mlsanityData to 14 Oct 2026

High-Dimension Human Value Representation in Large Language Models

arXiv 2404.07900 ↗First posted 11 Apr 2024cs.CL cs.AIcited by 20

Not on the map. Why?

Abstract (arXiv)

The widespread application of LLMs across various tasks and fields has necessitated the alignment of these models with human values and preferences. Given various approaches of human value alignment, there is an urgent need to understand the scope and nature of human values injected into these LLMs before their deployment and adoption. We propose UniVaR, a high-dimensional neural representation of symbolic human value distributions in LLMs, orthogonal to model architecture and training data. This is a continuous and scalable representation, self-supervised from the value-relevant output of 8 LLMs and evaluated on 15 open-source and commercial LLMs. Through UniVaR, we visualize and explore how LLMs prioritize different values in 25 languages and cultures, shedding light on complex interplay between human values and language modeling.

Key facts. Cited by 20 papers in the map (data through 14 Oct 2026). Builds on 94 of its 160 references that are in the map. First posted on arXiv 11 Apr 2024.

Builds on

94 of 160 references.

References · 94

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2023GPT-4 Technical Reportcited by 18,396
  3. 2022Training language models to follow instructions with human feedbackcited by 15,971
  4. 2017Proximal Policy Optimization Algorithmscited by 15,215
  5. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,372
  6. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  7. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  8. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  9. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  10. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  11. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  12. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
38 more
  1. 2013Efficient Estimation of Word Representations in Vector Spacecited by 4,375
  2. 2021On the Opportunities and Risks of Foundation Modelscited by 4,033
  3. 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,935
  4. 2017Deep reinforcement learning from human preferencescited by 3,912
  5. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,780
  6. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  7. 2013Distributed Representations of Words and Phrases and their Compositionalitycited by 3,474
  8. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 3,050
  9. 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,898
  10. 2015FaceNet: A Unified Embedding for Face Recognition and Clusteringcited by 2,849
  11. 2020Learning to summarize from human feedbackcited by 2,588
  12. 2021SimCSE: Simple Contrastive Learning of Sentence Embeddingscited by 2,173
  13. 2018SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processingcited by 1,989
  14. 2021Barlow Twins: Self-Supervised Learning via Redundancy Reductioncited by 1,838
  15. 2021WebGPT: Browser-assisted question-answering with human feedbackcited by 1,745
  16. 2021Multitask Prompted Training Enables Zero-Shot Task Generalizationcited by 1,470
  17. 2023The Rise and Potential of Large Language Model Based Agents: A Surveycited by 1,439
  18. 2020Big Self-Supervised Models are Strong Semi-Supervised Learnerscited by 1,380
  19. 2022No Language Left Behind: Scaling Human-Centered Machine Translationcited by 1,312
  20. 2017Neural Collaborative Filteringcited by 1,111
  21. 2023LIMA: Less Is More for Alignmentcited by 1,100
  22. 2021GLM: General Language Model Pretraining with Autoregressive Blank Infillingcited by 1,031
  23. 2020MPNet: Masked and Permuted Pre-training for Language Understandingcited by 981
  24. 2023A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivitycited by 894
  25. 2022GLM-130B: An Open Bilingual Pre-trained Modelcited by 843
  26. 2015Deep Learning and the Information Bottleneck Principlecited by 802
  27. 2023The Flan Collection: Designing Data and Methods for Effective Instruction Tuningcited by 739
  28. 2022Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learnedcited by 729
  29. 2020Aligning AI With Shared Human Valuescited by 723
  30. 2023OpenAssistant Conversations -- Democratizing Large Language Model Alignmentcited by 721
  31. 2024Yi: Open Foundation Models by 01.AIcited by 719
  32. 2024Self-Rewarding Language Modelscited by 681
  33. 2020Language-agnostic BERT Sentence Embeddingcited by 627
  34. 2020Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillationcited by 593
  35. 2023Textbooks Are All You Needcited by 569
  36. 2016Fully-Convolutional Siamese Networks for Object Trackingcited by 568
  37. 2023Textbooks Are All You Need II: phi-1.5 technical reportcited by 565
  38. 2022Improving alignment of dialogue agents via targeted human judgementscited by 531

50 most cited of 94 All 94 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 5; 2025: 8; 2026: 7.

20 papers.

Sort:

By area:

  1. 2024Survey of Cultural Awareness in Language Models: Text and Beyondcited by 103
  2. 2025Utility Engineering: Analyzing and Controlling Emergent Value Systems in AIscited by 67
  3. 2025Embodied AI Agents: Modeling the Worldcited by 61
  4. 2024Language Model Alignment in Multilingual Trolley Problemscited by 46
  5. 2026Lost in Simulation: LLM-Simulated Users are Unreliable Proxies for Human Users in Agentic Evaluationscited by 40
  6. 2025Internal Value Alignment in Large Language Models through Controlled Value Vector Activationcited by 16
  7. 2024Measuring Human and AI Values Based on Generative Psychometrics with Large Language Modelscited by 13
  8. 2024I Want to Break Free! Persuasion and Anti-Social Behavior of LLMs in Multi-Agent Settings with Social Hierarchycited by 12
  9. 2025The Multilingual Divide and Its Impact on Global AI Safetycited by 10
  10. 2025Revisiting LLM Value Probing Strategies: Are They Robust and Expressive?cited by 8
  11. 2026Knowing But Not Doing: Convergent Morality and Divergent Action in LLMscited by 7
  12. 2024SEAL: Systematic Error Analysis for Value ALignmentcited by 4
8 more
  1. 2025Generative Psycho-Lexical Approach for Constructing Value Systems in Large Language Modelscited by 4
  2. 2025AdAEM: An Adaptively and Automated Extensible Measurement of LLMs' Value Differencecited by 4
  3. 2025Beyond External Monitors: Enhancing Transparency of Large Language Models for Easier Monitoringcited by 2
  4. 2026CALM: Culturally Self-Aware Language Modelscited by 2
  5. 2026Truthfulness Despite Weak Supervision: Evaluating and Training LLMs Using Peer Predictioncited by 2
  6. 2026Anthropogenic Regional Adaptation in Multimodal Vision-Language Modelcited by 2
  7. 2026PAVE: A Cognitive Architecture for Legitimate Violation in Generative Agent Societiescited by 0
  8. 2026FineWeb-CLaR: Culture, Language, and Region Annotations for Benchmark-Aligned Corpus Auditingcited by 0