# TinyChemVL: Advancing Chemical Vision-Language Models via Efficient Visual Token Reduction and Complex Reaction Tasks

arXiv: https://arxiv.org/abs/2511.06283 · HTML: https://mlsanity.com/paper/2511.06283/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in Chart understanding and anomaly detection, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 25 of its 39 references that are in the map.
- First posted on arXiv 9 Nov 2025.

## Placed in

- Multimodal LLMs / vision-language models / [Chart understanding and anomaly detection](https://mlsanity.com/mllm/chart-understanding-and-anomaly-detection/)
- Efficient LLM inference and serving / [Visual token pruning in multimodal models](https://mlsanity.com/inference/visual-token-pruning-in-multimodal-models/)

## Builds on (most cited first, 25 of 25)

- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling](https://mlsanity.com/paper/2412.05271/) (2024)
- [SciBERT: A Pretrained Language Model for Scientific Text](https://mlsanity.com/paper/1903.10676/) (2019)
- [Token Merging: Your ViT But Faster](https://mlsanity.com/paper/2210.09461/) (2022)
- [ChemCrow: Augmenting large-language models with chemistry tools](https://mlsanity.com/paper/2304.05376/) (2023)
- [Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizations](https://mlsanity.com/paper/2202.07800/) (2022)
- [Evo-ViT: Slow-Fast Token Evolution for Dynamic Vision Transformer](https://mlsanity.com/paper/2108.01390/) (2021)
- [ChartCoder: Advancing Multimodal Large Language Model for Chart-to-Code Generation](https://mlsanity.com/paper/2501.06598/) (2025)
- [Scientific Large Language Models: A Survey on Biological & Chemical Domains](https://mlsanity.com/paper/2401.14656/) (2024)
- [SELFIES and the future of molecular string representations](https://mlsanity.com/paper/2204.00056/) (2022)
- [ChemVLM: Exploring the Power of Multimodal Large Language Models in Chemistry Area](https://mlsanity.com/paper/2408.07246/) (2024)
- [PRISM: Self-Pruning Intrinsic Selection Method for Training-Free Multimodal Data Selection](https://mlsanity.com/paper/2502.12119/) (2025)
- [MolScribe: Robust Molecular Structure Recognition with Image-To-Graph Generation](https://mlsanity.com/paper/2205.14311/) (2022)
- [PPT: Token Pruning and Pooling for Efficient Vision Transformers](https://mlsanity.com/paper/2310.01812/) (2023)
- [MolGrapher: Graph-based Visual Recognition of Chemical Structures](https://mlsanity.com/paper/2308.12234/) (2023)
- [UniMoT: Unified Molecule-Text Language Model with Discrete Token Representation](https://mlsanity.com/paper/2408.00863/) (2024)
- [LLaMo: Large Language Model-based Molecular Graph Assistant](https://mlsanity.com/paper/2411.00871/) (2024)
- [VinciCoder: Unifying Multimodal Code Generation via Coarse-to-fine Visual Reinforcement Learning](https://mlsanity.com/paper/2511.00391/) (2025)
- [A quantitative analysis of knowledge-learning preferences in large language models in molecular science](https://mlsanity.com/paper/2402.04119/) (2024)
- [ChemDFM-X: Towards Large Multimodal Model for Chemistry](https://mlsanity.com/paper/2409.13194/) (2024)
- [Token Transforming: A Unified and Training-Free Token Compression Framework for Vision Transformer Acceleration](https://mlsanity.com/paper/2506.05709/) (2025)
- [RFL: Simplifying Chemical Structure Recognition with Ring-Free Language](https://mlsanity.com/paper/2412.07594/) (2024)

## Built on by (most cited first, 6 of 6)

- [Beyond NL2Code: A Structured Survey of Multimodal Code Intelligence](https://mlsanity.com/paper/2606.15932/) (2026)
- [ChemVLR: Prioritizing Reasoning in Perception for Chemical Vision-Language Understanding](https://mlsanity.com/paper/2604.06685/) (2026)
- [Real Data Closes Synthetic-to-Real Gap in Optical Chemical Structure Recognition](https://mlsanity.com/paper/2608.09100/) (2026)
- [S3C-LLM: Skill-Code Guided Agentic Language Models for Spectrum-to-Structure Elucidation](https://mlsanity.com/paper/2608.30910/) (2026)
- [When Vision Meets Graphs: A Survey on Graph Reasoning and Learning](https://mlsanity.com/paper/2609.03816/) (2026)
- [Learning Transferable Reaction Mechanisms from Visual Chemical Knowledge](https://mlsanity.com/paper/2609.33608/) (2026)

## Abstract (arXiv)

While Vision Language Models (VLMs) have demonstrated remarkable capabilities in general visual understanding, their application in the chemical domain has been limited, with previous works predominantly focusing on text and thus overlooking critical visual information, such as molecular structures. Current approaches that directly adopt standard VLMs for chemical tasks suffer from two primary issues: (i) computational inefficiency of processing entire chemical images with non-informative backgrounds. (ii) a narrow scope on molecular-level tasks that restricts progress in chemical reasoning. In this work, we propose TinyChemVL, an efficient and powerful chemical VLM that leverages visual token reduction and reaction-level tasks to improve model efficiency and reasoning capacity. Also, we propose ChemRxn-V, a reaction-level benchmark for assessing vision-based reaction recognition and prediction tasks. Directly predicting reaction products from molecular images poses a non-trivial challenge, as it requires models to integrate both recognition and reasoning capacities. Our results demonstrate that with only 4B parameters, TinyChemVL achieves superior performance on both molecular and reaction tasks while demonstrating faster inference and training speeds compared to existing models. Notably, TinyChemVL outperforms ChemVLM while utilizing only 1/16th of the visual tokens. This work builds efficient yet powerful VLMs for chemical domains by co-designing model architecture and task complexity.