VeraRAN is introduced, which checks plans before actuation by modeling request, delivery, acceptance, application, completion, and observation for each action while exploring plausible delays and event orders, and MI-POR prunes independent interleavings using RAN lifecycle and resource footprints.
A two-layer evaluation framework that separates scorer-independent execution evidence, including termination, answer exposure, parseability, and completion length, from scorer-dependent correctness is introduced, demonstrating that accuracy conflates execution case mix with verification policy.
It is argued that LLM-as-judge reports should include dataset slices, bias probes, error-dependence estimates, error-dependence estimates, and protocol audit trails.
CompressAgent is introduced, an environment-verified benchmark for ACC compression across nine independently constructed ACCs, three task families, three fixed Qwen API model identifiers, six retained-context budgets, and 15,525 runs, uncovering a nonlinear, method-dependent reliability frontier.
Ying-Han Hou, Zong-You Yang· 3 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
