Martin Brümmer, Milan Dojchinovski, Sebastian Hellmann. Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC'16). 2016.
Documenting Large Webtext Corpora: A Case Study on the Colossal Clean Crawled Corpus
Large language models have led to remarkable progress on many NLP tasks, and researchers are turning to ever-larger text corpora to train them. Some of the largest corpora available are made by scraping significant portions of the internet, and are frequently introduced with only minimal documentation. In this work we provide some of the first documentation for the Colossal Clean Crawled Corpus (C4; Raffel et al., 2020), a dataset created by applying a set of filters to a single snapshot of Common Crawl. We begin by investigating where the data came from, and find a significant amount of text from unexpected sources like patents and US military websites. Then we explore the content of the text itself, and find machine-generated text (e.g., from machine translation systems) and evaluation examples from other benchmark NLP datasets. To understand the impact of the filters applied to create this dataset, we evaluate the text that was removed, and show that blocklist filtering disproportionately removes text from and about minority individuals. Finally, we conclude with some recommendations for how to created and document web-scale datasets from a scrape of the internet.
Wiki-40B: Multilingual Language Model Dataset
We propose a new multilingual language model benchmark that is composed of 40+ languages spanning several scripts and linguistic families. With around 40 billion characters, we hope this new resource will accelerate the research of multilingual modeling. We train monolingual causal language models using a state-of-the-art model (Transformer-XL) establishing baselines for many languages. We also introduce the task of multilingual causal language modeling where we train our model on the combined text of 40+ languages from Wikipedia with different vocabulary sizes and evaluate on the languages individually. We released the cleaned-up text of 40+ Wikipedia language editions, the corresponding trained monolingual language models, and several multilingual language models with different fixed vocabulary sizes.
Wikipedia, the free encyclopedia
Wikipedia is a free online encyclopedia, created and edited by volunteers around the world and hosted by the Wikimedia Foundation.
OKA – Disseminating free content on Wikipedia and open platforms through targeted funding
We are a non-profit organization dedicated to improving Wikipedia and other open platforms. We do so by providing monthly stipends to full-time contributors and translators. We leverage AI (Large Language Models) to automate most of the work.
Cross-Lingual Exploration for Parametric Knowledge | Idan Szpektor
Accepted to EMNLP Findings! While Large Language Models encode vast amounts of multicultural and factual information, this parametric knowledge is often unevenly accessible across languages. Standard inference frequently fails to surface localized facts, creating persistent gaps in cross-lingual knowledge transfer and consistency. We address this challenge in our paper "Cross-Lingual Exploration for Parametric Knowledge" (https://lnkd.in/drf36bV7), a collaboration between Elisha Diskind, Itamar Trainin, and Omri Abend from The Hebrew University of Jerusalem, Leshem Choshen from the Weizmann Institute of Science, alongside Uri Shaham and myself from our Google Research IL group. We formalize cross-lingual exploration as a structured search process across four core dimensions: language selection, exploration routing, answer aggregation, and inference budget. Evaluating across 17 typologically diverse languages on the ECLeKTic and CLIKE benchmarks, we show that allowing models to autonomously navigate alternative linguistic paths yields up to a 21% gain in knowledge transfer and a 16% boost in factual recall over native baselines, surpassing both standard English-pivot routing and native-language reasoning. Crucially, cross-lingual exploration defines a significantly more efficient compute Pareto frontier than scaling within the native query language, while driving intrinsic cross-lingual consistency gains beyond what accuracy improvements alone explain. Technically, this moves the needle for multilingual inference and knowledge elicitation pipelines, demonstrating that strategic language switching is a powerful, training-free mechanism for unlocking latent parametric knowledge.
Wikidata
the free knowledge base with 120,900,955 data entities that anyone can edit.
Evaluating Multilingual Metadata Quality in Crossref
Introduction: Scholarly research spans multiple languages, making multilingual metadata crucial for organizing and accessing knowledge across linguistic boundaries. These multilingual metadata already exist and are propagated throughout scholarly publishing infrastructure, but the extent to which they are correctly recorded, or how they affect metadata quality more broadly is little understood. Methods: Our study quantifies the prevalence of multilingual records across a sample of publisher metadata and offers an understanding of their completeness, quality, and alignment with metadata standards. Utilizing the Crossref API to generate a random sample of 519,665 journal article records, we categorize each record into four distinct language types: English monolingual, non-English monolingual, multilingual, and uncategorized. We then investigate the prevalence of programmatically-detectable errors and the prevalence of multilingual records within the sample to determine whether multilingualism influences the quality of article metadata. Results: We find that English-only records are still in the vast majority among metadata found in Crossref, but that, while non-English and multilingual records present unique challenges, they are not a source of significant metadata quality issues and, in few instances, are more complete or correct than English monolingual records. Discussion & Conclusion: Our findings contribute to discussions surrounding multilingualism in scholarly communication, serving as a resource for researchers, publishers, and information professionals seeking to enhance the global dissemination of knowledge and foster inclusivity in the academic landscape.
Named Entity Recognition with NLTK and SpaCy
NER is used in many fields in Natural Language Processing (NLP)
Jimmy Wales Says Wikipedia Could Use AI. Editors Call It the 'Antithesis of Wikipedia'
Wikipedia's founder said he used ChatGPT in the review process for an article and thought it could be helpful. Editors replied to point out it was full of mistakes.
OWNER — Toward Unsupervised Open-World Named Entity Recognition
Named Entity Recognition (NER) is a crucial task in Natural Language Processing (NLP), traditionally addressed through supervised learning, which requires extensive annotated corpora. This requirement poses challenges, particularly in specialized domains with limited labeled data. In response, the field has shifted towards lower-resource approaches, such as few-shot and zero-shot learning, which reduce the dependency on annotated data. However, even zero-shot models require prior knowledge of entity types, limiting their applicability in exploratory scenarios. In this context, we introduce OWNER, our unsupervised and open-world NER model, designed to operate without annotated documents or predefined entity types. OWNER leverages Encoder-only Language Models like BERT to infer and organize entities into dynamic entity types through a two-step process: mention detection and entity typing. Mention detection employs a BIO sequence labeling approach to locate entities, while entity typing uses BERT-based embeddings, refined through contrastive learning, for clustering and naming entity types. This method allows OWNER to automatically identify and structure unknown entity types, offering advantages for exploratory dataset analysis and knowledge graph construction. Our experimental evaluation on 13 domain-specific datasets demonstrates that OWNER surpasses existing LLM-based open-world NER models and remains competitive with more supervised and closed-world zero-shot models. OWNER’s architecture provides a lightweight, easily deployable solution that advances the state of the art in unsupervised and open-world NER. The source code of OWNER is publicly available at https://github.com/alteca/OWNER, facilitating future research in this domain.
KARMA: Leveraging Multi-Agent LLMs for Automated Knowledge Graph Enrichment
Maintaining comprehensive and up-to-date knowledge graphs (KGs) is critical for modern AI systems, but manual curation struggles to scale with the rapid growth of scientific literature. This paper presents KARMA, a novel framework employing multi-agent large language models (LLMs) to automate KG enrichment through structured analysis of unstructured text. Our approach employs nine collaborative agents, spanning entity discovery, relation extraction, schema alignment, and conflict resolution that iteratively parse documents, verify extracted knowledge, and integrate it into existing graph structures while adhering to domain-specific schema. Experiments on 1,200 PubMed articles from three different domains demonstrate the effectiveness of KARMA in knowledge graph enrichment, with the identification of up to 38,230 new entities while achieving 83.1% LLM-verified correctness and reducing conflict edges by 18.6% through multi-layer assessments.
TheoremGraph: Bridging Formal and Informal Mathematics
Mathematical knowledge is organized around statements and their dependencies, but this structure is exposed unevenly: informal papers cite mostly at the document level, while formal libraries record fine-grained dependencies over a much smaller body of mathematics. We introduce TheoremGraph, a unified statement-level dependency graph spanning both informal and formal mathematics. On the informal side, we parse 11.7M theorem-like environments from mathematics arXiv and recover 18.3M candidate directed dependencies, each labeled by the extractor that proposed it so downstream users can trade coverage for precision. On the formal side, we release LeanGraph, a Lean 4 elaborator-level extractor producing 388,105 declaration nodes and 11.3M typed edges across 25 Lean projects. We bridge the two graphs by embedding generated natural-language slogans into a shared semantic space, linking related statements across papers and across the informal/formal divide; an LLM judge affirms 47,952 such matches above a 0.8 cosine floor, with the judge-acceptance rate rising from 48% across the floor to 87% in the >=0.9 tier. On formal concept retrieval, our name-and-signature representation with graph expansion comes within 0.5pp of LeanSearch v2's reranked Recall@10 (0.775 vs. 0.780) without an LM reranker. We release the dataset, extractors, HTTP API, and MCP interface as infrastructure for mathematical search, attribution, and retrieval-augmented reasoning, available at theoremsearch.com and huggingface.co/datasets/uw-math-ai/theorem-matching.
Self-Driving Datasets: From 20 Million Papers to Nuanced Biomedical Knowledge at Scale
Manually curated biomedical repositories -- spanning bioactivity, genomics, and chemistry -- are expensive to maintain, lag behind primary literature, and discard experimental context, obscuring nuances needed to assess data correctness and coverage. We show that PubMed itself can be autonomously and cost-effectively turned into structured datasets that are larger, more nuanced, and more accurate than the curated databases they replace. We present three coupled contributions: (1) an LLM-based entity-tagging pipeline, grounded in nine biomedical ontologies, that tags 4.5B entities across 19 categories in a 22.5M-paper, 2.5T-token PubMed corpus; (2) hybrid sparse-dense retrieval supporting entity-filtered semantic queries over the tagged corpus; and (3) Starling, a multi-agent deep research system that, given only a natural-language task description, designs precision- and recall-targeted retrieval filters, induces an extraction schema, and emits structured records with nuance-rich fields and supporting passages. Across six tasks -- blood-brain barrier permeability, oral bioavailability, acute toxicity (LD50), gene-disease associations, protein subcellular localization, and chemical reactions -- Starling produces ~6.3M records (91K-3M per task); several are, to our knowledge, the largest public datasets for their property. Frontier-model rejection of our extractions is 0.6-7.7% across tasks, far below error rates we measure on widely used curated counterparts (e.g., 16.5% on BBB_Martins, 7.3% on Bioavailability_Ma). Beyond scale and accuracy, the supporting passages carry nuance tabular databases discard -- e.g., oral bioavailability may depend on fed vs. fasted state. Together, the corpus, retrieval, and agent establish a foundation for AI-driven therapeutic design. Code and datasets: https://github.com/starling-labs/starling.
lexicon.community
lexicon.community has 7 repositories available. Follow their code on GitHub.
