Related papers: CodeCompass: Navigating the Navigation Paradox in …
Autonomous AI agents executing multi-step tool sequences face semantic attacks that manifest in behavioral traces rather than isolated prompts. A critical challenge is cross-attack generalization: can detectors trained on known attack…
Agentic workflows built on low-code orchestration platforms enable rapid development of multi-agent systems, but they also introduce new and poorly understood failure modes that hinder reliability and maintainability. Unlike traditional…
PointGoal Navigation is an embodied task that requires agents to navigate to a specified point in an unseen environment. Wijmans et al. showed that this task is solvable but their method is computationally prohibitive, requiring 2.5 billion…
Multi-agent systems powered by Large Language Models face a critical challenge: agents communicate through natural language, leading to semantic drift, hallucination propagation, and inefficient token consumption. We propose G2CP…
We consider an Anonymous Multi-Agent Path-Finding (AMAPF) problem where the set of agents is confined to a graph, a set of goal vertices is given and each of these vertices has to be reached by some agent. The problem is to find an…
Code intelligence is an emerging domain in software engineering, aiming to improve the effectiveness and efficiency of various code-related tasks. Recent research suggests that incorporating contextual information beyond the basic original…
Modern retrieval systems, whether lexical or semantic, expose a corpus through a fixed similarity interface that compresses access into a single top-k retrieval step before reasoning. This abstraction is efficient, but for agentic search,…
Planning is a crucial task for agents in task oriented dialogs (TODs). Human agents typically resolve user issues by following predefined workflows, decomposing workflow steps into actionable items, and performing actions by executing APIs…
Web navigation represents a critical and challenging domain for evaluating artificial general intelligence (AGI), demanding complex decision-making within high-entropy, dynamic environments with combinatorially explosive action spaces.…
Autonomous science agents built on large language models (LLMs) are increasingly used to generate hypotheses, design experiments, and produce reports. However, prior work mainly targets open-ended scientific problems with subjective outputs…
As agentic coding systems decompose work across multiple model instances, a critical safety question is whether those instances can coordinate to achieve a hidden malicious objective while remaining aligned with user intent. We introduce…
Recent advances in large language model-powered multi-agent systems have demonstrated remarkable collective intelligence through effective communication. However, existing approaches face two primary challenges: (i) \textit{Ineffective…
Modern distributed systems rely on complex networks of interconnected services, creating direct or indirect dependencies that can propagate faults and cause cascading failures. To localize the root cause of performance degradation in these…
Recent advances in coding agents have shown remarkable progress in software issue resolution. In practice, real-world issues are typically bug fixes or feature requests in which human developers naturally incorporate refactoring as part of…
Predicting program behavior without execution is a critical task in software engineering. Existing models often fall short in capturing the dynamic dependencies among program elements. To address this, we present CodeFlow, a novel machine…
Effective tool pre-selection via retrieval is essential for AI agents to select from a vast array of tools when identifying and planning actions in the context of complex user queries. Despite its central role in planning, this aspect…
Agent frameworks increasingly encode tool-using behavior as explicit workflow graphs, yet safety enforcement remains a runtime concern. These frameworks expose analyzable graph structure through their APIs, enabling pre-deployment static…
In the first half of 2025, coding agents have emerged as a category of development tools that have very quickly transitioned to the practice. Unlike ''traditional'' code completion LLMs such as Copilot, agents like Cursor, Claude Code, or…
Neural Architecture Search (NAS) has been quite successful in constructing state-of-the-art models on a variety of tasks. Unfortunately, the computational cost can make it difficult to scale. In this paper, we make the first attempt to…
Vision-and-Language Scene navigation is a fundamental capability for embodied human-AI collaboration, requiring agents to follow natural language instructions to execute coherent action sequences in complex environments. Existing approaches…