The paper introduces RAC, a relational agent contract designed for engineering tasks governed by strict constraints. RAC separates semantic decisions from tool-based derivations, enabling agents to specify desired relations while specialized tools compute precise outcomes. Experiments on PCB placement, routing, repair, job-shop scheduling, and embedded-code tasks show that RAC consistently outperforms baseline methods, achieving higher clean-pass and quality pass rates across multiple models and domains.
arXiv:2605. 10555v2 Announce Type: replace Abstract: As AI agents transition from research prototypes to enterprise production systems, the tool interfaces they consume remain rooted in human-oriented CRUD paradigms.
Tool Calling and Structured Output are two core capabilities of modern Agent systems, yet their interaction under joint deployment conditions remains insufficiently understood. This paper reports a reproducible phenomenon observed in a production Agent system: when Tool Calling and JSON Schema constraints are simultaneously enabled, multiple open-weight models cease invoking tools despite maintaining high schema compliance.
The paper investigates whether locally deployed large language models can automate hardware design workflows that involve repetitive, dependency-ordered operations using specialized tools. A Model Context Protocol (MCP) server is created to emulate a proprietary hardware design tool, and a benchmark tests single and multi-step edits, invalid requests, misspelled prompts, and multi-server contexts. Seven open-source models are evaluated across different pipeline choices, revealing that strong models can nearly fully cover expected calls, but reliability hinges on task structure and agent configuration, with comprehensive tool descriptions reducing failures and multi-agent setups aiding weaker models at the cost of extra calls.
arXiv:2607. 10059v1 Announce Type: new Abstract: Agent systems based on large language models (LLMs) are increasingly deployed for autonomous tasks, yet existing evaluations mostly focus on task success rather than whether agents know when to abstain.
The paper introduces Skillware, an open architectural framework that separates an agent’s intent from its execution by providing modular, versioned software capability bundles. These bundles include typed JSON contracts, deterministic Python code, task directives, datasets, and unit tests, allowing agents to perform tasks reliably without relying on fragile prompt engineering. Experiments across 23 models and 368 benchmarks show that small local models equipped with Skillware outperform larger frontier models, reducing token usage by up to 84.8%, cutting inference costs by up to 97.3%, and eliminating parameter errors.
CCTU is a new benchmark designed to evaluate large language models (LLMs) on their ability to use tools under complex constraints. It includes 200 test cases that average seven constraint types and 4,700‑token prompts, covering resource, behavior, toolset, and response dimensions. An executable validation module performs step‑level checks, and nine state‑of‑the‑art LLMs were tested, revealing that none exceed a 20% task completion rate when strict constraints are enforced, with frequent violations and limited self‑refinement.
arXiv:2606. 31252v1 Announce Type: new Abstract: Large language models can write plausible CAD scripts, but reliable industrial CAD modeling requires more than syntactically valid code: every feature, placement, and assembly relation must be accepted by an exact geometric kernel while remaining editable as parametric boundary representation geometry.
arXiv:2606. 19787v1 Announce Type: new Abstract: Large language models are increasingly deployed as autonomous agents for multi-step tasks in executable environments, yet their ability to perform realistic operations research (OR) work remains unclear.
arXiv:2608. 02650v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly rely on external tools to complete complex real-world tasks.
arXiv:2606. 02965v2 Announce Type: replace Abstract: As large language models gain tool access and are deployed as autonomous agents capable of editing records, executing transactions, and modifying infrastructure, we still evaluate them based on the sole metric of task completion.
Harness-Bench is a diagnostic benchmark that evaluates how different harness configurations affect LLM agent performance in realistic workflows. It tests 106 sandboxed offline tasks across multiple model backends, recording artifacts, execution traces, and usage statistics to analyze completion, process quality, efficiency, and failure behavior. The study finds significant variation in outcomes across model-harness pairings, indicating that agent capability should be reported at the configuration level rather than solely by the base model.
The paper introduces URAI, a Universal Robot‑Agent Interface that separates robot control into two roles: a programming agent that writes reusable, task‑specific tools from intent, and an execution agent that calls these tools in a feedback loop. This design keeps high‑level decision making in the model while delegating low‑level motion to code, allowing tool revisions to persist across episodes without retraining the foundation model. Experiments on RoboDojo and AgileX tasks show significant gains in success rate, speed, and token efficiency compared to direct fingertip control and pre‑written programs.