Agents

The paper investigates how LLMs allocate their internal computational depth during multi-turn agentic planning, finding that agents progressively recruit deeper layers and shift toward corrective upda…

The paper introduces AgenticVBench, a comprehensive benchmark of 100 real-world video post-production tasks, and finds that even the best AI agents perform significantly worse than human experts on th…

The paper proposes an interaction-based legal framework for assigning tort liability when autonomous AI systems cause harm, categorizing liability based on the nature of the human-AI interaction.

The paper introduces extsc{Ptah}, a multi-agent harness designed to improve verifiable multimodal deep research by orchestrating the entire report generation process, ensuring factual grounding and v…

This survey provides a comprehensive, practical guide to ensuring the trustworthiness of complex, autonomous agentic AI systems by focusing on safety, robustness, privacy, and system security.

The paper surveys the use of LLMs for agentic NetOps and AIOps, arguing that operational reliability depends not on the model itself, but on robust surrounding machinery and workflow-centered evaluati…

The paper introduces Momento, a new benchmark that evaluates agentic AI's ability to maintain state and reason across multiple, disconnected sessions, revealing that current agents struggle with integ…

The paper introduces GAIATrace, a comprehensive token-level dataset, and Vidur-Agent, a simulator, to enable reproducible and detailed system-level characterization of complex multi-model agentic AI s…

This paper introduces a new benchmark to test Tool Description Poisoning (TDP) attacks on LLM agents, demonstrating that even advanced models like GPT-4o are highly vulnerable and that current defense…

The paper introduces VeriTrip, a new verifiable benchmark that evaluates travel planning agents' ability to perform evidence-grounded reasoning over complex, unstructured, and multimodal web data, rev…

This paper empirically demonstrates that the architectural design of multi-agent systems significantly impacts their security, finding that coordination mechanisms can introduce vulnerabilities greate…

DynaTree introduces a two-stage framework that pre-constructs a reusable retrieval tree offline using coordinated agents, allowing for efficient, structure-aware, and highly effective time-sensitive n…

The paper introduces ARGUS, a defense mechanism that uses provenance-aware decision auditing to protect LLM agents from sophisticated, context-aware prompt injection attacks, significantly reducing th…

The paper proposes the Secret-Use Delegation Protocol (SUDP) to solve the Agent Secret Use (ASU) problem, ensuring that autonomous agents can perform user-authorized operations without gaining reusabl…

The paper introduces AutoMedBench, a novel workflow-aware benchmark that evaluates autonomous medical-AI agents across a five-stage research process, revealing that agents struggle most with validatio…

The paper argues that current 'on-the-fly' AI agent design lacks necessary software engineering rigor and proposes an 'AI Workflow Store' to provide hardened, reusable, and reliable agent workflows.

The paper introduces a self-healing agentic orchestrator that significantly improves the reliability of tool-augmented LLM systems by treating failure as a bounded runtime control problem, achieving h…

The paper proposes a hybrid LLM-based assistance system that enhances traditional capability-based planning by providing natural language interaction, interpretability, and flexible knowledge model ad…

The paper introduces STRIATUM-CTF, a modular agentic framework that uses a standardized context protocol to enable LLMs to perform multi-step, stateful reasoning for general-purpose CTF solving, achie…

COMPASS introduces a Cognitive MCTS-Guided Process Alignment framework to ensure robust safety for LLM search agents by identifying and supervising risky intermediate steps in multi-step reasoning.