average prompt length via a length penalty, making it both cost-efficient and cost-aware. Even without few-shot examples, CAPO outperforms its competitors and generally remains robust to initial prompts. CAPO represents an important step toward making prompt optimization more powerful and accessible by improving cost-efficiency.
△ Less
Submitted 17 June, 2025; v1 submitted 22 April, 2025;
originally announced April 2025.