Tootfinder

Is Sequence Information All You Need for Bayesian Optimization of Antibodies?
Sebastian W. Ober, Calvin McCarter, Aniruddh Raghu, Yucen Lily Li, Alan N. Amin, Andrew Gordon Wilson, Hunter Elliott
https://arxiv.org/abs/2509.24933

Is Sequence Information All You Need for Bayesian Optimization of Antibodies?
Bayesian optimization is a natural candidate for the engineering of antibody therapeutic properties, which is often iterative and expensive. However, finding the optimal choice of surrogate model for optimization over the highly structured antibody space is difficult, and may differ depending on the property being optimized. Moreover, to the best of our knowledge, no prior works have attempted to incorporate structural information into antibody Bayesian optimization. In this work, we explore di…

Adaptive Pseudo-Marginal Algorithm
Sarra Abaoubida, Myl\`ene B\'edard, Florian Maire
https://arxiv.org/abs/2509.24820 https://arxiv.org/pdf/2509.24820

Adaptive Pseudo-Marginal Algorithm
The Pseudo-Marginal (PM) algorithm is a popular Markov chain Monte Carlo (MCMC) method used to sample from a target distribution when its density is inaccessible, but can be estimated with a non-negative unbiased estimator. Its performance depends on a key parameter, N, the number of iterations (or particles) used to approximate the target density. Larger values of N yield more accurate estimates but at increased running time. Previous studies has provided guidelines for selecting an optimal va…

I made nog again

Multi-Agent Debate for LLM Judges with Adaptive Stability Detection
Tianyu Hu, Zhen Tan, Song Wang, Huaizhi Qu, Tianlong Chen
https://arxiv.org/abs/2510.12697 https://

Multi-Agent Debate for LLM Judges with Adaptive Stability Detection
With advancements in reasoning capabilities, Large Language Models (LLMs) are increasingly employed for automated judgment tasks. While LLMs-as-Judges offer promise in automating evaluations, current approaches often rely on simplistic aggregation methods (e.g., majority voting), which can fail even when individual agents provide correct answers. To address this, we propose a multi-agent debate judge framework where agents collaboratively reason and iteratively refine their responses. We formal…

Digital ID – The New Chains of Capitalist Surveillance
[…] From passports to colonial passbooks, from welfare cards to border regimes, the apparatus of identification has always been tied to domination. Digital ID is simply the latest iteration of this long history, but with a scale and sophistication that makes its dangers even more profound. […]
👷

Digital ID – The New Chains of Capitalist Surveillance - Aotearoa Workers Solidarity Movement
The world is entering an era where identity is no longer a matter of personal relationships, lived experience, or even paperwork. Increasingly, it is reduced...

Flexible fixed-point iteration and its applications for nonsymmetric algebraic Riccati equations
Zhen-Chen Guo, Xin Liang
https://arxiv.org/abs/2509.25942 https://

Flexible fixed-point iteration and its applications for nonsymmetric algebraic Riccati equations
In this paper, we reveal the intrinsic Toeplitz structure in the unique stabilizing solution for nonsymmetric algebraic Riccati equations by employing a shift-involved fixed-point iteration, and propose an RADI-type method for computing this solution for large-scale equations of this type with sparse and low-rank structure by incorporating flexible shifts into the fixed-point iteration. We present a shift-selection strategy, termed Leja shifts, based on rational approximation theory, which is i…

Building an Open AIBOM Standard in the Wild
Gopi Krishnan Rajbahadur, Keheliya Gallaba, Elyas Rashno, Arthit Suriyawongkul, Karen Bennet, Kate Stewart, Ahmed E. Hassan
https://arxiv.org/abs/2510.07070 …

Building an Open AIBOM Standard in the Wild
Modern software engineering increasingly relies on open, community-driven standards, yet how such standards are created in fast-evolving domains like AI-powered systems remains underexplored. This paper presents a detailed experience report on the development of the AI Bill of Materials AIBOM specification, an extension of the ISO/IEC 5962:2021 Software Package Data Exchange (SPDX) software bill of materials (SBOM) standard, which captures AI components such as datasets and iterative training a…

DeMuon: A Decentralized Muon for Matrix Optimization over Graphs
Chuan He, Shuyi Ren, Jingwei Mao, Erik G. Larsson
https://arxiv.org/abs/2510.01377 https://

DeMuon: A Decentralized Muon for Matrix Optimization over Graphs
In this paper, we propose DeMuon, a method for decentralized matrix optimization over a given communication topology. DeMuon incorporates matrix orthogonalization via Newton-Schulz iterations-a technique inherited from its centralized predecessor, Muon-and employs gradient tracking to mitigate heterogeneity among local functions. Under heavy-tailed noise conditions and additional mild assumptions, we establish the iteration complexity of DeMuon for reaching an approximate stochastic stationary …

Human-Centered Evaluation of RAG outputs: a framework and questionnaire for human-AI collaboration
Aline Mangold, Kiran Hoffmann
https://arxiv.org/abs/2509.26205 https://…

Human-Centered Evaluation of RAG outputs: a framework and questionnaire for human-AI collaboration
Retrieval-augmented generation (RAG) systems are increasingly deployed in user-facing applications, yet systematic, human-centered evaluation of their outputs remains underexplored. Building on Gienapp's utility-dimension framework, we designed a human-centred questionnaire that assesses RAG outputs across 12 dimensions. We iteratively refined the questionnaire through several rounds of ratings on a set of query-output pairs and semantic discussions. Ultimately, we incorporated feedback from bo…

ProSEA: Problem Solving via Exploration Agents
William Nguyen, Vinh Luong, Christopher Nguyen
https://arxiv.org/abs/2510.07423 https://arxiv.org/pdf/2510.0…

ProSEA: Problem Solving via Exploration Agents
Large language models (LLMs) have empowered AI agents to tackle increasingly complex tasks. However, most existing agents remain limited to static planning and brittle interactions, falling short of true collaboration or adaptive reasoning. We introduce ProSEA, a modular, general-purpose multi-agent framework designed for iterative problem solving through exploration and plan evolution. ProSEA features a hierarchical architecture in which a Manager Agent orchestrates domain-specialized Expert A…