RehanaHasin/mistral-7b-instruct-v0.3-adjuvant-extractor

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:May 29, 2026License:apache-2.0Architecture:Transformer Open Weights Warm

RehanaHasin/mistral-7b-instruct-v0.3-adjuvant-extractor is a 7 billion parameter instruction-tuned model based on Mistral 7B Instruct v0.3. Developed by RehanaHasin, this model is specifically fine-tuned for the task of extracting vaccine adjuvant concepts and associated evidence snippets from infectious disease biomedical abstracts. It excels at structured information extraction, returning results in a precise JSON format. This model is primarily intended for research workflows in biomedical literature mining, focusing on vaccine adjuvant concept extraction and evidence-linked information.

Loading preview...

Model Overview

This model, mistral-7b-instruct-v0.3-adjuvant-extractor, is a specialized fine-tuned version of the Mistral 7B Instruct v0.3 base model. Its core purpose is to perform evidence-linked adjuvant extraction from biomedical abstracts, specifically focusing on infectious disease vaccine literature.

Key Capabilities

  • Task-Specific Extraction: Designed to identify and extract vaccine adjuvant concepts.
  • Evidence Snippet Linking: Provides supporting text snippets (evidence) directly from the input abstract for each extracted adjuvant.
  • Structured JSON Output: Guarantees output in a strict JSON format, containing adjuvant and evidence keys, facilitating automated parsing.
  • Biomedical Focus: Fine-tuned on a curated infectious disease adjuvant corpus, making it highly relevant for this domain.

Intended Use Cases

  • Infectious Disease Vaccine Literature Curation: Streamlining the process of identifying key components in vaccine research.
  • Vaccine Adjuvant Concept Extraction: Automating the discovery of adjuvant mentions in scientific texts.
  • Evidence-Linked Information Extraction: Supporting downstream manual review by providing direct textual evidence.

Limitations

  • Generalization beyond the focused infectious-disease adjuvant corpus is not guaranteed.
  • Performance can vary based on abstract quality and terminology.
  • Outputs may require post-processing and manual validation, and the model is not intended for clinical decision-making.

Finetunes
1 models