Tootfinder

Vision Transformers: the threat of realistic adversarial patches
Kasper Cools, Clara Maathuis, Alexander M. van Oers, Claudia S. H\"ubner, Nikos Deligiannis, Marijke Vandewal, Geert De Cubber
https://arxiv.org/abs/2509.21084

Vision Transformers: the threat of realistic adversarial patches
The increasing reliance on machine learning systems has made their security a critical concern. Evasion attacks enable adversaries to manipulate the decision-making processes of AI systems, potentially causing security breaches or misclassification of targets. Vision Transformers (ViTs) have gained significant traction in modern machine learning due to increased 1) performance compared to Convolutional Neural Networks (CNNs) and 2) robustness against adversarial perturbations. However, ViTs rem…

A Taxonomy of Data Risks in AI and Quantum Computing (QAI) - A Systematic Review
Grace Billiris, Asif Gill, Madhushi Bandara
https://arxiv.org/abs/2509.20418 https://

A Taxonomy of Data Risks in AI and Quantum Computing (QAI) - A Systematic Review
Quantum Artificial Intelligence (QAI), the integration of Artificial Intelligence (AI) and Quantum Computing (QC), promises transformative advances, including AI-enabled quantum cryptography and quantum-resistant encryption protocols. However, QAI inherits data risks from both AI and QC, creating complex privacy and security vulnerabilities that are not systematically studied. These risks affect the trustworthiness and reliability of AI and QAI systems, making their understanding critical. This…

Car bomb in Moscow kills senior Russian general
-- the third such attack in a year

Lt. Gen. Fanil Sarvarov, who headed the army’s
Directorate for Operational Training,
died as a result of his injuries Monday,
when a bomb planted under his car exploded,
making him the third general killed by a car bomb in the past year.
His department plays a critical role in planning military operations and ensuring the combat readiness of the Russian army.

Car bomb in Moscow kills senior Russian general
Lt. Gen. Fanil Sarvarov, 56, had headed the Russian army’s Directorate for Operational Training and participated in Moscow’s war in Ukraine, investigators said.

Jerry Jones highly critical of Matt Eberflus, making Cowboys DC's seat even hotter https://www.sportingnews.com/us/nfl/dallas-cowboys/news/jerry-jones-highly-critical-matt-eberflus-cowboys-seat-…

ASPI 12.11.2025
Resilient critical infrastructure is its own deterrence
"The ability to absorb disruption and keep operating isn’t just a technical virtue; it’s a signal. If an adversary believes that disruption will be temporary and costly to sustain, the incentive to strike is reduced."
ht…

Easy Adaptation: An Efficient Task-Specific Knowledge Injection Method for Large Models in Resource-Constrained Environments
Dong Chen, Zhengqing Hu, Shixing Zhao, Yibo Guo
https://arxiv.org/abs/2512.17771 https://arxiv.org/pdf/2512.17771 https://arxiv.org/html/2512.17771
arXiv:2512.17771v1 Announce Type: new
Abstract: While the enormous parameter scale endows Large Models (LMs) with unparalleled performance, it also limits their adaptability across specific tasks. Parameter-Efficient Fine-Tuning (PEFT) has emerged as a critical approach for effectively adapting LMs to a diverse range of downstream tasks. However, existing PEFT methods face two primary challenges: (1) High resource cost. Although PEFT methods significantly reduce resource demands compared to full fine-tuning, it still requires substantial time and memory, making it impractical in resource-constrained environments. (2) Parameter dependency. PEFT methods heavily rely on updating a subset of parameters associated with LMs to incorporate task-specific knowledge. Yet, due to increasing competition in the LMs landscape, many companies have adopted closed-source policies for their leading models, offering access only via Application Programming Interface (APIs). Whereas, the expense is often cost-prohibitive and difficult to sustain, as the fine-tuning process of LMs is extremely slow. Even if small models perform far worse than LMs in general, they can achieve superior results on particular distributions while requiring only minimal resources. Motivated by this insight, we propose Easy Adaptation (EA), which designs Specific Small Models (SSMs) to complement the underfitted data distribution for LMs. Extensive experiments show that EA matches the performance of PEFT on diverse tasks without accessing LM parameters, and requires only minimal resources.
toXiv_bot_toot

So happy some of the folks from Brooklyn Nine -Nine are making a PI show. They can be much more critical of police work with a show about that.
https://deadline.com/2025/11/brooklyn-nine-nine-duo-dan-goor-pi-comedy-nbc-pilot-1236614348/…

HALF: Harm-Aware LLM Fairness Evaluation Aligned with Deployment
Ali Mekky, Omar El Herraoui, Preslav Nakov, Yuxia Wang
https://arxiv.org/abs/2510.12217 https://

HALF: Harm-Aware LLM Fairness Evaluation Aligned with Deployment
Large language models (LLMs) are increasingly deployed across high-impact domains, from clinical decision support and legal analysis to hiring and education, making fairness and bias evaluation before deployment critical. However, existing evaluations lack grounding in real-world scenarios and do not account for differences in harm severity, e.g., a biased decision in surgery should not be weighed the same as a stylistic bias in text summarization. To address this gap, we introduce HALF (Harm-A…

Risk level dependent Minimax Quantile lower bounds for Interactive Statistical Decision Making
Raghav Bongole, Amirreza Zamani, Tobias J. Oechtering, Mikael Skoglund
https://arxiv.org/abs/2510.05808

Risk level dependent Minimax Quantile lower bounds for Interactive Statistical Decision Making
Minimax risk and regret focus on expectation, missing rare failures critical in safety-critical bandits and reinforcement learning. Minimax quantiles capture these tails. Three strands of prior work motivate this study: minimax-quantile bounds restricted to non-interactive estimation; unified interactive analyses that focus on expected risk rather than risk level specific quantile bounds; and high-probability bandit bounds that still lack a quantile-specific toolkit for general interactive prot…

Comparing Knowledge Source Integration Methods for Optimizing Healthcare Knowledge Fusion in Rescue Operation
Mubaris Nadeem, Madjid Fathi
https://arxiv.org/abs/2510.09223 https…

Comparing Knowledge Source Integration Methods for Optimizing Healthcare Knowledge Fusion in Rescue Operation
In the field of medicine and healthcare, the utilization of medical expertise, based on medical knowledge combined with patients' health information is a life-critical challenge for patients and health professionals. The within-laying complexity and variety form the need for a united approach to gather, analyze, and utilize existing knowledge of medical treatments, and medical operations to provide the ability to present knowledge for the means of accurate patient-driven decision-making. One wa…

OFP-Repair: Repairing Floating-point Errors via Original-Precision Arithmetic
Youshuai Tan, Zishuo Ding, Jinfu Chen, Weiyi Shang
https://arxiv.org/abs/2510.09938 https://…

OFP-Repair: Repairing Floating-point Errors via Original-Precision Arithmetic
Errors in floating-point programs can lead to severe consequences, particularly in critical domains such as military, aerospace, and financial systems, making their repair a crucial research problem. In practice, some errors can be fixed using original-precision arithmetic, while others require high-precision computation. Developers often avoid addressing the latter due to excessive computational resources required. However, they sometimes struggle to distinguish between these two types of erro…

The Ethics Engine: A Modular Pipeline for Accessible Psychometric Assessment of Large Language Models
Jake Van Clief, Constantine Kyritsopoulos
https://arxiv.org/abs/2510.11742 …

The Ethics Engine: A Modular Pipeline for Accessible Psychometric Assessment of Large Language Models
As Large Language Models increasingly mediate human communication and decision-making, understanding their value expression becomes critical for research across disciplines. This work presents the Ethics Engine, a modular Python pipeline that transforms psychometric assessment of LLMs from a technically complex endeavor into an accessible research tool. The pipeline demonstrates how thoughtful infrastructure design can expand participation in AI research, enabling investigators across cognitive…

CDE: Concept-Driven Exploration for Reinforcement Learning
Le Mao, Andrew H. Liu, Renos Zabounidis, Zachary Kingston, Joseph Campbell
https://arxiv.org/abs/2510.08851 https://…

CDE: Concept-Driven Exploration for Reinforcement Learning
Intelligent exploration remains a critical challenge in reinforcement learning (RL), especially in visual control tasks. Unlike low-dimensional state-based RL, visual RL must extract task-relevant structure from raw pixels, making exploration inefficient. We propose Concept-Driven Exploration (CDE), which leverages a pre-trained vision-language model (VLM) to generate object-centric visual concepts from textual task descriptions as weak, potentially noisy supervisory signals. Rather than direct…

The assumption is that one day large language models and other related AI technologies fostered by Google Gemini and OpenAI ChatGPT actually will be a great and infallible productivity tool for genuine work.
It already is decent for providing basic overviews of highly-covered, well-sourced topics, even as hallucinations and sycophancy continue to dog the tech, particularly in situations where accountability is more critical.
Despite today's downsides,
many companies ar…

Analysis: OpenAI is a loss-making machine, with estimates that it has no road to profitability by 2030 — and will need a further $207 billion in funding even if it gets there
Don't call it a bubble! Loss-making monster OpenAI is on the hook for $1.4 trillion (with a T) in compute commitments. How can this go on?

I can help but feel this "feature" should raise more concerns than it does alleviate them.
"Accelerated recovery for managing public DNS records addresses this need by targeting DNS changes that customers can make within 60 minutes of a service disruption in the US East (N. Virginia) Region."

Amazon Route 53 launches Accelerated recovery for managing public DNS records | Amazon Web Services
Amazon Route 53 now offers Accelerated recovery for managing public DNS records, a new business continuity feature designed to provide a 60-minute recovery time objective (RTO) during service disruptions in the US East (N. Virginia) Region. This enhancement enables customers to continue making critical DNS changes and provisioning infrastructure even during regional outages, ensuring greater resilience for mission-critical applications.

Universal Discrete-Domain Speech Enhancement
Fei Liu, Yang Ai, Ye-Xin Lu, Rui-Chen Zheng, Hui-Peng Du, Zhen-Hua Ling
https://arxiv.org/abs/2510.09974 https://

Universal Discrete-Domain Speech Enhancement
In real-world scenarios, speech signals are inevitably corrupted by various types of interference, making speech enhancement (SE) a critical task for robust speech processing. However, most existing SE methods only handle a limited range of distortions, such as additive noise, reverberation, or band limitation, while the study of SE under multiple simultaneous distortions remains limited. This gap affects the generalization and practical usability of SE methods in real-world environments.To add…

Two-dimensional superconducting diode effect in topological insulator/superconductor heterostructure
Soma Nagahama, Yuki Sato, Minoru Kawamura, Ilya Belopolski, Ryutaro Yoshimi, Atsushi Tsukazaki, Naoya Kanazawa, Kei S Takahashi, Masashi Kawasaki, Yoshinori Tokura
https://arxiv.org/abs/2510.09921

Two-dimensional superconducting diode effect in topological insulator/superconductor heterostructure
The superconducting diode effect (SDE) is characterized by the nonreciprocity of Cooper-pair motion with respect to current direction. In three-dimensional (3D) materials, SDE results in a critical current that varies with direction, making the effect distinctly observable: the material exhibits superconductivity in one direction while behaving as a resistive metal in the opposite direction. However, in genuinely two-dimensional (2D) materials, the critical current density is theoretically zero…

FeNOMS: Enhancing Open Modification Spectral Library Search with In-Storage Processing on Ferroelectric NAND (FeNAND) Flash
Sumukh Pinge, Ashkan Moradifirouzabadi, Keming Fan, Prasanna Venkatesan Ravindran, Tanvir H. Pantha, Po-Kai Hsu, Zheyu Li, Weihong Xu, Zihan Xia, Flavio Ponzina, Winston Chern, Taeyoung Song, Priyankka Ravikumar, Mengkun Tian, Lance Fernandes, Huy Tran, Hari Jayasankar, Hang Chen, Chinsung Park, Amrit Garlapati, Kijoon Kim, Jongho Woo, Suhwan Lim, Kwangsoo Kim, Wa…

FeNOMS: Enhancing Open Modification Spectral Library Search with In-Storage Processing on Ferroelectric NAND (FeNAND) Flash
The rapid expansion of mass spectrometry (MS) data, now exceeding hundreds of terabytes, poses significant challenges for efficient, large-scale library search - a critical component for drug discovery. Traditional processors struggle to handle this data volume efficiently, making in-storage computing (ISP) a promising alternative. This work introduces an ISP architecture leveraging a 3D Ferroelectric NAND (FeNAND) structure, providing significantly higher density, faster speeds, and lower volt…

Establishing assembly-oriented modular product architectures through Design for Assembly enhanced Modular Function Deployment
Fabio Marco Monetti, Adam Lundstr\"om, Colin de Kwant, Magnus Gyllenskepp, Antonio Maffei
https://arxiv.org/abs/2510.11089

Establishing assembly-oriented modular product architectures through Design for Assembly enhanced Modular Function Deployment
Modular product design has become a strategic enabler for companies seeking to balance product variety, operational efficiency, and market responsiveness, making the alignment between modular architecture and manufacturing considerations increasingly critical. Modular Function Deployment (MFD) is a widely adopted method for defining modular product architectures, yet it lacks systematic support for assembly considerations during early concept and system-level development. This limitation increa…

Why Sean McVay felt 'sick' over failed fourth-down play call in Rams' overtime loss to 49ers

https://www.cbssports.com/nfl/news/why-sea

Knowledge-Guided Machine Learning Models to Upscale Evapotranspiration in the U.S. Midwest
Aleksei Rozanov, Samikshya Subedi, Vasudha Sharma, Bryan C. Runck
https://arxiv.org/abs/2510.11505

Knowledge-Guided Machine Learning Models to Upscale Evapotranspiration in the U.S. Midwest
Evapotranspiration (ET) plays a critical role in the land-atmosphere interactions, yet its accurate quantification across various spatiotemporal scales remains a challenge. In situ measurement approaches, like eddy covariance (EC) or weather station-based ET estimation, allow for measuring ET at a single location. Agricultural uses of ET require estimates for each field over broad areas, making it infeasible to deploy sensing systems at each location. This study integrates tree-based and knowle…

AAR 2025, Yogācāra Studies Unit, upcoming panels
https://ift.tt/1yFJmv2
Critical Survey (Vol. 34, Issue 2) Dear Colleague, The latest issue of Critical Survey…
via Input 4 RELCFP https://

HackWorld: Evaluating Computer-Use Agents on Exploiting Web Application Vulnerabilities
Xiaoxue Ren, Penghao Jiang, Kaixin Li, Zhiyong Huang, Xiaoning Du, Jiaojiao Jiang, Zhenchang Xing, Jiamou Sun, Terry Yue Zhuo
https://arxiv.org/abs/2510.12200

HackWorld: Evaluating Computer-Use Agents on Exploiting Web Application Vulnerabilities
Web applications are prime targets for cyberattacks as gateways to critical services and sensitive data. Traditional penetration testing is costly and expertise-intensive, making it difficult to scale with the growing web ecosystem. While language model agents show promise in cybersecurity, modern web applications demand visual understanding, dynamic content handling, and multi-step interactions that only computer-use agents (CUAs) can perform. Yet, their ability to discover and exploit vulnera…

Anduril cofounder Palmer Lucked defended the use of AI technology to make life-and-death decisions in war on Sunday.
A group of defense tech startups that includes Anduril,
along with traditional defense companies,
is developing autonomous AI weapons and tools for use in conflicts around the world,
-- worrying some who say the technology is not ready for such high-stakes environments.
"When it comes to life and death decision-making, I think that it is too …

Anduril's Palmer Luckey makes an ethical case for using AI in war: 'There is no moral high ground in using inferior technology'
For "life and death decision-making," Anduril cofounder Palmer Luckey says it's "too critical" to not use the most advanced technology.

Beyond Canonical Rounds: Communication Abstractions for Optimal Byzantine Resilience
Hagit Attiya, Itay Flam, Jennifer L. Welch
https://arxiv.org/abs/2510.04310 https://

Beyond Canonical Rounds: Communication Abstractions for Optimal Byzantine Resilience
We study communication abstractions for asynchronous Byzantine fault tolerance with optimal failure resilience, where $n > 3f$. Two classic patterns -- canonical asynchronous rounds and communication-closed layers -- have long been considered as general frameworks for designing distributed algorithms, making asynchronous executions appear synchronous and enabling modular reasoning. We show that these patterns are inherently limited in the critical resilience regime $3f < n \le 5f$. Several ke…

Cracking CodeWhisperer: Analyzing Developers' Interactions and Patterns During Programming Tasks
Jeena Javahar, Tanya Budhrani, Manaal Basha, Cleidson R. B. de Souza, Ivan Beschastnikh, Gema Rodriguez-Perez
https://arxiv.org/abs/2510.11516

Cracking CodeWhisperer: Analyzing Developers' Interactions and Patterns During Programming Tasks
The use of AI code-generation tools is becoming increasingly common, making it important to understand how software developers are adopting these tools. In this study, we investigate how developers engage with Amazon's CodeWhisperer, an LLM-based code-generation tool. We conducted two user studies with two groups of 10 participants each, interacting with CodeWhisperer - the first to understand which interactions were critical to capture and the second to collect low-level interaction data using…

MAGIC-MASK: Multi-Agent Guided Inter-Agent Collaboration with Mask-Based Explainability for Reinforcement Learning
Maisha Maliha, Dean Hougen
https://arxiv.org/abs/2510.00274 ht…

MAGIC-MASK: Multi-Agent Guided Inter-Agent Collaboration with Mask-Based Explainability for Reinforcement Learning
Understanding the decision-making process of Deep Reinforcement Learning agents remains a key challenge for deploying these systems in safety-critical and multi-agent environments. While prior explainability methods like StateMask, have advanced the identification of critical states, they remain limited by computational cost, exploration coverage, and lack of adaptation to multi-agent settings. To overcome these limitations, we propose a mathematically grounded framework, MAGIC-MASK (Multi-Agen…

Red Lines and Grey Zones in the Fog of War: Benchmarking Legal Risk, Moral Harm, and Regional Bias in Large Language Model Military Decision-Making
Toby Drinkall
https://arxiv.org/abs/2510.03514

Red Lines and Grey Zones in the Fog of War: Benchmarking Legal Risk, Moral Harm, and Regional Bias in Large Language Model Military Decision-Making
As military organisations consider integrating large language models (LLMs) into command and control (C2) systems for planning and decision support, understanding their behavioural tendencies is critical. This study develops a benchmarking framework for evaluating aspects of legal and moral risk in targeting behaviour by comparing LLMs acting as agents in multi-turn simulated conflict. We introduce four metrics grounded in International Humanitarian Law (IHL) and military doctrine: Civilian Tar…

TeV Emission from PSR B1055-52 with HESS: Evidence for a Pulsar Halo
Tina Wach (for the H.E.S.S. Collaboration), Alison M. W. Mitchell (for the H.E.S.S. Collaboration)
https://arxiv.org/abs/2510.02802 …

TeV Emission from PSR B1055-52 with HESS: Evidence for a Pulsar Halo
Pulsar halos are a recently identified class of TeV $γ$-ray sources, offering valuable insights into the evolution of pulsar systems at the highest energies. However, only a handful of such sources have been detected so far, making each new identification critical for understanding the properties of the population as a whole. We report the first detection of extended very-high-energy (VHE) $γ$-ray emission around PSR~B1055$-$52 using observations from the H.E.S.S. array. This middle-aged puls…

Addressing Methodological Uncertainty in MCDM with a Systematic Pipeline Approach to Data Transformation Sensitivity Analysis
Juan B. Cabral, Alvaro Roy Schachner
https://arxiv.org/abs/2509.24996

Addressing Methodological Uncertainty in MCDM with a Systematic Pipeline Approach to Data Transformation Sensitivity Analysis
Multicriteria decision-making methods exhibit critical dependence on the choice of normalization techniques, where different selections can alter 20-40% of the final rankings. Current practice is characterized by the ad-hoc selection of methods without systematic robustness evaluation. We present a framework that addresses this methodological uncertainty through automated exploration of the scaling transformation space. The implementation leverages the existing Scikit-Criteria infrastructure to…

AutoPK: Leveraging LLMs and a Hybrid Similarity Metric for Advanced Retrieval of Pharmacokinetic Data from Complex Tables and Documents
Hossein Sholehrasa, Amirhossein Ghanaatian, Doina Caragea, Lisa A. Tell, Jim E. Riviere, Majid Jaberi-Douraki
https://arxiv.org/abs/2510.00039

AutoPK: Leveraging LLMs and a Hybrid Similarity Metric for Advanced Retrieval of Pharmacokinetic Data from Complex Tables and Documents
Pharmacokinetics (PK) plays a critical role in drug development and regulatory decision-making for human and veterinary medicine, directly affecting public health through drug safety and efficacy assessments. However, PK data are often embedded in complex, heterogeneous tables with variable structures and inconsistent terminologies, posing significant challenges for automated PK data retrieval and standardization. AutoPK, a novel two-stage framework for accurate and scalable extraction of PK da…

Graph Conditioned Diffusion for Controllable Histopathology Image Generation
Sarah Cechnicka, Matthew Baugh, Weitong Zhang, Mischa Dombrowski, Zhe Li, Johannes C. Paetzold, Candice Roufosse, Bernhard Kainz
https://arxiv.org/abs/2510.07129

Graph Conditioned Diffusion for Controllable Histopathology Image Generation
Recent advances in Diffusion Probabilistic Models (DPMs) have set new standards in high-quality image synthesis. Yet, controlled generation remains challenging, particularly in sensitive areas such as medical imaging. Medical images feature inherent structure such as consistent spatial arrangement, shape or texture, all of which are critical for diagnosis. However, existing DPMs operate in noisy latent spaces that lack semantic structure and strong priors, making it difficult to ensure meaningf…

Staged Event Trees for Transparent Treatment Effect Estimation
Gherardo Varando, Manuele Leonelli, Jordi Cerd\`a-Bautista, Vasileios Sitokonstantinou, Gustau Camps-Valls
https://arxiv.org/abs/2509.26265

Staged Event Trees for Transparent Treatment Effect Estimation
Average and conditional treatment effects are fundamental causal quantities used to evaluate the effectiveness of treatments in various critical applications, including clinical settings and policy-making. Beyond the gold-standard estimators from randomized trials, numerous methods have been proposed to estimate treatment effects using observational data. In this paper, we provide a novel characterization of widely used causal inference techniques within the framework of staged event trees, dem…

Polarization Domain Mapping From 4D-STEM Using Deep Learning
Fintan G. Hardy, Sinead M. Griffin, Mariana Palos, Yaqi Li, Geri Topore, Aron Walsh, Michele Shelly Conroy
https://arxiv.org/abs/2510.00693 …

Polarization Domain Mapping From 4D-STEM Using Deep Learning
Polarization in ferroelectric domains arises from atomic-scale structural variations that govern macroscopic functionalities. The interfaces between these domains known as domain walls host distinct physical responses, making their identification and control critical. Four dimensional scanning transmission electron microscopy (4DSTEM) enables simultaneous acquisition of real and reciprocal-space information at the atomic scale, offering a powerful platform for domain mapping. However, conventio…

Orochi: Versatile Biomedical Image Processor
Gaole Dai, Chenghao Zhou, Yu Zhou, Rongyu Zhang, Yuan Zhang, Chengkai Hou, Tiejun Huang, Jianxu Chen, Shanghang Zhang
https://arxiv.org/abs/2509.22583

Orochi: Versatile Biomedical Image Processor
Deep learning has emerged as a pivotal tool for accelerating research in the life sciences, with the low-level processing of biomedical images (e.g., registration, fusion, restoration, super-resolution) being one of its most critical applications. Platforms such as ImageJ (Fiji) and napari have enabled the development of customized plugins for various models. However, these plugins are typically based on models that are limited to specific tasks and datasets, making them less practical for biol…

StaMo: Unsupervised Learning of Generalizable Robot Motion from Compact State Representation
Mingyu Liu, Jiuhe Shu, Hui Chen, Zeju Li, Canyu Zhao, Jiange Yang, Shenyuan Gao, Hao Chen, Chunhua Shen
https://arxiv.org/abs/2510.05057

StaMo: Unsupervised Learning of Generalizable Robot Motion from Compact State Representation
A fundamental challenge in embodied intelligence is developing expressive and compact state representations for efficient world modeling and decision making. However, existing methods often fail to achieve this balance, yielding representations that are either overly redundant or lacking in task-critical information. We propose an unsupervised approach that learns a highly compressed two-token state representation using a lightweight encoder and a pre-trained Diffusion Transformer (DiT) decoder…

A-MemGuard: A Proactive Defense Framework for LLM-Based Agent Memory
Qianshan Wei, Tengchao Yang, Yaochen Wang, Xinfeng Li, Lijun Li, Zhenfei Yin, Yi Zhan, Thorsten Holz, Zhiqiang Lin, XiaoFeng Wang
https://arxiv.org/abs/2510.02373

A-MemGuard: A Proactive Defense Framework for LLM-Based Agent Memory
Large Language Model (LLM) agents use memory to learn from past interactions, enabling autonomous planning and decision-making in complex environments. However, this reliance on memory introduces a critical security risk: an adversary can inject seemingly harmless records into an agent's memory to manipulate its future behavior. This vulnerability is characterized by two core aspects: First, the malicious effect of injected records is only activated within a specific context, making them hard t…

AutoQual: An LLM Agent for Automated Discovery of Interpretable Features for Review Quality Assessment
Xiaochong Lan, Jie Feng, Yinxing Liu, Xinlei Shi, Yong Li
https://arxiv.org/abs/2510.08081

AutoQual: An LLM Agent for Automated Discovery of Interpretable Features for Review Quality Assessment
Ranking online reviews by their intrinsic quality is a critical task for e-commerce platforms and information services, impacting user experience and business outcomes. However, quality is a domain-dependent and dynamic concept, making its assessment a formidable challenge. Traditional methods relying on hand-crafted features are unscalable across domains and fail to adapt to evolving content patterns, while modern deep learning approaches often produce black-box models that lack interpretabili…

The Kremlin has heaped praise on Donald Trump’s latest national security strategy,
calling it an encouraging change of policy
that largely aligns with Russian thinking.
The remarks follow the publication of a White House document on Friday
💥that criticises the EU and says Europeis at risk of “civilisational erasure”,
💥while making clear the US is keen to establish better relations with Russia.
🔥“The adjustments that we see correspond in many ways to our vi…

MARLIN: Multi-Agent Reinforcement Learning with Murmuration Intelligence and LLM Guidance for Reservoir Management
Heming Fu, Guojun Xiong, Jian Li, Shan Lin
https://arxiv.org/abs/2509.25034

MARLIN: Multi-Agent Reinforcement Learning with Murmuration Intelligence and LLM Guidance for Reservoir Management
As climate change intensifies extreme weather events, water disasters pose growing threats to global communities, making adaptive reservoir management critical for protecting vulnerable populations and ensuring water security. Modern water resource management faces unprecedented challenges from cascading uncertainties propagating through interconnected reservoir networks. These uncertainties, rooted in physical water transfer losses and environmental variability, make precise control difficult.…

Cowboys’ George Pickens Claps Back At Ex All-Pro After Criticism https://heavy.com/sports/nfl/dallas-cowboys/george-pickens-responds-richard-sherman-criticism/?adt_ei=[email]

KnowGuard: Knowledge-Driven Abstention for Multi-Round Clinical Reasoning
Xilin Dang, Kexin Chen, Xiaorui Su, Ayush Noori, I\~naki Arango, Lucas Vittor, Xinyi Long, Yuyang Du, Marinka Zitnik, Pheng Ann Heng
https://arxiv.org/abs/2509.24816

KnowGuard: Knowledge-Driven Abstention for Multi-Round Clinical Reasoning
In clinical practice, physicians refrain from making decisions when patient information is insufficient. This behavior, known as abstention, is a critical safety mechanism preventing potentially harmful misdiagnoses. Recent investigations have reported the application of large language models (LLMs) in medical scenarios. However, existing LLMs struggle with the abstentions, frequently providing overconfident responses despite incomplete information. This limitation stems from conventional abste…

LLMs as Policy-Agnostic Teammates: A Case Study in Human Proxy Design for Heterogeneous Agent Teams
Aju Ani Justus, Chris Baber
https://arxiv.org/abs/2510.06151 https://

LLMs as Policy-Agnostic Teammates: A Case Study in Human Proxy Design for Heterogeneous Agent Teams
A critical challenge in modelling Heterogeneous-Agent Teams is training agents to collaborate with teammates whose policies are inaccessible or non-stationary, such as humans. Traditional approaches rely on expensive human-in-the-loop data, which limits scalability. We propose using Large Language Models (LLMs) as policy-agnostic human proxies to generate synthetic data that mimics human decision-making. To evaluate this, we conduct three experiments in a grid-world capture game inspired by Sta…

Private and public school efficiency gaps in Latin America-A combined DEA and machine learning approach based on PISA 2022
Marcos Delprato
https://arxiv.org/abs/2509.25353 https…

Private and public school efficiency gaps in Latin America-A combined DEA and machine learning approach based on PISA 2022
Latin America's education systems are fragmented and segregated, with substantial differences by school type. The concept of school efficiency (the ability of school to produce the maximum level of outputs given available resources) is policy relevant due to scarcity of resources in the region. Knowing whether private and public schools are making an efficient use of resources -and which are the leading drivers of efficiency- is critical, even more so after the learning crisis brought by the CO…

A former Clemson University professor is claiming his firing over a post about the killing of conservative activist Charlie Kirk was unconstitutional, according to a lawsuit filed in federal court Friday.
Joshua Bregy, an assistant professor in the Department of Environmental Engineering and Earth Sciences,
was one of three employees at Clemson fired for making social media posts critical of Kirk after the activist was shot and killed Sept. 10 during a speaking event at a Utah co…

Learning Stability Certificate for Robotics in Real-World Environments
Zhe Shen
https://arxiv.org/abs/2510.03123 https://arxiv.org/pdf/2510.03123

Learning Stability Certificate for Robotics in Real-World Environments
Stability certificates play a critical role in ensuring the safety and reliability of robotic systems. However, deriving these certificates for complex, unknown systems has traditionally required explicit knowledge of system dynamics, often making it a daunting task. This work introduces a novel framework that learns a Lyapunov function directly from trajectory data, enabling the certification of stability for autonomous systems without needing detailed system models. By parameterizing the Lyap…

Redesigning GROMACS Halo Exchange: Improving Strong Scaling with GPU-initiated NVSHMEM
Mahesh Doijade, Andrey Alekseenko, Ania Brown, Alan Gray, Szil\'ard P\'all
https://arxiv.org/abs/2509.21527

Redesigning GROMACS Halo Exchange: Improving Strong Scaling with GPU-initiated NVSHMEM
Improving time-to-solution in molecular dynamics simulations often requires strong scaling due to fixed-sized problems. GROMACS is highly latency-sensitive, with peak iteration rates in the sub-millisecond, making scalability on heterogeneous supercomputers challenging. MPI's CPU-centric nature introduces additional latencies on GPU-resident applications' critical path, hindering GPU utilization and scalability. To address these limitations, we present an NVSHMEM-based GPU kernel-initiated rede…

TopInG: Topologically Interpretable Graph Learning via Persistent Rationale Filtration
Cheng Xin, Fan Xu, Xin Ding, Jie Gao, Jiaxin Ding
https://arxiv.org/abs/2510.05102 https:/…

TopInG: Topologically Interpretable Graph Learning via Persistent Rationale Filtration
Graph Neural Networks (GNNs) have shown remarkable success across various scientific fields, yet their adoption in critical decision-making is often hindered by a lack of interpretability. Recently, intrinsically interpretable GNNs have been studied to provide insights into model predictions by identifying rationale substructures in graphs. However, existing methods face challenges when the underlying rationale subgraphs are complex and varied. In this work, we propose TopInG: Topologically Int…

ARMs: Adaptive Red-Teaming Agent against Multimodal Models with Plug-and-Play Attacks
Zhaorun Chen, Xun Liu, Mintong Kang, Jiawei Zhang, Minzhou Pan, Shuang Yang, Bo Li
https://arxiv.org/abs/2510.02677

ARMs: Adaptive Red-Teaming Agent against Multimodal Models with Plug-and-Play Attacks
As vision-language models (VLMs) gain prominence, their multimodal interfaces also introduce new safety vulnerabilities, making the safety evaluation challenging and critical. Existing red-teaming efforts are either restricted to a narrow set of adversarial patterns or depend heavily on manual engineering, lacking scalable exploration of emerging real-world VLM vulnerabilities. To bridge this gap, we propose ARMs, an adaptive red-teaming agent that systematically conducts comprehensive risk ass…

ZK-WAGON: Imperceptible Watermark for Image Generation Models using ZK-SNARKs
Aadarsh Anantha Ramakrishnan, Shubham Agarwal, Selvanayagam S, Kunwar Singh
https://arxiv.org/abs/2510.01967

ZK-WAGON: Imperceptible Watermark for Image Generation Models using ZK-SNARKs
As image generation models grow increasingly powerful and accessible, concerns around authenticity, ownership, and misuse of synthetic media have become critical. The ability to generate lifelike images indistinguishable from real ones introduces risks such as misinformation, deepfakes, and intellectual property violations. Traditional watermarking methods either degrade image quality, are easily removed, or require access to confidential model internals - making them unsuitable for secure and …

Uncertainty Quantification for Regression using Proper Scoring Rules
Alexander Fishkov, Kajetan Schweighofer, Mykyta Ielanskyi, Nikita Kotelevskii, Mohsen Guizani, Maxim Panov
https://arxiv.org/abs/2509.26610

Uncertainty Quantification for Regression using Proper Scoring Rules
Quantifying uncertainty of machine learning model predictions is essential for reliable decision-making, especially in safety-critical applications. Recently, uncertainty quantification (UQ) theory has advanced significantly, building on a firm basis of learning with proper scoring rules. However, these advances were focused on classification, while extending these ideas to regression remains challenging. In this work, we introduce a unified UQ framework for regression based on proper scoring r…

Profit over Proxies: A Scalable Bayesian Decision Framework for Optimizing Multi-Variant Online Experiments
Srijesh Pillai, Rajesh Kumar Chandrawat
https://arxiv.org/abs/2509.22677

Profit over Proxies: A Scalable Bayesian Decision Framework for Optimizing Multi-Variant Online Experiments
Online controlled experiments (A/B tests) are fundamental to data-driven decision-making in the digital economy. However, their real-world application is frequently compromised by two critical shortcomings: the use of statistically flawed heuristics like "p-value peeking", which inflates false positive rates, and an over-reliance on proxy metrics like conversion rates, which can lead to decisions that inadvertently harm core business profitability. This paper addresses these challenges by intro…

DS-STAR: Data Science Agent via Iterative Planning and Verification
Jaehyun Nam, Jinsung Yoon, Jiefeng Chen, Jinwoo Shin, Tomas Pfister
https://arxiv.org/abs/2509.21825 https://…

DS-STAR: Data Science Agent via Iterative Planning and Verification
Data science, which transforms raw data into actionable insights, is critical for data-driven decision-making. However, these tasks are often complex, involving steps for exploring multiple data sources and synthesizing findings to deliver insightful answers. While large language models (LLMs) show significant promise in automating this process, they often struggle with heterogeneous data formats and generate sub-optimal analysis plans, as verifying plan sufficiency is inherently difficult with…

fev-bench: A Realistic Benchmark for Time Series Forecasting
Oleksandr Shchur, Abdul Fatir Ansari, Caner Turkmen, Lorenzo Stella, Nick Erickson, Pablo Guerron, Michael Bohlke-Schneider, Yuyang Wang
https://arxiv.org/abs/2509.26468

fev-bench: A Realistic Benchmark for Time Series Forecasting
Benchmark quality is critical for meaningful evaluation and sustained progress in time series forecasting, particularly given the recent rise of pretrained models. Existing benchmarks often have narrow domain coverage or overlook important real-world settings, such as tasks with covariates. Additionally, their aggregation procedures often lack statistical rigor, making it unclear whether observed performance differences reflect true improvements or random variation. Many benchmarks also fail to…

Rearchitecting Datacenter Lifecycle for AI: A TCO-Driven Framework
Jovan Stojkovic, Chaojie Zhang, \'I\~nigo Goiri, Ricardo Bianchini
https://arxiv.org/abs/2509.26534 https:…

Rearchitecting Datacenter Lifecycle for AI: A TCO-Driven Framework
The rapid rise of large language models (LLMs) has been driving an enormous demand for AI inference infrastructure, mainly powered by high-end GPUs. While these accelerators offer immense computational power, they incur high capital and operational costs due to frequent upgrades, dense power consumption, and cooling demands, making total cost of ownership (TCO) for AI datacenters a critical concern for cloud providers. Unfortunately, traditional datacenter lifecycle management (designed for gen…

FuncPoison: Poisoning Function Library to Hijack Multi-agent Autonomous Driving Systems
Yuzhen Long, Songze Li
https://arxiv.org/abs/2509.24408 https://arx…

FuncPoison: Poisoning Function Library to Hijack Multi-agent Autonomous Driving Systems
Autonomous driving systems increasingly rely on multi-agent architectures powered by large language models (LLMs), where specialized agents collaborate to perceive, reason, and plan. A key component of these systems is the shared function library, a collection of software tools that agents use to process sensor data and navigate complex driving environments. Despite its critical role in agent decision-making, the function library remains an under-explored vulnerability. In this paper, we introd…