A focused look at this week’s most significant advances in agents & tool use \u2014 14 papers surveyed from arXiv and leading AI labs.<\/p>\n
—<\/p>\n
Agentic AI is maturing fast. This week’s papers explore how agents plan, execute, collaborate, and \u2014 crucially \u2014 how they fail. Several papers address the reliability gap that’s emerged as a central challenge.<\/p>\n
ShareLock: A Stealthy Multi-Tool Threshold Poisoning Attack Against MCP<\/strong> \u2014 Liwei Liu, Tianzhu Han, Zijian Liu, Zishu Dong, Na Ruan<\/em><\/p>\n With the rapid evolution of LLM-driven agents, Model Context Protocol (MCP), an open protocol bridging LLMs with external tools, has quickly become foundational to modern agent ecosystems. However, th…<\/p>\n arXiv<\/a><\/p>\n The Red Queen G\u00f6del Machine: Co-Evolving Agents and Their Evaluators<\/strong> \u2014 Alex Iacob, Andrej Jovanovi\u0107, William F. Shen, Daniel Burkhardt, Meghdad Kurmanji, Nurbek Tastan, Lorenzo Sani, Niccol\u00f2 <\/em><\/p>\n Self-improving agents are state-of-the-art (SOTA) on agentic coding benchmarks and have recently been extended to general domains. However, their search methods generally assume a stationary evaluatio…<\/p>\n arXiv<\/a><\/p>\n Empowering GUI Agents via Autonomous Experience Exploration and Hindsight Experience Utilization for Task Planning<\/strong> \u2014 Tianyi Men, Zhuoran Jin, Pengfei Cao, Yubo Chen, Kang Liu, Jun Zhao<\/em><\/p>\n Multimodal web agents can assist humans in operating repetitive GUI tasks, where effective task planning is essential for decomposing complex tasks into executable actions. While small open source MLL…<\/p>\n arXiv<\/a><\/p>\n OpenRCA 2.0: From Outcome Labels to Causal Process Supervision<\/strong><\/p>\n Aoyang Fang, Yifan Yang, Jin’ao Shang, Qisheng Lu, Junjielung Xu, Rui Wang, Songhan Zhang, Yuzhong Z<\/em><\/p>\n Root cause analysis (RCA) poses a holistic test of LLM agentic capabilities, such as long-context understanding, multi-step reasoning, and tool use. However, existing datasets suffer from a fundamental gap: they label only the root cause, not the…<\/p>\n arXiv<\/a><\/p>\n Bridging Talk and Thought: Understanding Dialogue Dynamics Across Collaborative Problem-Solving Contexts<\/strong><\/p>\n Zhengyuan Liu, Stella Xin Yin, Min-Yen Kan, Nancy F. Chen<\/em><\/p>\n We present a conceptual framework for analyzing dialogue in collaborative problem-solving contexts, with an emphasis on the emerging dynamics of human-AI and multi-agent collaboration. As intelligent systems become active agents capable of…<\/p>\n arXiv<\/a><\/p>\n The Spec Growth Engine: Spec-Anchored, Code-Coupled, Drift-Enforced Architecture for AI-Assisted Software Development<\/strong><\/p>\n Hartwig Grabowski<\/em><\/p>\n AI coding agents dramatically accelerate implementation speed but introduce two structural failure modes that existing spec-driven approaches do not fully solve: (1) context explosion — the agent must reason over an entire repository at once,…<\/p>\n arXiv<\/a><\/p>\n Surviving by Serving: Functional Relevance Drives Self-Organization in Complex Adaptive Systems<\/strong><\/p>\n Claus Metzner, Ali Ghebleh, Achim Schilling, Andreas Maier, Thomas Kinfe, Patrick Krauss<\/em><\/p>\n Complex adaptive systems often develop organized structures without centralized control. Yet the local mechanisms by which functional organization emerges and persists remain incompletely understood. Here we propose Surviving by Serving (SBS) as a…<\/p>\n arXiv<\/a><\/p>\n AgentX: Towards Agent-Driven Self-Iteration of Industrial Recommender Systems<\/strong><\/p>\n Changxin Lao, Fei Pan, Guozhuang Ma, Han Li, Huihuang Lin, Jijun Shi, Kangzhi Zhao, Kun Gai, Mo Zhou<\/em><\/p>\n Recommendation algorithm iteration is moving from an artisanal, engineer-bound process toward an industrialized research loop, but this transition remains blocked by a structural execution bottleneck: the idea-to-launch cycle still depends on human…<\/p>\n arXiv<\/a><\/p>\n Proposal-Conditioned Latent Diffusion for Closed-Loop Traffic Scenario Generation<\/strong><\/p>\n Shubham Vaijanath Phoolari, Aleyna Kara, Christoph Lauer, Steven Peters<\/em><\/p>\n Closed-loop traffic simulation remains challenging because it must generate interactive multi-agent behaviors that are scene-consistent and controllable throughout rollout. Prior diffusion-based approaches achieve strong realism, but their…<\/p>\n arXiv<\/a><\/p>\n Qwen-Image-Agent: Bridging the Context Gap in Real-World Image Generation<\/strong><\/p>\n Zekai Zhang, Jiahao Li, Jie Zhang, Kaiyuan Gao, Kun Yan, Lihan Jiang, Ningyuan Tang, Shengming Yin, <\/em><\/p>\n While text-to-image (T2I) models have achieved remarkable progress, they struggle with real-world requests that are often underspecified, implicit, or dependent on up-to-date knowledge. We identify this challenge as the Context Gap: the mismatch…<\/p>\n arXiv<\/a><\/p>\n EvoFlock: evolved inverse design of multi-agent motion<\/strong><\/p>\n Craig Reynolds<\/em><\/p>\n This paper describes an automatic method for adjusting or tuning models of multi-agent motion. Simulating the motion of bird flocks, human crowds, vehicle traffic, and other multi-agent systems is a widely used technique. These simulations model…<\/p>\n arXiv<\/a><\/p>\n Decomposing Financial Market Dynamics via Mechanism Analysis in an Evolutionary Multi-Agent Simulation<\/strong><\/p>\n Zhibao Chen<\/em><\/p>\n Evolutionary agent-based markets (ABMs) couple several mechanisms — who reproduces, how price forms, how biased the agents are, how consensus propagates — yet these are usually fixed by convention, so it is unclear which mechanism controls which…<\/p>\n arXiv<\/a><\/p>\n Look-Before-Move: Narrative-Grounded World Visual Attention in Dynamic 3D Story Worlds<\/strong><\/p>\n Jiaming Bian, Bingliang Li, Yuehao Wu, Pichao Wang, Zhi Wang, Hailan Ma, Huadong Mo, Zhenhong Sun<\/em><\/p>\n As embodied AI and world models increasingly operate in dynamic 3D environments, visual perception must move beyond passively interpreting given observations toward actively deciding what to observe. We study this problem through camera planning in…<\/p>\nReasoning & Inference<\/h2>\n
Agent Systems<\/h2>\n
Additional Research<\/h2>\n