Downloads · 30 days
5
15% of all-time downloads
Academica/AcademicaMed-FATE-ClinicalTrials-Outcome-256
AcademicaMed-FATE-ClinicalTrials-Outcome-256 is a sentence similarity model from Academica. Use it when you need a score for how close two texts are. It is set up for sentence-transformers. The card lists the license as cc-by-nc-nd-4.0.
FATE is the historical outcome-pattern embedding model in the AcademicaMed clinical-trial model family.
Downloads · 30 days
5
15% of all-time downloads
All-time downloads
34
Public
Repo size
—
Likes
0
Public
Click a slice to open those files.
.md2 KB · 46%
From the Hugging Face model README
FATE is the historical outcome-pattern embedding model in the AcademicaMed clinical-trial model family.
The model represents trial text as a 256-dimensional vector. Its contrastive objective groups completed trials through verified outcome labels.
| Property | Value |
|---|---|
| Base model | NeuML/pubmedbert-base-embeddings |
| Output dimension | 256 |
| Training method | Triplet-loss contrastive learning |
| Training data | 18,132 completed trials with outcome labels |
| Positive class | 10,609 trials with confirmed primary-endpoint success |
| Negative class | 7,523 completed trials with negative primary endpoints |
| Epochs | 10 |
| Batch size | 32 |
| Learning rate | 2e-5 |
| Loss margin | 0.5 |
| Training hardware | NVIDIA DGX Spark, GB10 Blackwell |
FATE supports neighborhood analysis against historical outcome patterns. It is not an independent success predictor.
Trial text -> PubMedBERT embedding -> dense 384-to-256 projection -> L2-normalized vector
Outcome labels reflect available ClinicalTrials.gov records. Reporting gaps and historical selection effects can affect the embedding space.
This repository contains the original published model card and configuration scope. The source repository did not publish executable weight files.