File size: 1,583 Bytes
01c1ff6 4afbca2 0cecf3a 4afbca2 01c1ff6 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 | ---
license: apache-2.0
pipeline_tag: text-ranking
library_name: lightning-ir
base_model:
- google/electra-base-discriminator
---
# Set-Encoder
This repository contains the code for the paper: [`Set-Encoder: Permutation-Invariant Inter-Passage Attention for Listwise Passage Re-Ranking with Cross-Encoders`](https://arxiv.org/abs/2404.06912).
We use [`lightning-ir`](https://github.com/webis-de/lightning-ir) to train and fine-tune models. Download and install the library to use the code in this repository.
## Model Zoo
We provide the following pre-trained models:
| Model Name | TREC DL 19 (BM25) | TREC DL 20 (BM25) | TREC DL 19 (ColBERTv2) | TREC DL 20 (ColBERTv2) |
| ------------------------------------------------------------------- | ----------------- | ----------------- | ---------------------- | ---------------------- |
| [set-encoder-base](https://huggingface.co/webis/set-encoder-base) | 0.724 | 0.710 | 0.788 | 0.777 |
| [set-encoder-large](https://huggingface.co/webis/set-encoder-large) | 0.727 | 0.735 | 0.789 | 0.790 |
## Inference
We recommend using the `lightning-ir` cli to run inference. The following command can be used to run inference using the `set-encoder-base` model on the TREC DL 19 and TREC DL 20 datasets:
```bash
lightning-ir re_rank --config configs/re-rank.yaml --config configs/set-encoder-finetuned.yaml --config configs/trec-dl.yaml
```
## Fine-Tuning
WIP |