Search: “tools”

arXiv:2605.08368v1 Announce Type: new Abstract: Debates about large language model post-training often treat supervised fine-tuning (SFT) as imitation and reinforcement learning (RL) as discovery. But

local-aiarxiv-cs-ai

Local Ai

PAAC: Privacy-Aware Agentic Device-Cloud Collaboration

DGX agent

arXiv:2605.08646v1 Announce Type: cross Abstract: Large language model (LLM) agents face a structural tension: cloud agents provide strong reasoning but expose user data, while on-device agents preser

local-aiarxiv-cs-cl

12 May 2026

PaperFit: Vision-in-the-Loop Typesetting Optimization for Scientific Documents

arXiv:2605.10341v1 Announce Type: new Abstract: A LaTeX manuscript that compiles without error is not necessarily publication-ready. The resulting PDFs frequently suffer from misplaced floats, overflo

Pix2Fact: When Vision Is Not Enough -- Benchmarking Fine-Grained VQA with Web Verification on High-Resolution Real-World Scenes

model-releasesarxiv-cs-cv

arXiv:2602.00593v2 Announce Type: replace Abstract: Despite progress on general tasks, vision-language models (VLMs) still struggle with challenges that demand both fine-grained visual grounding and e

Position: Life-Logging Video Streams Make the Privacy-Utility Trade-off Inevitable

arXiv:2605.10404v1 Announce Type: new Abstract: With the growing prevalence of always-on hardware such as smart glasses, body cameras, and home security systems, life-logging visual sensing is becomin

researcharxiv-cs-cv

PRISM: Fast Online LLM Serving via Scheduling-Memory Co-design

arXiv:2605.08581v1 Announce Type: new Abstract: Modern online large language model (LLM) services, such as Retrieval-Augmented Generation (RAG) and agent systems, increasingly expose two prominent cha

agentsarxiv-cs-lg

Privacy Auditing Synthetic Data Release through Local Likelihood Attacks

arXiv:2508.21146v2 Announce Type: replace Abstract: Auditing the privacy leakage of synthetic data is an important but unresolved problem. Existing privacy auditing frameworks for synthetic data rely

Revisiting Mixture Policies in Entropy-Regularized Actor-Critic

arXiv:2605.09157v1 Announce Type: cross Abstract: Mixture policies theoretically offer greater flexibility than unimodal policies in continuous action reinforcement learning, but the practical benefit

researcharxiv-cs-ai

RubricEM: Meta-RL with Rubric-guided Policy Decomposition beyond Verifiable Rewards

arXiv:2605.10899v1 Announce Type: new Abstract: Training deep research agents, namely systems that plan, search, evaluate evidence, and synthesize long-form reports, pushes reinforcement learning beyo

safetyarxiv-cs-cl

SalesSim: Benchmarking and Aligning Multimodal Language Models as Retail User Simulators

arXiv:2605.08334v1 Announce Type: new Abstract: We present SalesSim, a framework and testbed for evaluating the ability of Multimodal Large Language Models (MLLMs) to simulate realistic, persona-drive

safetyarxiv-cs-cl

SAR-RAG: ATR Visual Question Answering by Semantic Search, Retrieval, and MLLM Generation

arXiv:2602.04712v2 Announce Type: replace-cross Abstract: We present a visual-context image-retrieval-augmented generation (ImageRAG)- assisted AI agent for automatic target recognition (ATR) of synth

SimWorld Studio: Automatic Environment Generation with Evolving Coding Agent for Embodied Agent Learning

arXiv:2605.09423v1 Announce Type: new Abstract: LLM/VLM-based digital agents have advanced rapidly thanks to scalable sandboxes for coding, web navigation, and computer use, which provide rich interac

SmartEval: A Benchmark for Evaluating LLM-Generated Smart Contracts from Natural Language Specifications

arXiv:2605.09610v1 Announce Type: cross Abstract: We introduce SmartEval, a benchmark for systematically evaluating the quality of Solidity smart contracts generated by large language models (LLMs) fr

Industry

Source: Anthropic is in advanced talks to acquire New York-based Stainless, which helps developers generate SDKs from APIs, for at least $300M (The Information)

The Information: Source: Anthropic is in advanced talks to acquire New York-based Stainless, which helps developers generate SDKs from APIs, for at least 300M — Anthropic is in advanced talks to acqui

industrytechmeme

Industry

Sources: Wispr Flow developer Wispr AI is in talks to raise a round that could more than double its valuation to 2B; source: the round is set to total ~260M (Bloomberg)

Bloomberg: Sources: Wispr Flow developer Wispr AI is in talks to raise a round that could more than double its valuation to 2B; source: the round is set to total ~260M — Wispr AI Inc., the developer b

industrytechmeme

Spherical Boltzmann machines: a solvable theory of learning and generation in energy-based models

arXiv:2605.09031v1 Announce Type: new Abstract: Energy-based models (EBMs) are flexible generative architectures inspired by statistical physics, but their learning and generative properties remain po

Steerable but Not Decodable: Function Vectors Operate Beyond the Logit Lens

arXiv:2604.02608v2 Announce Type: replace Abstract: Activation steering presupposes that task-relevant behaviors correspond to linear directions in activation space -- directions that should both stee

Strategic Exploitation in LLM Agent Markets: A Simulation Framework for E-Commerce Trust

arXiv:2605.10059v1 Announce Type: new Abstract: Agent-based modeling (ABM) has long been used in economics to study human behavior, and large language model (LLM) agents now enable new forms of social

Survey on Disaster Management Datasets for Remote Sensing Based Emergency Applications

arXiv:2605.08196v1 Announce Type: new Abstract: Recent natural disasters have highlighted the urgent need for efficient data-driven approaches to disaster management. Machine learning (ML) and deep le

researcharxiv-cs-cv

Industry

The 9 biggest new features in Android 17

Would it shock you to hear that Android 17 is filled with new AI-enabled features, like improved dictation and vibe-coded widgets? Fortunately, that's not all. The platform is getting non-AI updates t

industrythe-verge-ai

The Agent Use of Agent Beings: Agent Cybernetics Is the Missing Science of Foundation Agents

arXiv:2605.10754v1 Announce Type: new Abstract: LLM-based foundation agents that perceive, reason, and act across thousands of reasoning steps are rapidly becoming the dominant paradigm for deploying

The Association of Transformer-based Sentiment Analysis with Symptom Distress and Deterioration in Routine Psychotherapy Care

arXiv:2605.09838v1 Announce Type: new Abstract: Sentiment analysis has been of long-standing interest in psychotherapy research. Recently, the Transformer deep learning architecture has produced text-

researcharxiv-cs-cl

the Mini Shai-Hulud attack is scary because it attacks new AI coding workflows like CI, editor hooks, agent configs, etc

The Mini Shai-Hulud attack targets emerging AI-assisted development workflows by compromising multiple integration points including continuous integration systems, code editor hooks, and AI agent conf

agentsyohei-nakajima--x

The Pokemon Theorem and other Fairness Impossibility Results

arXiv:2605.09221v1 Announce Type: cross Abstract: Fairness impossibility results often look like distinct scalar incompatibility statements. We show that several share one RKHS geometry: fairness crit

There will be no AI jobpocalypse. The story that AI will lead to massive unemployment is stoking unnecessary fear. AI — like any other techn…

There will be no AI jobpocalypse. The story that AI will lead to massive unemployment is stoking unnecessary fear. AI — like any other technology — does affect jobs, but telling overblown stories of l

safetyandrew-ng--x

thoughts after doing a bunch of synthetic data gen for eval + environment building - LLMs are incredible projections of the world bundled in…

thoughts after doing a bunch of synthetic data gen for eval + environment building - LLMs are incredible projections of the world bundled into a set of weights - but doing targeted extraction of certa

agentsharrison-chase--x

Applications

Topological Data Analysis Applications in Natural Language Processing: A Survey

arXiv:2411.10298v5 Announce Type: replace Abstract: The surge of data available on the Internet has driven the adoption of a wide range of computational methods for analyzing and extracting insights f

applicationsarxiv-cs-cl

Towards a Certificate of Trust: Task-Aware OOD Detection for Scientific AI

arXiv:2509.25080v3 Announce Type: replace Abstract: Data-driven models are increasingly adopted in critical scientific fields like weather forecasting and fluid dynamics. These methods can fail on out

researcharxiv-cs-lg

Towards a Virtual Neuroscientist: Autonomous Neuroimaging Analysis via Multi-Agent Collaboration

arXiv:2605.09366v1 Announce Type: new Abstract: Transforming neuroimaging data into clinically actionable biomarkers is a knowledge-intensive and labor-intensive process. Standardized workflows such a

Trajectory-Consistent Flow Matching for Robust Visuomotor Policy Learning

arXiv:2605.08511v1 Announce Type: new Abstract: Flow matching policies learn continuous velocity fields that transport noise to actions, enabling fast deterministic inference for robot manipulation. H

safetyarxiv-cs-ro

TrajPrism: A Multi-Task Benchmark for Language-Grounded Urban Trajectory Understanding

arXiv:2605.10782v1 Announce Type: new Abstract: Urban mobility is naturally expressed both as trajectories in space and as natural-language descriptions of travel intent, constraints, and preferences.

Applications

Transformer autoencoder with local attention for sparse and irregular time series with application on risk estimation

arXiv:2605.08914v1 Announce Type: cross Abstract: This paper introduces a framework specifically designed for sparse and irregular time series {risk estimation}. It is based on a Transformer Autoencod

applicationsarxiv-cs-ai

Transforming the Use of Earth Observation Data: Exascale Training of a Generative Compression Model with Historical Priors for up to 10,000x Data Reduction

arXiv:2605.08633v1 Announce Type: cross Abstract: Earth observation is becoming one of the largest data-producing activities in science, yet current pipelines still treat compression as a storage and

researcharxiv-cs-cv

Trustworthy AI: Ensuring Reliability and Accountability from Models to Agents

arXiv:2605.08964v1 Announce Type: new Abstract: In this thesis, we develop algorithms with theoretical guarantees for ensuring reliability and accountability of Machine Learning (ML) systems. As ML sy

safetyarxiv-cs-lg

Universal Feature Selection with Noisy Observations and Weak Symmetry Conditions

arXiv:2605.09396v1 Announce Type: cross Abstract: This paper relaxes the restrictive symmetry conditions adopted in [4], [5] and extends their universal feature selection framework to accommodate nois

researcharxiv-cs-lg

V-ABS: Action-Observer Driven Beam Search for Dynamic Visual Reasoning

arXiv:2605.10172v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) have achieved remarkable success in general perception, yet complex multi-step visual reasoning remains a per

safetyarxiv-cs-cl

Voice choice shapes how an agent feels to users, from fintech support to healthcare intake to entertainment. Try voice finder: https://findt…

Voice selection significantly impacts user perception and experience across various applications, including financial services support, healthcare intake processes, and entertainment platforms. Togeth

agentstogether-ai--x

VT-Bench: A Unified Benchmark for Visual-Tabular Multi-Modal Learning

arXiv:2605.08146v1 Announce Type: cross Abstract: Multi-model learning has attracted great attention in visual-text tasks. However, visual-tabular data, which plays a pivotal role in high-stakes domai

What Software Engineering Looks Like to AI Agents? -- An Empirical Study of AI-Only Technical Discourse on MoltBook

arXiv:2605.08380v1 Announce Type: cross Abstract: AI agents are increasingly framed as software-engineering teammates, yet most research studies them inside human-centered workflows. Little is known a

When Agents Overtrust Environmental Evidence: An Extensible Agentic Framework for Benchmarking Evidence-Grounding Defects in LLM Agents

arXiv:2605.08828v1 Announce Type: new Abstract: Large language model agents increasingly operate through environment-facing scaffolds that expose files, web pages, APIs, and logs. These observations i

When Child Inherits: Modeling and Exploiting Subagent Spawn in Multi-Agent Networks

arXiv:2605.08460v1 Announce Type: cross Abstract: Since the official release of ChatGPT in 2022, large language models (LLMs) have rapidly evolved from chatbot-style interfaces into agentic systems th

123D: Unifying Multi-Modal Autonomous Driving Data at Scale

arXiv:2605.08084v1 Announce Type: cross Abstract: The pursuit of autonomous driving has produced one of the richest sensor data collections in all of robotics. However, its scale and diversity remain

agentsarxiv-cs-cv

A Reproducible Optimisation Protocol for Calibrating Prompt-Based Large Language Model Workflows in Evidence Synthesis

arXiv:2605.06937v1 Announce Type: new Abstract: This methods article presents a reproducible calibration workflow for prompt-based large language models (LLMs) in structured evidence-synthesis tasks.

A Systematic Investigation of The RL-Jailbreaker in LLMs

arXiv:2605.07032v1 Announce Type: cross Abstract: The evolution of generative models from next-token predictors to autonomous engines of complex systems necessitates rigorous safety hardening. Adversa

Activation Differences Reveal Backdoors: A Comparison of SAE Architectures

arXiv:2605.07324v1 Announce Type: cross Abstract: Backdoor attacks on language models pose a significant threat to AI safety, where models behave normally on most inputs but exhibit harmful behavior w

AffineLens: Capturing the Continuous Piecewise Affine Functions of Neural Networks

arXiv:2605.06218v2 Announce Type: replace Abstract: Piecewise affine neural networks (PANNs) provide a principled geometric perspective on neural network expressivity by characterizing the input--outp

researcharxiv-cs-lg

Agent engineering is hard Deep Agents hides a lot of the weird systems complexity, but still gives you more room to customize the harness th…

Agent engineering is hard Deep Agents hides a lot of the weird systems complexity, but still gives you more room to customize the harness than almost any agent SDK I've used This is harder to build th

agentsharrison-chase--x

Agentic Coding Needs Proactivity, Not Just Autonomy

arXiv:2605.06717v1 Announce Type: cross Abstract: Coding agents are rapidly changing the landscape of software development, moving from inline completion to autonomous systems that edit repositories,