The paper introduces a calibrated instrument for rigorously measuring how inference optimizations—such as quantization, early‑exit, and speculative decoding—affect the output quality of large language models. It uses a formally calibrated LLM judge that verifies no systematic bias between statistically equivalent outputs and includes a null condition to ensure measured differences are zero. Applying this method, the authors find that a 4‑bit model is indistinguishable from its 16‑bit counterpart, while 3‑bit quantization and early‑exit techniques incur measurable quality losses that vary by language and task, and that token‑certainty‑based acceptance rules cannot reliably identify impactful errors.
By Jerry Kaplan
arXiv:2606. 23767v1 Announce Type: new Abstract: Headline accuracies on the Tuebingen cause-effect pairs are routinely compared across papers even though each is measured under its authors' own protocol -- different pair subsets, weightings, model-selection, and decision rates.
By Wietse Stienstra
arXiv:2609.00654v1 Announce Type: new
Abstract: We describe the SciTrue team's participation in both subtasks of the NTCIR-19 SciClaimEval task~\cite{sciclaimeval}, which asks systems to verify scien...
By Qiming Bao, Ne\c{s}et \"Ozkan Tan, Siyuan Wang, Mark Gahegan
The paper introduces Janus, a method for validating error patterns in language models by comparing error rates across predefined yes/no properties and using shuffled decoy labels to set significance thresholds. Janus requires that a pattern’s error difference surpasses the decoy-derived threshold and is replicated on held‑out data before reporting. Experiments on a controlled code‑finding task confirm several meaningful error patterns, while on MuSiQue and LongBench v2 Janus reports no confirmed patterns for the tested properties, contrasting with standard shuffling tests that sometimes confirm patterns.
By Vyzantinos Repantis, Ameya Gawde, Harshvardhan Singh
The study analyzes 373,019 judgments from LLM‑scored benchmarks, decomposing variance into system, item, judge, and interaction components via generalizability theory. It finds that with a single judge, generalizability converges to a ceiling determined by the system‑by‑judge variance, which is substantially lower in pairwise preference settings, allowing one judge to suffice. The research also reveals significant biases in presentation order and highlights that many published win‑rate claims fall below the measured floor of the benchmarks.
By Atul Anand
Large language model optimization is an active research area, spanning quantization of model weights, early-exit methods for skipping layers, and speculative decoding. Each track uses its own quality...
arXiv:2607. 08065v1 Announce Type: new Abstract: LLM-as-judge (Zheng et al.
By Kaihua Ding
arXiv:2604. 18245v3 Announce Type: replace Abstract: Large language models operate in protocols containing multiple calls, yet added calls are usually evaluated only by their net effect.
By Fernando Reitich
The paper investigates whether the object selected in a grounded language‑model pipeline actually reaches the reader, a failure that can break the handoff between stages. By auditing 600 HybridQA questions across three selector families, the authors find that exact key lookup and title matching recover the selected object in all 1,463 resolvable records, but body‑only BM25 omits it in 26.6% of cases at cutoff five, while hybrid retrieval with reranking omits it only 1.0%. The study also shows that misalignment between selected and retrieved objects can reduce exact match scores by up to 31 points, and introduces the Returned‑Object Profile (ROP) as a tool for reproducible auditing.
By Siddharth Vohra, Runmin Jiang, Xiaomo Li, Min Xu
arXiv:2607. 04429v1 Announce Type: cross Abstract: The dominant practice in language model evaluation is to report a single accuracy number per model and declare the higher one better, without testing whether the gap could plausibly be sampling noise.
By Shreyas K Chandrahas
The paper investigates item-sensitivity—whether a language model’s choice depends on the specific input—in a forced-choice signalling task derived from the board game Deception: Murder in Hong Kong. Across seven models, two families, a post‑training ablation, and three scoring rules, every tested cell shows item‑sensitivity, yet many are statistically indistinguishable from random choice and some perform worse than random. The authors term this phenomenon "consistency without alignment" and argue it undermines evaluations that rely solely on item‑sensitivity, permutation consistency, or self‑consistency without an independent reference.
By Cris Huynh
arXiv:2609.09245v1 Announce Type: new
Abstract: Repeated-sampling evaluations increasingly extrapolate pass@k far beyond the number n of samples collected per problem. We show that, in the pooled/ran...
By Pranav Singh, Prashant Singh