README.md

---
base_model: allenai/scibert_scivocab_uncased
library_name: peft
pipeline_tag: feature-extraction
tags:
- embeddings
- m6
- paper-fulltext-embedding
- papers
- repository-library
- research-library
- scientific-papers
- t3_paper_text
---

# Paper Fulltext Embedding

Produces embeddings over full paper text for retrieval and clustering tasks.

## Model Details

- Artifact type: LoRA adapter
- Base model: `allenai/scibert_scivocab_uncased`
- Backbone type: `encoder`
- Model ID: `M6`
- Tier: `T3_paper_text`
- Role in stack: specialized Repository Library component

This model is part of the Repository Library stack, a research system for indexing, retrieving, aligning, and reasoning over scientific papers, structured paper content, repositories, and cross-domain links between them.

## Model Sources

- Hugging Face repo: `https://huggingface.co/PeytonT/paper-fulltext-embedding`
- Hugging Face collection: `https://huggingface.co/collections/PeytonT/research-library-6a49c589ef4d763f7539b50d`
- GitHub repository: `https://github.com/peytontolbert/research_library`
- Experiment config: `https://github.com/peytontolbert/research_library/blob/main/models/experiments/m6_paper_fulltext_embedding.json`
- Models directory: `https://github.com/peytontolbert/research_library/tree/main/models`

## Intended Use

- Primary use: Produces embeddings over full paper text for retrieval and clustering tasks.
- Downstream use: retrieval, ranking, planning, paper understanding, or cross-domain reasoning inside the broader Repository Library system, depending on the model family.
- Out of scope: production safety claims, benchmark claims beyond the tracked experiment config, or deployment without task-specific validation.

## Training Data

The training inputs for this package were assembled from the following Repository Library data sources:

- Primary published dataset: `local/paper_text_2m_dedup_v1`
- Source `paper_text_parquet`: full-text paper corpus records prepared for model training.

## Training Procedure

- Sources: `paper_text_parquet`
- Input fields: `title, abstract, text`
- Target fields: `fulltext_embedding`
- Train/val/test split: `[0.8, 0.1, 0.1]`
- Max samples: `0`
- Batch size: `4`
- Precision: `bf16`
- Objective: `contrastive`
- Learning rate: `0.0001`
- Max source tokens: `512`
- Max target tokens: `128`
- Fine-tune strategy: `peft_lora`
- Max steps: `1000`

## Compute

- Hardware: 4x RTX_3090 (24 GB)
- Distributed strategy: `ddp`
- Estimated GPU hours in config: `0`

## Evaluation

- Declared metrics: `recall_at_10, ndcg_at_10`
- Status: this card reflects the current tracked experiment configuration and packaged weights in the Repository Library model stack.

## Usage

```python
from transformers import AutoModel, AutoTokenizer
from peft import PeftModel

repo_id = "PeytonT/paper-fulltext-embedding"
base_id = "allenai/scibert_scivocab_uncased"

tokenizer = AutoTokenizer.from_pretrained(repo_id)
base = AutoModel.from_pretrained(base_id)
model = PeftModel.from_pretrained(base, repo_id)
```

## Limitations

- These cards are generated from tracked experiment metadata and packaged artifacts, not from a separate benchmark report or external audit.
- Several training sources are pipeline outputs from the Repository Library codebase rather than standalone public datasets.
- These models are components of a larger research system and should be validated in their target workflow before deployment.

## Project Context

- GitHub repository: `https://github.com/peytontolbert/research_library`
- Model collection: `https://huggingface.co/collections/PeytonT/research-library-6a49c589ef4d763f7539b50d`
- Publisher: `PeytonT`