AI Agents

Deepmind's 'AI Agent Traps' Paper Maps How Hackers Could Weaponize AI Agents Against Users

Claude, OpenClaw and the new reality: AI agents are here — and so is the chaos

Agentic AI in Beauty: How ChatGPT Is Reshaping Discovery, Trust, and Conversion

[2602.18419] Benchmarking Graph Neural Networks in Solving Hard Constraint Satisfaction Problems

This paper evaluates the performance of Graph Neural Networks (GNNs) in solving hard constraint satisfaction problems, comparing them aga...

[2510.26752] The Oversight Game: Learning to Cooperatively Balance an AI Agent's Safety and Autonomy

The paper explores a framework for balancing AI agent autonomy and human oversight through a cooperative game model, ensuring safety with...

[2602.18386] Learning to Tune Pure Pursuit in Autonomous Racing: Joint Lookahead and Steering-Gain Control with PPO

This article presents a reinforcement learning approach to optimize Pure Pursuit parameters in autonomous racing, enhancing path tracking...

[2602.18372] "How Do I ...?": Procedural Questions Predominate Student-LLM Chatbot Conversations

This paper investigates the predominance of procedural questions in student interactions with LLM chatbots, analyzing data from various l...

[2602.18374] Zero-shot Interactive Perception

The paper presents Zero-Shot Interactive Perception (ZS-IP), a framework that enhances robotic manipulation through a memory-driven Visio...

[2602.18346] Vichara: Appellate Judgment Prediction and Explanation for the Indian Judicial System

Vichara is a framework designed to predict and explain appellate judgments in the Indian judicial system, leveraging AI to enhance legal ...

[2602.18351] Validating Political Position Predictions of Arguments

This article presents a dual-scale validation framework for assessing political position predictions in argumentative discourse, utilizin...

[2602.18319] Robo-Saber: Generating and Simulating Virtual Reality Players

The paper presents Robo-Saber, a motion generation system designed for playtesting virtual reality games, specifically focusing on genera...

[2602.18283] HyTRec: A Hybrid Temporal-Aware Attention Architecture for Long Behavior Sequential Recommendation

HyTRec introduces a Hybrid Temporal-Aware Attention architecture designed to enhance long behavior sequential recommendations, improving ...

[2602.18262] Simplifying Outcomes of Language Model Component Analyses with ELIA

The paper presents ELIA, an interactive web application designed to simplify the analysis of Large Language Models (LLMs) for non-experts...

[2602.18172] Can AI Lower the Barrier to Cybersecurity? A Human-Centered Mixed-Methods Study of Novice CTF Learning

This study explores how AI can facilitate novice learning in cybersecurity Capture-the-Flag (CTF) competitions by lowering entry barriers...

[2602.17917] Interactions that reshape the interfaces of the interacting parties

This paper introduces polynomial trees to model dynamic systems where interactions reshape interfaces, enhancing understanding of state-d...

[2602.17894] Learning from Biased and Costly Data Sources: Minimax-optimal Data Collection under a Budget

This paper explores optimal data collection strategies from biased and costly sources, focusing on maximizing effective sample size under...

[2602.18137] Agentic Adversarial QA for Improving Domain-Specific LLMs

The paper presents an adversarial question-generation framework aimed at enhancing the performance of domain-specific large language mode...

[2602.18104] MeanVoiceFlow: One-step Nonparallel Voice Conversion with Mean Flows

MeanVoiceFlow introduces a one-step nonparallel voice conversion model that enhances speech quality and speaker similarity while reducing...

[2602.17787] Market Games for Generative Models: Equilibria, Welfare, and Strategic Entry

This paper explores market dynamics in generative model ecosystems, focusing on equilibria, welfare implications, and strategic entry by ...

[2602.18029] Towards More Standardized AI Evaluation: From Models to Agents

This paper discusses the evolution of AI evaluation from static models to dynamic agents, emphasizing the need for standardized evaluatio...

[2602.18026] Mean-Field Reinforcement Learning without Synchrony

This paper presents a new framework for Mean-Field Reinforcement Learning (MF-RL) that addresses the challenges of asynchrony in multi-ag...

[2602.18022] Dual-Channel Attention Guidance for Training-Free Image Editing Control in Diffusion Transformers

This paper introduces Dual-Channel Attention Guidance (DCAG), a novel training-free method for enhancing image editing control in Diffusi...

[2602.17773] Learning Flow Distributions via Projection-Constrained Diffusion on Manifolds

The paper presents a novel generative modeling framework for synthesizing physically feasible two-dimensional incompressible flows, addre...