arXiv:2608. 17719v1 Announce Type: cross Abstract: Context: Software systems that depend on commercial large language model APIs must migrate to successor versions when vendors deprecate older models.
arXiv:2511. 04689v3 Announce Type: replace-cross Abstract: Evaluating large language models (LLMs) typically requires thousands of benchmark items, making the process expensive, slow, and increasingly impractical at scale.
arXiv:2608. 05086v1 Announce Type: new Abstract: Language models differ in how safely they behave and these differences are measured by safety benchmarks.
arXiv:2606. 05104v1 Announce Type: new Abstract: Knowledge benchmarks for LLMs face three issues: scaling-driven designs that do not operationalize disciplinary representativeness; flat-payment annotation that permits lazy consensus; and unaudited ranking instability under bounded test budgets.
arXiv:2606. 26429v1 Announce Type: new Abstract: Current LLM evaluation relies on two complementary but often disconnected signals: static benchmarks with objective correctness labels and arena-style preference data that better reflect open-ended user interactions.
arXiv:2510. 00492v3 Announce Type: replace Abstract: The reliability of large language models (LLMs) during test-time scaling is often assessed with \emph{external verifiers} or \emph{reward models} that distinguish correct reasoning from flawed logic.
arXiv:2607. 09739v1 Announce Type: new Abstract: We study LLM benchmark coreset selection: selecting a small subset of prompts over multiple benchmarks whose induced model scores and rankings approximate those obtained from the full benchmark suite.
arXiv:2608. 15046v1 Announce Type: new Abstract: A fraction of a point of benchmark accuracy is the usual evidence that a compressed model is equivalent to its original.
arXiv:2606. 11686v1 Announce Type: cross Abstract: End-to-end task-success is the dominant way to evaluate LLM agents, but one aggregate number tells you that an agent regressed, not where.
Newly developed items must ordinarily be field tested before their psychometric properties are known, creating a cold start problem for item calibration. Predicting item parameters from features is a long standing measurement problem dating back to the Linear Logistic Test Model; modern text embeddings now automate the design matrices traditionally specified by hand.
arXiv:2606. 03650v1 Announce Type: cross Abstract: Choosing or ranking language models for a specific application is hardest when no task-specific labeled data exists, and standard public benchmarks cannot be trusted, their items having likely leaked into pretraining, so scores reflect memorization rather than fitness.
arXiv:2607. 15190v1 Announce Type: new Abstract: AI benchmarks increasingly leverage item-level statistical models, particularly item response theory (IRT), to estimate model capabilities, rank systems, select informative examples, and diagnose benchmark quality.
arXiv:2606. 26836v1 Announce Type: new Abstract: Existing benchmarks typically report accuracy for a single model on a single run.