OpenSanctions Pairs is the first large‑scale public benchmark for entity matching on sanctions and OSINT data, comprising 755,540 expert‑labeled pairs drawn from over 1 million entities across 293 source datasets and 45 jurisdictions. The dataset spans multiple languages and writing systems, inconsistent structures, and time‑varying provenance, making it far more heterogeneous than prior benchmarks. Baseline experiments show a rule‑based matcher achieving 91.3 % F1, GPT‑4o reaching 99.0 % F1, and a locally deployable open‑source model scoring 98.2 % F1, with complementary failure modes that highlight the need to focus on downstream pipeline components.
By Chandler Smith, Magnus Sesodia, Friedrich Lindenberg, Christian Schroeder de Witt
arXiv:2606. 04382v1 Announce Type: cross Abstract: Automated subject cataloging assigns controlledvocabulary headings to bibliographic records, but LCSH has no standard public benchmark.
By Kwok Leong Tang
arXiv:2609.25298v1 Announce Type: new
Abstract: Cultural evaluation coverage and robustness in language models are difficult to diagnose because pretraining corpora and cultural benchmarks are rarely...
By Yusser Al Ghussin, Eva Gavaller, Cristina Espa\~na-Bonet, Josef van Genabith, Simon Ostermann
arXiv:2610.27043v1 Announce Type: cross
Abstract: Financial institutions increasingly deploy large language models (LLMs) in multilingual markets, raising a practical question: can pipelines develope...
By Joong Ho (Joshua), Choi, Sherine George, Jiayang Zhao, Jayesh Rathi, Ashwin Kandath, Ananya Sane, Vishvak Srinivasan
arXiv:2607. 24688v1 Announce Type: cross Abstract: Entity matching identifies records that refer to the same real-world entity.
By Zeyu Zhang, Xue Li, Iacer Calixto, Paul Groth, Sebastian Schelter
The paper evaluates large language models (LLMs) as data quality annotators on two e-commerce tasks: entity matching and brand mislabeling. In entity matching, a simple rule-based baseline matched the LLM’s zero-shot performance (F1≈0.95), and a few-shot prompt actually lowered performance, highlighting the risk of small-sample prompt tuning. For brand mislabeling, the LLM outperformed a naive rule baseline (F1 0.833 vs 0.721) by leveraging background knowledge, and demonstrated high consistency across repeated runs (99.7% agreement).
By Praphulla Lal Shrestha
The paper reports that in English all‑words word sense disambiguation (WSD), the scarcity of high‑quality labels—not the models—has become the limiting factor. The authors introduce lexEN, a human‑adjudicated correction layer over the Maru2022 ALL_NEW benchmark, and SenseBench, a living leaderboard for LLM WSD evaluation. They show that frontier large language models reach about 95 % accuracy on lexEN‑v1, that relabeling corpora with these models improves downstream systems, and that fine‑grained WordNet senses are often ill‑posed, with coarsening improving both annotator agreement and model performance.
"whyItMatters":"The study highlights that improving label quality and managing annotation costs are now the critical challenges for advancing WSD performance, as model accuracy is already near its theoretical ceiling."
By Vassili Philippov, Amro Salman, Dmitrii Andreev, Penny Hands, Emil Kaiumov, Pavel Katunin, Anton Nikolaev
arXiv:2607. 07469v1 Announce Type: cross Abstract: Fine-tuning large language models (LLMs) for e-commerce attribute extraction requires labeled data representative across thousands of product types, attributes, and multiple languages.
By Andrea Scarinci, Virginia Negri, Brayan Impata, Suleiman Khan, Victor Martinez, Marcello Federico
arXiv:2610.18415v1 Announce Type: cross
Abstract: Large Language Models (LLMs) are aligned to human preferences, after pre-training and supervised fine-tuning steps. However, standard RL alignment st...
By Pratuat Amatya, Vinay Setty, V Venktesh, Avishek Anand
arXiv:2609.13847v1 Announce Type: new
Abstract: In this paper, we extend FLORES+ with Portuguese-source evaluation sets for three Mozambican Bantu varieties: Xichangana, Mozambican Nyanja, and Sena....
By Felermino D. M. A. Ali, Delfina L\'azaro Mateus, Manuel Valente Mangue
The paper introduces Multilingual GSM‑Symbolic, a large multilingual math dataset with 30,000 question‑answer pairs across 15 languages, designed to evaluate cross‑lingual capability transfer using symbolic templates that generate millions of variations. It quantifies key determinants of transfer—model size, language resource level, reasoning ability, and typological distance—showing that larger models and better reasoning reduce performance gaps between low‑ and high‑resource languages, while typological distance has limited impact. The authors present a predictive framework that explains 92 % of language‑to‑language variation and can estimate unseen language performance within 6 pp, improving to 4.2 pp with only ten target‑language templates.
By Kenneth Enevoldsen, Riley Herchert, Sofie Mosegaard, Dan Saattrup Smart, Simon Enni, Isaac Chung, Sofie Bruun, Ayush Sunil Munot, Max M\"uller-Eberstein, Adnan El-Assadi, Elisa Bassignana, Gianluca Barmina, Hafsteinn Einarsson, Iben Nyholm Debess, Linda Freienthal, Lukas Galke Poech, Mike Zhang, Nicolas Legrand, Vladimir Salnikov, Yevhen Kostiuk, Zafar Hussain, Sagandeep Kaur, Agnes Toftg{\aa}rd, Marie Mattson, Kristoffer Nielbo
Compass‑v3 is a 245B‑parameter Mixture‑of‑Experts language model tailored for Southeast Asian e‑commerce, featuring 71B active parameters per token and hardware‑efficient expert parallelism. It is trained on 12 trillion multilingual tokens and synthetic e‑commerce instructions, and incorporates Optimal‑Transport Direct Preference Optimization to improve instruction adherence. Benchmarks show it outperforms GPT‑4, DeepSeek‑V3.1, and Qwen3‑235B, and it is already deployed at scale on Shopee, handling over 70% of the platform’s LLM traffic.
By Sophia Maria