datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
arxiv-funding-entity-extractions
arxiv-funding-entity-extractions
Funder/award entity extractions over cometadata/arxiv-funding-statements.
Extractor: funding-entity-extractor (vLLM + LoRA)
Base model: meta-llama/Llama-3.1-8B-Instruct
LoRA: cometadata/funding-extraction-llama-3.1-8b-instruct-artifact-data-mix-grpo-mixed-reward
Hardware: A100-large bf16, concurrency 256
Total rows: 1,823,650
Configs
predictions (default) — original extractions, no ROR enrichment.
predictions_with_ror — same rows… See the full description on the dataset page: https://huggingface.co/datasets/cometadata/arxiv-funding-entity-extractions.funding-entity-extraction-baguette-600m-predictions
