Fine-tuning DeepSeek-OCR-2 for Molecular Structure Recognition

• URL: http://arxiv.org/abs/2604.03476v1
• Date: Fri, 03 Apr 2026 21:42:54 GMT
• Title: Fine-tuning DeepSeek-OCR-2 for Molecular Structure Recognition
• Abstract summary: We adapt DeepSeek-OCR-2 for molecular optical recognition by formulating the task as image-conditioned SMILES generation.<n>We train our model on a large-scale corpus combining synthetic renderings from PubChem and realistic patent images from USPTO-MOL.<n>Our model, MolSeek-OCR, demonstrates competitive capabilities, achieving exact matching accuracies comparable to the best-performing image-to-sequence model.
• Score: 1.957558771641347
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Optical Chemical Structure Recognition (OCSR) is critical for converting 2D molecular diagrams from printed literature into machine-readable formats. While Vision-Language Models have shown promise in end-to-end OCR tasks, their direct application to OCSR remains challenging, and direct full-parameter supervised fine-tuning often fails. In this work, we adapt DeepSeek-OCR-2 for molecular optical recognition by formulating the task as image-conditioned SMILES generation. To overcome training instabilities, we propose a two-stage progressive supervised fine-tuning strategy: starting with parameter-efficient LoRA and transitioning to selective full-parameter fine-tuning with split learning rates. We train our model on a large-scale corpus combining synthetic renderings from PubChem and realistic patent images from USPTO-MOL to improve coverage and robustness. Our fine-tuned model, MolSeek-OCR, demonstrates competitive capabilities, achieving exact matching accuracies comparable to the best-performing image-to-sequence model. However, it remains inferior to state-of-the-art image-to-graph modelS. Furthermore, we explore reinforcement-style post-training and data-curation-based refinement, finding that they fail to improve the strict sequence-level fidelity required for exact SMILES matching.

Related papers

• Molecular Identifier Visual Prompt and Verifiable Reinforcement Learning for Chemical Reaction Diagram Parsing [52.825281124618535]
  Reaction diagram parsing (RxnDP) is critical for extracting chemical synthesis information from literature.<n>Recent Vision-Language Models (VLMs) have emerged as a promising paradigm to automate this complex visual reasoning task.<n>This work enhances VLM-based RxnDP from two complementary perspectives: prompting representation and learning paradigms.
  arXiv  Detail & Related papers  (2026-03-16T09:17:05Z)
• IAR2: Improving Autoregressive Visual Generation with Semantic-Detail Associated Token Prediction [77.06211178777939]
  IAR2 is an advanced autoregressive framework that enables a hierarchical semantic-detail synthesis process.<n>We show that IAR2 sets a new state-of-the-art for autoregressive image generation, achieving a FID of 1.50 on ImageNet.
  arXiv  Detail & Related papers  (2025-10-08T12:08:21Z)
• DianJin-OCR-R1: Enhancing OCR Capabilities via a Reasoning-and-Tool Interleaved Vision-Language Model [9.557159109747372]
  Large vision-language models (LVLMs) are prone to hallucinations--generating words that do not exist in input images.<n>We propose DianJin-OCR-R1, a reasoning-and-tool interleaved VLMs trained on domain-specific datasets.
  arXiv  Detail & Related papers  (2025-08-18T03:28:57Z)
• Vision Foundation Models as Effective Visual Tokenizers for Autoregressive Image Generation [52.261584726401686]
  We present a novel direction to build an image tokenizer directly on top of a frozen vision foundation model.<n>Based on these designs, our proposed image tokenizer, VFMTok, achieves substantial improvements in image reconstruction and generation quality.
  arXiv  Detail & Related papers  (2025-07-11T09:32:45Z)
• GTR-CoT: Graph Traversal as Visual Chain of Thought for Molecular Structure Recognition [60.76623665324548]
  GTR-Mol-VLM is a novel framework featuring two key innovations.<n>It emulates human reasoning by incrementally parsing molecular graphs through sequential atom-bond predictions.<n>MolRec-Bench is the first benchmark designed for a fine-grained evaluation of graph-parsing accuracy in OCSR.
  arXiv  Detail & Related papers  (2025-06-09T08:47:10Z)
• Referee Can Play: An Alternative Approach to Conditional Generation via Model Inversion [35.21106030549071]
  Diffusion Probabilistic Models (DPMs) are dominant force in text-to-image generation tasks. We propose an alternative view of state-of-the-art DPMs as a way of inverting advanced Vision-Language Models (VLMs) By directly optimizing images with the supervision of discriminative VLMs, the proposed method can potentially achieve a better text-image alignment.
  arXiv  Detail & Related papers  (2024-02-26T05:08:40Z)
• Universal Generative Modeling for Calibration-free Parallel Mr Imaging [13.875986147033002]
  We present an unsupervised deep learning framework for calibration-free parallel MRI. We make use of the merits of both wavelet transform and the adaptive iteration strategy in a unified framework. We train a powerful noise conditional score network by forming wavelet tensor as the network input.
  arXiv  Detail & Related papers  (2022-01-25T10:05:39Z)
• Two-Stage Self-Supervised Cycle-Consistency Network for Reconstruction of Thin-Slice MR Images [62.4428833931443]
  The thick-slice magnetic resonance (MR) images are often structurally blurred in coronal and sagittal views. Deep learning has shown great potential to re-construct the high-resolution (HR) thin-slice MR images from those low-resolution (LR) cases. We propose a novel Two-stage Self-supervised Cycle-consistency Network (TSCNet) for MR slice reconstruction.
  arXiv  Detail & Related papers  (2021-06-29T13:29:18Z)
• Characteristic Regularisation for Super-Resolving Face Images [81.84939112201377]
  Existing facial image super-resolution (SR) methods focus mostly on improving artificially down-sampled low-resolution (LR) imagery. Previous unsupervised domain adaptation (UDA) methods address this issue by training a model using unpaired genuine LR and HR data. This renders the model overstretched with two tasks: consistifying the visual characteristics and enhancing the image resolution. We formulate a method that joins the advantages of conventional SR and UDA models.
  arXiv  Detail & Related papers  (2019-12-30T16:27:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.