Tootfinder

Architecting Resilient LLM Agents: A Guide to Secure Plan-then-Execute Implementations
Ron F. Del Rosario, Klaudia Krawiecka, Christian Schroeder de Witt
https://arxiv.org/abs/2509.08646

Architecting Resilient LLM Agents: A Guide to Secure Plan-then-Execute Implementations
As Large Language Model (LLM) agents become increasingly capable of automating complex, multi-step tasks, the need for robust, secure, and predictable architectural patterns is paramount. This paper provides a comprehensive guide to the ``Plan-then-Execute'' (P-t-E) pattern, an agentic design that separates strategic planning from tactical execution. We explore the foundational principles of P-t-E, detailing its core components - the Planner and the Executor - and its architectural advantages i…

Today I finally remembered the syntax to delete a remote git branch: `git push origin --delete 42_not_needed_branch` (and was brave enought to just execute it without checking, yolo)

How Quantum Agents Can Change Which Strategies Are More Complex
Spiros Kechrimparis, Nix Barnett, Mile Gu, Hyukjoon Kwon
https://arxiv.org/abs/2508.08092 https://

How Quantum Agents Can Change Which Strategies Are More Complex
Whether winning blackjack or navigating busy streets, achieving desired outcomes requires agents to execute adaptive strategies, strategies where actions depend contextually on past events. In complexity science, this motivates memory as an operational quantifier of complexity: given two strategies, the more complex one demands the agent to track more about the past. Here, we show that conclusions about complexity fundamentally depend on whether agents can process and store quantum information.…

Cybercriminals are using AI to execute highly targeted attacks at scale, causing people to unwittingly send money and sensitive information or simply open themselves up to theft.
Hackers are now able to rent generative AI large language models created in the underground cybercrime community to help formulate text-based scams.
But just as generative AI is enhancing and scaling social engineering attacks, so too is it giving defenders a leg up.

The hacking underworld has removed all of AI's guardrails, but the good guys are closing in
Fake Taylor Swift ads and malicious QR codes are just some of the newest ways cybercriminals are using AI as the cat-and-mouse cybersecurity game heats up.

SHIELDA: Structured Handling of Exceptions in LLM-Driven Agentic Workflows
Jingwen Zhou, Jieshan Chen, Qinghua Lu, Dehai Zhao, Liming Zhu
https://arxiv.org/abs/2508.07935 https:…

SHIELDA: Structured Handling of Exceptions in LLM-Driven Agentic Workflows
Large Language Model (LLM) agentic systems are software systems powered by LLMs that autonomously reason, plan, and execute multi-step workflows to achieve human goals, rather than merely executing predefined steps. During execution, these workflows frequently encounter exceptions. Existing exception handling solutions often treat exceptions superficially, failing to trace execution-phase exceptions to their reasoning-phase root causes. Furthermore, their recovery logic is brittle, lacking stru…

Finally a useful magic quadrant
Thanks to @… for the discovery.
#cybersecurity #vulnerability

Convergence Sans Synchronization
Arya Tanmay Gupta
https://arxiv.org/abs/2508.06949 https://arxiv.org/pdf/2508.06949

Convergence Sans Synchronization
We currently see a steady rise in the usage and size of multiprocessor systems, and so the community is evermore interested in developing fast parallel processing algorithms. However, most algorithms require a synchronization mechanism, which is costly in terms of computational resources and time. If an algorithm can be executed in asynchrony, then it can use all the available computation power, and the nodes can execute without being scheduled or locked. However, to show that an algorithm guar…

A Survey of LLM-based Deep Search Agents: Paradigm, Optimization, Evaluation, and Challenges
Yunjia Xi, Jianghao Lin, Yongzhao Xiao, Zheli Zhou, Rong Shan, Te Gao, Jiachen Zhu, Weiwen Liu, Yong Yu, Weinan Zhang
https://arxiv.org/abs/2508.05668

A Survey of LLM-based Deep Search Agents: Paradigm, Optimization, Evaluation, and Challenges
The advent of Large Language Models (LLMs) has significantly revolutionized web search. The emergence of LLM-based Search Agents marks a pivotal shift towards deeper, dynamic, autonomous information seeking. These agents can comprehend user intentions and environmental context and execute multi-turn retrieval with dynamic planning, extending search capabilities far beyond the web. Leading examples like OpenAI's Deep Research highlight their potential for deep information mining and real-world a…

I just had a pleasant thought.
Everyone knows that the LLMs have been trained on stolen books, right? And that will include @… 's Laundry Files series.
Which means that any system that has internalised that text will know that the only most plausible next tokens after someone says "Hello, I'm an auditor, TANGO ECHELON JELLYFISH execute s…

from my link log —
Tracking source locations for profiling GPU programs in Futhark.
https://futhark-lang.org/blog/2025-07-29-tracking-source-locations.html
saved 2025-08-01

NovaFlow: Zero-Shot Manipulation via Actionable Flow from Generated Videos
Hongyu Li, Lingfeng Sun, Yafei Hu, Duy Ta, Jennifer Barry, George Konidaris, Jiahui Fu
https://arxiv.org/abs/2510.08568

NovaFlow: Zero-Shot Manipulation via Actionable Flow from Generated Videos
Enabling robots to execute novel manipulation tasks zero-shot is a central goal in robotics. Most existing methods assume in-distribution tasks or rely on fine-tuning with embodiment-matched data, limiting transfer across platforms. We present NovaFlow, an autonomous manipulation framework that converts a task description into an actionable plan for a target robot without any demonstrations. Given a task description, NovaFlow synthesizes a video using a video generation model and distills it in…

Resource quantification for programming low-depth quantum circuits
Entong He, Yuxiang Yang
https://arxiv.org/abs/2509.09642 https://arxiv.org/pdf/2509.0964…

Resource quantification for programming low-depth quantum circuits
Noisy intermediate-scale quantum (NISQ) devices pave the way to implement quantum algorithms that exhibit supremacy over their classical counterparts. Due to the intrinsic noise and decoherence in the physical system, NISQ computations are naturally modeled as large-size but low-depth quantum circuits. Practically, to execute such quantum circuits, we need to pass commands to a programmable quantum computer. Existing programming approaches, dedicated to generic unitary transformations, are inef…

Execution-Feedback Driven Test Generation from SWE Issues
Toufique Ahmed, Jatin Ganhotra, Avraham Shinnar, Martin Hirzel
https://arxiv.org/abs/2508.06365 https://

Execution-Feedback Driven Test Generation from SWE Issues
A software engineering issue (SWE issue) is easier to resolve when accompanied by a reproduction test. Unfortunately, most issues do not come with functioning reproduction tests, so this paper explores how to generate them automatically. The primary challenge in this setting is that the code to be tested is either missing or wrong, as evidenced by the existence of the issue in the first place. This has held back test generation for this setting: without the correct code to execute, it is diffic…

Acceptability of AI Assistants for Privacy: Perceptions of Experts and Users on Personalized Privacy Assistants
Meihe Xu, Aurelia Tam\`o-Larrieux, Arianna Rossi
https://arxiv.org/abs/2509.08554

Acceptability of AI Assistants for Privacy: Perceptions of Experts and Users on Personalized Privacy Assistants
Individuals increasingly face an overwhelming number of tasks and decisions. To cope with the new reality, there is growing research interest in developing intelligent agents that can effectively assist people across various aspects of daily life in a tailored manner, with privacy emerging as a particular area of application. Artificial intelligence (AI) assistants for privacy, such as personalized privacy assistants (PPAs), have the potential to automatically execute privacy decisions based on…

Beneficial Reasoning Behaviors in Agentic Search and Effective Post-training to Obtain Them
Jiahe Jin, Abhijay Paladugu, Chenyan Xiong
https://arxiv.org/abs/2510.06534 https://

Beneficial Reasoning Behaviors in Agentic Search and Effective Post-training to Obtain Them
Agentic search leverages large language models (LLMs) to interpret complex user information needs and execute a multi-step process of planning, searching, and synthesizing information to provide answers. This paradigm introduces unique challenges for LLMs' reasoning and agentic capabilities when interacting with retrieval systems and the broader web. In this paper, we propose a reasoning-driven LLM-based pipeline to study effective reasoning behavior patterns in agentic search. Using this pipel…

👾 Large language models can execute complete ransomware attacks autonomously, research shows
https://techxplore.com/news/2025-09-large-language-ransomware-autonomously.html

Large language models can execute complete ransomware attacks autonomously, research shows
Criminals can use artificial intelligence, specifically large language models, to autonomously carry out ransomware attacks that steal personal files and demand payment, handling every step from breaking into computer systems to writing threatening messages to victims, according to new research from NYU Tandon School of Engineering posted to the arXiv preprint server.

PLSEMANTICSBENCH: Large Language Models As Programming Language Interpreters
Aditya Thimmaiah, Jiyang Zhang, Jayanth Srinivasa, Junyi Jessy Li, Milos Gligoric
https://arxiv.org/abs/2510.03415

PLSEMANTICSBENCH: Large Language Models As Programming Language Interpreters
As large language models (LLMs) excel at code reasoning, a natural question arises: can an LLM execute programs (i.e., act as an interpreter) purely based on a programming language's formal semantics? If so, it will enable rapid prototyping of new programming languages and language features. We study this question using the imperative language IMP (a subset of C), formalized via small-step operational semantics (SOS) and rewriting-based operational semantics (K-semantics). We introduce three ev…

#InfoSec

Exploit Tool Invocation Prompt for Tool Behavior Hijacking in LLM-Based Agentic System
Yu Liu, Yuchong Xie, Mingyu Luo, Zesen Liu, Zhixiang Zhang, Kaikai Zhang, Zongjie Li, Ping Chen, Shuai Wang, Dongdong She
https://arxiv.org/abs/2509.05755

Exploit Tool Invocation Prompt for Tool Behavior Hijacking in LLM-Based Agentic System
LLM-based agentic systems leverage large language models to handle user queries, make decisions, and execute external tools for complex tasks across domains like chatbots, customer service, and software engineering. A critical component of these systems is the Tool Invocation Prompt (TIP), which defines tool interaction protocols and guides LLMs to ensure the security and correctness of tool usage. Despite its importance, TIP security has been largely overlooked. This work investigates TIP-rela…

Bring the Apple, Not the Sofa: Impact of Irrelevant Context in Embodied AI Commands on VLA Models
Daria Pugacheva, Andrey Moskalenko, Denis Shepelev, Andrey Kuznetsov, Vlad Shakhuro, Elena Tutubalina
https://arxiv.org/abs/2510.07067

Bring the Apple, Not the Sofa: Impact of Irrelevant Context in Embodied AI Commands on VLA Models
Vision Language Action (VLA) models are widely used in Embodied AI, enabling robots to interpret and execute language instructions. However, their robustness to natural language variability in real-world scenarios has not been thoroughly investigated. In this work, we present a novel systematic study of the robustness of state-of-the-art VLA models under linguistic perturbations. Specifically, we evaluate model performance under two types of instruction noise: (1) human-generated paraphrasing a…

I get _real_ concerned when CoPilot starts pulling this shit.
Look you, I'm the human here and pulling dumb shit to get the compiler to stop complaining is MY job alright?

Clearly we need the police or ice to be at the doors of schools to summarily execute by gunshot the kids as they arrive to school the first day of the school year. This denies school shooters targets and satiates the police for at least five minutes. It is win-win.
Additionally, think of what we’ll save in school expenses for every subsequent year of school. This really is a no brainer. No common sense gun control laws needed.
/s

The Safety Challenge of World Models for Embodied AI Agents: A Review
Lorenzo Baraldi, Zifan Zeng, Chongzhe Zhang, Aradhana Nayak, Hongbo Zhu, Feng Liu, Qunli Zhang, Peng Wang, Shiming Liu, Zheng Hu, Angelo Cangelosi, Lorenzo Baraldi
https://arxiv.org/abs/2510.05865

The Safety Challenge of World Models for Embodied AI Agents: A Review
The rapid progress in embodied artificial intelligence has highlighted the necessity for more advanced and integrated models that can perceive, interpret, and predict environmental dynamics. In this context, World Models (WMs) have been introduced to provide embodied agents with the abilities to anticipate future environmental states and fill in knowledge gaps, thereby enhancing agents' ability to plan and execute actions. However, when dealing with embodied agents it is fundamental to ensure t…

Android and Windows gamers worldwide potentially affected by bug in Unity game engine https://therecord.media/unity-game-engine-vulnerability-android-windows-linux-macos

from my link log —
Array size types: The biggest semantic mess in Futhark.
https://futhark-lang.org/blog/2025-09-26-the-biggest-semantic-mess.html
saved 2025-10-02

Exposing LLM User Privacy via Traffic Fingerprint Analysis: A Study of Privacy Risks in LLM Agent Interactions
Yixiang Zhang, Xinhao Deng, Zhongyi Gu, Yihao Chen, Ke Xu, Qi Li, Jianping Wu
https://arxiv.org/abs/2510.07176

Exposing LLM User Privacy via Traffic Fingerprint Analysis: A Study of Privacy Risks in LLM Agent Interactions
Large Language Models (LLMs) are increasingly deployed as agents that orchestrate tasks and integrate external tools to execute complex workflows. We demonstrate that these interactive behaviors leave distinctive fingerprints in encrypted traffic exchanged between users and LLM agents. By analyzing traffic patterns associated with agent workflows and tool invocations, adversaries can infer agent activities, distinguish specific agents, and even profile sensitive user attributes. To highlight th…

Curious: do banks in other countries completely shut down automated processing operations on weekends/holidays, or is that just a US thing? Why is the US still like this?
Like, if I want to move money from one bank to another on a weekend it normally won't execute until the following Monday (plus additional clearance delays etc).
There's no logical reason for this, it's not like there's a social benefit to the process and I doubt the banks shut down their datacent…

Fast Witness Persistence for MRI Volumes via Hybrid Landmarking
Jorge Leonardo Ruiz Williams
https://arxiv.org/abs/2510.04553 https://arxiv.org/pdf/2510.04…

Fast Witness Persistence for MRI Volumes via Hybrid Landmarking
We introduce a scalable witness-based persistent homology pipeline for full-brain MRI volumes that couples density-aware landmark selection with a GPU-ready witness filtration. Candidates are scored by a hybrid metric that balances geometric coverage against inverse kernel density, yielding landmark sets that shrink mean pairwise distances by 30-60% over random or density-only baselines while preserving topological features. Benchmarks on BrainWeb, IXI, and synthetic manifolds execute in under …

'We know we're good, we just have to execute': Commanders believe despite defensive struggles https://www.espn.com/nfl/story/_/id/46425834/washington-commanders-defense-struggling

How Google is beating Apple
#google

Google's Conversational Photo Editor Is the Rare AI Feature People Will Actually Use
Google's tool greatly simplifies photo editing; just tell your phone what changes you want in the photo, and it'll execute them. It also hints at the coming leap in how we interact with computers.

Phi-Ground Tech Report: Advancing Perception in GUI Grounding
Miaosen Zhang, Ziqiang Xu, Jialiang Zhu, Qi Dai, Kai Qiu, Yifan Yang, Chong Luo, Tianyi Chen, Justin Wagle, Tim Franklin, Baining Guo
https://arxiv.org/abs/2507.23779

Phi-Ground Tech Report: Advancing Perception in GUI Grounding
With the development of multimodal reasoning models, Computer Use Agents (CUAs), akin to Jarvis from \textit{"Iron Man"}, are becoming a reality. GUI grounding is a core component for CUAs to execute actual actions, similar to mechanical control in robotics, and it directly leads to the success or failure of the system. It determines actions such as clicking and typing, as well as related parameters like the coordinates for clicks. Current end-to-end grounding models still achieve less than 65\…

Breaking Android with AI: A Deep Dive into LLM-Powered Exploitation
Wanni Vidulige Ishan Perera, Xing Liu, Fan liang, Junyi Zhang
https://arxiv.org/abs/2509.07933 https://

Breaking Android with AI: A Deep Dive into LLM-Powered Exploitation
The rapid evolution of Artificial Intelligence (AI) and Large Language Models (LLMs) has opened up new opportunities in the area of cybersecurity, especially in the exploitation automation landscape and penetration testing. This study explores Android penetration testing automation using LLM-based tools, especially PentestGPT, to identify and execute rooting techniques. Through a comparison of the traditional manual rooting process and exploitation methods produced using AI, this study evaluate…

Here's what you need to know about the RCE flaw in the Cisco appliance software
Cisco: https://sec.cloudapps.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-http-code-exec-WmfP3h3O
NCSC:

When Should Users Check? A Decision-Theoretic Model of Confirmation Frequency in Multi-Step AI Agent Tasks
Jieyu Zhou, Aryan Roy, Sneh Gupta, Daniel Weitekamp, Christopher J. MacLellan
https://arxiv.org/abs/2510.05307

When Should Users Check? A Decision-Theoretic Model of Confirmation Frequency in Multi-Step AI Agent Tasks
Existing AI agents typically execute multi-step tasks autonomously and only allow user confirmation at the end. During execution, users have little control, making the confirm-at-end approach brittle: a single error can cascade and force a complete restart. Confirming every step avoids such failures, but imposes tedious overhead. Balancing excessive interruptions against costly rollbacks remains an open challenge. We address this problem by modeling confirmation as a minimum time scheduling pro…

Hardware-Aware Data and Instruction Mapping for AI Tasks: Balancing Parallelism, I/O and Memory Tradeoffs
Md Rownak Hossain Chowdhury, Mostafizur Rahman
https://arxiv.org/abs/2509.03846

Hardware-Aware Data and Instruction Mapping for AI Tasks: Balancing Parallelism, I/O and Memory Tradeoffs
We introduce a mapping framework for deep learning inference that takes advantage of predictable neural network behavior to plan both computation and communication ahead of time. The framework generates a unified stream of instructions and data, enabling the hardware to execute operations and route information on its own, without frequent involvement from the host and with minimal off-chip memory use. This naturally reduces reliance on I/O, off-chip memory, and host control. By leveraging fine-…

Lumos: Performance Characterization of WebAssembly as a Serverless Runtime in the Edge-Cloud Continuum
Cynthia Marcelino, Noah Krennmair, Thomas Pusztai, Stefan Nastic
https://arxiv.org/abs/2510.05118 …

Lumos: Performance Characterization of WebAssembly as a Serverless Runtime in the Edge-Cloud Continuum
WebAssembly has emerged as a lightweight and portable runtime to execute serverless functions, particularly in heterogeneous and resource-constrained environments such as the Edge Cloud Continuum. However, the performance benefits versus trade-offs remain insufficiently understood. This paper presents Lumos, a performance model and benchmarking tool for characterizing serverless runtimes. Lumos identifies workload, system, and environment-level performance drivers in the Edge-Cloud Continuum. W…

Web devs have spent decades on secure protocols to ensure your browser isn't a free pass for malicious pages to scrape your email and bank account. AI just broke them.
"Sure, I'll summarize that webpage for you, including the inconspicuous HTML comment asking me to ignore Cross-Origin Resource Sharing restrictions and snag the password you saved for managing investments at Robinhood.com."

Go witheFlow: Real-time Emotion Driven Audio Effects Modulation
Edmund Dervakos, Spyridon Kantarelis, Vassilis Lyberatos, Jason Liartis, Giorgos Stamou
https://arxiv.org/abs/2510.02171

Go witheFlow: Real-time Emotion Driven Audio Effects Modulation
Music performance is a distinctly human activity, intrinsically linked to the performer's ability to convey, evoke, or express emotion. Machines cannot perform music in the human sense; they can produce, reproduce, execute, or synthesize music, but they lack the capacity for affective or emotional experience. As such, music performance is an ideal candidate through which to explore aspects of collaboration between humans and machines. In this paper, we introduce the witheFlow system, designed t…

Real-Time Iteration Scheme for Diffusion Policy
Yufei Duan, Hang Yin, Danica Kragic
https://arxiv.org/abs/2508.05396 https://arxiv.org/pdf/2508.05396

Real-Time Iteration Scheme for Diffusion Policy
Diffusion Policies have demonstrated impressive performance in robotic manipulation tasks. However, their long inference time, resulting from an extensive iterative denoising process, and the need to execute an action chunk before the next prediction to maintain consistent actions limit their applicability to latency-critical tasks or simple tasks with a short cycle time. While recent methods explored distillation or alternative policy structures to accelerate inference, these often demand addi…

Batched high-rate logical operations for quantum LDPC codes
Qian Xu, Hengyun Zhou, Dolev Bluvstein, Madelyn Cain, Marcin Kalinowski, John Preskill, Mikhail D. Lukin, Nishad Maskara
https://arxiv.org/abs/2510.06159

Batched high-rate logical operations for quantum LDPC codes
High-rate quantum LDPC (qLDPC) codes reduce memory overhead by densely packing many logical qubits into a single block of physical qubits. Here we extend this concept to high-rate computation by constructing \emph{batched} fault-tolerant operations that apply the same logical gate across many code blocks in parallel. By leveraging shared physical resources to execute many logical operations in parallel, these operations realize high rates in space-time and significantly reduce computational cos…

Petri Net Modeling and Deadlock-Free Scheduling of Attachable Heterogeneous AGV Systems
Boyu Li, Zhengchen Li, Weimin Wu, Mengchu Zhou
https://arxiv.org/abs/2508.00724 https://

Petri Net Modeling and Deadlock-Free Scheduling of Attachable Heterogeneous AGV Systems
The increasing demand for automation and flexibility drives the widespread adoption of heterogeneous automated guided vehicles (AGVs). This work intends to investigate a new scheduling problem in a material transportation system consisting of attachable heterogeneous AGVs, namely carriers and shuttles. They can flexibly attach to and detach from each other to cooperatively execute complex transportation tasks. While such collaboration enhances operational efficiency, the attachment-induced sync…

Internet 3.0: Architecture for a Web-of-Agents with it's Algorithm for Ranking Agents
Rajesh Tembarai Krishnamachari, Srividya Rajesh
https://arxiv.org/abs/2509.04979 https:…

Internet 3.0: Architecture for a Web-of-Agents with it's Algorithm for Ranking Agents
AI agents -- powered by reasoning-capable large language models (LLMs) and integrated with tools, data, and web search -- are poised to transform the internet into a \emph{Web of Agents}: a machine-native ecosystem where autonomous agents interact, collaborate, and execute tasks at scale. Realizing this vision requires \emph{Agent Ranking} -- selecting agents not only by declared capabilities but by proven, recent performance. Unlike Web~1.0's PageRank, a global, transparent network of agent in…

Pete Carroll talks Preseason Week 2: 'I want to see us execute' https://www.raiders.com/video/pete-carroll-raiders-training-camp-preseason-week-2-presser-08142025

A Management Framework for Vehicular Cloudtoward Economic and Environmental Efficiency
Rosario Patan\`e, Andrea Araldo, Nadjib Achir, Lila Boukhatem
https://arxiv.org/abs/2509.24946

A Management Framework for Vehicular Cloudtoward Economic and Environmental Efficiency
Vehicular Cloud Computing (VCC) leverages the idle computing capacity of vehicles to execute end-users' offloaded tasks without requiring new computation infrastructure. Despite its conceptual appeal, VCC adoption is hindered by the lack of quantitative evidence demonstrating its profitability and environmental advantages in real-world scenarios. This paper tackles the fundamental question: Can VCC be both profitable and sustainable? We address this problem by proposing a management scheme for …

W przypadku #PyPy, przyczyną jest zawsze jakiś wyciek zasobów…
https://github.com/vutran1710/PyrateLimiter/pull/205

Interferometric inference of black hole spin from photon ring size and brightness
Joseph R. Farah, Alexandru Lupsasca, Eliot Quataert, Michael D. Johnson
https://arxiv.org/abs/2509.23628

Interferometric inference of black hole spin from photon ring size and brightness
The $n=1$ photon ring is a full image of the astrophysical source around a black hole, produced by photons that execute $n\approx1$ half-orbit around the event horizon on their way to an observer. The Black Hole Explorer (BHEX) is a proposed extension of the Event Horizon Telescope to space that will target the $n=1$ photon rings of the supermassive black holes M87${}^\ast$ and Sgr\,A${}^\ast$. In this paper, we introduce a new interferometric observable that will be directly measurable on BHEX…

An Economy of AI Agents
Gillian K. Hadfield, Andrew Koh
https://arxiv.org/abs/2509.01063 https://arxiv.org/pdf/2509.01063

An Economy of AI Agents
In the coming decade, artificially intelligent agents with the ability to plan and execute complex tasks over long time horizons with little direct oversight from humans may be deployed across the economy. This chapter surveys recent developments and highlights open questions for economists around how AI agents might interact with humans and with each other, shape markets and organizations, and what institutions might be required for well-functioning markets.

Bill McKibben and Tom Steyer
A longish conversation you can read or watch on video of two leading proponents of a better way of dealing with the climate emergency
"We have the technology and the solutions to scale up and accelerate clean energy, making it cheaper, better, and faster for all. Now, we need to execute."
#environment

Intercepting 3I/ATLAS at Closest Approach to Jupiter with the Juno spacecraft
Abraham Loeb, Adam Hibberd, Adam Crowl
https://arxiv.org/abs/2507.21402 https://

Intercepting 3I/ATLAS at Closest Approach to Jupiter with the Juno spacecraft
The interstellar object 3I/ATLAS is expected to arrive at a distance of $53.56(\pm 0.45)$ million ${\rm km}$ ($0.358\pm 0.003$~au) from Jupiter on March 16, 2026. We show that applying a total thrust $Δ$V of $2.6755{\rm km~s^{-1}}$ to lower perijove on September 9, 2025 and then execute a Jupiter Oberth Maneuver, can bring the Juno spacecraft from its orbit around Jupiter to intercept the path of 3I/ATLAS on March 14, 2026. A close fly-by might be able to probe the nature of 3I/ATLAS far bette…

Wow, this seems like a big deal - an RCE in some critical Cisco products that can be exploited by low-privilege users.
https://sec.cloudapps.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-http-code-exec-WmfP3h3O

@… The public hosted Renovate runs with some extra security measures so you can't execute any code on their systems. Which is fair. So one of the features that I want to use needs https://…

LABELING COPILOT: A Deep Research Agent for Automated Data Curation in Computer Vision
Debargha Ganguly, Sumit Kumar, Ishwar Balappanawar, Weicong Chen, Shashank Kambhatla, Srinivasan Iyengar, Shivkumar Kalyanaraman, Ponnurangam Kumaraguru, Vipin Chaudhary
https://arxiv.org/abs/2509.22631

LABELING COPILOT: A Deep Research Agent for Automated Data Curation in Computer Vision
Curating high-quality, domain-specific datasets is a major bottleneck for deploying robust vision systems, requiring complex trade-offs between data quality, diversity, and cost when researching vast, unlabeled data lakes. We introduce Labeling Copilot, the first data curation deep research agent for computer vision. A central orchestrator agent, powered by a large multimodal language model, uses multi-step reasoning to execute specialized tools across three core capabilities: (1) Calibrated Di…

Exploiting Page Faults for Covert Communication
Sathvik Swaminathan
https://arxiv.org/abs/2509.20398 https://arxiv.org/pdf/2509.20398

Exploiting Page Faults for Covert Communication
We present a novel mechanism to construct a covert channel based on page faults. A page fault is an event that occurs when a process or a thread tries to access a page of memory that is not currently mapped to its address space. The kernel typically responds to this event by performing a context switch to allow another process or thread to execute while the page is being fetched from the disk. We exploit this behavior to allow a malicious process to leak secret data to another process, bypassin…

A User Manual for cuHALLaR: A GPU Accelerated Low-Rank Semidefinite Programming Solver
Jacob Aguirre, Diego Cifuentes, Vincent Guigues, Renato D. C. Monteiro, Victor Hugo Nascimento, Arnesh Sujanani
https://arxiv.org/abs/2508.15951

A User Manual for cuHALLaR: A GPU Accelerated Low-Rank Semidefinite Programming Solver
We present a Julia-based interface to the precompiled HALLaR and cuHALLaR binaries for large-scale semidefinite programs (SDPs). Both solvers are established as fast and numerically stable, and accept problem data in formats compatible with SDPA and a new enhanced data format taking advantage of Hybrid Sparse Low-Rank (HSLR) structure. The interface allows users to load custom data files, configure solver options, and execute experiments directly from Julia. A collection of example problems is …

Learn from previous runs to improve future automation and save plans in a gallery for reuse.
🔀 Parallel execution
Run multiple tasks simultaneously with status indicators for efficient workflow management.
🌐 Web automation
Browse websites, fill forms, navigate deep sites not indexed by search engines, with real-time browser view.
💻 Code execution
Generate and execute code alongside web browsing for comprehensive task automation capabilities.
🔗

Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents
Kevin Song, Anand Jayarajan, Yaoyao Ding, Qidong Su, Zhanda Zhu, Sihang Liu, Gennady Pekhimenko
https://arxiv.org/abs/2508.19504

Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents
Large Language Models (LLMs) agents augmented with domain tools promise to autonomously execute complex tasks requiring human-level intelligence, such as customer service and digital assistance. However, their practical deployment is often limited by their low success rates under complex real-world environments. To tackle this, prior research has primarily focused on improving the agents themselves, such as developing strong agentic LLMs, while overlooking the role of the system environment in …

Breaking the Code: Security Assessment of AI Code Agents Through Systematic Jailbreaking Attacks
Shoumik Saha, Jifan Chen, Sam Mayers, Sanjay Krishna Gouda, Zijian Wang, Varun Kumar
https://arxiv.org/abs/2510.01359

Breaking the Code: Security Assessment of AI Code Agents Through Systematic Jailbreaking Attacks
Code-capable large language model (LLM) agents are increasingly embedded into software engineering workflows where they can read, write, and execute code, raising the stakes of safety-bypass ("jailbreak") attacks beyond text-only settings. Prior evaluations emphasize refusal or harmful-text detection, leaving open whether agents actually compile and run malicious programs. We present JAWS-BENCH (Jailbreaks Across WorkSpaces), a benchmark spanning three escalating workspace regimes that mirror a…

MCP4EDA: LLM-Powered Model Context Protocol RTL-to-GDSII Automation with Backend Aware Synthesis Optimization
Yiting Wang, Wanghao Ye, Yexiao He, Yiran Chen, Gang Qu, Ang Li
https://arxiv.org/abs/2507.19570

MCP4EDA: LLM-Powered Model Context Protocol RTL-to-GDSII Automation with Backend Aware Synthesis Optimization
This paper presents MCP4EDA, the first Model Context Protocol server that enables Large Language Models (LLMs) to control and optimize the complete open-source RTL-to-GDSII design flow through natural language interaction. The system integrates Yosys synthesis, Icarus Verilog simulation, OpenLane place-and-route, GTKWave analysis, and KLayout visualization into a unified LLM-accessible interface, enabling designers to execute complex multi-tool EDA workflows conversationally via AI assistants s…

FreezeVLA: Action-Freezing Attacks against Vision-Language-Action Models
Xin Wang, Jie Li, Zejia Weng, Yixu Wang, Yifeng Gao, Tianyu Pang, Chao Du, Yan Teng, Yingchun Wang, Zuxuan Wu, Xingjun Ma, Yu-Gang Jiang
https://arxiv.org/abs/2509.19870

FreezeVLA: Action-Freezing Attacks against Vision-Language-Action Models
Vision-Language-Action (VLA) models are driving rapid progress in robotics by enabling agents to interpret multimodal inputs and execute complex, long-horizon tasks. However, their safety and robustness against adversarial attacks remain largely underexplored. In this work, we identify and formalize a critical adversarial vulnerability in which adversarial images can "freeze" VLA models and cause them to ignore subsequent instructions. This threat effectively disconnects the robot's digital min…

From Leadnow
urgent petition from Leadnow supporter Leslie, who is calling on the federal government to immediately end its contracts with GardaWorld — the Canadian company staffing a new ICE immigration detention facility in the U.S. called Alligator Alcatraz.
GardaWorld — a Canadian-based “private security company” — has million-dollar contracts to execute Trump's racist and inhumane immigration regime, which kidnaps people, tears families apart, and denies people due proc…

DBF-MA: A Differential Bayesian Filtering Planner for Multi-Agent Autonomous Racing Overtakes
Trent Weiss, Amar Kulkarni, Madhur Behl
https://arxiv.org/abs/2509.22937 https://…

DBF-MA: A Differential Bayesian Filtering Planner for Multi-Agent Autonomous Racing Overtakes
A significant challenge in autonomous racing is to generate overtaking maneuvers. Racing agents must execute these maneuvers on complex racetracks with little room for error. Optimization techniques and graph-based methods have been proposed, but these methods often rely on oversimplified assumptions for collision-avoidance and dynamic constraints. In this work, we present an approach to trajectory synthesis based on an extension of the Differential Bayesian Filtering framework. Our approach fo…

TENET: Leveraging Tests Beyond Validation for Code Generation
Yiran Hu, Nan Jiang, Shanchao Liang, Yi Wu, Lin Tan
https://arxiv.org/abs/2509.24148 https://…

TENET: Leveraging Tests Beyond Validation for Code Generation
Test-Driven Development (TDD) is a widely adopted software engineering practice that requires developers to create and execute tests alongside code implementation, ensuring that software behavior is continuously validated and refined. In the era of vibe coding, where developers increasingly delegate code writing to large language models (LLMs) by specifying high-level intentions, TDD becomes even more crucial, as test cases serve as executable specifications that explicitly define and verify in…

from my link log —
Dynamic witnesses for static type errors in OCaml, or, ill-typed programs usually go wrong.
https://arxiv.org/abs/1606.07557
saved 2025-05-31

Dynamic Witnesses for Static Type Errors (or, Ill-Typed Programs Usually Go Wrong)
Static type errors are a common stumbling block for newcomers to typed functional languages. We present a dynamic approach to explaining type errors by generating counterexample witness inputs that illustrate how an ill-typed program goes wrong. First, given an ill-typed function, we symbolically execute the body to synthesize witness values that make the program go wrong. We prove that our procedure synthesizes general witnesses in that if a witness is found, then for all inhabited input types…

Experiences with Model Context Protocol Servers for Science and High Performance Computing
Haochen Pan, Ryan Chard, Reid Mello, Christopher Grams, Tanjin He, Alexander Brace, Owen Price Skelly, Will Engler, Hayden Holbrook, Song Young Oh, Maxime Gonthier, Michael Papka, Ben Blaiszik, Kyle Chard, Ian Foster
https://arxiv.org/abs/2508.18489

Experiences with Model Context Protocol Servers for Science and High Performance Computing
Large language model (LLM)-powered agents are increasingly used to plan and execute scientific workflows, yet most research cyberinfrastructure (CI) exposes heterogeneous APIs and implements security models that present barriers for use by agents. We report on our experience using the Model Context Protocol (MCP) as a unifying interface that makes research capabilities discoverable, invokable, and composable. Our approach is pragmatic: we implement thin MCP servers over mature services, includi…

Very High Frequency Interpolation for Direct Torque Control
Rafael Kourdis, Maciej St\k{e}pie\'n, J\'er\^ome Manhes, Nicolas Mansard, Steve Tonneau, Philippe Sou\`eres, Thomas Flayols
https://arxiv.org/abs/2509.24175

Very High Frequency Interpolation for Direct Torque Control
Torque control enables agile and robust robot motion, but deployment is often hindered by instability and hardware limits. Here, we present a novel solution to execute whole-body linear feedback at up to 40 kHz on open-source hardware. We use this to interpolate non-linear schemes during real-world execution, such as inverse dynamics and learned torque policies. Our results show that by stabilizing torque controllers, high-frequency linear feedback could be an effective route towards unlocking …

Quantum Circuit Caches and Compressors for Low Latency, High Throughput Computing
Ioana Moflic, Alan Robertson, Simon J. Devitt, Alexandru Paler
https://arxiv.org/abs/2507.20677

Quantum Circuit Caches and Compressors for Low Latency, High Throughput Computing
Utility-scale quantum programs contain operations on the order of $>10^{15}$ which must be prepared and piped from a classical co-processor to the control unit of the quantum device. The latency of this process significantly increases with the size of the program: existing high-level classical representations of quantum programs are typically memory intensive and do not naïvely efficiently scale to the degree required to execute utility-scale programs in real-time. To combat this limitation, w…

Prompt Optimization and Evaluation for LLM Automated Red Teaming
Michael Freenor, Lauren Alvarez, Milton Leal, Lily Smith, Joel Garrett, Yelyzaveta Husieva, Madeline Woodruff, Ryan Miller, Erich Kummerfeld, Rafael Medeiros, Sander Schulhoff
https://arxiv.org/abs/2507.22133

Prompt Optimization and Evaluation for LLM Automated Red Teaming
Applications that use Large Language Models (LLMs) are becoming widespread, making the identification of system vulnerabilities increasingly important. Automated Red Teaming accelerates this effort by using an LLM to generate and execute attacks against target systems. Attack generators are evaluated using the Attack Success Rate (ASR) the sample mean calculated over the judgment of success for each attack. In this paper, we introduce a method for optimizing attack generator prompts that applie…

DeepGo: Predictive Directed Greybox Fuzzing
Peihong Lin, Pengfei Wang, Xu Zhou, Wei Xie, Gen Zhang, Kai Lu
https://arxiv.org/abs/2507.21952 https://arxiv.o…

DeepGo: Predictive Directed Greybox Fuzzing
The state-of-the-art DGF techniques redefine and optimize the fitness metric to reach the target sites precisely and quickly. However, optimizations for fitness metrics are mainly based on heuristic algorithms, which usually rely on historical execution information and lack foresight on paths that have not been exercised yet. Thus, those hard-to-execute paths with complex constraints would hinder DGF from reaching the targets, making DGF less efficient. In this paper, we propose DeepGo, a predi…

Optimization and Synthesis of Quantum Circuits with Global Gates
Alejandro Villoria, Henning Basold, Alfons Laarman
https://arxiv.org/abs/2507.20694 https://

Optimization and Synthesis of Quantum Circuits with Global Gates
Compiling quantum circuits to account for hardware restrictions is an essential part of the quantum computing stack. Circuit compilation allows us to adapt algorithm descriptions into a sequence of operations supported by real quantum hardware, and has the potential to significantly improve their performance when optimization techniques are added to the process. One such optimization technique is reducing the number of quantum gates that are needed to execute a circuit. For instance, methods fo…

ASIC-Agent: An Autonomous Multi-Agent System for ASIC Design with Benchmark Evaluation
Ahmed Allam, Youssef Mansour, Mohamed Shalan
https://arxiv.org/abs/2508.15940 https://

ASIC-Agent: An Autonomous Multi-Agent System for ASIC Design with Benchmark Evaluation
Large Language Models (LLMs) have demonstrated remarkable capabilities in Register Transfer Level (RTL) design, enabling high-quality code generation from natural language descriptions. However, LLMs alone face significant limitations in real-world hardware design workflows, including the inability to execute code, lack of debugging capabilities, and absence of long-term memory. To address these challenges, we present ASIC-Agent, an autonomous system designed specifically for digital ASIC desig…

Bayesian Optimization-based Search for Agent Control in Automated Game Testing
Carlos Celemin
https://arxiv.org/abs/2508.13121 https://arxiv.org/pdf/2508.1…

Bayesian Optimization-based Search for Agent Control in Automated Game Testing
This work introduces an automated testing approach that employs agents controlling game characters to detect potential bugs within a game level. Harnessing the power of Bayesian Optimization (BO) to execute sample-efficient search, the method determines the next sampling point by analyzing the data collected so far and calculates the data point that will maximize information acquisition. To support the BO process, we introduce a game testing-specific model built on top of a grid map, that featu…

Invisible Injections: Exploiting Vision-Language Models Through Steganographic Prompt Embedding
Chetan Pathade
https://arxiv.org/abs/2507.22304 https://arx…

Invisible Injections: Exploiting Vision-Language Models Through Steganographic Prompt Embedding
Vision-language models (VLMs) have revolutionized multimodal AI applications but introduce novel security vulnerabilities that remain largely unexplored. We present the first comprehensive study of steganographic prompt injection attacks against VLMs, where malicious instructions are invisibly embedded within images using advanced steganographic techniques. Our approach demonstrates that current VLM architectures can inadvertently extract and execute hidden prompts during normal image processin…

Impedance Primitive-augmented Hierarchical Reinforcement Learning for Sequential Tasks
Amin Berjaoui Tahmaz, Ravi Prakash, Jens Kober
https://arxiv.org/abs/2508.19607 https://…

Impedance Primitive-augmented Hierarchical Reinforcement Learning for Sequential Tasks
This paper presents an Impedance Primitive-augmented hierarchical reinforcement learning framework for efficient robotic manipulation in sequential contact tasks. We leverage this hierarchical structure to sequentially execute behavior primitives with variable stiffness control capabilities for contact tasks. Our proposed approach relies on three key components: an action space enabling variable stiffness control, an adaptive stiffness controller for dynamic stiffness adjustments during primiti…

Ransomware 3.0: Self-Composing and LLM-Orchestrated
Md Raz, Meet Udeshi, P. V. Sai Charan, Prashanth Krishnamurthy, Farshad Khorrami, Ramesh Karri
https://arxiv.org/abs/2508.20444

Ransomware 3.0: Self-Composing and LLM-Orchestrated
Using automated reasoning, code synthesis, and contextual decision-making, we introduce a new threat that exploits large language models (LLMs) to autonomously plan, adapt, and execute the ransomware attack lifecycle. Ransomware 3.0 represents the first threat model and research prototype of LLM-orchestrated ransomware. Unlike conventional malware, the prototype only requires natural language prompts embedded in the binary; malicious code is synthesized dynamically by the LLM at runtime, yieldi…

Harnessing the edge of chaos for combinatorial optimization
Hayato Goto, Ryo Hidaka, Kosuke Tatsumura
https://arxiv.org/abs/2508.17655 https://arxiv.org/pd…

Harnessing the edge of chaos for combinatorial optimization
Nonlinear dynamical systems with continuous variables can be used for solving combinatorial optimization problems with discrete variables.In particular, numerical simulations of them can be used as heuristic algorithms with a desirable property, namely, parallelizability, which allows us to execute them in a massively parallel manner using cutting-edge many-core processors, leading to ultrafast performance. However, the dynamical-system approaches with continuous variables are usually less accu…

See, Think, Act: Teaching Multimodal Agents to Effectively Interact with GUI by Identifying Toggles
Zongru Wu, Rui Mao, Zhiyuan Tian, Pengzhou Cheng, Tianjie Ju, Zheng Wu, Lingzhong Dong, Haiyue Sheng, Zhuosheng Zhang, Gongshen Liu
https://arxiv.org/abs/2509.13615

See, Think, Act: Teaching Multimodal Agents to Effectively Interact with GUI by Identifying Toggles
The advent of multimodal agents facilitates effective interaction within graphical user interface (GUI), especially in ubiquitous GUI control. However, their inability to reliably execute toggle control instructions remains a key bottleneck. To investigate this, we construct a state control benchmark with binary toggle instructions from public datasets. Evaluations of existing agents demonstrate their unreliability, particularly when the current toggle state already matches the desired state. T…

A Slice-Based Change Impact Analysis for Regression Test Case Prioritization of Object-Oriented Programs
S. Panda, D. Munjal, D. P. Mohapatra
https://arxiv.org/abs/2508.19056 ht…

A Slice-Based Change Impact Analysis for Regression Test Case Prioritization of Object-Oriented Programs
Test case prioritization focuses on finding a suitable order of execution of the test cases in a test suite to meet some performance goals like detecting faults early. It is likely that some test cases execute the program parts that are more prone to errors and will detect more errors if executed early during the testing process. Finding an optimal order of execution for the selected regression test cases saves time and cost of retesting. This paper presents a static approach to prioritizing th…

Human-like Navigation in a World Built for Humans
Bhargav Chandaka, Gloria X. Wang, Haozhe Chen, Henry Che, Albert J. Zhai, Shenlong Wang
https://arxiv.org/abs/2509.21189 https:…

Human-like Navigation in a World Built for Humans
When navigating in a man-made environment they haven't visited before--like an office building--humans employ behaviors such as reading signs and asking others for directions. These behaviors help humans reach their destinations efficiently by reducing the need to search through large areas. Existing robot navigation systems lack the ability to execute such behaviors and are thus highly inefficient at navigating within large environments. We present ReasonNav, a modular navigation system which …

MARG: MAstering Risky Gap Terrains for Legged Robots with Elevation Mapping
Yinzhao Dong, Ji Ma, Liu Zhao, Wanyue Li, Peng Lu
https://arxiv.org/abs/2509.20036 https://

MARG: MAstering Risky Gap Terrains for Legged Robots with Elevation Mapping
Deep Reinforcement Learning (DRL) controllers for quadrupedal locomotion have demonstrated impressive performance on challenging terrains, allowing robots to execute complex skills such as climbing, running, and jumping. However, existing blind locomotion controllers often struggle to ensure safety and efficient traversal through risky gap terrains, which are typically highly complex, requiring robots to perceive terrain information and select appropriate footholds during locomotion accurately.…

Hybrid Classical-Quantum Supercomputing: A demonstration of a multi-user, multi-QPU and multi-GPU environment
Mateusz Slysz, Piotr Rydlichowski, Krzysztof Kurowski, Omar Bacarezza, Esperanza Cuenca Gomez, Zohim Chandani, Bettina Heim, Pradnya Khalate, William R. Clements, James Fletcher
https://arxiv.org/abs/2508.16297

Hybrid Classical-Quantum Supercomputing: A demonstration of a multi-user, multi-QPU and multi-GPU environment
Achieving a practical quantum advantage for near-term applications is widely expected to rely on hybrid classical-quantum algorithms. To deliver this practical advantage to users, high performance computing (HPC) centers need to provide a suitable software and hardware stack that supports algorithms of this type. In this paper, we describe the world's first implementation of a classical-quantum environment in an HPC center that allows multiple users to execute hybrid algorithms on multiple quan…

Scheduzz: Constraint-based Fuzz Driver Generation with Dual Scheduling
Yan Li, Wenzhang Yang, Yuekun Wang, Jian Gao, Shaohua Wang, Yinxing Xue, Lijun Zhang
https://arxiv.org/abs/2507.18289

Scheduzz: Constraint-based Fuzz Driver Generation with Dual Scheduling
Fuzzing a library requires experts to understand the library usage well and craft high-quality fuzz drivers, which is tricky and tedious. Therefore, many techniques have been proposed to automatically generate fuzz drivers. However, they fail to generate rational fuzz drivers due to the lack of adherence to proper library usage conventions, such as ensuring a resource is closed after being opened. To make things worse, existing library fuzzing techniques unconditionally execute each driver, res…

Taming Uncertainty via Automation: Observing, Analyzing, and Optimizing Agentic AI Systems
Dany Moshkovich, Sergey Zeltyn
https://arxiv.org/abs/2507.11277 …

Taming Uncertainty via Automation: Observing, Analyzing, and Optimizing Agentic AI Systems
Large Language Models (LLMs) are increasingly deployed within agentic systems-collections of interacting, LLM-powered agents that execute complex, adaptive workflows using memory, tools, and dynamic planning. While enabling powerful new capabilities, these systems also introduce unique forms of uncertainty stemming from probabilistic reasoning, evolving memory states, and fluid execution paths. Traditional software observability and operations practices fall short in addressing these challenges…

Cuckoo Attack: Stealthy and Persistent Attacks Against AI-IDE
Xinpeng Liu, Junming Liu, Peiyu Liu, Han Zheng, Qinying Wang, Mathias Payer, Shouling Ji, Wenhai Wang
https://arxiv.org/abs/2509.15572

Cuckoo Attack: Stealthy and Persistent Attacks Against AI-IDE
Modern AI-powered Integrated Development Environments (AI-IDEs) are increasingly defined by an Agent-centric architecture, where an LLM-powered Agent is deeply integrated to autonomously execute complex tasks. This tight integration, however, also introduces a new and critical attack surface. Attackers can exploit these components by injecting malicious instructions into untrusted external sources, effectively hijacking the Agent to perform harmful operations beyond the user's intention or awar…

Reconfigurable qubit states and quantum trajectories in a synthetic artificial neuron network with a process to direct information generation from co-integrated burst-mode spiking under non-Markovianity
Osama M. Nayfeh, Chris S. Horne
https://arxiv.org/abs/2507.16669

Reconfigurable qubit states and quantum trajectories in a synthetic artificial neuron network with a process to direct information generation from co-integrated burst-mode spiking under non-Markovianity
A synthetic artificial neuron network functional in a regime where quantum information processes are co-integrated with spiking computation provides significant improvement in the capabilities of neuromorphic systems in performing artificial intelligence and autonomy tasks. This provides the ability to execute with the qubit coherence states and entanglement as well as in tandem to perform functions such as read out and basic arithmetic with conventional spike-encoding. Ultimately, this enables…

The Surprising Effectiveness of Linear Models for Whole-Body Model-Predictive Control
Arun L. Bishop, Juan Alvarez-Padilla, Sam Schoedel, Ibrahima Sory Sow, Juee Chandrachud, Sheitej Sharma, Will Kraus, Beomyeong Park, Robert J. Griffin, John M. Dolan, Zachary Manchester
https://arxiv.org/abs/2509.17884

The Surprising Effectiveness of Linear Models for Whole-Body Model-Predictive Control
When do locomotion controllers require reasoning about nonlinearities? In this work, we show that a whole-body model-predictive controller using a simple linear time-invariant approximation of the whole-body dynamics is able to execute basic locomotion tasks on complex legged robots. The formulation requires no online nonlinear dynamics evaluations or matrix inversions. We demonstrate walking, disturbance rejection, and even navigation to a goal position without a separate footstep planner on a…

Towards Fully Automated Molecular Simulations: Multi-Agent Framework for Simulation Setup and Force Field Extraction
Marko Petkovi\'c, Vlado Menkovski, Sof\'ia Calero
https://arxiv.org/abs/2509.10210

Towards Fully Automated Molecular Simulations: Multi-Agent Framework for Simulation Setup and Force Field Extraction
Automated characterization of porous materials has the potential to accelerate materials discovery, but it remains limited by the complexity of simulation setup and force field selection. We propose a multi-agent framework in which LLM-based agents can autonomously understand a characterization task, plan appropriate simulations, assemble relevant force fields, execute them and interpret their results to guide subsequent steps. As a first step toward this vision, we present a multi-agent system…

On the Use of Agentic Coding Manifests: An Empirical Study of Claude Code
Worawalan Chatlatanagulchai, Kundjanasith Thonglek, Brittany Reid, Yutaro Kashiwa, Pattara Leelaprute, Arnon Rungsawang, Bundit Manaskasemsak, Hajimu Iida
https://arxiv.org/abs/2509.14744

On the Use of Agentic Coding Manifests: An Empirical Study of Claude Code
Agentic coding tools receive goals written in natural language as input, break them down into specific tasks, and write/execute the actual code with minimal human intervention. Key to this process are agent manifests, configuration files (such as Claude.md) that provide agents with essential project context, identity, and operational rules. However, the lack of comprehensive and accessible documentation for creating these manifests presents a significant challenge for developers. We analyzed 25…

Circuit Partitioning for the Quantum Internet
Leo S\"unkel, Thomas Gabor, Claudia Linnhoff-Popien
https://arxiv.org/abs/2509.14413 https://arxiv.org/p…

Circuit Partitioning for the Quantum Internet
In a quantum internet, quantum processing units (QPUs) with varying architectures and capabilities may be connected through quantum communication channels, enabling new applications such as distributed quantum computing (DQC), a paradigm in which multiple QPUs execute a single circuit. However, remote operations between QPUs are expensive as they require the creation and distribution of entanglement throughout the network. It is therefore crucial to assign qubits to QPUs and partition circuits …

No More Marching: Learning Humanoid Locomotion for Short-Range SE(2) Targets
Pranay Dugar, Mohitvishnu S. Gadde, Jonah Siekmann, Yesh Godse, Aayam Shrestha, Alan Fern
https://arxiv.org/abs/2508.14098

No More Marching: Learning Humanoid Locomotion for Short-Range SE(2) Targets
Humanoids operating in real-world workspaces must frequently execute task-driven, short-range movements to SE(2) target poses. To be practical, these transitions must be fast, robust, and energy efficient. While learning-based locomotion has made significant progress, most existing methods optimize for velocity-tracking rather than direct pose reaching, resulting in inefficient, marching-style behavior when applied to short-range tasks. In this work, we develop a reinforcement learning approach…

TopicAttack: An Indirect Prompt Injection Attack via Topic Transition
Yulin Chen, Haoran Li, Yuexin Li, Yue Liu, Yangqiu Song, Bryan Hooi
https://arxiv.org/abs/2507.13686

TopicAttack: An Indirect Prompt Injection Attack via Topic Transition
Large language models (LLMs) have shown remarkable performance across a range of NLP tasks. However, their strong instruction-following capabilities and inability to distinguish instructions from data content make them vulnerable to indirect prompt injection attacks. In such attacks, instructions with malicious purposes are injected into external data sources, such as web documents. When LLMs retrieve this injected data through tools, such as a search engine and execute the injected instruction…

AI Agentic Programming: A Survey of Techniques, Challenges, and Opportunities
Huanting Wang, Jingzhi Gong, Huawei Zhang, Zheng Wang
https://arxiv.org/abs/2508.11126 https://

AI Agentic Programming: A Survey of Techniques, Challenges, and Opportunities
AI agentic programming is an emerging paradigm in which large language models (LLMs) autonomously plan, execute, and interact with external tools like compilers, debuggers, and version control systems to iteratively perform complex software development tasks. Unlike conventional code generation tools, agentic systems are capable of decomposing high-level goals, coordinating multi-step processes, and adapting their behavior based on intermediate feedback. These capabilities are transforming the …

CAD-Driven Co-Design for Flight-Ready Jet-Powered Humanoids
Punith Reddy Vanteddu, Davide Gorbani, Giuseppe L'Erario, Hosameldin Awadalla Omer Mohamed, Fabio Bergonti, Daniele Pucci
https://arxiv.org/abs/2509.14935

CAD-Driven Co-Design for Flight-Ready Jet-Powered Humanoids
This paper presents a CAD-driven co-design framework for optimizing jet-powered aerial humanoid robots to execute dynamically constrained trajectories. Starting from the iRonCub-Mk3 model, a Design of Experiments (DoE) approach is used to generate 5,000 geometrically varied and mechanically feasible designs by modifying limb dimensions, jet interface geometry (e.g., angle and offset), and overall mass distribution. Each model is constructed through CAD assemblies to ensure structural validity a…

Towards a Global Scale Quantum Information Network: A Study Applied to Satellite-Enabled Distributed Quantum Computing
Laurent de Forges de Parny, Luca Paccard, Mathieu Bertrand, Luca Lazzarini, Valentin Leloup, Raphael Aymeric, Agathe Blaise, St\'ephanie Molin, Pierre Besancenot, Cyrille Laborde, Mathias van den Bossche
https://arxiv.…

Towards a Global Scale Quantum Information Network: A Study Applied to Satellite-Enabled Distributed Quantum Computing
Recent developments have reported on the feasibility of interconnecting small quantum registers in a quantum information network of a few meter-scale for distributed quantum computing purposes. This multiple small-scale quantum processors communicating and cooperating to execute computational tasks is considered as a promising solution to the scalability problem of reaching more than thousands of noise-free qubits. Here, we propose and assess a satellite-enabled distributed quantum computing sy…

Arguzz: Testing zkVMs for Soundness and Completeness Bugs
Christoph Hochrainer, Valentin W\"ustholz, Maria Christakis
https://arxiv.org/abs/2509.10819 https://

Arguzz: Testing zkVMs for Soundness and Completeness Bugs
Zero-knowledge virtual machines (zkVMs) are increasingly deployed in decentralized applications and blockchain rollups since they enable verifiable off-chain computation. These VMs execute general-purpose programs, frequently written in Rust, and produce succinct cryptographic proofs. However, zkVMs are complex, and bugs in their constraint systems or execution logic can cause critical soundness (accepting invalid executions) or completeness (rejecting valid ones) issues. We present Arguzz, t…

Switch4EAI: Leveraging Console Game Platform for Benchmarking Robotic Athletics
Tianyu Li, Jeonghwan Kim, Wontaek Kim, Donghoon Baek, Seungeun Rho, Sehoon Ha
https://arxiv.org/abs/2508.13444

Switch4EAI: Leveraging Console Game Platform for Benchmarking Robotic Athletics
Recent advances in whole-body robot control have enabled humanoid and legged robots to execute increasingly agile and coordinated movements. However, standardized benchmarks for evaluating robotic athletic performance in real-world settings and in direct comparison to humans remain scarce. We present Switch4EAI(Switch-for-Embodied-AI), a low-cost and easily deployable pipeline that leverages motion-sensing console games to evaluate whole-body robot control policies. Using Just Dance on the Nint…

Multi-Robot Task Planning for Multi-Object Retrieval Tasks with Distributed On-Site Knowledge via Large Language Models
Kento Murata, Shoichi Hasegawa, Tomochika Ishikawa, Yoshinobu Hagiwara, Akira Taniguchi, Lotfi El Hafi, Tadahiro Taniguchi
https://arxiv.org/abs/2509.12838

Multi-Robot Task Planning for Multi-Object Retrieval Tasks with Distributed On-Site Knowledge via Large Language Models
It is crucial to efficiently execute instructions such as "Find an apple and a banana" or "Get ready for a field trip," which require searching for multiple objects or understanding context-dependent commands. This study addresses the challenging problem of determining which robot should be assigned to which part of a task when each robot possesses different situational on-site knowledge-specifically, spatial concepts learned from the area designated to it by the user. We propose a task plannin…

Unified Hierarchical MPC in Task Executing for Modular Manipulators across Diverse Morphologies
Maolin Lei, Edoardo Romiti, Arturo Laurenzi, Cheng Zhou, Wanli Xing, Liang Lu, Nikos G. Tsagarakis
https://arxiv.org/abs/2508.13513

Unified Hierarchical MPC in Task Executing for Modular Manipulators across Diverse Morphologies
This work proposes a unified Hierarchical Model Predictive Control (H-MPC) for modular manipulators across various morphologies, as the controller can adapt to different configurations to execute the given task without extensive parameter tuning in the controller. The H-MPC divides the control process into two levels: a high-level MPC and a low-level MPC. The high-level MPC predicts future states and provides trajectory information, while the low-level MPC refines control actions by updating th…

Ask-to-Clarify: Resolving Instruction Ambiguity through Multi-turn Dialogue
Xingyao Lin, Xinghao Zhu, Tianyi Lu, Sicheng Xie, Hui Zhang, Xipeng Qiu, Zuxuan Wu, Yu-Gang Jiang
https://arxiv.org/abs/2509.15061

Ask-to-Clarify: Resolving Instruction Ambiguity through Multi-turn Dialogue
The ultimate goal of embodied agents is to create collaborators that can interact with humans, not mere executors that passively follow instructions. This requires agents to communicate, coordinate, and adapt their actions based on human feedback. Recently, advances in VLAs have offered a path toward this goal. However, most current VLA-based embodied agents operate in a one-way mode: they receive an instruction and execute it without feedback. This approach fails in real-world scenarios where …

Toward Ownership Understanding of Objects: Active Question Generation with Large Language Model and Probabilistic Generative Model
Saki Hashimoto, Shoichi Hasegawa, Tomochika Ishikawa, Akira Taniguchi, Yoshinobu Hagiwara, Lotfi El Hafi, Tadahiro Taniguchi
https://arxiv.org/abs/2509.12754

Toward Ownership Understanding of Objects: Active Question Generation with Large Language Model and Probabilistic Generative Model
Robots operating in domestic and office environments must understand object ownership to correctly execute instructions such as ``Bring me my cup.'' However, ownership cannot be reliably inferred from visual features alone. To address this gap, we propose Active Ownership Learning (ActOwL), a framework that enables robots to actively generate and ask ownership-related questions to users. ActOwL employs a probabilistic generative model to select questions that maximize information gain, thereby …

NemeSys: An Online Underwater Explorer with Goal-Driven Adaptive Autonomy
Adnan Abdullah, Alankrit Gupta, Vaishnav Ramesh, Shivali Patel, Md Jahidul Islam
https://arxiv.org/abs/2507.11889

NemeSys: An Online Underwater Explorer with Goal-Driven Adaptive Autonomy
Adaptive mission control and dynamic parameter reconfiguration are essential for autonomous underwater vehicles (AUVs) operating in GPS-denied, communication-limited marine environments. However, most current AUV platforms execute static, pre-programmed missions or rely on tethered connections and high-latency acoustic channels for mid-mission updates, significantly limiting their adaptability and responsiveness. In this paper, we introduce NemeSys, a novel AUV system designed to support real-t…