Skip To Main Content
van Stein on Muck Rack

van Stein

As seen in: arXiv
Is this you? As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work. Claim your profile
Back to van Stein’s profile

Articles by van Stein

Computer Science > Artificial Intelligence

|
By van Duijn, van der Meer, van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (351) Issue with article?

Computer Science > Computation and Language

|
By van Stein
| arXiv
In this paper, we evaluate open-source generative LLMs on legal Natural Language Inference (NLI). Legal inspectorial processes take place in specific domains and often deal with confidential data. This creates a need for working with local models that do not require labeled training data. We evaluate our models on the ContractNLI benchmark and two NLI4Wills datasets.
Open in Who Shared Issue with article?

Computer Science > Neural and Evolutionary Computing

|
By van Stein
| arXiv
[Submitted on 6 Jul 2026] Title:LLM-Driven Evolutionary Generation of Multi-Objective Bayesian Optimization Algorithms View a PDF of the paper titled LLM-Driven Evolutionary Generation of Multi-Objective Bayesian Optimization Algorithms, by Georgios Laskaris and 5 other authors View PDF HTML (experimental) Abstract:Designing effective multi-objective Bayesian optimization (MOBO) algorithms requires balancing many interdependent design choices whose optimal configuration is problem-dependent...
Open in Who Shared Issue with article?

Computer Science > Computer Vision and Pattern Recognition

|
By van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared Issue with article?

Computer Science > Artificial Intelligence

|
By Mohammad Mohammadi Amiri, van Stein, van Leeuwen
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (635) Issue with article?

Computer Science > Neural and Evolutionary Computing

|
By van Stein, Frank Neumann, Thomas G. Back, Thomas H.W. Bäck
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (599) Issue with article?

From Heuristic Selection to Automated Algorithm Design: LLMs Benefit from Strong Priors

|
By van Stein
| arXiv
Open in Who Shared Issue with article?

[2602.06742] Structural bias in multi-objective optimisation

|
By van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (527) Issue with article?

Assessing Reproducibility in Evolutionary Computation: A Case Study using Human- and LLM-based Assessment

|
By Da Ros, van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (641) Issue with article?

Landscape-aware Automated Algorithm Design: An Efficient Framework for Real-world Optimization

|
By Jian Cheng, van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (321) Issue with article?

LLaMEA-SAGE: Guiding Automated Algorithm Design with Structural Feedback from Explainable AI

|
By van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (330) Issue with article?

Investigating the Interplay of Parameterization and Optimizer in Gradient-Free Topology Optimization: A Cantilever Beam Case Study

|
By Iván Olarte, van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (631) Issue with article?

[2601.18846] LLM Driven Design of Continuous Optimization Problems with Controllable High-level Properties

|
By van Stein, Niki van Stein, Britta Grimme, Moritz Seiler
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (630) Issue with article?

State Design Matters: How Representations Shape Dynamic Reasoning in Large Language Models

|
By van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (472) Issue with article?

[2511.21514] Mechanistic Interpretability for Transformer-based Time Series Classification

|
By van Stein, Thomas G. Back, Thomas H.W. Bäck, Niki van Stein, Sofoklis Kitharidis
| arXiv
[Submitted on 26 Nov 2025] Title:Mechanistic Interpretability for Transformer-based Time Series Classification View a PDF of the paper titled Mechanistic Interpretability for Transformer-based Time Series Classification, by Mat\=iss Kaln\=are and 3 other authors View PDF HTML (experimental) Abstract:Transformer-based models have become state-of-the-art tools in various machine learning tasks, including time series classification, yet their complexity makes understanding their internal...
Open in Who Shared Issue with article?

From Performance to Understanding: A Vision for Explainable Automated Algorithm Design

|
By van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (549) Issue with article?

Benchmarking that Matters: Rethinking Benchmarking for Practical Impact

|
By van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (624) Issue with article?

[2510.22209] Visual Model Selection using Feature Importance Clusters in Fairness-Performance Similarity Optimized Space

|
By van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (658) Issue with article?

[2510.17899] Automated Algorithm Design for Auto-Tuning Optimizers

|
By van Stein, van Werkhoven, Niki van Stein, Ben van Werkhoven
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (517) Issue with article?

CoComposer: LLM Multi-agent Collaborative Music Composition

|
By van Stein, Aske Plaat, Niki van Stein
| arXiv
arXiv:2509.00132 (cs) View PDF HTML (experimental) Subjects: Sound (cs.SD); Artificial Intelligence (cs.AI); Multimedia (cs.MM); Audio and Speech Processing (eess.AS) Cite as: arXiv:2509.00132 [cs.SD] (or arXiv:2509.00132v1 [cs.SD] for this version) https://doi.org/10.48550/arXiv.2509.00132 Submission history From: Aske Plaat [ view email] [v1] Fri, 29 Aug 2025 14:15:12 UTC (34 KB) Bibliographic Tools Bibliographic Explorer Toggle Bibliographic Explorer () Connected Papers Toggle Connected...
Open in Who Shared View Duplicates (118) Issue with article?

[2507.22928] How does Chain of Thought Think? Mechanistic Interpretability of Chain-of-Thought Reasoning with Sparse Autoencoding

|
By van Stein
| arXiv
Open in Who Shared Issue with article?

[2507.03605] Behaviour Space Analysis of LLM-driven Meta-heuristic Discovery

|
By van Stein
| arXiv
Open in Who Shared Issue with article?

[2506.19692] Leveraging Lightweight Generators for Memory Efficient Continual Learning

|
By Ahmed Nabil, van Stein
| arXiv
[Submitted on 24 Jun 2025] Title:Leveraging Lightweight Generators for Memory Efficient Continual Learning View a PDF of the paper titled Leveraging Lightweight Generators for Memory Efficient Continual Learning, by Christiaan Lamers and 3 other authors View PDF HTML (experimental) Abstract:Catastrophic forgetting can be trivially alleviated by keeping all data from previous tasks in memory.
Open in Who Shared Issue with article?

[2505.10543] Towards a Deeper Understanding of Reasoning Capabilities in Large Language Models

|
By van Stein
| arXiv
Work on one of the world's most important websites and make an impact on open science.
Open in Who Shared View Duplicates (548) Issue with article?

[2505.06030] Why Are You Wrong? Counterfactual Explanations for Language Grounding with 3D Objects

|
By van Stein
| arXiv
Open in Who Shared Issue with article?

[2503.23037] Agentic Large Language Models, a survey

|
By van Duijn, van Stein, van der Putten
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (409) Issue with article?

[2503.19742] Optimizing Photonic Structures with Large Language Model Driven Algorithm Discovery

|
By van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (412) Issue with article?

Code Evolution Graphs: Understanding Large Language Model Driven Design of Algorithms

|
By van Stein, Anna V. Kononova, Lars Kotthoff, Thomas G. Back, Thomas H.W. Bäck
| arXiv
Open in Who Shared Issue with article?

[2412.03250] Controlling the Mutation in Large Language Models for the Efficient Evolution of Algorithms

|
By van Stein
| arXiv
Open in Who Shared Issue with article?

EconoJax: A Fast & Scalable Economic Simulation in Jax

|
By van Stein
| arXiv
arXiv:2410.22165 (cs) View PDF HTML (experimental) Comments: Subjects: Multiagent Systems (cs.MA); Machine Learning (cs.LG); General Economics (econ.GN) Cite as: arXiv:2410.22165 [cs.MA] (or arXiv:2410.22165v1 [cs.MA] for this version) https://doi.org/10.48550/arXiv.2410.22165 Submission history From: Koen Ponse [ view email] [v1] Tue, 29 Oct 2024 16:02:50 UTC (792 KB) Bibliographic Tools Bibliographic Explorer Toggle Bibliographic Explorer () Litmaps Toggle Litmaps (What is Litmaps?)...
Open in Who Shared View Duplicates (319) Issue with article?

[2410.13657] Selection of Filters for Photonic Crystal Spectrometer Using Domain-Aware Evolutionary Algorithms

|
By van Stein, Thomas H. W
| arXiv
Open in Who Shared Issue with article?

[2410.16309] In-the-loop Hyper-Parameter Optimization for LLM-Based Automated Design of Heuristics

|
By van Stein
| arXiv
Open access is only possible with YOUR support. Give to arXiv this week to help keep science open for all.
Open in Who Shared View Duplicates (284) Issue with article?

TX-Gen: Multi-Objective Optimization for Sparse Counterfactual Explanations for Time-Series Classification

|
By van Stein
| arXiv
Open in Who Shared Issue with article?

Beyond the Veil of Similarity: Quantifying Semantic Continuity in Explainable AI

|
By van Stein, Shadnan Azwad, Emanuele Mezzi, Osman Mutlu
| arXiv
Open in Who Shared Issue with article?

[2407.11511] Reasoning with Large Language Models, a Survey

|
By van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (416) Issue with article?

Evolutionary Computation and Explainable AI: A Roadmap to Transparent Intelligent Systems

|
By van Stein
| arXiv
Open in Who Shared Issue with article?

[2406.06629] A Survey of Meta-features Used for Automated Selection of Algorithms for Black-box Single-objective Continuous Optimization

|
By van Stein
| arXiv
Open in Who Shared Issue with article?

LLaMEA: A Large Language Model Evolutionary Algorithm for Automatically Generating Metaheuristics

|
By van Stein
| arXiv
Open in Who Shared Issue with article?

[2405.10271] Automated Federated Learning via Informed Pruning

|
By van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (647) Issue with article?

[2404.17323] A Deep Dive into Effects of Structural Bias on CMA-ES Performance along Affine Trajectories

|
By van Stein
| arXiv
Open in Who Shared Issue with article?

[2402.06299] A Functional Analysis Approach to Symbolic Regression

|
By van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (695) Issue with article?

[2401.17842] Explainable Benchmarking for Iterative Optimization Heuristics

|
By van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (690) Issue with article?

[2309.12078] Clustering-based Domain-Incremental Learning

|
By van Stein
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (539) Issue with article?

DoE2Vec: Deep-learning Based Features for Exploratory Landscape Analysis

|
By van Stein, Fu Xing
| arXiv
arXiv:2304.01219 (math) Download a PDF of the paper titled DoE2Vec: Deep-learning Based Features for Exploratory Landscape Analysis, by Bas van Stein and 5 other authors Download PDF Submission history From: Bas van Stein [ view email] [v1] Fri, 31 Mar 2023 09:38:44 UTC (24,821 KB) Bibliographic Tools Bibliographic Explorer Toggle Bibliographic Explorer () Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media DagsHub...
Open in Who Shared View Duplicates (56) Issue with article?
Show More