arXiv subjects

Jens Sjölund

Publications and source records attributed to Jens Sjölund.

At least 19 recordsLinked to original sources

arXiv2026

Can LLMs Reliably Annotate Bioassay Metadata to Improve Data Readiness?

The emergence of foundation models for molecular property prediction requires a high degree of AI data readiness, including reliable metadata annotation. However, both public repositories and industrial screening databases suffer from missing, inconsistent, or conflated assay annotations. In this work, we quantify the extent of missing annotations in PubChem for the BioAssay Ontology (BAO) assay format and physical detection method fields and investigate whether open-source and proprietary large language models (LLMs) can reliably predict and audit metadata annotations directly from the assay text. In our assessment, we found that the annotation coverage across PubChem's $\sim$2 million bioassays is critically sparse, 36\% lacking an assay format, 89\% a BioAssay type, and >99.9\% any BAO-mapped assay format or detection technology term. This motivates the need for automated test-metadata curation. Using evaluation sets derived from PubChem and ChEMBL, we assess the agreement of seven open-source and proprietary LLMs with existing silver labels. Recall is at least 0.96 for biochemical and cell-based assay formats, with a similar pattern for detection technology, although disagreements increase on under-represented classes. Manual inspection shows that many of these disagreements trace back to inconsistencies between silver sources rather than to LLM error. Moreover, in a qualitative study with a senior industrial curator, LLM-generated evidence prompted the expert to revise some of their own labels, showing LLMs can flag potentially mislabeled assays. Across the study, performance differences between proprietary and open-source models were small. Together, these results suggest LLMs can support the large-scale annotation and auditing of assay metadata, though per-class reliability estimates and targeted human review remain necessary before such labels enter downstream ML pipelines.

cs.CL↗

arXiv2026

Interacting particle guidance for sampling reward-tilted generative priors

Inference-time steering adapts pretrained diffusion and flow-based models to new tasks, e.g., to generate samples from a conditional distribution or samples with desired properties, without retraining. This can be formalized as sampling from a reward-tilted generative prior. As exact sampling from this distribution is intractable, guidance-based methods rely on approximations producing biased samples, and sequential Monte Carlo (SMC) methods correct for this bias using importance weights. However, while exact in the large particle limit, SMC suffers from weight degeneracy and particle collapse in practice. We propose interacting particle guidance (IPG), which replaces reweighting with transport. The particles interact through an additional drift, derived from the Feynman--Kac PDE to cancel the reweighting term, and remain unweighted. Choosing the drift in a reproducing kernel Hilbert space yields a closed-form solution that is cheap to compute, with negligible overhead compared to SMC. We demonstrate the method on Gaussian mixtures with known posteriors, and on high-dimensional image inpainting and protein structure inference tasks.

cs.LG↗

arXiv2026

Sampling Meets Interaction: Sequentially-Controlled Multi-Particle Flow-Maps for Efficient Inference-time Search

While generative models have enabled training-free reward alignment, existing particle-based methods are fundamentally constrained by their propensity to local exploration within narrow regions of the underlying distribution, severely restricting sample diversity. This limitation becomes especially acute under tight reward-feedback budgets, where effective search demands broad, strategic exploration to uncover high-utility regions. To address this, we propose Sequentially-Controlled Interactive Multi-Particle Flow-Maps (IMPFM), a framework for feedback-efficient search. IMPFM progressively transports a group of interactive particles toward the target distribution, maintaining the broad coverage essential for heterogeneous preference alignment. IMPFM leverages a principled and efficient posterior sample-sharing mechanism across particles powered by flow maps. By correcting individual particle drift with the collective value gradient from the entire ensemble's posterior samples at each correction step, the framework maximizes sample utility to enable global exploration while actively mitigating reward over-optimization, typical of standard control frameworks. Paired with a principled exploration-exploitation reweighting mechanism involving multi-particle interaction, this sequentially corrected multi-particle dynamics explicitly preserves structural diversity and overcomes the weight degeneracy inherent to standard Sequential Monte Carlo (SMC) samplers. Crucially, we prove that the resulting sampling framework yields a multi-particle interaction-aware Feynman-Kac corrector that progressively steers the multi-particle system toward a KL-tilted target distribution, facilitating global exploration and preventing mode collapse. Extensive empirical evaluations and ablations across diverse search and alignment tasks confirm the efficacy of IMPFM over existing baselines.

cs.LG↗

arXiv2026

BFMT: Enhancing Search Capabilities of Tree Sampler via Bootstrap Flow-Map Tree

The ability to efficiently navigate high-dimensional spaces to identify optimal candidates-whether designing targeted molecular structures or generating images with preferred attributes-stands as a central pillar of modern scientific progress. Recent tree-based samplers enable highly scalable inference-time search by eliminating the reward-gradient bottleneck inherent to particle-based methods. Despite their potential, existing tree-based samplers are fundamentally constrained by the massive number of function evaluations (NFEs) required for node valuation, crippling their utility in exploration-heavy tasks. Furthermore, their reliance on small, uniform transitions at each depth precludes dynamic, adaptive search capabilities. To address this, we introduce Bootstrap Flow-Map-Tree (a.k.a BFMT), a computationally efficient sampling framework that enables full tree-path construction from any tree depth using a single function evaluation, drastically reducing computational overhead while providing critical foresight for sequential sampling. By enabling dynamic transition time-step scheduling, BFMT efficiently allocates its sampling budget, smoothly transitioning from broad global exploration to fine-grained local refinement of high-utility modes discovered through exploration. Extensive experiments and ablation studies across diverse domains-from high-dimensional images to large-scale molecules-demonstrate BFMT's superiority over baseline approaches in both search and alignment tasks.

cs.LG↗

arXiv2026

Correcting Boundary Bias and Observation Independence in Bayesian Experimental Design

In many experimental settings, active learning can improve sample efficiency by sequentially selecting where to measure, which is particularly valuable when experiments are expensive. Gaussian processes with variance-based acquisition criteria are widely used for this purpose, but have two limitations. First, they are observation-independent: their posterior variance depends only on where samples are acquired, not on what is measured, impairing their sensitivity to the structure of the acquired data. Second, they inflate the variance near boundaries, leading to excessive sampling at the edges of the space compared to the interior. These limitations undermine the gains in sampling efficiency expected from sequential acquisition. We address both limitations. We derive a reconstruction-driven design density and use the posterior mean to build a training-free warp that places more measurements where the target function varies rapidly. A geometric equalizer separately corrects boundary bias. Across sixteen synthetic and two real-data benchmarks, the geometric equalizer consistently improves function reconstruction by correcting boundary bias, while the reconstruction warp provides further gains by concentrating measurements where the posterior mean varies rapidly.

cs.LG↗

arXiv2026

Efficient Image Restoration with State-Dependent Forward Diffusion

This paper proposes to perform image restoration through a state-dependent mean-reverting forward diffusion (FoD) process. In contrast to traditional diffusion-based approaches that rely on a coupled forward-backward diffusion scheme, FoD directly learns image restoration through a single forward diffusion process, yielding a simple yet efficient framework. The core of FoD is a state-dependent stochastic differential equation (SDE) that involves a mean-reverting term in both the drift and diffusion functions. This mean-reverting structure drives the low-quality data toward the clean endpoint with controlled stochastic variation, therefore simulating a stochastic interpolation between source and target distributions. More importantly, FoD is analytically tractable and is trained using a simple stochastic flow matching objective, enabling few-step sampling during inference. The proposed FoD model, despite its simplicity, achieves strong overall performance on various image restoration tasks compared to representative diffusion, diffusion bridge, and flow matching approaches. The code is available at https://github.com/Algolzw/FoD.

cs.LG↗

arXiv2026

Improving Sample Efficiency in Multi-Agent Reinforcement Learning for Simulated Football Games via Exploration

Multi-agent reinforcement learning has shown promise in learning cooperative behaviors in team-based environments. However, such methods often demand extensive training time, which inhibits their application for game-AI in standard game development. For instance, the state-of-the-art method TiZero takes 40 days to train high-quality policies for a football environment. In this paper, we hypothesize that better exploration mechanisms can improve the sample efficiency of multi-agent methods. Thereby, we propose utilizing a random network distillation bonus within the multi-agent TiZero framework, aiming to promote exploration. Additionally, we introduce architectural modifications to the original algorithm to enhance TiZero's computational efficiency. We evaluate the sample efficiency of our approach against original TiZero through extensive experiments. Our results show that random network distillation improves the sample efficiency per training phase by 13.3% compared with the original TiZero, enhancing generalization and adaptability to previously difficult scenarios. This highlights the better applicability of our variant in practical game development settings. Lastly, we qualitatively evaluate the gameplay of the produced models against a heuristic AI. We find that random network distillation leads to a higher accuracy in shooting, and it achieves higher behavioral stability as shown by the lower standard deviation achieved in gameplay metrics. The code is available at https://github.com/electronicarts/marling.

cs.LG↗

arXiv2026

Boundary Variance Inflation Causes Acquisition Bias in Gaussian Processes

Gaussian processes with stationary kernels on bounded domains exhibit inflated posterior variance near the boundary. Despite being a long-recognized artifact in geostatistics and a source of over-exploration in Bayesian optimization, the causes and effects of boundary-induced acquisition bias are underexplored. We trace the root cause to a simple geometric mechanism: the truncation of the kernel correlation neighborhood at the domain boundary creates an observation-independent distortion that worsens with dimensionality. We show how this distortion manifests across three acquisition classes: variance maximization concentrates selections at the corners, whereas negative integrated posterior variance and expected predictive information gain move selections inward to axis-aligned interior shells. These patterns arise without reference to any objective function, meaning that acquisition behavior can be dominated by kernel geometry rather than the desired task-specific uncertainty. To quantify this, we introduce a function-free selection-profile diagnostic for arbitrary acquisitions, kernels, and bounded-domain geometries.

cs.LG↗

arXiv2026

Warm-starting active-set solvers using graph neural networks

Quadratic programming (QP) solvers are widely used in real-time control and optimization, but their computational cost often limits applicability in time-critical settings. To resolve this, we propose a learning-to-optimize approach using graph neural networks (GNNs) to predict active constraints in the dual active-set solver DAQP. Our method exploits the structural properties of QPs by representing them as bipartite graphs and learns to approximate the optimal active set for effectively warm-starting the solver. Across varying problem sizes, the GNN consistently reduces the number of solver iterations compared to cold-starting, while performance is comparable to a multilayer perceptron baseline. In contrast to the baseline, our GNN-based approach trained on varying problem sizes generalizes to unseen dimensions, demonstrating flexibility and scalability. These results highlight the potential of structure-aware learning to accelerate optimization in real-time applications such as model predictive control.

cs.LG↗

arXiv2026

Learning to accelerate distributed ADMM using graph neural networks

Distributed optimization is fundamental to large-scale machine learning and control applications. Among existing methods, the alternating direction method of multipliers (ADMM) has gained popularity due to its strong convergence guarantees and suitability for decentralized computation. However, ADMM can suffer from slow convergence and high sensitivity to hyperparameter choices. In this work, we show that distributed ADMM iterations can be naturally expressed within the message-passing framework of graph neural networks (GNNs). Building on this connection, we propose learning adaptive step sizes and communication weights through a GNN that predicts these yperparameters based on the current iterates. By unrolling ADMM for a fixed number of iterations, we train the network end-to-end to minimize the solution distance after these iterations for a given problem class, while preserving the algorithm's convergence properties. Numerical experiments demonstrate that our learned variant consistently improves convergence speed and solution quality compared to standard ADMM, both within the trained computational budget and beyond. The code is available at https://github.com/paulhausner/learning-distributed-admm.

cs.LG↗

arXiv2026

Machine learning for in-situ composition mapping in a self-driving magnetron sputtering system

Self-driving labs (SDLs), employing automation and machine learning (ML) to accelerate experimental procedures, have enormous potential in the discovery of new materials. However, in thin film science, SDLs are mainly restricted to solution-based synthetic methods which are easier to automate but cannot access the broad chemical space of inorganic materials. This work presents an SDL based on magnetron co-sputtering. We are using combinatorial frameworks, obtaining accurate composition maps on multi-element, compositionally graded thin films. This normally requires time-consuming ex-situ analysis prone to systematic errors. We present a rapid and calibration-free in-situ, ML driven approach to produce composition maps for arbitrary source combinations and sputtering conditions. We develop a method to predict the composition distribution in a multi-element combinatorial thin film, using in-situ measurements from quartz-crystal microbalance sensors placed in a sputter chamber. For a given source, the sensor readings are learned as a function of the sputtering pressure and magnetron power, through active learning using Gaussian processes (GPs). The final GPs are combined with a geometric model of the deposition flux distribution in the chamber, which allows interpolation of the deposition rates from each source, at any position across the sample. We investigate several acquisition functions for the ML procedure. A fully Bayesian GP - BALM (Bayesian active learning MacKay) - achieved the best performance, learning the deposition rates for a single source in 10 experiments. Prediction accuracy for co-sputtering composition distributions was verified experimentally. Our framework dramatically increases throughput by avoiding the need for extensive characterisation or calibration, thus demonstrating the potential of ML-guided SDLs to accelerate materials exploration.

cs.LG↗

arXiv2025

Taming Diffusion Models for Image Restoration: A Review

Diffusion models have achieved remarkable progress in generative modelling, particularly in enhancing image quality to conform to human preferences. Recently, these models have also been applied to low-level computer vision for photo-realistic image restoration (IR) in tasks such as image denoising, deblurring, dehazing, etc. In this review paper, we introduce key constructions in diffusion models and survey contemporary techniques that make use of diffusion models in solving general IR tasks. Furthermore, we point out the main challenges and limitations of existing diffusion-based IR frameworks and provide potential directions for future work.

cs.CV↗

arXiv2025

ESS-Flow: Training-free guidance of flow-based models as inference in source space

Guiding pretrained flow-based generative models for conditional generation or to produce samples with desired target properties enables solving diverse tasks without retraining on paired data. We present ESS-Flow, a gradient-free method that leverages the typically Gaussian prior of the source distribution in flow-based models to perform Bayesian inference directly in the source space using Elliptical Slice Sampling. ESS-Flow only requires forward passes through the generative model and observation process, no gradient or Jacobian computations, and is applicable even when gradients are unreliable or unavailable, such as with simulation-based observations or quantization in the generation or observation process. We demonstrate its effectiveness on designing materials with desired target properties and predicting protein structures from sparse inter-residue distance measurements.

cs.LG↗

arXiv2025

A parallel algorithm for generating Pareto-optimal radiosurgery treatment plans

Using inverse planning tools to create radiotherapy treatment plans is an iterative process, where clinical trade-offs are explored by changing the relative importance of different objectives and rerunning the optimizer until a desirable plan is found. We seek to optimize hundreds of radiosurgery treatment plans, corresponding to different weightings of objectives, fast enough to incorporate interactive Pareto navigation of clinical trade-offs into the clinical workflow. We apply the alternating direction method of multipliers (ADMM) to the linear-program formulation of the optimization problem used in Lightning. We implement both a CPU and a GPU version of ADMM in Matlab and compare them to Matlab's built-in, single-threaded dual-simplex solver. The ADMM implementation is adapted to the optimization procedure used in the clinical software, with a bespoke algorithm for maximizing overlap between low-dose points for different objective weights. The method is evaluated on a test dataset consisting of 20 cases from three different indications, with between one and nine targets and total target volumes ranging from 0.66 to 52 cm3, yielding speedups of 1.6-97 and 54-1500 times on CPU and GPU, respectively, compared to simplex. Plan quality was evaluated by rerunning the ADMM optimization 20 times, each with a different random seed, for each test case and for nine objective weightings per case. The resulting clinical metrics closely mimicked those obtained when rerunning the simplex solver, verifying the validity of the method. In conclusion, we show how ADMM can be adapted for radiosurgery plan optimization, allowing hundreds of high-quality Gamma Knife treatment plans to be created in under two minutes on a single GPU, also for very large cases.

physics.med-ph↗

arXiv2025

Minimal positive Markov realizations

Finding a positive state-space realization with the minimum dimension for a given transfer function is an open problem in control theory. In this paper, we focus on positive realizations in Markov form and propose a linear programming approach that computes them with a minimum dimension. Such minimum dimension of positive Markov realizations is an upper bound of the minimal positive realization dimension. However, we show that these two dimensions are equal for certain systems.

eess.SY↗

arXiv2025

Exploring Modularity of Agentic Systems for Drug Discovery

Large-language models (LLMs) and agentic systems present exciting opportunities to accelerate drug discovery. In this study, we examine the modularity of LLM-based agentic systems for drug discovery, i.e., whether parts of the system such as the LLM and type of agent are interchangeable, a topic that has received limited attention in drug discovery. We compare the performance of different LLMs and the effectiveness of tool-calling agents versus code-generating agents. Our case study, comparing performance in orchestrating tools for chemistry and drug discovery using an LLM-as-a-judge score, shows that Claude-3.5-Sonnet, Claude-3.7-Sonnet and GPT-4o outperform alternative language models such as Llama-3.1-8B, Llama-3.1-70B, GPT-3.5-Turbo, and Nova-Micro. Although we confirm that code-generating agents outperform the tool-calling ones on average, we show that this is highly question- and model-dependent. Furthermore, the impact of replacing system prompts is dependent on the question and model, underscoring that even in this particular domain one cannot just replace components of the system without re-engineering. Our study highlights the necessity of further research into the modularity of agentic systems to enable the development of reliable and modular solutions for real-world problems.

cs.LG↗

arXiv2025

Navigating chemical design spaces for metal-ion batteries via machine-learning-guided phase-field simulations

Metal anodes provide the highest energy density in batteries. However, they still suffer from electrode/electrolyte interface side reactions and dendrite growth, especially under fast-charging conditions. In this paper, we consider a phase-field model of electrodeposition in metal-anode batteries and provide a scalable, versatile framework for optimizing its chemical parameters. Our approach is based on Bayesian optimization and explores the parameter space with a high sample efficiency and a low computation complexity. We use this framework to find the optimal cell for suppressing dendrite growth and accelerating charging speed under constant voltage. We identify interfacial mobility as a key parameter, which should be maximized to inhibit dendrites without compromising the charging speed. The results are verified using extended simulations of dendrite evolution in charging half cells with lithium-metal anodes.

cond-mat.mtrl-sci↗

arXiv2025

Probabilistic Zeeman-Doppler imaging of stellar magnetic fields: I. Analysis of tau Scorpii in the weak-field limit

Zeeman-Doppler imaging (ZDI) is used to study the surface magnetic field topology of stars, based on high-resolution spectropolarimetric time series observations. Multiple ZDI inversions have been conducted for the early B-type star tau Sco, which has been found to exhibit a weak but complex non-dipolar surface magnetic field. The classical ZDI framework suffers from a significant limitation in that it provides little to no reliable uncertainty quantification for the reconstructed magnetic field maps, with essentially all published results being confined to point estimates. To fill this gap, we propose a Bayesian framework for probabilistic ZDI. Here, the proposed framework is demonstrated on tau Sco in the weak-field limit. We propose three distinct statistical models, and use archival ESPaDOnS high-resolution Stokes V observations to carry out the probabilistic magnetic inversion in closed form. The surface magnetic field is parameterised by a high-dimensional spherical-harmonic expansion. By comparing three different prior distributions over the latent variables in the spherical-harmonic decomposition, our results showcase the ZDI sensitivity to various hyperparameters. The mean magnetic field maps are qualitatively similar to previously published point estimates, but analysis of the magnetic energy distribution indicates high uncertainty and higher energy content at low angular degrees l. Our results effectively demonstrate that, for stars in the weak-field regime, reliable uncertainty quantification of recovered magnetic field maps can be obtained in closed form with natural assumptions on the statistical model. Future work will explore extending this framework beyond the weak-field approximation and incorporating prior uncertainty over multiple stellar parameters in more complex magnetic inversion problems.

astro-ph.SR↗