mlsanityData to 14 Oct 2026

Embedding-Aligned Language Models

arXiv 2406.00024 ↗First posted 24 May 2024cs.CL cs.AI cs.ETcited by 9

On the map: LLM alignment / RLHF / preference optimisation → LLM preference optimization methods

Abstract (arXiv)

We propose a novel approach for training large language models (LLMs) to adhere to objectives defined within a latent embedding space. Our method leverages reinforcement learning (RL), treating a pre-trained LLM as an environment. Our embedding-aligned guided language (EAGLE) agent is trained to iteratively steer the LLM's generation towards optimal regions of the latent embedding space, w.r.t. some predefined criterion. We demonstrate the effectiveness of the EAGLE agent using the MovieLens 25M and Amazon Review datasets to surface content gaps that satisfy latent user demand. We also demonstrate the benefit of using an optimal design of a state-dependent action set to improve EAGLE's efficiency. Our work paves the way for controlled and grounded text generation using LLMs, ensuring consistency with domain-specific knowledge and data representations.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 39 of its 55 references that are in the map. First posted on arXiv 24 May 2024.

Builds on

39 of 55 references.

Closest ancestors · 2

Same direction, newest first

  1. 2023Aligning Language Models with Human Preferences via a Bayesian Approach9 Oct
  2. 2022Training language models to follow instructions with human feedback4 Mar

Other references · 37

  1. 2017Attention Is All You Needcited by 60,361
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  3. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  4. 2023GPT-4 Technical Reportcited by 18,396
  5. 2017Proximal Policy Optimization Algorithmscited by 15,215
  6. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  7. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  8. 2022Video Diffusion Modelscited by 2,305
  9. 2019Fine-Tuning Language Models from Human Preferencescited by 2,150
  10. 2023ImageBind: One Embedding Space To Bind Them Allcited by 1,175
  11. 2023BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Datasetcited by 860
  12. 2016Automatic chemical design using a data-driven continuous representation of moleculescited by 765
25 more
  1. 2020NVAE: A Deep Hierarchical Variational Autoencodercited by 637
  2. 2023RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedbackcited by 625
  3. 2014Semi-Supervised Learning with Deep Generative Modelscited by 496
  4. 2023HaluEval: A Large-Scale Hallucination Evaluation Benchmark for Large Language Modelscited by 341
  5. 2023Chain-of-Verification Reduces Hallucination in Large Language Modelscited by 324
  6. 2018Variational Autoencoders for Collaborative Filteringcited by 310
  7. 2023Detecting and Preventing Hallucinations in Large Vision Language Modelscited by 256
  8. 2020Accelerating Reinforcement Learning with Learned Skill Priorscited by 223
  9. 2019MineRL: A Large-Scale Dataset of Minecraft Demonstrationscited by 215
  10. 2022Fine-tuning language models to find agreement among humans with diverse preferencescited by 189
  11. 2017Latent Space Oddity: on the Curvature of Deep Generative Modelscited by 174
  12. 2017Improved Variational Autoencoders for Text Modeling using Dilated Convolutionscited by 81
  13. 2020Action Space Shaping in Deep Reinforcement Learningcited by 46
  14. 2020Geometrically Enriched Latent Spacescited by 45
  15. 2022Contextual Bandits with Large Action Spaces: Made Practicalcited by 33
  16. 2019The Natural Language of Actionscited by 28
  17. 2020Safe Reinforcement Learning with Natural Language Constraintscited by 23
  18. 2023Embedding in Recommender Systems: A Surveycited by 20
  19. 2019Enabling hyperparameter optimization in sequential autoencoders for spiking neural datacited by 18
  20. 2023Demystifying Embedding Spaces using Large Language Modelscited by 17
  21. 2018Fast Approximate Geodesics for Deep Generative Modelscited by 13
  22. 2021Uncertainty Estimation Using Riemannian Model Dynamics for Offline Reinforcement Learningcited by 10
  23. 2023Representation-Driven Reinforcement Learningcited by 4
  24. 2023Factual and Personalized Recommendations using Language Models and Reinforcement Learningcited by 4
  25. 2016Enabling Dark Energy Science with Deep Generative Models of Galaxy Imagescited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 2; 2025: 4; 2026: 3.

9 papers.

  1. 2025Rethinking Diverse Human Preference Learning through Principal Component Analysiscited by 10
  2. 2024Preference Adaptive and Sequential Text-to-Image Generationcited by 8
  3. 2024Comparing Few to Rank Many: Active Human Preference Learning using Randomized Frank-Wolfecited by 5
  4. 2026Incentive-Aware Multi-Fidelity Optimization for Generative Advertising in Large Language Modelscited by 1
  5. 2025Descriptive History Representations: Learning Representations by Answering Questionscited by 0
  6. 2025FuDoBa: Fusing Document and Knowledge Graph-based Representations with Bayesian Optimisationcited by 0
  7. 2025RLRF: Competitive Search Agent Design via Reinforcement Learning from Ranker Feedbackcited by 0
  8. 2026Do Reasoning Models Enhance Embedding Models?cited by 0
  9. 2026OLLM: Options-based Large Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Reinforcement Learning for Aligning Large Language Models Agents with Interactive Environments: Quantifying and Mitigating Prompt Overfitting
  2. 2025Controlling Large Language Model with Latent Actions
  3. 2024Aligning Large Language Models via Fine-grained Supervision
  4. 2025REFINE-AF: A Task-Agnostic Framework to Align Language Models via Self-Generated Instructions using Reinforcement Learning from Automated Feedback
  5. 2025Embedding Domain Knowledge for Large Language Models via Reinforcement Learning from Augmented Generation
  6. 2022Is Reinforcement Learning (Not) for Natural Language Processing: Benchmarks, Baselines, and Building Blocks for Natural Language Policy Optimization
  7. 2024Input Conditioned Graph Generation for Language Agents
  8. 2025Reusing Embeddings: Reproducible Reward Model Research in Large Language Model Alignment without GPUs