The paper investigates how sentence‑specificity scores can guide the selection of revisions in collaborative technical documentation. It compares two predictors—SpeciTeller and a target‑adapted model by Ko et al.—across Wikipedia and three technical corpora, finding that the predictors rank sentences differently and that SpeciTeller can improve direction‑valid selection rates in certain datasets. The study also shows that filtering and token‑length adjustments alter but do not reconcile these ranking differences.
arXiv:2606. 05970v1 Announce Type: cross Abstract: Large language models are increasingly used for structured extraction from clinical free-text notes, but the sensitivity of their output to upstream configuration choices is less understood than their accuracy on fixed benchmarks.
arXiv:2609.13611v1 Announce Type: new Abstract: The WMT26 General MT task evaluates systems on 10 language pairs that have no human references (neither translated from scratch nor post-edited from MT...
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...
Large Language Model judges are commonly used to rank texts via pairwise comparison, with reliability traditionally measured by position bias, transitivity, and pairwise agreement. This paper argues that these proxies are misleading because they are dominated by close‑rank‑gap pairs, which contribute little to the overall ranking, while far‑gap pairs carry the true ranking signal. Experiments on simulations and human‑rated corpora show weak correlation between the proxies and actual ranking accuracy, suggesting judges should be evaluated using rank‑gap‑conditional metrics against human rankings.
arXiv:2606. 02255v1 Announce Type: cross Abstract: Human annotation is the empirical foundation of much NLP research, from dataset construction to model evaluation, but papers often leave unclear who produced the annotations and how the annotation process was controlled.
arXiv:2608. 12138v1 Announce Type: cross Abstract: General-purpose large language models (LLMs) have recently been reported to match or exceed specialized clinical AI tools on medical benchmarks, but such comparisons draw on a narrow set of systems and on benchmarks developed largely in high-income settings.
The paper introduces three retrieval methods for Polish statutory law that use language‑model annotations attached to articles as surrogates. The methods—ASCR, ASCR‑H, and DTF—vary in cost and quality, with ASCR‑H achieving the highest rank‑one accuracy on bar exam questions, while DTF offers competitive performance with lower latency and cost. Extensive evaluation against 14 baselines on 300 exam questions demonstrates significant improvements in head‑rank accuracy and discusses limitations such as coverage asymmetry and negative results for lemmatisation, pseudo‑relevance feedback, and query rewriting.
arXiv:2608.21374v1 Announce Type: new Abstract: Literature reviews are essential to scientific progress, but rigorously evaluating automatically generated reviews remains difficult because many aspec...
The paper investigates how benchmark contamination—leakage of test items into training data—affects large language model (LLM) leaderboards. By comparing original test items with semantically equivalent paraphrases, the authors measure contamination as a violation of anchor-item invariance and find that it inflates absolute scores but rarely changes model rankings. Across 47 public models and 74 finetuned models on four benchmarks, the rank correlation between standard and paraphrase-controlled leaderboards is 0.997, with only a handful of cases showing differential contamination that could alter rankings.
arXiv:2609.37472v1 Announce Type: cross Abstract: Behavioral tests measure how a language model reads evidence. We ask whether those measurements help choose a recommendation interface. We evaluate s...
The paper introduces Intelligent Target Locator (ITL), a method that measures how well a document aligns with concepts in a Structured Reference Document (SRD) by creating concept‑specific term profiles and computing a textual‑unit–concept affinity matrix. ITL assigns importance weights to terms based on concept membership, term specificity, and discriminability, enabling traceable, quantitative alignment scores at multiple granularity levels. An internal consistency test on the 17 Sustainable Development Goals showed that each goal statement achieved its highest affinity with its corresponding concept, demonstrating ITL’s ability to distinguish conceptual profiles.
arXiv:2608.20385v1 Announce Type: new Abstract: Systematic reviews rely on quality appraisal of included studies, a process that is time-consuming and sensitive to ambiguity in checklist criteria. Al...