The 1st Sci-ImageMiner Competition: Information Extraction from Scientific Figures in Materials Science Focus: Quantitative plots from Atomic Layer Deposition and Etching (ALD/E) research

Organized as part of the ICDAR 2026 Competition track https://icdar2026.org/index.php/call-for-competitions/ https://icdar2026.org/index.php/competitions/ https://icdar2026.org/index.php/competitions/ https://icdar2026.org/index.php/call-for-competitions/ ICDAR 2026 - The 20th International Conference on Document Analysis and Recognition 30 Aug - 04 Sep 2026 | Vienna, Austria

Sci-ImageMiner Competition Website: https://sites.google.com/view/sci-imageminer/ https://sites.google.com/view/sci-imageminer/

Overview

Scientific figures often contain critical results that never appear explicitly in the text. Despite recent advances in multimodal large language models, most existing benchmarks rely on generic or synthetic visuals. Sci-ImageMiner addresses this gap by introducing a curated benchmark grounded in authentic scientific figures from a specialized scientific domain.

Tasks

The competition hosts four tasks:

1. Figure Classification - identify the chart or figure type 2. Data Table Extraction - reconstruct the underlying tabular data from quantitative plots 3. Figure Summarization - generate concise, factual summaries of key trends 4. Visual Question Answering (VQA) - answer scientific questions requiring reasoning over figure content

Teams may participate in any subset of tasks (including all four).

Data & Evaluation

* A trial dataset is available now to familiarize participants with the data and annotations. * The full dataset will be released in stages (training, development, blind test). * All evaluations will be conducted on Codabench, with per-task leaderboards.

Important Dates (selected)

* Trial data release: 8 December 2025 * Evaluation start: 3 March 2026 * Evaluation end: 3 April 2026 * Paper submission deadline: 17 April 2026 * Camera-ready deadline: 4 May 2026