mlsanityData to 14 Oct 2026

The Culture Funnel: You Can't Align What isn't in the Data

arXiv 2606.13808 ↗First posted 11 Jun 2026cs.CLcited by 2

Not on the map. Why?

Abstract (arXiv)

Current cultural alignment approaches focus on inference-time interventions, assuming models already contain sufficient cultural knowledge. We argue modern LLM pipelines suffer from a cultural data funnel. Using a multidimensional tagging framework across pretraining, fine-tuning, alignment, and reasoning datasets, we show explicit cultural signals decline sharply during post-training, while geographically concentrated, task-specialized data dominates. Multilinguality enhances geographic diversity of cultural knowledge but does not ensure balanced representation. Our tags improve downstream cultural benchmark performance, demonstrating that advances require shifting focus in training data pipelines. To facilitate future research, we release our culturally tagged dataset with 5.6M samples at https://huggingface.co/datasets/CohereLabs/CultureMarkers.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 53 of its 74 references that are in the map. First posted on arXiv 11 Jun 2026.

Builds on

53 of 74 references.

References · 53

  1. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  2. 2020mT5: A massively multilingual pre-trained text-to-text transformercited by 1,818
  3. 2020The State and Fate of Linguistic Diversity and Inclusion in the NLP Worldcited by 798
  4. 2021BBQ: A Hand-Built Bias Benchmark for Question Answeringcited by 701
  5. 2022Language Models are Multilingual Chain-of-Thought Reasonerscited by 530
  6. 2023UltraFeedback: Boosting Language Models with Scaled AI Feedbackcited by 406
  7. 2025OpenThoughts: Data Recipes for Reasoning Modelscited by 304
  8. 2021Quality at a Glance: An Audit of Web-Crawled Multilingual Datasetscited by 243
  9. 2022Challenges and Strategies in Cross-Cultural NLPcited by 218
  10. 2024Global MMLU: Understanding and Addressing Cultural and Linguistic Biases in Multilingual Evaluationcited by 191
  11. 2024Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuningcited by 177
  12. 2023CulturaX: A Cleaned, Enormous, and Multilingual Dataset for Large Language Models in 167 Languagescited by 174
38 more
  1. 2022Data Cards: Purposeful and Transparent Dataset Documentation for Responsible AIcited by 171
  2. 2024Towards Measuring and Modeling "Culture" in LLMs: A Surveycited by 167
  3. 2023Cultural Bias and Cultural Alignment of Large Language Modelscited by 149
  4. 2024The PRISM Alignment Dataset: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Modelscited by 144
  5. 2024Investigating Cultural Alignment of Large Language Modelscited by 143
  6. 2024Survey of Cultural Awareness in Language Models: Text and Beyondcited by 103
  7. 2024CultureBank: An Online Community-Driven Knowledge Base Towards Culturally Aware Language Technologiescited by 98
  8. 2022Language Contamination Helps Explain the Cross-lingual Capabilities of English Pretrained Modelscited by 92
  9. 2023Searching for Needles in a Haystack: On the Role of Incidental Bilingualism in PaLM's Translation Capabilitycited by 62
  10. 2024AI Suggestions Homogenize Writing Toward Western Styles and Diminish Cultural Nuancescited by 60
  11. 2023Breaking Language Barriers in Multilingual Mathematical Reasoning: Insights and Observationscited by 59
  12. 2022Some Languages are More Equal than Others: Probing Deeper into the Linguistic Disparity in the NLP Worldcited by 55
  13. 2023Benchmarking Machine Translation with Cultural Awarenesscited by 53
  14. 2024NormAd: A Framework for Measuring the Cultural Adaptability of Large Language Modelscited by 48
  15. 2025The State of Multilingual LLM Safety Research: From Measuring the Language Gap to Mitigating Itcited by 45
  16. 2025Multilingual != Multicultural: Evaluating Gaps Between Multilingual Capabilities and Cultural Alignment in LLMscited by 39
  17. 2024A Shocking Amount of the Web is Machine Translated: Insights from Multi-Way Parallelismcited by 35
  18. 2025Scaling Laws for Forgetting during Finetuning with Pretraining Data Injectioncited by 35
  19. 2026Reward Hacking in the Era of Large Models: Mechanisms, Emergent Misalignment, Challengescited by 33
  20. 2024Culturally Aware and Adapted NLP: A Taxonomy and a Survey of the State of the Artcited by 26
  21. 2025Learn Globally, Speak Locally: Bridging the Gaps in Multilingual Reasoningcited by 24
  22. 2021Dataset Geography: Mapping Language Data to Language Userscited by 22
  23. 2025On The Origin of Cultural Biases in Language Models: From Pre-training Data to Linguistic Phenomenacited by 22
  24. 2024Do Large Language Models Have an English Accent? Evaluating and Improving the Naturalness of Multilingual LLMscited by 19
  25. 2024CIVICS: Building a Dataset for Examining Culturally-Informed Values in Large Language Modelscited by 16
  26. 2025MultiNRC: A Challenging and Native Multilingual Reasoning Evaluation Benchmark for LLMscited by 16
  27. 2025Hire Your Anthropologist! Rethinking Culture Benchmarks Through an Anthropological Lenscited by 14
  28. 2025CultureScope: A Dimensional Lens for Probing Cultural Understanding in LLMscited by 12
  29. 2025The Multilingual Divide and Its Impact on Global AI Safetycited by 10
  30. 2025Fluent but Foreign: Even Regional LLMs Lack Cultural Alignmentcited by 10
  31. 2025Culture is Everywhere: A Call for Intentionally Cultural Evaluationcited by 10
  32. 2025Mind the Gap... or Not? How Translation Errors and Evaluation Details Skew Multilingual Resultscited by 10
  33. 2025Lost in Cultural Translation: Do LLMs Struggle with Math Across Cultural Contexts?cited by 9
  34. 2025From Surveys to Narratives: Rethinking Cultural Value Adaptation in LLMscited by 9
  35. 2024The ShareLM Collection and Plugin: Contributing Human-Model Chats for the Benefit of the Communitycited by 8
  36. 2025Rethinking Cross-lingual Alignment: Balancing Transfer and Cultural Erasure in Multilingual LLMscited by 7
  37. 2025The Art of Asking: Multilingual Prompt Optimization for Synthetic Datacited by 5
  38. 2026"Be My Cheese?": Cultural Nuance Benchmarking for Machine Translation in Multilingual LLMscited by 3

50 most cited of 53 All 53 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Benchmarking Frontier LLMs on Arabic Cultural and Sociolinguistic Knowledge: A Cross-Evaluation Framework with Human SME Ground Truthcited by 1
  2. 2026Building Multilingual Bridges: Data Mixing as the Pillar of Generalization for In-Language Reasoningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Aligned but Flattened: Analyzing the Trade-off between Cultural Alignment and Diversity in LLMs
  2. 2026Cross-Lingual Consensus: Aligning Multilingual Cultural Knowledge via Multilingual Self-Consistency
  3. 2026Steering LLMs for Culturally Localized Generation
  4. 2025CulFiT: A Fine-grained Cultural-aware LLM Training Paradigm via Multilingual Critique Data Synthesis
  5. 2023CDEval: A Benchmark for Measuring the Cultural Dimensions of Large Language Models
  6. 2024Self-Alignment: Improving Alignment of Cultural Values in LLMs via In-Context Learning
  7. 2026Progressing beyond Art Masterpieces or Touristic Clich'es: how to assess your LLMs for cultural alignment?
  8. 2024CRAFT: Extracting and Tuning Cultural Instructions from the Wild