mlsanityData to 7 Oct 2026

Latent JEPA: Abstract Future Prediction for Latent Reasoning in Chemistry

arXiv 2610.01947 ↗First posted 1 Oct 2026cs.LG cs.CLcited by 0

On the map: LLM reasoning → Clinical and biomedical applications also fits: Test-time compute and efficient thinking

Abstract (arXiv)

Large language models offer a promising foundation for chemical reasoning, bringing together chemical knowledge and multistep problem solving. Chemical intuition can provide an initial sense of plausible outcomes before the details of a solution are fully worked out. Inspired by how such expectations complement explicit analysis, we study how continuous latent thoughts can be trained to anticipate informative aspects of future solutions without verbalizing every intermediate step. We introduce Latent JEPA, a framework that combines autoregressive learning with joint-embedding prediction of one or more future views. For chemical reasoning, we develop textual and molecular prediction objectives that connect latent thoughts to both subsequent reasoning and molecular outcomes. Experiments on ChemCoTBench show gains in molecular optimization and on several editing and reaction metrics. Representation analyses show that future prediction makes latent thoughts more informative about molecular outcomes and strengthens their correspondence with chemical structure. These findings support abstract future prediction as a learning principle for connecting continuous latent reasoning with scientific outcomes.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning. Builds on 28 of its 37 references that are in the map. First posted on arXiv 1 Oct 2026.

Builds on

28 of 37 references.

Closest ancestors · 3

Same direction, newest first

  1. 2026LatentChem: From Textual CoT to Latent Thinking in Chemical Reasoning6 Feb
  2. 2025Training a Scientific Reasoning Model for Chemistry4 Jun
  3. 2025Beyond Chemical QA: Evaluating LLM's Chemical Reasoning with Modular Chemical Operations27 May

Other references · 25

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2025Qwen3 Technical Reportcited by 8,195
  3. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  4. 2021VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learningcited by 1,013
  5. 2023Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecturecited by 982
  6. 2024Training Large Language Models to Reason in a Continuous Latent Spacecited by 763
  7. 2022data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Languagecited by 718
  8. 2024Revisiting Feature Prediction for Learning Visual Representations from Videocited by 490
  9. 2024LLM2Vec: Large Language Models Are Secretly Powerful Text Encoderscited by 396
  10. 2020ChemBERTa: Large-Scale Self-Supervised Pretraining for Molecular Property Predictioncited by 332
  11. 2025Reasoning with Latent Thoughts: On the Power of Looped Transformerscited by 223
  12. 2025CODI: Compressing Chain-of-Thought into Continuous Space via Self-Distillationcited by 216
13 more
  1. 2021Large-Scale Chemical Language Representations Capture Molecular Structure and Propertiescited by 203
  2. 2022Multi-modal Molecule Structure-text Model for Text-based Retrieval and Editingcited by 165
  3. 2025SoftCoT: Soft Chain-of-Thought for Efficient Reasoning with LLMscited by 110
  4. 2025Intuitive physics understanding emerges from self-supervised pretraining on natural videoscited by 66
  5. 2025Paper2Poster: Towards Multimodal Poster Automation from Scientific Paperscited by 56
  6. 2023Introduction to Latent Variable Energy-Based Models: A Path Towards Autonomous Machine Intelligencecited by 52
  7. 2025LLM-JEPA: Large Language Models Meet Joint Embedding Predictive Architecturescited by 35
  8. 2025Chem-R: Learning to Reason as a Chemistcited by 16
  9. 2025Beyond Multi-Token Prediction: Pretraining LLMs with Future Summariescited by 14
  10. 2025AOT*: Efficient Synthesis Planning via LLM-Empowered AND-OR Tree Searchcited by 13
  11. 2024Semformer: Transformer Language Models with Semantic Planningcited by 9
  12. 2026MolEmb: Multimodal Large Language Models Can Be Strong Molecular Embedding Modelscited by 3
  13. 2026RetroAgent: Harnessing LLMs to Search Over Structured Memory for Agentic Retrosynthesis Planningcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025MiST: Understanding the Role of Mid-Stage Scientific Training in Developing Chemical Reasoning Models
  2. 2025ChemATP: A Training-Free Chemical Reasoning Framework for Large Language Models
  3. 2026Learning Mechanistic Reasoning for Chemical Reactions with Large Language Models
  4. 2025ChemDFM-R: A Chemical Reasoning LLM Enhanced with Atomized Chemical Knowledge
  5. 2023Monte Carlo Thought Search: Large Language Model Querying for Complex Scientific Reasoning in Catalyst Design
  6. 2025Reasoning-Enhanced Large Language Models for Molecular Property Prediction
  7. 2025Socrates-Mol: Self-Oriented Cognitive Reasoning through Autonomous Trial-and-Error with Empirical-Bayesian Screening for Molecules
  8. 2025A Scientific Reasoning Model for Organic Synthesis Procedure Generation