Tootfinder

Real-Time AI-Driven Pipeline for Automated Medical Study Content Generation in Low-Resource Settings: A Kenyan Case Study
Emmanuel Korir, Eugene Wechuli
https://arxiv.org/abs/2507.05212

Real-Time AI-Driven Pipeline for Automated Medical Study Content Generation in Low-Resource Settings: A Kenyan Case Study
Juvenotes is a real-time AI-driven pipeline that automates the transformation of academic documents into structured exam-style question banks, optimized for low-resource medical education settings in Kenya. The system combines Azure Document Intelligence for OCR and Azure AI Foundry (OpenAI o3-mini) for question and answer generation in a microservices architecture, with a Vue/TypeScript frontend and AdonisJS backend. Mobile-first design, bandwidth-sensitive interfaces, institutional tagging, a…

OK, this feels deliberate now. Updated my Mac laptop and that also caused settings to be wiped. WTF?

Bluesky rolls out Activity Notifications, letting users receive notifications of new posts and replies from selected accounts, as it expands its sports presence (Richard Lawler/The Verge)
https://www.theverge.com/news/699671/bluesky-push-n…

Bluesky can really keep up with the news now that it has activity notifications
Bluesky’s latest update added more notification settings, as well as activity notifications to see whenever certain accounts have a new post.

We’re closing in on this month’s release of OmniFocus 4.7, with planned dates, mutually-exclusive tags, and improved repeats!
Updated test builds are now available, adding new settings for Forecast counts and summary dots, adding support for keyboard modifiers when using drag-and-drop in Forecast on iPad and visionOS (bringing parity with the support already on Mac), and making other improvements and bug fixes.

Bluesky rolls out Activity Notifications, letting users receive notifications of new posts and replies from selected accounts, as it expands its sports presence (Richard Lawler/The Verge)
https://www.theverge.com/news/699671/bluesky-push-n…

Bluesky can really keep up with the news now that it has activity notifications
Bluesky’s latest update added more notification settings, as well as activity notifications to see whenever certain accounts have a new post.

ThumbnailTruth: A Multi-Modal LLM Approach for Detecting Misleading YouTube Thumbnails Across Diverse Cultural Settings
Wajiha Naveed, Zartash Afzal Uzmi, Zafar Ayyub Qazi
https://arxiv.org/abs/2509.04714

ThumbnailTruth: A Multi-Modal LLM Approach for Detecting Misleading YouTube Thumbnails Across Diverse Cultural Settings
Misleading video thumbnails on platforms like YouTube are a pervasive problem, undermining user trust and platform integrity. This paper proposes a novel multi-modal detection pipeline that uses Large Language Models (LLMs) to flag misleading thumbnails. We first construct a comprehensive dataset of 2,843 videos from eight countries, including 1,359 misleading thumbnail videos that collectively amassed over 7.6 billion views -- providing a unique cross-cultural perspective on this global issue.…

Comparison of Parametric versus Machine-learning Multiple Imputation in Clinical Trials with Missing Continuous Outcomes
Mia S. Tackney, Jonathan W. Bartlett, Elizabeth Williamson, Kim May Lee
https://arxiv.org/abs/2510.03512

Comparison of Parametric versus Machine-learning Multiple Imputation in Clinical Trials with Missing Continuous Outcomes
The use of flexible machine-learning (ML) models to generate imputations of missing data within the framework of Multiple Imputation (MI) has recently gained traction, particularly in observational settings. For randomised controlled trials (RCTs), it is unclear whether ML approaches to MI provide valid inference, and whether they outperform parametric MI approaches under complex data generating mechanisms. We conducted two simulations in RCT settings that have incomplete continuous outcomes bu…

So, if you want to start qBittorrent hidden in the system tray (v5.1.2, KDE), you must:
* check "Minimise qBittorrent to notification area" in Settings/Behaviour (which, oddly, doesn't hide it from the taskbar)
* open ~/config/qbittorrent/qBittorrent.conf and set StartUpWindowState to Minimized

COGITAO: A Visual Reasoning Framework To Study Compositionality & Generalization
Yassine Taoudi-Benchekroun, Klim Troyan, Pascal Sager, Stefan Gerber, Lukas Tuggener, Benjamin Grewe
https://arxiv.org/abs/2509.05249

COGITAO: A Visual Reasoning Framework To Study Compositionality & Generalization
The ability to compose learned concepts and apply them in novel settings is key to human intelligence, but remains a persistent limitation in state-of-the-art machine learning models. To address this issue, we introduce COGITAO, a modular and extensible data generation framework and benchmark designed to systematically study compositionality and generalization in visual domains. Drawing inspiration from ARC-AGI's problem-setting, COGITAO constructs rule-based tasks which apply a set of transfor…

Listened in to the Academy Software Foundation's virtual townhall about the Open Review Initiative. I knew that project because they provide very useful ffmpeg command lines and they're doing all the encoding tests to figure out good quality settings and obscure parameters.
Learnings from the presentation: prores by ffmpeg is sadly pretty crappy compared to the real prores from Apple. There's the OpenAPV codec that is gearing up to be an open-source alternative:

On the Normalization of Confusion Matrices: Methods and Geometric Interpretations
Johan Erbani, Pierre-Edouard Portier, Elod Egyed-Zsigmond, Sonia Ben Mokhtar, Diana Nurbakova
https://arxiv.org/abs/2509.04959

On the Normalization of Confusion Matrices: Methods and Geometric Interpretations
The confusion matrix is a standard tool for evaluating classifiers by providing insights into class-level errors. In heterogeneous settings, its values are shaped by two main factors: class similarity -- how easily the model confuses two classes -- and distribution bias, arising from skewed distributions in the training and test sets. However, confusion matrix values reflect a mix of both factors, making it difficult to disentangle their individual contributions. To address this, we introduce b…

Collaboration and Conflict between Humans and Language Models through the Lens of Game Theory
Mukul Singh, Arjun Radhakrishna, Sumit Gulwani
https://arxiv.org/abs/2509.04847 htt…

Collaboration and Conflict between Humans and Language Models through the Lens of Game Theory
Language models are increasingly deployed in interactive online environments, from personal chat assistants to domain-specific agents, raising questions about their cooperative and competitive behavior in multi-party settings. While prior work has examined language model decision-making in isolated or short-term game-theoretic contexts, these studies often neglect long-horizon interactions, human-model collaboration, and the evolution of behavioral patterns over time. In this paper, we investig…

Not only Gmail has been infested with bad AI, but also other Google products. Chrome has a "help me write" feature. You can [disable it in the AI innovations settings](chrome://settings/ai). Unfortunately the intrusive popup is still showing up in Gmail and I can't find any way to turn it off. (See image)
There's also a Chrome setting that allows better browser history search. I've wanted that for years so I've turned that one on for now. Maybe it's better than all the other Gemini crap?

There was a time when, if you were lucky enough to get a piece of software stable and working well, it continued to be stable and work well.
Now we have relentless updates, whose problems are exemplified by an overnight update of my Mac to OS X 15.6 doing a factory reset of the settings of many of my programs.
Let's not even go into the behavioral changes that get pushed that invalidate all your muscle memory in order to boost some product manager's performance review.
I like stability, but it's just another casualty of this endless churn.

Resource-Efficient Fine-Tuning of LLaMA-3.2-3B for Medical Chain-of-Thought Reasoning
Imran Mansha
https://arxiv.org/abs/2510.05003 https://arxiv.org/pdf/2…

Resource-Efficient Fine-Tuning of LLaMA-3.2-3B for Medical Chain-of-Thought Reasoning
Large Language Models (LLMs) such as GPT-4 and LLaMA have demonstrated remarkable reasoning abilities but require significant computational resources for fine-tuning. This paper presents a resource-efficient fine-tuning approach for LLaMA-3.2-3B to enhance medical chain-of-thought reasoning while operating under constrained GPU and memory settings. Using parameter-efficient tuning techniques such as LoRA and QLoRA, we adapt the base model on publicly available medical reasoning datasets. The mo…

Set to Be Fair: Demographic Parity Constraints for Set-Valued Classification
Eyal Cohen (LPSM), Christophe Denis (SAMM), Mohamed Hebiri (LAMA)
https://arxiv.org/abs/2510.04926 h…

Set to Be Fair: Demographic Parity Constraints for Set-Valued Classification
Set-valued classification is used in multiclass settings where confusion between classes can occur and lead to misleading predictions. However, its application may amplify discriminatory bias motivating the development of set-valued approaches under fairness constraints. In this paper, we address the problem of set-valued classification under demographic parity and expected size constraints. We propose two complementary strategies: an oracle-based method that minimizes classification risk while…

AgentTypo: Adaptive Typographic Prompt Injection Attacks against Black-box Multimodal Agents
Yanjie Li, Yiming Cao, Dong Wang, Bin Xiao
https://arxiv.org/abs/2510.04257 https://…

AgentTypo: Adaptive Typographic Prompt Injection Attacks against Black-box Multimodal Agents
Multimodal agents built on large vision-language models (LVLMs) are increasingly deployed in open-world settings but remain highly vulnerable to prompt injection, especially through visual inputs. We introduce AgentTypo, a black-box red-teaming framework that mounts adaptive typographic prompt injection by embedding optimized text into webpage images. Our automatic typographic prompt injection (ATPI) algorithm maximizes prompt reconstruction by substituting captioners while minimizing human det…

Walking, Rolling, and Beyond: First-Principles and RL Locomotion on a TARS-Inspired Robot
Aditya Sripada, Abhishek Warrier
https://arxiv.org/abs/2510.05001 https://

Walking, Rolling, and Beyond: First-Principles and RL Locomotion on a TARS-Inspired Robot
Robotic locomotion research typically draws from biologically inspired leg designs, yet many human-engineered settings can benefit from non-anthropomorphic forms. TARS3D translates the block-shaped 'TARS' robot from Interstellar into a 0.25 m, 0.99 kg research platform with seven actuated degrees of freedom. The film shows two primary gaits: a bipedal-like walk and a high-speed rolling mode. For TARS3D, we build reduced-order models for each, derive closed-form limit-cycle conditions, and valid…

from my link log —
SQLite's durability settings are a mess.
https://www.agwa.name/blog/post/sqlite_durability
saved 2025-08-29 https://

AI-Driven Grading and Moderation for Collaborative Projects in Computer Science Education
Songmei Yu, Andrew Zagula
https://arxiv.org/abs/2510.03998 https://

AI-Driven Grading and Moderation for Collaborative Projects in Computer Science Education
Collaborative group projects are integral to computer science education, as they foster teamwork, problem-solving skills, and industry-relevant competencies. However, assessing individual contributions within group settings has long been a challenge. Traditional assessment strategies, such as the equal distribution of grades or subjective peer assessments, often fall short in terms of fairness, objectivity, and scalability, particularly in large classrooms. This paper introduces a semi-automate…

Continual Recommender Systems
Hyunsik Yoo, SeongKu Kang, Hanghang Tong
https://arxiv.org/abs/2507.03861 https://arxiv.org/pdf/2507.03…

Continual Recommender Systems
Modern recommender systems operate in uniquely dynamic settings: user interests, item pools, and popularity trends shift continuously, and models must adapt in real time without forgetting past preferences. While existing tutorials on continual or lifelong learning cover broad machine learning domains (e.g., vision and graphs), they do not address recommendation-specific demands-such as balancing stability and plasticity per user, handling cold-start items, and optimizing recommendation metrics…

I see… color-coding 🤔
https://toot.fedilab.app/@apps/115152776458975591

Fedilab Apps (@apps@toot.fedilab.app)
Attached: 4 images We are adding a visual indicator concerning the visibility of messages when composing. Each color related to visibility can be customized (public, unlisted, private and direct). This option can be disabled in settings. #Fedilab (1/2)

AURAD: Anatomy-Pathology Unified Radiology Synthesis with Progressive Representations
Shuhan Ding, Jingjing Fu, Yu Gu, Naiteek Sangani, Mu Wei, Paul Vozila, Nan Liu, Jiang Bian, Hoifung Poon
https://arxiv.org/abs/2509.04819

AURAD: Anatomy-Pathology Unified Radiology Synthesis with Progressive Representations
Medical image synthesis has become an essential strategy for augmenting datasets and improving model generalization in data-scarce clinical settings. However, fine-grained and controllable synthesis remains difficult due to limited high-quality annotations and domain shifts across datasets. Existing methods, often designed for natural images or well-defined tumors, struggle to generalize to chest radiographs, where disease patterns are morphologically diverse and tightly intertwined with anatom…

Adaptive Parallel Downloader for Large Genomic Datasets
Rasman Mubtasim Swargo, Engin Arslan, Md Arifuzzaman
https://arxiv.org/abs/2508.05511 https://arxiv…

Adaptive Parallel Downloader for Large Genomic Datasets
Modern next-generation sequencing (NGS) projects routinely generate terabytes of data, which researchers commonly download from public repositories such as SRA or ENA. Existing download tools often employ static concurrency settings, leading to inefficient bandwidth utilization and prolonged download times due to their inability to adapt to dynamic network conditions. We introduce FastBioDL, a parallel file downloader designed for large biological datasets, featuring an adaptive concurrency con…

On the global linear Zarankiewicz problem
Pantelis E. Eleftheriou, Aris Papadopoulos
https://arxiv.org/abs/2510.03546 https://arxiv.org/pdf/2510.03546

On the global linear Zarankiewicz problem
The `global' Zarankiewicz problem for hypergraphs asks for an upper bound on the number of edges of a finite $r$-hypergraph $V$ in terms of the number $|V|$ of its vertices, assuming the edge relation is induced by a fixed $K_{k, \dots, k}$-free $r$-hypergraph $E$, for some $k\in\mathbb N$. In [4], such bounds of size $O(|V|^{r-1})$ were achieved for a semilinear $E$, namely, definable in a linear o-minimal structure. We establish the same bounds in five new settings: when $…

Introduction to Interpolation-Based Optimization
Lindon Roberts
https://arxiv.org/abs/2510.04473 https://arxiv.org/pdf/2510.04473

Introduction to Interpolation-Based Optimization
The field of derivative-free optimization (DFO) studies algorithms for nonlinear optimization that do not rely on the availability of gradient or Hessian information. It is primarily designed for settings when functions are black-box, expensive to evaluate and/or noisy. A widely used and studied class of DFO methods for local optimization is interpolation-based optimization (IBO), also called model-based DFO, where the general principles from derivative-based nonlinear optimization algorithms a…

Maximizing the Margin between Desirable and Undesirable Elements in a Covering Problem
Sophie Boileau, Andrew Hong, David Liben-Nowell, Alistair Pattison, Anna N. Rafferty, Charlie Roslansky
https://arxiv.org/abs/2507.03817

Maximizing the Margin between Desirable and Undesirable Elements in a Covering Problem
In many covering settings, it is natural to consider the simultaneous presence of desirable elements (that we seek to include) and undesirable elements (that we seek to avoid). This paper introduces a novel combinatorial problem formalizing this tradeoff: from a collection of sets containing both "desirable" and "undesirable" items, pick the subcollection that maximizes the margin between the number of desirable and undesirable elements covered. We call this the Target Approximation Problem (TA…

OneDB: A Distributed Multi-Metric Data Similarity Search System
Tang Qian, Yifan Zhu, Lu Chen, Xiangyu Ke, Jingwen Zhao, Tianyi Li, Yunjun Gao, Christian S. Jensen
https://arxiv.org/abs/2507.04256

OneDB: A Distributed Multi-Metric Data Similarity Search System
Increasingly massive volumes of multi-modal data are being accumulated in many {real world} settings, including in health care and e-commerce. This development calls for effective general-purpose data management solutions for multi-modal data. Such a solution must facilitate user-friendly and accurate retrieval of any multi-modal data according to diverse application requirements. Further, such a solution must be capable of efficient and scalable retrieval. To address this need, we present On…

Deceptive Planning Exploiting Inattention Blindness
Mustafa O. Karabag, Jesse Milzman, Ufuk Topcu
https://arxiv.org/abs/2510.02714 https://arxiv.org/pdf/25…

Deceptive Planning Exploiting Inattention Blindness
We study decision-making with rational inattention in settings where agents have perception constraints. In such settings, inaccurate prior beliefs or models of others may lead to inattention blindness, where an agent is unaware of its incorrect beliefs. We model this phenomenon in two-player zero-sum stochastic games, where Player 1 has perception constraints and Player 2 deceptively deviates from its security policy presumed by Player 1 to gain an advantage. We formulate the perception constr…

I made a repository on #codeberg to document my own #FreshRSS settings to get fulltext feeds of different websites: #RSS

Deep Inverse Rosenblatt Transport for Structural Reliability Analysis
Aryan Tyagi, Jan N. Fuhg
https://arxiv.org/abs/2509.05061 https://arxiv.org/pdf/2509.…

Deep Inverse Rosenblatt Transport for Structural Reliability Analysis
Accurately estimating the probability of failure in engineering systems under uncertainty is a fundamental challenge, particularly in high-dimensional settings and for rare events. Conventional reliability analysis methods often become computationally intractable or exhibit high estimator variance when applied to problems with hundreds of uncertain parameters or highly concentrated failure regions. In this work, we investigate the use of the recently proposed Deep Inverse Rosenblatt Transport (…

On the new weighted geometric inequalities near the sphere in space forms
Weimin Sheng, Yinhang Wang
https://arxiv.org/abs/2508.04067 https://arxiv.org/pdf…

On the new weighted geometric inequalities near the sphere in space forms
In this paper, we first investigate weighted Minkowski type inequalities for nearly spherical sets in space forms, focusing on the sets that are $C^1$-close to geodesic spheres. Our results generalize the work of \cite{G22} by incorporating broader geometric settings and convex weight functions. Additionally, we establish quantitative stability estimates for weighted Alexandrov-Fenchel type inequalities in $\mathbb{R}^{n+1}$ and $\mathbb{H}^{n+1}$, extending the earlier results of \cite{VW24} a…

Bonded braids and the Markov theorem
Paolo Cavicchioli, Bo\v{s}tjan Gabrov\v{s}ek, Matic Simoni\v{c}
https://arxiv.org/abs/2507.04565 https://

Bonded braids and the Markov theorem
Bonded knots arise naturally in topological protein modeling, where intramolecular interactions such as disulfide bridges stabilize folded configurations. These structures extend classical knot theory by incorporating embedded graphs, and have been formalized as bonded knots. In this paper, we develop the algebraic theory of bonded braids, introducing the bonded braid monoid in the topological and rigid settings, which encodes both classical braid crossings and (rigid) bonded connections. We pr…

Ich hoffe übrigens, ihr habt das alle ausgeschaltet, oder?
#chatgpt #settings

I've summarized my settings and customization of my desktop using #GNOME48 (from #Debian13 #Trixie) on

Causal Mediation in Natural Experiments
Senan Hogan-Hennessy
https://arxiv.org/abs/2508.05449 https://arxiv.org/pdf/2508.05449

Causal Mediation in Natural Experiments
Natural experiments are a cornerstone of applied economics, providing settings for estimating causal effects with a compelling argument for treatment randomisation, but give little indication of the mechanisms behind causal effects. Causal Mediation (CM) provides a framework to analyse mechanisms by identifying the average direct and indirect effects (CM effects), yet conventional CM methods require the relevant mediator is as-good-as-randomly assigned. When people choose the mediator based on …

The Spatial Regime Conversion Method
Charles G. Cameron, Cameron A. Smith, Christian A. Yates
https://arxiv.org/abs/2507.04937 https://

The Spatial Regime Conversion Method
We present the spatial regime conversion method (SRCM), a novel hybrid modelling framework for simulating reaction-diffusion systems that adaptively combines stochastic discrete and deterministic continuum representations. Extending the regime conversion method (RCM) to spatial settings, the SRCM employs a discrete reaction-diffusion master equation (RDME) representation in regions of low concentration and continuum partial differential equations (PDEs) where concentrations are high, dynamicall…

@CoMaps@floss.social
The settings offer the option to use ‘Google Fused Location Services’. I assume this is so that this Google service can still be used even if it is disabled in the Android settings. @… offers an alternative #locationService

Pure many-body interactions in colloidal systems by artificial random light fields
Augustin Muster, Luis S. Froufe-P\'erez
https://arxiv.org/abs/2510.03503 https://

Pure many-body interactions in colloidal systems by artificial random light fields
We propose a method to generate pure many-body interactions in colloidal systems by using optical forces induced by random optical fields with an optimized spectral energy density. To assess the feasibility in general settings, we develop a simple model for Lorentzian electric and magnetic dipole response. An optimization procedure is then introduced to design the spectral energy density of the random field that minimizes pair interactions at constant electromagnetic energy density. We conclude…

16WW Heat-pump Control - Settings and control of the heat pump at 16WW. #heatPump #controls #DHW #spaceHeat

16WW Heat-pump Control
Settings and control of the heat pump at 16WW. #heatPump #controls #DHW #spaceHeat

A note on last passage percolation and Schur processes
Evgeni Dimitrov, Zongrui Yang
https://arxiv.org/abs/2510.04713 https://arxiv.org/pdf/2510.04713

A note on last passage percolation and Schur processes
In this note we provide a short proof of the distributional equality between last passage percolation with geometric weights along a general down-right path and Schur processes. We do this in both the full-space and half-space settings, and for general parameters. The main inputs for our arguments are generalizations of the Robinson-Schensted-Knuth correspondence and Greene's theorem due to Krattenthaler, which are based on Fomin's growth diagrams.

Local Fr'echet Regression via RKHS embedding and Its Applications to Data Analysis on Manifolds
Yuki Iida, Hiroshi Shiraishi, Hiroaki Ogata
https://arxiv.org/abs/2507.03288

Local Fr'echet Regression via RKHS embedding and Its Applications to Data Analysis on Manifolds
Local Fr'echet Regression (LFR) is a nonparametric regression method for settings in which the explanatory variable lies in a Euclidean space and the response variable lies in a metric space. It is used to estimate smooth trajectories in general metric spaces from noisy observations of random objects taking values in such spaces. Since metric spaces form a broad class of spaces that often lack algebraic structures such as addition or scalar multiplication characteristics typical of vector space…

Looks like I have this to investigate now because a warning is appearing on my admin dashboard. Whatever the problem is seems to affect Elasticsearch 7.17.23 and Mastodon 4.4.2 , because this never appeared on Mastodon 4.3.x and the suggested command only returns this error:
[400] {"error":{"root_cause":[{"type":"illegal_argument_exception","reason":"unknown setting [index.analysis] please check that any required plugins are installed, or check the breaking changes documentation for removed settings"}],"type":"illegal_argument_exception","reason":"unknown setting [index.analysis] please check that any required plugins are installed, or check the breaking changes documentation for removed settings"},"status":400}
#MastoAdmin

Topological pumping of light governed by Fibonacci numbers
Ruihan Peng, Kai Yang, Qidong Fu, Yanli Chen, Peng Wang, Yaroslav V. Kartashov, Vladimir V. Konotop, Fangwei Ye
https://arxiv.org/abs/2509.04910

Topological pumping of light governed by Fibonacci numbers
Topological pumping refers to transfer of a physical quantity governed by the systemtopology, resulting in quantized amounts of the transferred quantities. It is a ubiqui-tous wave phenomenon typically considered subject to exactly periodic adiabatic vari-ation of the system parameters. Recently, proposals for generalizing quasi-periodictopological pumping and identifying possible physical settings for its implementa-tion have emerged. In a strict sense, pumping with incommensurate frequencies …

Frage an alle #GitLab Admins: gibt es eine Möglichkeit, einen User automatisch auf alle Projekte Instanz-weit in GitLab self hosted zuzulassen?
Für eine Überprüfung reicht ein Audit User nicht aus und ich möchte das so weit wir möglich automatisieren, auch für neue Projekte.

Quaternion Toeplitz matrices and their fundamental properties
Muhammad Ahsan Khan, Sohail Khan
https://arxiv.org/abs/2510.04297 https://arxiv.org/pdf/2510.…

Quaternion Toeplitz matrices and their fundamental properties
Toeplitz matrices are characterized by their constant diagonals, have been extensively studied in various settings, including over real and complex numbers. However, their study over quaternions is quite sparse. In this paper, we investigate the structure and the algebraic properties of quaternion Toeplitz matrices. Most importantly, we established a complete characterization of all normal Toeplitz matrices having entries commutative quaternions.

Toward Accessible Dermatology: Skin Lesion Classification Using Deep Learning Models on Mobile-Acquired Images
Asif Newaz, Masum Mushfiq Ishti, A Z M Ashraful Azam, Asif Ur Rahman Adib
https://arxiv.org/abs/2509.04800

Toward Accessible Dermatology: Skin Lesion Classification Using Deep Learning Models on Mobile-Acquired Images
Skin diseases are among the most prevalent health concerns worldwide, yet conventional diagnostic methods are often costly, complex, and unavailable in low-resource settings. Automated classification using deep learning has emerged as a promising alternative, but existing studies are mostly limited to dermoscopic datasets and a narrow range of disease classes. In this work, we curate a large dataset of over 50 skin disease categories captured with mobile devices, making it more representative o…

Fundamental Limits of Crystalline Equivariant Graph Neural Networks: A Circuit Complexity Perspective
Yang Cao, Zhao Song, Jiahao Zhang, Jiale Zhao
https://arxiv.org/abs/2510.05494

Fundamental Limits of Crystalline Equivariant Graph Neural Networks: A Circuit Complexity Perspective
Graph neural networks (GNNs) have become a core paradigm for learning on relational data. In materials science, equivariant GNNs (EGNNs) have emerged as a compelling backbone for crystalline-structure prediction, owing to their ability to respect Euclidean symmetries and periodic boundary conditions. Despite strong empirical performance, their expressive power in periodic, symmetry-constrained settings remains poorly understood. This work characterizes the intrinsic computational and expressive…

Coalitional stability under myopic expectations and externalities
Agustin G. Bonifacio, Maria Haydee Fonseca-Mairena, Pablo Neme
https://arxiv.org/abs/2507.03259

Coalitional stability under myopic expectations and externalities
We study coalition formation problems in the presence of externalities, focusing on settings where agents exhibit myopic expectations, that is, they evaluate potential deviations based solely on the immediate outcome, assuming no further reactions or reorganization by others. First, we establish a sufficient condition for the non-emptiness of both the core and the stable set. In the case of the core, our condition for ensuring non-emptiness also provides a characterization of all core partition…

I’ve set up modes on my Pixel 9a running GrapheneOS. I’ll share an update in a few weeks with my experience using it.
#MentalHealth #Modes #Pixel9a

Teaching Introductory Functional Programming Using Haskelite
Pedro Vasconcelos (University of Porto)
https://arxiv.org/abs/2508.03640 https://arxiv.org/pdf…

Teaching Introductory Functional Programming Using Haskelite
Learning functional programming requires learning a substitution-based computational model. While substitution should be a familiar concept from high-school algebra, students often have difficulty applying it to new settings, such as recursive definitions, algebraic data types and higher-order functions. Step-by-step interpreters have been shown to help beginners by clarifying misconceptions and improving understanding. This paper reports on the experience of using a step-by-step tracing inte…

TIL: Apple has enhanced the built in calculator app massively with conversions of all sorts and advanced calculator and math functionality. I never noticed this before. When was that introduced?

Lightweight Transformers for Zero-Shot and Fine-Tuned Text-to-SQL Generation Using Spider
Chirag Seth, Utkarsh Singh
https://arxiv.org/abs/2508.04623 https://

Lightweight Transformers for Zero-Shot and Fine-Tuned Text-to-SQL Generation Using Spider
Text-to-SQL translation enables non-expert users to query relational databases using natural language, with applications in education and business intelligence. This study evaluates three lightweight transformer models - T5-Small, BART-Small, and GPT-2 - on the Spider dataset, focusing on low-resource settings. We developed a reusable, model-agnostic pipeline that tailors schema formatting to each model's architecture, training them across 1000 to 5000 iterations and evaluating on 1000 test sam…

A Note on Code Quality Score: LLMs for Maintainable Large Codebases
Sherman Wong, Jalaj Bhandari, Leo Zhou Fan Yang, Xylan Xu, Yi Zhuang, Cem Cayiroglu, Payal Bhuptani, Sheela Yadawad, Hung Duong
https://arxiv.org/abs/2508.02732

A Note on Code Quality Score: LLMs for Maintainable Large Codebases
Maintaining code quality in large-scale software systems presents significant challenges, particularly in settings where a large numbers of engineers work concurrently on a codebase. This paper introduces Code Quality Score (CQS) system to automatically detect issues with a set of code changes and provide actionable insights. At its core, the CQS system is powered by two Llama3 models, fine-tuned (with SFT and offline RL approaches), to a) detect common code quality issues related to coding bes…

Automaton Constrained Q-Learning
Anastasios Manganaris, Vittorio Giammarino, Ahmed H. Qureshi
https://arxiv.org/abs/2510.05061 https://arxiv.org/pdf/2510.0…

Automaton Constrained Q-Learning
Real-world robotic tasks often require agents to achieve sequences of goals while respecting time-varying safety constraints. However, standard Reinforcement Learning (RL) paradigms are fundamentally limited in these settings. A natural approach to these problems is to combine RL with Linear-time Temporal Logic (LTL), a formal language for specifying complex, temporally extended tasks and safety constraints. Yet, existing RL methods for LTL objectives exhibit poor empirical performance in compl…

P2P: A Poison-to-Poison Remedy for Reliable Backdoor Defense in LLMs
Shuai Zhao, Xinyi Wu, Shiqian Zhao, Xiaobao Wu, Zhongliang Guo, Yanhao Jia, Anh Tuan Luu
https://arxiv.org/abs/2510.04503

P2P: A Poison-to-Poison Remedy for Reliable Backdoor Defense in LLMs
During fine-tuning, large language models (LLMs) are increasingly vulnerable to data-poisoning backdoor attacks, which compromise their reliability and trustworthiness. However, existing defense strategies suffer from limited generalization: they only work on specific attack types or task settings. In this study, we propose Poison-to-Poison (P2P), a general and effective backdoor defense algorithm. P2P injects benign triggers with safe alternative labels into a subset of training samples and fi…

Quantile-Scaled Bayesian Optimization Using Rank-Only Feedback
Tunde Fahd Egunjobi
https://arxiv.org/abs/2510.03277 https://arxiv.org/pdf/2510.03277…

Quantile-Scaled Bayesian Optimization Using Rank-Only Feedback
Bayesian Optimization (BO) is widely used for optimizing expensive black-box functions, particularly in hyperparameter tuning. However, standard BO assumes access to precise objective values, which may be unavailable, noisy, or unreliable in real-world settings where only relative or rank-based feedback can be obtained. In this study, we propose Quantile-Scaled Bayesian Optimization (QS-BO), a principled rank-based optimization framework. QS-BO converts ranks into heteroscedastic Gaussian targe…

🎉 New Kitten¹ Release: A little housekeeping 🧹
Today’s release only concerns production servers:
• Kitten no longer counts all *hits* in its stats. You can still see which of your *pages* are most popular, etc., and see stats for missing URLs, etc., as before from either the web interface or the interactive shell, but not every hit is logged. Instead, you can see the latest 25 served routes in Kitten’s Settings (at /🐱/settings/state/requests/ via the web on your server).
- …

app/CHANGELOG.md at main
app - A web development kit that’s small, purrs, and loves you.

VirtLab: An AI-Powered System for Flexible, Customizable, and Large-scale Team Simulations
Mohammed Almutairi, Charles Chiang, Haoze Guo, Matthew Belcher, Nandini Banerjee, Maria Milkowski, Svitlana Volkova, Daniel Nguyen, Tim Weninger, Michael Yankoski, Trenton W. Ford, Diego Gomez-Zara
https://arxiv.org/abs/2508.04634

VirtLab: An AI-Powered System for Flexible, Customizable, and Large-scale Team Simulations
Simulating how team members collaborate within complex environments using Agentic AI is a promising approach to explore hypotheses grounded in social science theories and study team behaviors. We introduce VirtLab, a user-friendly, customizable, multi-agent, and scalable team simulation system that enables testing teams with LLM-based agents in spatial and temporal settings. This system addresses the current frameworks' design and technical limitations that do not consider flexible simulation s…

#EroticMusings August 2025
Wie angekündigt, darf ich mit Erlaubnis von @… das Hashtaggame ins Deutsche übersetzen und mache das um einen Monat versetzt. Hier sind die Fragen für den August:
Ich hänge sie gleich noch in Einzelposts an diesen hier dran, z…

An Adaptive Responsible AI Governance Framework for Decentralized Organizations
Kiana Jafari Meimandi, Anka Reuel, Gabriela Aranguiz-Dias, Hatim Rahama, Ala-Eddine Ayadi, Xavier Boullier, J\'er\'emy Verdo, Louis Montanie, Mykel Kochenderfer
https://arxiv.org/abs/2510.03368

An Adaptive Responsible AI Governance Framework for Decentralized Organizations
This paper examines the assessment challenges of Responsible AI (RAI) governance efforts in globally decentralized organizations through a case study collaboration between a leading research university and a multinational enterprise. While there are many proposed frameworks for RAI, their application in complex organizational settings with distributed decision-making authority remains underexplored. Our RAI assessment, conducted across multiple business units and AI use cases, reveals four key …

MoME: Mixture of Matryoshka Experts for Audio-Visual Speech Recognition
Umberto Cappellazzo, Minsu Kim, Pingchuan Ma, Honglie Chen, Xubo Liu, Stavros Petridis, Maja Pantic
https://arxiv.org/abs/2510.04136

MoME: Mixture of Matryoshka Experts for Audio-Visual Speech Recognition
Large language models (LLMs) have recently shown strong potential in audio-visual speech recognition (AVSR), but their high computational demands and sensitivity to token granularity limit their practicality in resource-constrained settings. Token compression methods can reduce inference cost, but they require fixing a compression rate in advance and produce a single fixed-length output, offering no flexibility to balance information density and efficiency at inference time. Matryoshka represen…

Characterizing Compute-Communication Overlap in GPU-Accelerated Distributed Deep Learning: Performance and Power Implications
Seonho Lee, Jihwan Oh, Junkyum Kim, Seokjin Go, Jongse Park, Divya Mahajan
https://arxiv.org/abs/2507.03114

Characterizing Compute-Communication Overlap in GPU-Accelerated Distributed Deep Learning: Performance and Power Implications
This paper provides an in-depth characterization of GPU-accelerated systems, to understand the interplay between overlapping computation and communication which is commonly employed in distributed training settings. Due to the large size of models, distributing them across multiple devices is required. Overlapping strategies, which enable concurrent computation and communication, are critical for mitigating communication bottlenecks and maximizing GPU utilization. However, the current consensus…

Riemannian Inexact Gradient Descent for Quadratic Discrimination
Uday Talwar, Meredith K. Kupinski, Afrooz Jalilzadeh
https://arxiv.org/abs/2507.04670 http…

Riemannian Inexact Gradient Descent for Quadratic Discrimination
We propose an inexact optimization algorithm on Riemannian manifolds, motivated by quadratic discrimination tasks in high-dimensional, low-sample-size (HDLSS) imaging settings. In such applications, gradient evaluations are often biased due to limited sample sizes. To address this, we introduce a novel Riemannian optimization algorithm that is robust to inexact gradient information and prove an $\mathcal O(1/K)$ convergence rate under standard assumptions. We also present a line search variant …

Discriminating Distal Ischemic Stroke from Seizure-Induced Stroke Mimics Using Dynamic Susceptibility Contrast MRI
Marijn Borghouts, Richard McKinley, Josien Pluim, Manuel K\"ostner, Roland Wiest, Ruisheng Su
https://arxiv.org/abs/2508.04404

Discriminating Distal Ischemic Stroke from Seizure-Induced Stroke Mimics Using Dynamic Susceptibility Contrast MRI
Distinguishing acute ischemic strokes (AIS) from stroke mimics (SMs), particularly in cases involving medium and small vessel occlusions, remains a significant diagnostic challenge. While computed tomography (CT) based protocols are commonly used in emergency settings, their sensitivity for detecting distal occlusions is limited. This study explores the potential of magnetic resonance perfusion (MRP) imaging as a tool for differentiating distal AIS from epileptic seizures, a prevalent SM. Using…

Streaming Max-Cut in General Metrics
Shaofeng H. -C. Jiang, Pan Peng, Haoze Wang
https://arxiv.org/abs/2510.04435 https://arxiv.org/pdf/2510.04435

Streaming Max-Cut in General Metrics
Max-Cut is a fundamental combinatorial optimization problem that has been studied in various computational settings. In this work, we initiate the study of its streaming complexity in general metric spaces with access to distance oracles. We give a $(1 + ε)$-approximation algorithm for estimating the Max-Cut value sliding-window streams using only poly-logarithmic space. This is the first sliding-window algorithm for Max-Cut even in Euclidean spaces, and it achieves a similar error-space trade…

KuaiLive: A Real-time Interactive Dataset for Live Streaming Recommendation
Changle Qu, Sunhao Dai, Ke Guo, Liqin Zhao, Yanan Niu, Xiao Zhang, Jun Xu
https://arxiv.org/abs/2508.05633

KuaiLive: A Real-time Interactive Dataset for Live Streaming Recommendation
Live streaming platforms have become a dominant form of online content consumption, offering dynamically evolving content, real-time interactions, and highly engaging user experiences. These unique characteristics introduce new challenges that differentiate live streaming recommendation from traditional recommendation settings and have garnered increasing attention from industry in recent years. However, research progress in academia has been hindered by the lack of publicly available datasets …

Penalized mixed models to adjust for batch effects and unobserved confounding in high dimensional regression
Yujing Lu, Patrick Breheny
https://arxiv.org/abs/2510.03531 https://…

Penalized mixed models to adjust for batch effects and unobserved confounding in high dimensional regression
Confounding can lead to spurious associations. Typically, one must observe confounders in order to adjust for them, but in high-dimensional settings, recent research has shown that it becomes possible to adjust even for unobserved confounders. The methods for carrying out these adjustments, however, have not been thoroughly investigated. In this study, we derive a confounding framework in which the signal, bias, and variability can be cleanly partitioned and quantified, thereby enabling simulat…

Inequality in the Age of Pseudonymity
Aviv Yaish, Nir Chemaya, Lin William Cong, Dahlia Malkhi
https://arxiv.org/abs/2508.04668 https://arxiv.org/pdf/2508.…

Inequality in the Age of Pseudonymity
Inequality measures such as the Gini coefficient are used to inform and motivate policymaking, and are increasingly applied to digital platforms. We analyze how measures fare in pseudonymous settings, as common to internet-based or blockchain-based platforms. One key challenge that arises is the ability of actors to create multiple fake identities under fictitious false names, also known as ``Sybils.'' While some actors may do so to preserve their privacy, we show that this can inadvertently di…

SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs
Dachuan Shi, Abedelkadir Asi, Keying Li, Xiangchi Yuan, Leyan Pan, Wenke Lee, Wen Xiao
https://arxiv.org/abs/2510.05069

SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs
Recent work shows that, beyond discrete reasoning through explicit chain-of-thought steps, which are limited by the boundaries of natural languages, large language models (LLMs) can also reason continuously in latent space, allowing richer information per step and thereby improving token efficiency. Despite this promise, latent reasoning still faces two challenges, especially in training-free settings: 1) purely latent reasoning broadens the search distribution by maintaining multiple implicit …

In-Context Policy Adaptation via Cross-Domain Skill Diffusion
Minjong Yoo, Woo Kyung Kim, Honguk Woo
https://arxiv.org/abs/2509.04535 https://arxiv.org/pdf…

In-Context Policy Adaptation via Cross-Domain Skill Diffusion
In this work, we present an in-context policy adaptation (ICPAD) framework designed for long-horizon multi-task environments, exploring diffusion-based skill learning techniques in cross-domain settings. The framework enables rapid adaptation of skill-based reinforcement learning policies to diverse target domains, especially under stringent constraints on no model updates and only limited target domain data. Specifically, the framework employs a cross-domain skill diffusion scheme, where domai…

ERDES: A Benchmark Video Dataset for Retinal Detachment and Macular Status Classification in Ocular Ultrasound
Pouyan Navard, Yasemin Ozkut, Srikar Adhikari, Elaine Situ-LaCasse, Josie Acu\~na, Adrienne Yarnish, Alper Yilmaz
https://arxiv.org/abs/2508.04735

ERDES: A Benchmark Video Dataset for Retinal Detachment and Macular Status Classification in Ocular Ultrasound
Retinal detachment (RD) is a vision-threatening condition that requires timely intervention to preserve vision. Macular involvement -- whether the macula is still intact (macula-intact) or detached (macula-detached) -- is the key determinant of visual outcomes and treatment urgency. Point-of-care ultrasound (POCUS) offers a fast, non-invasive, cost-effective, and accessible imaging modality widely used in diverse clinical settings to detect RD. However, ultrasound image interpretation is limite…

Refereed Learning
Ran Canetti, Ephraim Linder, Connor Wagaman
https://arxiv.org/abs/2510.05440 https://arxiv.org/pdf/2510.05440

Refereed Learning
We initiate an investigation of learning tasks in a setting where the learner is given access to two competing provers, only one of which is honest. Specifically, we consider the power of such learners in assessing purported properties of opaque models. Following prior work that considers the power of competing provers in different settings, we call this setting refereed learning. After formulating a general definition of refereed learning tasks, we show refereed learning protocols that obtai…

Are AI-Generated Fixes Secure? Analyzing LLM and Agent Patches on SWE-bench
Amirali Sajadi, Kostadin Damevski, Preetha Chatterjee
https://arxiv.org/abs/2507.02976

Are AI-Generated Fixes Secure? Analyzing LLM and Agent Patches on SWE-bench
Large Language Models (LLMs) and their agentic frameworks are increasingly adopted to automate software development tasks such as issue resolution and program repair. While prior work has identified security risks in LLM-generated code, most evaluations have focused on synthetic or isolated settings, leaving open questions about the security of these systems in real-world development contexts. In this study, we present the first large-scale security analysis of LLM-generated patches using 20,00…

Long-Term Experiences From Working with Extended Reality in the Wild
Verena Biener, Florian Jack Winston, Dieter Schmalstieg, Alexander Plopski
https://arxiv.org/abs/2509.05067 …

Long-Term Experiences From Working with Extended Reality in the Wild
Extended Reality (XR) is increasingly used as a productivity tool and recent commercial XR devices have even been specifically designed as productivity tools, or, at least, are heavily advertised for such purposes, such as the Apple Vision Pro (AVP), which has now been available for more than one year. In spite of what marketing suggests, research still lacks an understanding of the long-term usage of such devices in ecologically valid everyday settings, as most studies are conducted in very co…

Uncovering Synergistic Educational Injustices of COVID-19 and AI
Ahmad Banyasady (Malayer university)
https://arxiv.org/abs/2507.03095 https://

Uncovering Synergistic Educational Injustices of COVID-19 and AI
Grounded in critical realism and using narrative inquiry, this article explores this article explores the long-term consequences of the COVID-19 pandemic and the rapid proliferation of artificial intelligence within higher education. Through the analysis of student narratives collected in Iranian university settings, the study reveals that learning experiences during and after the pandemic, coupled with unprepared exposure to AI tools, have generated hidden yet impactful layers of educational i…

Also a big thanks to the unknown-to-me person who already made the first PRs to add more configs over the weekend: https://codeberg.org/gedankenstuecke/freshrss-fulltext-settings
I've also expanded the README a bit to hopefully provide more context 🙂

CogBench: A Large Language Model Benchmark for Multilingual Speech-Based Cognitive Impairment Assessment
Feng Rui, Zhiyao Luo, Wei Wang, Yuting Song, Yong Liu, Tingting Zhu, Jianqing Li, Xingyao Wang
https://arxiv.org/abs/2508.03360

CogBench: A Large Language Model Benchmark for Multilingual Speech-Based Cognitive Impairment Assessment
Automatic assessment of cognitive impairment from spontaneous speech offers a promising, non-invasive avenue for early cognitive screening. However, current approaches often lack generalizability when deployed across different languages and clinical settings, limiting their practical utility. In this study, we propose CogBench, the first benchmark designed to evaluate the cross-lingual and cross-site generalizability of large language models (LLMs) for speech-based cognitive impairment assessme…

Optimality and computational barriers in variable selection under dependence
Ming Gao, Bryon Aragam
https://arxiv.org/abs/2510.03990 https://arxiv.org/pdf/…

Optimality and computational barriers in variable selection under dependence
We study the optimal sample complexity of variable selection in linear regression under general design covariance, and show that subset selection is optimal while under standard complexity assumptions, efficient algorithms for this problem do not exist. Specifically, we analyze the variable selection problem and provide the optimal sample complexity with exact dependence on the problem parameters for both known and unknown sparsity settings. Moreover, we establish a sample complexity lower boun…

Non-Stationary Bandit Convex Optimization: An Optimal Algorithm with Two-Point Feedback
Chang He, Bo Jiang, Shuzhong Zhang
https://arxiv.org/abs/2508.04654 https://

Non-Stationary Bandit Convex Optimization: An Optimal Algorithm with Two-Point Feedback
This paper studies bandit convex optimization in non-stationary environments with two-point feedback, using dynamic regret as the performance measure. We propose an algorithm based on bandit mirror descent that extends naturally to non-Euclidean settings. Let $T$ be the total number of iterations and $\mathcal{P}_{T,p}$ the path variation with respect to the $\ell_p$-norm. In Euclidean space, our algorithm matches the optimal regret bound $\mathcal{O}(\sqrt{dT(1+\mathcal{P}_{T,2})})$, improving…

Semi-supervised inference for treatment heterogeneity
Yilizhati Anniwaer, Yuqian Zhang
https://arxiv.org/abs/2509.05048 https://arxiv.org/pdf/2509.05048

Semi-supervised inference for treatment heterogeneity
In causal inference, measuring treatment heterogeneity is crucial as it provides scientific insights into how treatments influence outcomes and guides personalized decision-making. In this work, we study semi-supervised settings where a labeled dataset is accompanied by a large unlabeled dataset, and develop semi-supervised estimators for two measures of treatment heterogeneity: the total treatment heterogeneity (TTH) and the explained treatment heterogeneity (ETH) of a simplified working model…

PLA: Prompt Learning Attack against Text-to-Image Generative Models
Xinqi Lyu, Yihao Liu, Yanjie Li, Bin Xiao
https://arxiv.org/abs/2508.03696 https://arxi…

PLA: Prompt Learning Attack against Text-to-Image Generative Models
Text-to-Image (T2I) models have gained widespread adoption across various applications. Despite the success, the potential misuse of T2I models poses significant risks of generating Not-Safe-For-Work (NSFW) content. To investigate the vulnerability of T2I models, this paper delves into adversarial attacks to bypass the safety mechanisms under black-box settings. Most previous methods rely on word substitution to search adversarial prompts. Due to limited search space, this leads to suboptimal p…

Benchmarking Uncertainty and its Disentanglement in multi-label Chest X-Ray Classification
Simon Baur, Wojciech Samek, Jackie Ma
https://arxiv.org/abs/2508.04457 https://…

Benchmarking Uncertainty and its Disentanglement in multi-label Chest X-Ray Classification
Reliable uncertainty quantification is crucial for trustworthy decision-making and the deployment of AI models in medical imaging. While prior work has explored the ability of neural networks to quantify predictive, epistemic, and aleatoric uncertainties using an information-theoretical approach in synthetic or well defined data settings like natural image classification, its applicability to real life medical diagnosis tasks remains underexplored. In this study, we provide an extensive uncerta…

Distributed Download from an External Data Source in Asynchronous Faulty Settings
John Augustine, Soumyottam Chatterjee, Valerie King, Manish Kumar, Shachar Meir, David Peleg
https://arxiv.org/abs/2509.03755

Distributed Download from an External Data Source in Asynchronous Faulty Settings
The distributedData Retrieval (DR) model consists of $k$ peers connected by a complete peer-to-peer communication network, and a trusted external data source that stores an array $\textbf{X}$ of $n$ bits ($n \gg k$). Up to $βk$ of the peers might fail in any execution (for $β\in [0, 1)$). Peers can obtain the information either by inexpensive messages passed among themselves or through expensive queries to the source array $\textbf{X}$. In the DR model, we focus on designing protocols that mi…

HyperSumm-RL: A Dialogue Summarization Framework for Modeling Leadership Perception in Social Robots
Subasish Das
https://arxiv.org/abs/2507.04160 https://…

HyperSumm-RL: A Dialogue Summarization Framework for Modeling Leadership Perception in Social Robots
This paper introduces HyperSumm-RL, a hypertext-aware summarization and interaction analysis framework designed to investigate human perceptions of social robot leadership through long-form dialogue. The system utilizes a structured Natural Language Processing (NLP) workflow that combines transformer-based long dialogue summarization, leadership style modeling, and user response analysis, enabling scalable evaluation of social robots in complex human-robot interaction (HRI) settings. Unlike pri…

I hate it that without even noticing, I've somehow managed to agree to this option being on 😡 ->
How to stop Microsoft Edge from monitoring your Chrome browser history and settings
https://www.pocnetwork.net/tips/how-to-sto

How to stop Microsoft Edge from monitoring your Chrome browser history and settings | Poc Network // Tech
It shouldn't be surprising that any company might be interested in spying on you. Analytics is one of the many ways these companies thrive by learning everything they can about their customers. Including their browsing habits within other browsers, which is exactly what Microsoft Edge browser does. You'd think a browser would be limited to

Speech LLMs in Low-Resource Scenarios: Data Volume Requirements and the Impact of Pretraining on High-Resource Languages
Seraphina Fong, Marco Matassoni, Alessio Brutti
https://arxiv.org/abs/2508.05149

Speech LLMs in Low-Resource Scenarios: Data Volume Requirements and the Impact of Pretraining on High-Resource Languages
Large language models (LLMs) have demonstrated potential in handling spoken inputs for high-resource languages, reaching state-of-the-art performance in various tasks. However, their applicability is still less explored in low-resource settings. This work investigates the use of Speech LLMs for low-resource Automatic Speech Recognition using the SLAM-ASR framework, where a trainable lightweight projector connects a speech encoder and a LLM. Firstly, we assess training data volume requirements t…

Beyond Regularization: Inherently Sparse Principal Component Analysis
Jan O. Bauer
https://arxiv.org/abs/2510.03729 https://arxiv.org/pdf/2510.03729…

Beyond Regularization: Inherently Sparse Principal Component Analysis
Sparse principal component analysis (sparse PCA) is a widely used technique for dimensionality reduction in multivariate analysis, addressing two key limitations of standard PCA. First, sparse PCA can be implemented in high-dimensional low sample size settings, such as genetic microarrays. Second, it improves interpretability as components are regularized to zero. However, over-regularization of sparse singular vectors can cause them to deviate greatly from the population singular vectors, pote…

EoH-S: Evolution of Heuristic Set using LLMs for Automated Heuristic Design
Fei Liu, Yilu Liu, Qingfu Zhang, Xialiang Tong, Mingxuan Yuan
https://arxiv.org/abs/2508.03082 https:…

EoH-S: Evolution of Heuristic Set using LLMs for Automated Heuristic Design
Automated Heuristic Design (AHD) using Large Language Models (LLMs) has achieved notable success in recent years. Despite the effectiveness of existing approaches, they only design a single heuristic to serve all problem instances, often inducing poor generalization across different distributions or settings. To address this issue, we propose Automated Heuristic Set Design (AHSD), a new formulation for LLM-driven AHD. The aim of AHSD is to automatically generate a small-sized complementary heur…

In the FreshRSS discussion section on GH someone suggested also including OPML files that can be easily imported by others.
I didn't even know FreshRSS has a custom OPML extension to include those configs, but that's amazing!
Added some already now! https://codeberg.org/gedankenstuecke/freshrss-fulltext-settings

LRQ-DiT: Log-Rotation Post-Training Quantization of Diffusion Transformers for Text-to-Image Generation
Lianwei Yang, Haokun Lin, Tianchen Zhao, Yichen Wu, Hongyu Zhu, Ruiqi Xie, Zhenan Sun, Yu Wang, Qingyi Gu
https://arxiv.org/abs/2508.03485

LRQ-DiT: Log-Rotation Post-Training Quantization of Diffusion Transformers for Text-to-Image Generation
Diffusion Transformers (DiTs) have achieved impressive performance in text-to-image generation. However, their high computational cost and large parameter sizes pose significant challenges for usage in resource-constrained scenarios. Post-training quantization (PTQ) is a promising solution to reduce memory usage and accelerate inference, but existing PTQ methods suffer from severe performance degradation under extreme low-bit settings. We identify two key obstacles to low-bit post-training quan…

Inverse Probability Weighting for Recurrent Event Models
Jiren Sun, Tobias Mutze, Richard Cook, Tianmeng Lyu
https://arxiv.org/abs/2507.04567 https://

Inverse Probability Weighting for Recurrent Event Models
Recurrent events are common and important clinical trial endpoints in many disease areas, e.g., cardiovascular hospitalizations in heart failure, relapses in multiple sclerosis, or exacerbations in asthma. During a trial, patients may experience intercurrent events, that is, events after treatment assignment which affect the interpretation or existence of the outcome of interest. In many settings, a treatment effect in the scenario in which the intercurrent event would not occur is of clinical …

AMANDA: Agentic Medical Knowledge Augmentation for Data-Efficient Medical Visual Question Answering
Ziqing Wang, Chengsheng Mao, Xiaole Wen, Yuan Luo, Kaize Ding
https://arxiv.org/abs/2510.02328

AMANDA: Agentic Medical Knowledge Augmentation for Data-Efficient Medical Visual Question Answering
Medical Multimodal Large Language Models (Med-MLLMs) have shown great promise in medical visual question answering (Med-VQA). However, when deployed in low-resource settings where abundant labeled data are unavailable, existing Med-MLLMs commonly fail due to their medical reasoning capability bottlenecks: (i) the intrinsic reasoning bottleneck that ignores the details from the medical image; (ii) the extrinsic reasoning bottleneck that fails to incorporate specialized medical knowledge. To addr…

Beyond Policy Optimization: A Data Curation Flywheel for Sparse-Reward Long-Horizon Planning
Yutong Wang, Pengliang Ji, Kaixin Li, Baolong Bi, Tao Feng, Guillaume Sartoretti
https://arxiv.org/abs/2508.03018

Beyond Policy Optimization: A Data Curation Flywheel for Sparse-Reward Long-Horizon Planning
Large Language Reasoning Models have demonstrated remarkable success on static tasks, yet their application to multi-round agentic planning in interactive environments faces two fundamental challenges. First, the intractable credit assignment problem renders conventional reinforcement learning ineffective in sparse-reward settings. Second, the computational overhead of verbose, step-by-step reasoning histories is prohibitive. To address these challenges, we propose BPO, a three-stage framework …

Fine-Tuning Jailbreaks under Highly Constrained Black-Box Settings: A Three-Pronged Approach
Xiangfang Li, Yu Wang, Bo Li
https://arxiv.org/abs/2510.01342 https://

Fine-Tuning Jailbreaks under Highly Constrained Black-Box Settings: A Three-Pronged Approach
With the rapid advancement of large language models (LLMs), ensuring their safe use becomes increasingly critical. Fine-tuning is a widely used method for adapting models to downstream tasks, yet it is vulnerable to jailbreak attacks. However, most existing studies focus on overly simplified attack scenarios, limiting their practical relevance to real-world defense settings. To make this risk concrete, we present a three-pronged jailbreak attack and evaluate it against provider defenses under a…

Practical considerations for Gaussian Process modeling for causal inference quasi-experimental studies with panel data
Sofia L. Vega, Rachel C. Nethery
https://arxiv.org/abs/2507.05128

Practical considerations for Gaussian Process modeling for causal inference quasi-experimental studies with panel data
Estimating causal effects in quasi-experiments with spatio-temporal panel data often requires adjusting for unmeasured confounding that varies across space and time. Gaussian Processes (GPs) offer a flexible, nonparametric modeling approach that can account for such complex dependencies through carefully chosen covariance kernels. In this paper, we provide a practical and interpretable framework for applying GPs to causal inference in panel data settings. We demonstrate how GPs generalize popul…

ToolVQA: A Dataset for Multi-step Reasoning VQA with External Tools
Shaofeng Yin, Ting Lei, Yang Liu
https://arxiv.org/abs/2508.03284 https://arxiv.org/pdf…

ToolVQA: A Dataset for Multi-step Reasoning VQA with External Tools
Integrating external tools into Large Foundation Models (LFMs) has emerged as a promising approach to enhance their problem-solving capabilities. While existing studies have demonstrated strong performance in tool-augmented Visual Question Answering (VQA), recent benchmarks reveal significant gaps in real-world tool-use proficiency, particularly in functionally diverse multimodal settings requiring multi-step reasoning. In this work, we introduce ToolVQA, a large-scale multimodal dataset compri…

FairLangProc: A Python package for fairness in NLP
Arturo P\'erez-Peralta, Sandra Ben\'itez-Pe\~na, Rosa E. Lillo
https://arxiv.org/abs/2508.03677 https://

FairLangProc: A Python package for fairness in NLP
The rise in usage of Large Language Models to near ubiquitousness in recent years has risen societal concern about their applications in decision-making contexts, such as organizational justice or healthcare. This, in turn, poses questions about the fairness of these models in critical settings, which leads to the developement of different procedures to address bias in Natural Language Processing. Although many datasets, metrics and algorithms have been proposed to measure and mitigate harmful …

MissDDIM: Deterministic and Efficient Conditional Diffusion for Tabular Data Imputation
Youran Zhou, Mohamed Reda Bouadjenek, Sunil Aryal
https://arxiv.org/abs/2508.03083 https:…

MissDDIM: Deterministic and Efficient Conditional Diffusion for Tabular Data Imputation
Diffusion models have recently emerged as powerful tools for missing data imputation by modeling the joint distribution of observed and unobserved variables. However, existing methods, typically based on stochastic denoising diffusion probabilistic models (DDPMs), suffer from high inference latency and variable outputs, limiting their applicability in real-world tabular settings. To address these deficiencies, we present in this paper MissDDIM, a conditional diffusion framework that adapts Deno…

Full-History Graphs with Edge-Type Decoupled Networks for Temporal Reasoning
Osama Mohammed, Jiaxin Pan, Mojtaba Nayyeri, Daniel Hern\'andez, Steffen Staab
https://arxiv.org/abs/2508.03251

Full-History Graphs with Edge-Type Decoupled Networks for Temporal Reasoning
Modeling evolving interactions among entities is critical in many real-world tasks. For example, predicting driver maneuvers in traffic requires tracking how neighboring vehicles accelerate, brake, and change lanes relative to one another over consecutive frames. Likewise, detecting financial fraud hinges on following the flow of funds through successive transactions as they propagate through the network. Unlike classic time-series forecasting, these settings demand reasoning over who interacts…