Searcharxiv⌕ Search

arXiv subjects

Sebastian Pokutta

Publications and source records attributed to Sebastian Pokutta.

At least 55 records · Page 3Linked to original sources
arXiv2026

Stable Personas: Dual-Assessment of Temporal Stability in LLM-Based Human Simulation

Large Language Models (LLMs) acting as artificial agents offer the potential for scalable behavioral research, yet their validity depends on whether LLMs can maintain stable personas across extended conversations. We address this point using a dual-assessment framework measuring both self-reported characteristics and observer-rated persona expression. Across two experiments testing four persona conditions (default, high, moderate, and low ADHD presentations), seven LLMs, and three semantically equivalent persona prompts, we examine between-conversation stability (3,473 conversations) and within-conversation stability (1,370 conversations and 18 turns). Self-reports remain highly stable both between and within conversations. However, observer ratings reveal a tendency for persona expressions to decline during extended conversations. These findings suggest that persona-instructed LLMs produce stable, persona-aligned self-reports, an important prerequisite for behavioral research, while identifying this regression tendency as a boundary condition for multi-agent social simulation.

cs.HC↗
arXiv2026

FACET: Teacher-Centred LLM-Based Multi-Agent Systems-Towards Personalized Educational Worksheets

The increasing heterogeneity of student populations poses significant challenges for teachers, particularly in mathematics education, where cognitive, motivational, and emotional differences strongly influence learning outcomes. While AI-driven personalization tools have emerged, most remain performance-focused, offering limited support for teachers and neglecting broader pedagogical needs. This paper presents the FACET framework, a teacher-facing, large language model (LLM)-based multi-agent system designed to generate individualized classroom materials that integrate both cognitive and motivational dimensions of learner profiles. The framework comprises three specialized agents: (1) learner agents that simulate diverse profiles incorporating topic proficiency and intrinsic motivation, (2) a teacher agent that adapts instructional content according to didactical principles, and (3) an evaluator agent that provides automated quality assurance. We tested the system using authentic grade 8 mathematics curriculum content and evaluated its feasibility through a) automated agent-based assessment of output quality and b) exploratory feedback from K-12 in-service teachers. Results from ten internal evaluations highlighted high stability and alignment between generated materials and learner profiles, and teacher feedback particularly highlighted structure and suitability of tasks. The findings demonstrate the potential of multi-agent LLM architectures to provide scalable, context-aware personalization in heterogeneous classroom settings, and outline directions for extending the framework to richer learner profiles and real-world classroom trials.

cs.HC↗
arXiv2026

Abs-Smooth Frank-Wolfe Method: Primal-Dual Analysis, Heavy Ball Momentum, and Inexact Oracles

We study projection-free optimization for convex objectives that satisfy abs-smoothness, a structural property that captures many non-smooth yet piecewise smooth functions arising, e.g., in modern machine learning models. We develop a unified framework for Abs-Smooth Frank-Wolfe methods, establishing a clean primal-dual analysis that guarantees convergence without requiring classical smoothness assumptions. Our framework extends the available results in two important directions. First, we introduce a heavy ball momentum variant and show that momentum can be incorporated naturally under abs-smoothness while preserving convergence guarantees. Second, we analyze inexact minimization oracles, demonstrating robustness to approximate inner solutions. Moreover, we relax the full convexity assumption and study the case where convexity holds only for the piecewise linear approximations of the objective, further broadening the applicability of conditional gradient methods to a wider class of non-smooth problems.

math.OC↗
arXiv2026

What Do Evolutionary Coding Agents Evolve?

Recent work pairs LLMs with evolutionary search to iteratively generate, modify, and select code using task-specific feedback. These systems have produced strong results in mathematical discovery and algorithm design, yet a fundamental question remains: what do they actually evolve? Progress is typically summarized by the best score a run reaches under a task-specific evaluator, but that score can reflect several different mechanisms: new algorithmic structure, re-tuning an existing strategy, recombining ideas already in the model's internal knowledge, or overfitting to the evaluator. Distinguishing these mechanisms requires inspecting the search process itself, not only its final outcome. We introduce EvoTrace, a dataset of evolutionary coding traces spanning four evolutionary frameworks, reasoning and non-reasoning models, and 16 tasks across mathematics and algorithm design. To analyze these traces, we develop EvoReplay, a replay-based methodology that reconstructs the local search states behind high-scoring solutions and tests controlled interventions, including adjusting constants, removing program components and substituting models or prompting contexts. We annotate every code edit in EvoTrace with one of nine recurring edit types using an LLM-as-judge pipeline validated against blind human re-annotation. Across EvoTrace, most score gains come from a small subset of these edit types. We further find a deterministic cycling pattern: about 30% of code lines added during search are byte-identical re-introductions of previously-deleted lines, present throughout nearly every run. These results show that benchmark gains in evolutionary coding agents can arise from qualitatively different mechanisms, only some of which correspond to new algorithmic structure. EvoTrace enables more diagnostic evaluation of evolutionary coding agents beyond final benchmark scores.

cs.NE↗
arXiv2026

Sustainability via LLM Right-sizing

Large language models (LLMs) have become increasingly embedded in organizational workflows. This has raised concerns over their energy consumption, financial costs, and data sovereignty. While performance benchmarks often celebrate cutting-edge models, real-world deployment decisions require a broader perspective: when is a smaller, locally deployable model "good enough"? This study offers an empirical answer by evaluating eleven proprietary and open-weight LLMs across ten everyday occupational tasks, including summarizing texts, generating schedules, and drafting emails and proposals. Using a dual-LLM-based evaluation framework, we automated task execution and standardized evaluation across ten criteria related to output quality, factual accuracy, and ethical responsibility. Results show that GPT-4o delivers consistently superior performance but at a significantly higher cost and environmental footprint. Notably, smaller models like Gemma-3 and Phi-4 achieved strong and reliable results on most tasks, suggesting their viability in contexts requiring cost-efficiency, local deployment, or privacy. A cluster analysis revealed three model groups -- premium all-rounders, competent generalists, and limited but safe performers -- highlighting trade-offs between quality, control, and sustainability. Significantly, task type influenced model effectiveness: conceptual tasks challenged most models, while aggregation and transformation tasks yielded better performances. We argue for a shift from performance-maximizing benchmarks to task- and context-aware sufficiency assessments that better reflect organizational priorities. Our approach contributes a scalable method to evaluate AI models through a sustainability lens and offers actionable guidance for responsible LLM deployment in practice.

cs.CL↗
arXiv2026

Beyond Static Responses: Multi-Agent LLM Systems as a New Paradigm for Social Science Research

As large language models (LLMs) transition from static tools to fully agentic systems, their potential for transforming social science research has become increasingly evident. This paper introduces a structured framework for understanding the diverse applications of LLM-based agents, ranging from simple data processors to complex, multi-agent systems capable of simulating emergent social dynamics. By mapping this developmental continuum across six levels, the paper clarifies the technical and methodological boundaries between different agentic architectures, providing a comprehensive overview of current capabilities and future potential. It highlights how lower-tier systems streamline conventional tasks like text classification and data annotation, while higher-tier systems enable novel forms of inquiry, including the study of group dynamics, norm formation, and large-scale social processes. However, these advancements also introduce significant challenges, including issues of reproducibility, ethical oversight, and the risk of emergent biases. The paper critically examines these concerns, emphasizing the need for robust validation protocols, interdisciplinary collaboration, and standardized evaluation metrics. It argues that while LLM-based agents hold transformative potential for the social sciences, realizing this promise will require careful, context-sensitive deployment and ongoing methodological refinement. The paper concludes with a call for future research that balances technical innovation with ethical responsibility, encouraging the development of agentic systems that not only replicate but also extend the frontiers of social science, offering new insights into the complexities of human behavior.

cs.MA↗
arXiv2026

Curvature-Dependent Lower Bounds for Frank-Wolfe

The Frank-Wolfe algorithm achieves a convergence rate of $\mathcal{O}(1/T)$ for smooth convex optimization over compact convex domains, accelerating to $\mathcal{O}(1/T^2)$ when both the objective and the feasible set are strongly convex. This acceleration extends beyond strong convexity: Kerdreux et al. (2021a) proved rates of $\mathcal{O}(T^{-p/(p-1)})$ over $p$-uniformly convex feasible sets, a class that interpolates between strongly convex sets and more general curved domains such as $\ell_p$ balls. In this work, we establish a matching $Ω(T^{-p/(p-1)})$ lower bound for every $p\ge 3$ under exact line search or short steps, and extend the lower bound to objectives satisfying a Hölderian error bound. The proofs analyze the dynamics of Frank-Wolfe iterates on simple instances and hence are not limited to the high-dimensional setting, unlike information-theoretic lower bounds.

math.OC↗
arXiv2026

Simulating Eating Disorder Patients with LLMs: Evaluating Psychological Persona Stability in Multi-Turn Conversations

Large language model (LLM)-based simulations of clinical patients are increasingly used for research and training, yet their validity requires persona stability: coherent maintenance of an assigned psychological profile across and within conversations. We evaluate this prerequisite using eating disorder personas grounded in five published case vignettes, a dual-assessment framework (self-report + independent observer ratings), and validated psychometric instruments (EDE-Q) with known ground-truth scores. Across six LLMs and two experiments (between-conversation stability (Exp. I) and within-conversation stability (Exp. II)), we find that LLMs are paradoxically too stable and too inaccurate: variability is negligible, yet all models systematically overshoot ground-truth severity by 12-30% of the scale range (0.7-1.8 points on a 0-6 scale). The mechanism is selective stereotyping: models differentiate cases on behavioural items (dietary restraint) but maximise cognitive-affective items (body dissatisfaction, weight preoccupation) at ceiling regardless of case severity. Additional conversational context does not improve accuracy; it compounds the overshoot. LLMs can portray severe eating pathology but lack a representation of moderate clinical presentations, a "missing middle".

cs.CY↗
arXiv2026

Agentic MIP Research: Accelerated Constraint Handler Generation

Mixed-integer programming (MIP) research is both mathematically sophisticated and engineering-intensive: testing an algorithmic hypothesis within a branch-and-cut solver requires substantial implementation, debugging, tuning, and large-scale benchmarking. We propose an agentic MIP research framework that shortens this feedback loop by embedding LLM agents into a solver-aware harness for generating, verifying, and evaluating plugins for the open-source solver SCIP. Propagation methods play a central role in accelerating MIP solving by exploiting global constraints. We instantiate our framework on the semantic lifting of MIP formulations into global constraints and the automatic construction of propagation-only SCIP constraint handlers. On the MIPLIB 2017 benchmark set, the framework successfully recovers global constraint structures from constraint programming and generates executable constraint detectors and propagation-only constraint handlers. Furthermore, the framework naturally extends to in-context learning within a sandboxed environment, enabling agents not only to tune and debug generated constraint handlers on real instances, but also to explore global constraint patterns in MIP problems and discover novel propagation strategies not yet implemented in SCIP. This framework allows us to systematically distinguish meaningful algorithmic improvements from low-value or overly costly candidates: the novel propagation methods successfully solved five additional instances within the explored benchmark. Overall, this framework demonstrates that LLM agents can autonomously navigate the complex MIP research loop, paving the way for a more automated solver development process.

cs.AI↗
arXiv2026

RECON: Robust symmetry discovery via Explicit Canonical Orientation Normalization

Real world data often exhibits unknown, instance-specific symmetries that rarely exactly match a transformation group $G$ fixed a priori. Class-pose decompositions aim to create disentangled representations by factoring inputs into invariant features and a pose $g\in G$ defined relative to a training-dependent, arbitrary canonical representation. We introduce RECON, a class-pose agnostic canonical orientation normalization that corrects arbitrary canonicals via a simple right translation, yielding natural, data-aligned canonicalizations. This enables (i) unsupervised discovery of instance-specific pose distributions, (ii) detection of out-of-distribution poses and (iii) a plug-and-play test-time canonicalization layer. This layer can be attached on top of any pre-trained model to infuse group invariance, improving its performance without retraining. We validate on images and molecular ensembles, demonstrating accurate symmetry discovery, and matching or outperforming other canonicalizations in downstream classification.

cs.LG↗
arXiv2026

Out-of-the-Box Global Optimization for Packing Problems: New Models and Improved Solutions

Recent LLM-driven discoveries have renewed interest in geometric packing problems. In this paper, we study several classes of such packing problems through the lens of modern global nonlinear optimization. Starting from comparatively direct nonlinear formulations, we consider packing circles in squares and fixed-perimeter rectangles, packing circles into minimum-area ellipses, packing regular polygons into regular polygons, and packing Platonic solids into Platonic solids. For ellipse packing, we derive a novel containment formulation based on the S-lemma. For polygon and Platonic solid packing, we develop compact non-overlap formulations based on the Farkas lemma and study several natural modeling variants computationally. Using off-the-shelf global optimization solvers, namely FICO Xpress and SCIP, we obtain numerous new incumbent solutions as well as first solutions for previously unstudied variants without any problem-specific solver modifications beyond writing down the models. Our computational results analyze the impact of various formulation choices. Beyond the individual packing results, the paper illustrates a broader point. It provides further evidence that global nonlinear optimization has matured into an increasingly practical model-and-solve technology for highly nonconvex problems.

math.OC↗
arXiv2026

CHAP: A Hybrid GPU-CPU Heuristic for MIP

We present CHAP (Coordinating Heuristics Across Platforms) a GPU-CPU-hybrid primal heuristic framework for mixed-integer programming. CHAP adopts a portfolio approach where it coordinates a set of primal heuristics, including Local Search, Fix-and-Propagate, and Feasibility Pump, via a shared solution pool. The solution pool is used to exchange feasible incumbent solutions, LP solutions, along with promising infeasible solution candidates, enabling a more comprehensive exploration of the solution space. On the GPU side, we implement a native tabu search featuring a novel best-shift algorithm built on sort, scan, and reduce primitives, along with specialized kernel designs. We additionally leverage cuPDLPx as an approximate LP solver. On the CPU side, we employ various Fix-and-Propagate strategies, guided by information from the solution pool, complemented by a CPU-based tabu search and a Feasibility Pump. All components operate collaboratively, iteratively repairing and improving candidate solutions maintained in the pool. We evaluate our framework on the 50-instances benchmark from the 2026 Land-Doig MIP Competition under competition constraints, including a five-minute time limit. In these settings, CHAP finds solutions to 47 instances outperforming both Gurobi (44) in default mode and NVIDIA cuOpt (43) in heuristics-only mode. The results demonstrate that coordinated cross-platform portfolios offer a promising direction for the integration of GPU heuristics into modern high-performance MIP solvers. The code will be made available on GitHub.

math.OC↗
arXiv2026

Fast Isotopy Computation for T-Curves

A T-curve of degree $d$ is given by a regular unimodular triangulation of $d \cdot Δ_2$ together with a sign distribution on its lattice points. By Viro's Patchworking Theorem, this determines the ambient isotopy type (a.k.a. real scheme) of a smooth real plane projective algebraic curve of the same degree. We present a near-quadratic time algorithm for extracting that isotopy type from the triangulation and the signs. Through a GPU-accelerated implementation, this allows one to compute billions of real schemes per second, enabling exhaustive enumeration at scale. This algorithm was essential for our recent construction of all 121 real schemes of degree seven by T-curves.

math.AG↗
arXiv2026

Alternating Linear Minimization: Revisiting von Neumann's alternating projections

In 1933 von Neumann proved a beautiful result that one can approximate a point in the intersection of two convex sets by alternating projections, i.e., successively projecting on one set and then the other. This algorithm assumes that one has access to projection operators for both sets. In this note, we consider the much weaker setup where we have only access to linear minimization oracles over the convex sets and present an algorithm to find a point in the intersection of two convex sets.

math.OC↗
arXiv2026

The Agentic Researcher: A Practical Guide to AI-Assisted Research in Mathematics and Machine Learning

AI tools and agents are reshaping how researchers work, from proving theorems to training neural networks. Yet for many, it remains unclear how these tools fit into everyday research practice. This paper is a practical guide to AI-assisted research in mathematics and machine learning: We discuss how researchers can use modern AI systems productively, where these systems help most, and what kinds of guardrails are needed to use them responsibly. It is organized into three parts: (I) a five-level taxonomy of AI integration, (II) an open-source framework that, through a set of methodological rules formulated as agent prompts, turns CLI coding agents (e.g., Claude Code, Codex CLI, OpenCode) into autonomous research assistants, and (III) case studies from deep learning and mathematics. The framework runs inside a sandboxed container, works with any frontier LLM through existing CLI agents, is simple enough to install and use within minutes, and scales from personal-laptop prototyping to multi-node, multi-GPU experimentation across compute clusters. In practice, our longest autonomous session ran for over 20 hours, dispatching independent experiments across multiple nodes without human intervention. We stress that our framework is not intended to replace the researcher in the loop, but to augment them. Our code is publicly available at https://github.com/ZIB-IOL/The-Agentic-Researcher.

cs.LG↗
arXiv2026

Beyond Static Instruction: A Multi-agent AI Framework for Adaptive Augmented Reality Robot Training

Augmented Reality (AR) offers powerful visualization capabilities for industrial robot training, yet current interfaces remain predominantly static, failing to account for learners' diverse cognitive profiles. In this paper, we present an AR application for robot training and propose a multi-agent AI framework for future integration that bridges the gap between static visualization and pedagogical intelligence. We report on the evaluation of the baseline AR interface with 36 participants performing a robotic pick-and-place task. While overall usability was high, notable disparities in task duration and learner characteristics highlighted the necessity for dynamic adaptation. To address this, we propose a multi-agent framework that orchestrates multiple components to perform complex preprocessing of multimodal inputs (e.g., voice, physiology, robot data) and adapt the AR application to the learner's needs. By utilizing autonomous Large Language Model (LLM) agents, the proposed system would dynamically adapt the learning environment based on advanced LLM reasoning in real-time.

cs.RO↗
arXiv2026

Estimating Canopy Height at Scale

We propose a framework for global-scale canopy height estimation based on satellite data. Our model leverages advanced data preprocessing techniques, resorts to a novel loss function designed to counter geolocation inaccuracies inherent in the ground-truth height measurements, and employs data from the Shuttle Radar Topography Mission to effectively filter out erroneous labels in mountainous regions, enhancing the reliability of our predictions in those areas. A comparison between predictions and ground-truth labels yields an MAE / RMSE of 2.43 / 4.73 (meters) overall and 4.45 / 6.72 (meters) for trees taller than five meters, which depicts a substantial improvement compared to existing global-scale maps. The resulting height map as well as the underlying framework will facilitate and enhance ecological analyses at a global scale, including, but not limited to, large-scale forest and biomass monitoring.

cs.CV↗
arXiv2026

Capturing Temporal Dynamics in Large-Scale Canopy Tree Height Estimation

With the rise in global greenhouse gas emissions, accurate large-scale tree canopy height maps are essential for understanding forest structure, estimating above-ground biomass, and monitoring ecological disruptions. To this end, we present a novel approach to generate large-scale, high-resolution canopy height maps over time. Our model accurately predicts canopy height over multiple years given Sentinel-1 composite and Sentinel~2 time series satellite data. Using GEDI LiDAR data as the ground truth for training the model, we present the first 10m resolution temporal canopy height map of the European continent for the period 2019-2022. As part of this product, we also offer a detailed canopy height map for 2020, providing more precise estimates than previous studies. Our pipeline and the resulting temporal height map are publicly available, enabling comprehensive large-scale monitoring of forests and, hence, facilitating future research and ecological analyses.

cs.LG↗