arXiv:2607. 05031v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used to produce test oracles, the part of a test that decides whether observed behavior is correct.
arXiv:2607. 19442v1 Announce Type: cross Abstract: Machine unlearning is commonly evaluated by matching a retrained oracle on trained probes.
arXiv:2607. 20537v1 Announce Type: cross Abstract: We introduce ReliableTableQA, a framework for training an LLM to annotate the statistical reliability of tabular QA results, not whether the query is answerable, but whether the computed answer is statistically meaningful.
The article examines how classical test theory reliability statistics misrepresent the performance of large language model (LLM) judges. It shows that internal‑consistency coefficients, the dependability index, and Livingston‑Lewis accuracy each conflate judge error with item design or criterion validity, making it impossible to attribute a single reliability value to the judge alone. The authors argue that such misattribution can influence deployment decisions and documentation.
arXiv:2608. 07946v1 Announce Type: cross Abstract: Text-to-SQL benchmarks ship schemas whose column names already say what the columns mean.
arXiv:2608. 09254v1 Announce Type: new Abstract: LLM analytics agents are evaluated on SQL syntax accuracy, but production failures look different: questions with two valid business definitions, questions the warehouse cannot answer, deprecated columns after a schema change, and queries that execute successfully while returning the wrong business number.
arXiv:2608. 14617v1 Announce Type: cross Abstract: A recurring proposal in legal AI is to improve case-outcome prediction by fusing uncertainty tools (evidence graphs with belief propagation, sequential Bayesian odds updating, Dempster-Shafer combination, and conformal prediction) into one pipeline.
arXiv:2608. 14639v1 Announce Type: cross Abstract: Per-field accept/review with selective risk at most alpha -- accept a field only if the error rate among accepted fields is controlled -- is the trust contract document-extraction systems need, and the natural procedure silently violates it on real documents.
The paper introduces a distribution‑free certification layer that can be applied to any crash‑severity prediction model without modifying the model itself. It provides guarantees for ordinal outcomes, per‑class validity, transfer of coverage to unobserved severities, and one‑sided certificates under deployment shift, all grounded in a functional of the true data law. The framework is evaluated on 5.2 million Texas records, demonstrating a model‑independent lower bound on set width for vulnerable road users and is released as an open‑source package with theorem‑level tests.
ESQ‑Bench is an Oracle‑first NL2SQL benchmark that introduces systematic complexity tiers and silent‑divergence evaluation across three enterprise schema levels. It provides six populated schemas (465 tables, 164,682 rows) on Oracle, PostgreSQL, MySQL, and SQL Server, along with 550 gold‑validated question‑query pairs and a four‑metric evaluation harness. The benchmark reveals that state‑of‑the‑art models such as GPT‑4o and Claude Sonnet 4.6 experience significant drops in execution accuracy and high silent‑divergence rates as schema complexity increases, highlighting a gap between closed‑API models and open‑weight baselines on enterprise Oracle schemas.
The article examines how classical test theory statistics—Kuder‑Richardson coefficient, dependability index, and Livingston‑Lewis accuracy—can mislead when applied to large language model (LLM) judges that are evaluated with a single prompt and no gold labels. Using Claude Haiku 4.5 on 210 short‑answer items, the authors show that these metrics fail to isolate the judge’s performance because the judge’s single administration provides no variance component. They argue that reliable statements about an LLM judge require gold labels or varied scorer facets, and that bank design heavily influences reliability estimates.
The paper introduces Counterfactual Fragility Certificates (CFC), a model‑agnostic audit protocol that maps each prediction to an evidence‑failure trajectory, summarizing it with metrics such as greedy flip budget, margin‑collapse area, degradation thresholds, and fragility dominance score. CFC is shown to identify brittle high‑confidence predictions on seven tabular benchmarks with an AUROC of 0.915, outperforming existing scalar scores by up to +0.405. The method remains effective across various perturbation and review‑budget scenarios, and can also inform fragility‑aware regularization and temperature correction.