FineWeb-CLaR: Culture, Language, and Region Annotations for Benchmark-Aligned Corpus Auditing

arXiv:2609.25298 · cs.CL · Submitted 2026-09-21 · Read on arXiv

cs.CL

Submitted: 2026-09-21

Updated: 2026-09-21

Comments: accepted to EMNLP 2026 (Main)

Code: https://github.com/Yusser96/FineWeb-CLaRhttps:

Project page: https://behavior-in-the-wild.github.io/align-via-actions.html

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Benchmark — 112× in this paper · explained in 2 episodes
• Empirical Methods in Natural Language Processing — 14× in this paper
• Alignment — 13× in this paper · explained in 1 episode
• Embeddings — 10× in this paper · explained in 2 episodes
• Locale Topic Distributions — 10× in this paper
• Provenance — 9× in this paper · explained in 4 episodes
• Aggregation — 5× in this paper · explained in 1 episode
• Robustness — 5× in this paper · explained in 5 episodes
• Adaptation — 4× in this paper · explained in 1 episode
• alpha — 4× in this paper · explained in 1 episode
• Adam — 2× in this paper · explained in 2 episodes
• Benchmarking — 2× in this paper · explained in 3 episodes
• Conference on Language Modeling — 2× in this paper
• Cosine Similarity — 2× in this paper · explained in 3 episodes
• Transformer — 2× in this paper · explained in 2 episodes
• Accuracy — 1× in this paper · explained in 1 episode
• Adaptability — 1× in this paper · explained in 1 episode
• Annotator Agreement — 1× in this paper · explained in 1 episode
• Clustering — 1× in this paper · explained in 2 episodes
• Distillation — 1× in this paper · explained in 2 episodes
• Following the Cultural Taxonomy — 1× in this paper
• Instruction Tuning — 1× in this paper · explained in 2 episodes
• Knowledge Distillation — 1× in this paper · explained in 11 episodes
• Low-Resource Languages — 1× in this paper · explained in 4 episodes
• Multimodal Dataset — 1× in this paper · explained in 2 episodes
• Representation Learning — 1× in this paper · explained in 2 episodes
• Vision-language models — 1× in this paper · explained in 2 episodes

Sources

• The Multilingual Alignment Prism: Aligning Global and Local Preferences to Reduce Harm
• SEADialogues: A Multilingual Culturally Grounded Multi-turn Dialogue Dataset on Southeast Asian Languages
• SaudiCulture: A Benchmark for Evaluating Large Language Models Cultural Competence within Saudi Arabia
• CULEMO: Cultural Lenses on Emotion -- Benchmarking LLMs for Cross-Cultural Emotion Understanding
• Cultural Adaptation of Recipes
• M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation
• CulturalBench: A Robust, Diverse, and Challenging Cultural Benchmark by Human-AI CulturalTeaming
• TyDi QA: A Benchmark for Information-Seeking Question Answering in Typologically Diverse Languages
• Towards Measuring the Representation of Subjective Global Opinions in Language Models
• MASIVE: Open-Ended Affective State Identification in English and Spanish
• MASSIVE: A 1M-Example Multilingual Natural Language Understanding Dataset with 51 Typologically-Diverse Languages
• BERTopic: Neural topic modeling with a class-based TF-IDF procedure
• CARE: Multilingual Human Preference Learning for Cultural Awareness
• Hope Speech detection in under-resourced Kannada language
• KoBBQ: Korean Bias Benchmark for Question Answering
• CultureGuard: Towards Culturally-Aware Dataset and Guard Model for Multilingual Safety Applications
• BnMMLU: Measuring Massive Multitask Language Understanding in Bengali
• Are All Spanish Doctors Male? Evaluating Gender Bias in German Machine Translation
• LLM-GLOBE: A Benchmark Evaluating the Cultural Values Embedded in LLM Output
• The PRISM Alignment Dataset: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Models

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering