Agents & Tool Use A

Honda has deployed AI agents in new-car sales to help create higher-quality, opportunity-capturing deals as buyer behavior shifts. The system helps salespeople move beyond passive 'wait-and-see' selling, and has already produced closed sales.

Ajinomoto's finance arm has begun running an accounting AI agent that autonomously handles expense-approval work, cutting workload by 76%. The move makes it an early mover in a field where intolerance for errors had bred caution about adopting AI.

The new AI model "Claude Mythos" makes AI-driven attacks feel imminent, shifting the timeline from months to hours. As vulnerability discovery grows more capable, corporate AI rules and governance lag behind. The article outlines defenses for the AI agent era.

The paper presents Probe-and-Refine, a method for tuning the repository guidance (such as AGENTS.md files) that LLM-based coding agents rely on. It targets the higher-level operational knowledge—file layout, test workflows, and error-prone patterns—that is not contained in the code itself.

Securing AI agents that operate in complex digital environments has become critical, motivating runtime verification. This paper presents an efficient and sound probabilistic verification approach for AI agents.

The paper characterizes when streaming tool use helps in streaming retrieval-augmented generation, which issues tool queries in parallel with ongoing user input to cut perceived latency. It argues the benefit is query-intrinsic and studies how tool intent stabilizes before an utterance is complete.

The paper investigates over-privileged tool selection in LLM agents, which autonomously choose among tools with different privilege levels. It addresses a gap in prior tool-selection research, which focuses on safety-agnostic metadata preferences, by studying when lower-privilege tools would suffice.

The paper presents Connect the Dots (CoD), a reinforcement-learning framework for training large language models as long-lifecycle agents. It targets the meta-capability of solving a long sequence of tasks while continuously exploring an environment, aiming for cross-domain generalization.

Japan Post Insurance, serving 17 million customers, has embedded AI agents into its sales workflow, turning offhand remarks at post offices into insurance proposals. A demonstration shows how the technology changes frontline staff preparing for client meetings.

Reinforcement learning with verifiable rewards, such as GRPO, dominates post-training for complex LLM reasoning but often suffers policy entropy collapse. STARE introduces surprisal-guided token-level advantage reweighting to stabilize policy entropy and preserve exploration during training.

The Web was built on a three-decade assumption that its primary content consumer is human, which permeates every layer of its access model. This work argues for an agent-first web, redesigning the Web for AI agents and rethinking access, structure, and interaction for an agent-driven era.

Multi-turn tool-use RL is bottlenecked by the rapid depletion of informative samples in static datasets. Observing that GRPO's gradient signal concentrates on certain tasks, RODS performs reward-driven online data synthesis to continually supply informative samples for multi-turn tool-use agents.

Production LLM agents increasingly depend on real-time search but get locked into vendor-specific grounding. This work decouples search from reasoning with a vendor-agnostic grounding architecture, letting search backends be swapped while preserving reasoning quality.

Long-context reasoning is essential for large language models. Rather than relying on reward engineering, this work presents a data recipe for long-context reinforcement learning that drives effective training.

A development contest has begun for AI agents that play the Pokemon Trading Card Game. Unlike chess or shogi, it is an "imperfect-information game" where the opponent's hand is hidden, testing how well AI can handle strategic uncertainty.

NVIDIA introduced NVIDIA XR AI, a framework for developers to build AI agents for AR glasses and wearable XR devices. It targets the gap between ready hardware and the work of integrating live, real-time AI experiences. Capabilities are per NVIDIA's own announcement; third-party verification pending.

GitLab announced Project Switch, a next-generation Git-compatible source code management service aimed at AI agents, at its GitLab Transcend event in London. Reports cite up to 50x speed and roughly half the token usage; figures reflect the announcement and remain unverified.

AI agents are shifting from advisors to actors that book travel and run procurement. Existing animal-welfare benchmarks grade only text answers, so this work introduces an agentic benchmark testing whether implicit animal-welfare reasoning transfers to agent actions in frontier models.

Google DeepMind presents an AI Control Roadmap for securing the future of AI agents, combining traditional safeguards with real-time monitoring to protect internal systems. The framework lays out layered defenses against agent misuse and unsafe behavior as agents proliferate.

LLM agents rely on reusable tool specifications (skills), but real tasks require composing multiple skills. The paper formalizes compositional skill routing: decomposing a complex query into atomic sub-tasks, retrieving relevant skills, and composing them.

Tool-using LLM agents use the Model Context Protocol to answer from heterogeneous sources like search, APIs, databases and clinical records. ProvenanceGuard provides source-aware factuality verification to catch provenance-sensitive failure modes that standard metrics miss.

The paper introduces LLM Consumer Behavior Theory, a proposed field analyzing consumer behavior in agentic markets where LLMs make consumption decisions on behalf of users, drawing on classical and behavioral economics alongside NLP.

LLM web agents are usually deployed as tool callers that read a fresh page observation each turn and emit a structured action. The paper proposes reusing web skills across domains via transferable interaction patterns rather than domain-specific behaviors.

Simon Willison released datasette-agent 0.3a0, adding a new 'execute_write_sql' tool that requests user approval before writing to a database while respecting user permissions. It extends the approval mechanism introduced in the prior 0.2a0 release, enabling agent-driven write operations under explicit user consent.

Stack Overflow has launched a beta of 'Stack Overflow for Agents,' a service where AI agents share technical solutions and other information on an open message board. The move appears aimed at extending its human Q&A knowledge base into information exchange among agents.

The paper introduces GIST-CMTF, which augments Causal Minimal Tool Filtering with goal-state inference for tool-augmented LLM agents. It addresses wrong-goal execution, where ambiguous requests such as "handle my appointment" map to multiple goals and an agent may follow a valid causal tool path toward an unintended objective.

The paper proposes Collective Skill Tree Search (CSTS), a tree-search framework that automatically builds reusable skills for LLM agents via iterative collective generation and assessment across multiple models. Claims reflect the abstract.

An arXiv paper reportedly examining preference collapse in multimodal evaluators and its cross-modal contagion within self-evolving agent systems. The source excerpt was unavailable (content filter), so this summary is based on the title only; see the original for methods and findings.

This arXiv paper addresses tool selection for LLM agents whose harnesses connect to hundreds or thousands of APIs, where exhaustive tool-schema injection is costly and imposes a closed-world assumption. Noting that one-shot retrieval often fails to align isolated tool descriptions with the agent's true intent—especially in long-horizon tasks—the authors propose SING, a Synthetic Intention Graph for scalable, active tool discovery.

Sakana AI has launched Sakana Marlin, an AI research agent, commercializing the beta it had offered since April. Ahead of the release it held a press hands-on, showing reporters reports the AI generated from pre-collected themes.