PGC Psychiatric GWAS Summary Statistics Collection ~1 billion rows of genome-wide association study (GWAS) NOTE: We are in the process to transfer these datasets to the Psychiatric Genomics Consortiu • 12 items • Updated about 15 hours ago • 70
view article Article SynthVision: Building a 110K Synthetic Medical VQA Dataset with Cross-Model Validation 18 days ago • 16
Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation Paper • 2603.19220 • Published 22 days ago • 66
SynthVision Collection Medical VQA datasets and fine-tuned models from the SynthVision pipeline. • 8 items • Updated 23 days ago • 6
Dutch PII & De-Identification Collection 35 open-source Dutch PII detection models. 54 entity types. Best: DeBERTa-v3-large F1=94.2%. Apache 2.0. • 35 items • Updated Mar 9 • 2
Hindi PII & De-Identification Collection 35 open-source Hindi PII detection models. 54 entity types. Best F1: 96.6%. Apache 2.0. • 35 items • Updated Mar 10 • 4
Telugu PII & De-Identification Collection 35 open-source Telugu PII detection models. 54 entity types. Best F1: 95.3%. Apache 2.0. • 35 items • Updated Mar 10 • 4
Spanish PII & De-Identification Collection 33 models for Spanish PII detection & de-identification. 55+ entity types. HIPAA & GDPR compliant. Apache 2.0. • 35 items • Updated Feb 17 • 4
French PII & De-Identification Collection 33 models for French PII detection & de-identification. 55+ entity types. HIPAA & GDPR compliant. Apache 2.0. • 35 items • Updated Feb 17 • 3
Italian PII & De-Identification Collection 33 models for Italian PII detection & de-identification. 55+ entity types. HIPAA & GDPR compliant. Apache 2.0. • 35 items • Updated Feb 19 • 2
German PII & De-Identification Collection 33 models for German PII detection & de-identification. 55+ entity types. HIPAA & GDPR compliant. Apache 2.0. • 35 items • Updated Feb 17 • 3