mlsanityData to 14 Oct 2026

BiMol-Diff: A Unified Diffusion Framework for Molecular Generation and Captioning

arXiv 2604.24089 ↗First posted 27 Apr 2026cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Bridging molecular structures and natural language is essential for controllable design. Autoregressive models struggle with long-range dependencies, while standard diffusion processes apply uniform corruption across positions, which can distort structurally informative tokens. We present BiMol-Diff, a unified diffusion framework for the paired tasks of text-conditioned molecule generation and molecule captioning. Our key component is a token-aware noise schedule that assigns position-dependent corruption based on token recovery difficulty, preserving harder-to-recover substructures during the forward process. On ChEBI-20 and M3-20M, BiMol-Diff improves molecule reconstruction with a 15.4% relative gain in Exact Match and achieves strong captioning results, attaining best BLEU and BERTScore among compared baselines. These results indicate token-aware noising improves fidelity in molecular structure-language modelling.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 24 of its 49 references that are in the map. First posted on arXiv 27 Apr 2026.

Builds on

24 of 49 references.

References · 24

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  4. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  5. 2019BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehensioncited by 5,893
  6. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  7. 2022Diffusion-LM Improves Controllable Text Generationcited by 1,027
  8. 2022DiGress: Discrete Denoising diffusion for graph generationcited by 520
  9. 2022DiffuSeq: Sequence to Sequence Text Generation with Diffusion Modelscited by 450
  10. 2022Translation between Molecules and Natural Languagecited by 177
  11. 2023MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adaptercited by 128
  12. 2023Unifying Molecular and Textual Representations via Multi-task Language Modellingcited by 93
12 more
  1. 2023GIT-Mol: A Multi-modal Large Language Model for Molecular Science with Graph, Image, and Textcited by 82
  2. 2023MolXPT: Wrapping Molecules with Text for Generative Pre-trainingcited by 81
  3. 2023Structure-informed Language Models Are Protein Designerscited by 68
  4. 2024Graph Diffusion Transformers for Multi-Conditional Molecular Generationcited by 45
  5. 2024Text-Guided Molecule Generation with Diffusion Language Modelcited by 31
  6. 2024UniMoT: Unified Molecule-Text Language Model with Discrete Token Representationcited by 23
  7. 2024LDMol: A Text-to-Molecule Diffusion Model with Structurally Informative Latent Space Surpasses AR Modelscited by 11
  8. 20243M-Diffusion: Latent Multi-Modal Diffusion for Language-Guided Molecular Structure Generationcited by 8
  9. 2023Interactive Molecular Discovery with Natural Languagecited by 5
  10. 2024Instruction-Based Molecular Graph Generation with Unified Text-Graph Diffusion Modelcited by 3
  11. 2025GraphT5: Unified Molecular Graph-Language Modeling via Multi-Modal Cross-Token Attentioncited by 2
  12. 2025Training Text-to-Molecule Models with Context-Aware Tokenizationcited by 1

Built on by

Not cited yet.