gold-standard dataset for table extraction evalution

[jd-coderepos]

please consider these five domains (50 papers in each, in total 250 papers)

1. Biomedicine & Health
   (clinical research, public health, neuroscience, genomics, drug discovery, medical imaging, digital health)

2. Agriculture, Food & Environmental Systems
   (crop/soil science, agroecology, livestock, food science, sustainable farming, climate adaptation, biodiversity)

3. Computer Science, AI & Data Science
   (machine learning, NLP, computer vision, databases, cybersecurity, human-AI interaction)

4. Energy, Materials & Chemical Sciences
   (batteries, catalysis, semiconductors, polymers, green chemistry, process engineering, nanotech)

5. Engineering, Robotics & Built Infrastructure
   (robotics, manufacturing, sensors/IoT, civil infrastructure, transportation, aerospace, control systems)

[jd-coderepos]

for Energy, Materials & Chemical Sciences, you can reuse the papers from here to get started . The gold-standard tables are released as ORKG comparisons which can be exported in csv format for the evaluation.