Skip to content
ModelListedExternal
Download

Trialmatchai Gemma2 Reranker LoRA

A LoRA adapter for the google/gemma-2-2b family, fine-tuned for the second-level criterion reranking stage of TrialMatchAI, an end-to-end system that matches patients to the clinical trials they are eligible for.

KindOtherBase modelgoogle/gemma-2-2bVersionRevision 3118ba7Bgrade

Model source

Kind
Other
File size
24.4 MB
Base model
google/gemma-2-2b
Version
vRevision 3118ba7
Updated
Oct 20, 2026
Downloads
19
Source
Hugging Face

Source description

TrialMatchAI · gemma-2-2b criterion reranker adapter (LoRA)

A LoRA adapter for the google/gemma-2-2b family, fine-tuned for the second-level criterion reranking stage of TrialMatchAI, an end-to-end system that matches patients to the clinical trials they are eligible for.

After first-level retrieval surfaces candidate trials, this adapter scores each (patient, trial criterion) pair so the most relevant trials are reranked to the top and shortlisted for the downstream eligibility-reasoning stage.

Usage

TrialMatchAI loads this adapter automatically. In config.json it sits on top of the instruct base:

"reranker_model_path": "google/gemma-2-2b-it",
"reranker_adapter_path": "majdabd33/trialmatchai-gemma2-reranker-lora"

It downloads on first use, so no separate model download step is needed. See the TrialMatchAI repository for the full pipeline and configuration.

Base model and license

This is an adapter of google/gemma-2-2b, governed by the Gemma license; accept it on the base model page before use. For research and informational use only. It is not medical advice, not a medical device, and must not replace review by qualified healthcare professionals.

Citation

Abdallah, M. et al. TrialMatchAI: an end-to-end AI-powered clinical trial recommendation system to streamline patient-to-trial matching. Nature Communications 17, 4472 (2026). https://doi.org/10.1038/s41467-026-70509-w

Sources

1 source
Hugging FacePrimaryVersion 3118ba76d545f71f3aaa7952d2f031ad5fdef8c9
huggingface.co/majdabd33/trialmatchai-gemma2-reranker-lora/tree/3118ba76d545f71f3aaa7952d2f031ad5fdef8c9

Verified Oct 20

Model artifacts

1 artifact

adapter_model.safetensors

safetensors · 24.4 MB · SHA-256 851aeb9f82db…770a · Hugging Face

Download file

Source excerpts

3 excerpts

--- base_model: google/gemma-2-2b library_name: peft pipeline_tag: text-generation license: gemma tags: - clinical-trials - patient-matching - reranking - lora - trialmatchai --- # TrialMatchAI · gemma-2-2b criterion reranker adapter (LoRA) A LoRA adapter for the [google/gemma-2-2b](https://huggingface.co/google/gemma-2-2b) family, fine-tuned for the second-level **criterion reranking** stage of [TrialMatchAI](https://github.com/cbib/TrialMatchAI), an end-to-end system that matches patients to the clinical trials they are eligible for. After first-level retrieval surfaces candidate trials, this adapter scores each (patient, trial criterion) pair so the most relevant trials are reranked to the top and shortlisted for the downstream eligibility-reasoning stage. ## Usage TrialMatchAI loads this adapter automatically. In `config.json` it sits on top of the instruct base: ```json "reranker_model_path": "google/gemma-2-2b-it", "reranker_adapter_path": "majdabd33/trialmatchai-gemma2-reranker-lora" ``` It downloads on first use, so no separate model download step is needed. See the [TrialMatchAI repository](https://github.com/cbib/TrialMatchAI) for the full pipeline and configuration. ## Base model and license This is an adapter of `google/gemma-2-2b`, governed by the [Gemma license](https://ai.google.dev/gemma/terms); accept it on the base model page before use. For **research and informational use only**. It is not medical advice, not a medical device, and must not replace review by qualified healthcare professionals. ## Citation > Abdallah, M. *et al.* TrialMatchAI: an end-to-end AI-powered clinical trial recommendation system to > streamline patient-to-trial matching. *Nature Communications* **17**, 4472 (2026). > https://doi.org/10.1038/s41467-026-70509-w

Source context: 19 downloads · 0 likes · Pipeline text-generation · Library peft · Repo majdabd33/trialmatchai-gemma2-reranker-lora

Source context: 32 downloads · 0 likes · Pipeline text-generation · Library peft · Repo majdabd33/trialmatchai-gemma2-reranker-lora