Multimodal-CoT with vision features achieves higher validation accuracy at early training epochs (epoch 1-3) compared to one-stage and two-stage language-only baselines on ScienceQA

{
  "doc": "zhang-2023-multimodal.md",
  "context": "Evidence that multimodal information accelerates convergence speed during training.",
  "norm_label": "Multimodal-CoT with vision features achieves higher validation accuracy at early training epochs (epoch 1-3) compared to one-stage and two-stage language-only baselines on ScienceQA",
  "graphify_id": "finding_convergence_boost",
  "source_file": "zhang-2023-multimodal.md",
  "imported_from": "/Users/antonborzov/Documents/Research.nosync/papers/extract_typed_out/zhang-2023-multimodal/graph.json",
  "extracted_type": "finding",
  "source_location": "§6.1",
  "graphify_file_type": "finding"
}