Accelerating the inference of string generation-based chemical reaction models for industrial applications
Not on the map. Why?
Abstract (arXiv)
Template-free SMILES-to-SMILES translation models for reaction prediction and single-step retrosynthesis are of interest for industrial applications in computer-aided synthesis planning systems due to their state-of-the-art accuracy. However, they suffer from slow inference speed. We present a method to accelerate inference in autoregressive SMILES generators through speculative decoding by copying query string subsequences into target strings in the right places. We apply our method to the molecular transformer implemented in Pytorch Lightning and achieve over 3X faster inference in reaction prediction and single-step retrosynthesis, with no loss in accuracy.
Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 16 of its 30 references that are in the map. First posted on arXiv 12 Jul 2024.
Builds on
16 of 30 references.
References · 16
- 2017Attention Is All You Needcited by 59,932
- 2020Language Models are Few-Shot Learnerscited by 32,199
- 2015Distilling the Knowledge in a Neural Networkcited by 10,076
- 2023A Survey of Large Language Modelscited by 2,537
- 2022Fast Inference from Transformers via Speculative Decodingcited by 1,615
- 2024Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Headscited by 821
- 2020Efficient Transformers: A Surveycited by 784
- 2018Molecular Transformer - A Model for Uncertainty-Calibrated Chemical Reaction Predictioncited by 225
- 2017Learning to Plan Chemical Synthesescited by 175
- 2022Speculative Decoding: Exploiting Speculative Execution for Accelerating Seq2seq Generationcited by 169
- 2020Molecule Edit Graph Attention Network: Modeling Chemical Reactions as Sequences of Graph Editscited by 87
- 2022Root-aligned SMILES: A Tight Representation for Chemical Reaction Predictioncited by 49
4 more
- 2018A Generative Model For Electron Pathscited by 30
- 2018OpenNMT: Neural Machine Translation Toolkitcited by 27
- 2023Models Matter: The Impact of Single-Step Retrosynthesis on Synthesis Planningcited by 15
- 2023A Unified View of Deep Learning for Reaction and Retrosynthesis Prediction: Current Status and Future Challengescited by 4
Built on by
Citing papers per year: 2024: 1; 2025: 2; 2026: 1.
4 papers.
- 2025CopySpec: Accelerating LLMs with Speculative Copy-and-Paste Without Compromising Qualitycited by 6
- 2024Dynamic-Width Speculative Beam Decoding for Efficient LLM Inferencecited by 5
- 2025Fast and scalable retrosynthetic planning with a transformer neural network and speculative beam searchcited by 2
- 2026MolPaQ: Modular Quantum-Classical Patch Learning for Interpretable Molecular Generationcited by 0
- 2026MolPaQ: Modular Quantum-Classical Patch Learning for Interpretable Molecular Generation27 Mar 2026
- 2025Fast and scalable retrosynthetic planning with a transformer neural network and speculative beam search2 Aug 2025
- 2025CopySpec: Accelerating LLMs with Speculative Copy-and-Paste Without Compromising Quality13 Feb 2025
- 2024Dynamic-Width Speculative Beam Decoding for Efficient LLM Inference25 Sep 2024
Similar, not cited
Close in wording, no citation either way
- 2025RETRO SYNFLOW: Discrete Flow Matching for Accurate and Diverse Single-Step Retrosynthesis
- 2026Order Matters in Retrosynthesis: Structure-aware Generation via Reaction-Center-Guided Discrete Flow Matching
- 2025DiffER: Categorical Diffusion for Chemical Retrosynthesis
- 2022Retroformer: Pushing the Limits of Interpretable End-to-end Retrosynthesis Transformer
- 2026RetroAgent: Harnessing LLMs to Search Over Structured Memory for Agentic Retrosynthesis Planning
- 2026When Single Answer Is Not Enough: Rethinking Single-Step Retrosynthesis Benchmarks for LLMs
- 2025AOT*: Efficient Synthesis Planning via LLM-Empowered AND-OR Tree Search
- 2021Non-Autoregressive Electron Redistribution Modeling for Reaction Prediction