arXiv:2608. 01004v1 Announce Type: new Abstract: Platform teams hosting agent-extensibility surfaces face a regression-economics paradox: every onboarding customer ships an evaluation set tuned to their domain, but the platform's regression set must live under a hard query-count ceiling bounded by release cadence.
Agent evaluations tell us that a model picked the wrong tool, but rarely why. We introduce canary tools: diagnostic probe tools planted in an agent's Model Context Protocol (MCP) tool set, each engineered to probe one specific tool-selection weakness.
arXiv:2608. 04719v1 Announce Type: new Abstract: Agent evaluations tell us that a model picked the wrong tool, but rarely why.
arXiv:2607. 25891v1 Announce Type: new Abstract: Evaluating AI agents in interactive environments is hindered by fragmented tasks, scaffolds, verifiers, and scoring rules.
arXiv:2607. 16345v1 Announce Type: cross Abstract: Modern agentic systems increasingly rely on skills: installable packages of natural language and code that teach an LLM agent to perform a domain task.
Semantic Feature Analysis (SFA) is a method that refines agent specifications without performing any rollout-based search. It analyzes existing execution traces, clusters workflow node outputs, extracts semantic feature classes via an extended subject‑verb‑object schema, ranks these features with a decision tree, and injects the most impactful features back into the system prompt. Evaluations on four benchmarks show that SFA consistently outperforms five prompt‑optimisation algorithms and a single‑reflection baseline, especially when rollout costs are high.
arXiv:2607. 16122v1 Announce Type: new Abstract: Evaluations should do more than measure a models current performance.
arXiv:2607. 00011v2 Announce Type: replace-cross Abstract: Reusable agent skills are emerging as a service-oriented capability layer for Large Language Model (LLM) agents.
JudgeStealer is a query‑efficient framework that extracts the judging capabilities of large language models across pointwise scoring, pairwise comparison, and listwise ranking protocols. It leverages cross‑protocol agreement to convert pointwise scores into higher‑order supervision, dynamically selects informative inputs, and applies score smoothing and multi‑protocol review to preserve ordinal structure and avoid catastrophic forgetting. Experiments show it outperforms existing baselines, achieving up to 73.3% accuracy on pointwise, 87.0% on pairwise, and 71.6% on listwise evaluation, while remaining robust against common extraction defenses.
arXiv:2606. 02109v1 Announce Type: new Abstract: Enterprise AI systems that translate natural language into SQL queries and orchestrate multi-step agentic reasoning pipelines require evaluation approaches fundamentally different from academic benchmarks.
The paper describes a new approach for two‑sided service marketplaces that replaces fixed request forms with AI‑native probabilistic matching using large language models. It introduces an autoresearch loop that generates a provider‑side preference taxonomy for each occupation, iteratively refining candidate tag sets through a six‑rubric LLM judge and a seven‑critic panel. The system also maps legacy form questions back to the new taxonomy, enabling coverage assessment and human quality assurance.
arXiv:2608. 11241v1 Announce Type: new Abstract: Deploying LLM agents into industrial recommender operations exposes a three-way tension we frame as the autonomy-determinism-efficiency trilemma: general autonomy (interpreting operator intent, generating glue code zero-shot), industrial determinism (schema-conforming feature extraction, non-crashing A/B, zero compliance-path hallucination), and end-to-end efficiency.
arXiv:2607. 16387v1 Announce Type: cross Abstract: An agent system's execution traces record how it fails, and procedures that improve such a system without changing model weights (trajectory selection, prompt and workflow optimization, runtime monitoring) read these traces for feedback.