Model Card for passage-ranker-v1-L-en

This model is a passage ranker developed by Sinequa. It produces a relevance score given a query-passage pair and is used to order search results.

Model name: passage-ranker-v1-L-en

Supported Languages

The model was trained and tested in the following languages:

• English

Scores

Metric	Value
Relevance (NDCG@10)	0.466

Note that the relevance score is computed as an average over 14 retrieval datasets (see details below).

Inference Times

GPU	Quantization type	Batch size 1	Batch size 32
NVIDIA A10	FP16	2 ms	27 ms
NVIDIA A10	FP32	4 ms	82 ms
NVIDIA T4	FP16	3 ms	63 ms
NVIDIA T4	FP32	13 ms	342 ms
NVIDIA L4	FP16	2 ms	39 ms
NVIDIA L4	FP32	5 ms	119 ms

Gpu Memory usage

Quantization type	Memory
FP16	550 MiB
FP32	1100 MiB

Note that GPU memory usage only includes how much GPU memory the actual model consumes on an NVIDIA T4 GPU with a batch size of 32. It does not include the fix amount of memory that is consumed by the ONNX Runtime upon initialization which can be around 0.5 to 1 GiB depending on the used GPU.

Requirements

• Minimal Sinequa version: 11.10.0
• Minimal Sinequa version for using FP16 models and GPUs with CUDA compute capability of 8.9+ (like NVIDIA L4): 11.11.0
• Cuda compute capability: above 5.0 (above 6.0 for FP16 use)

Model Details

Overview

• Number of parameters: 109 million
• Base language model: English BERT-Base
• Insensitive to casing and accents
• Training procedure: MonoBERT

Training Data

• Probably-Asked Questions (Paper, Official Page)

Evaluation Metrics

To determine the relevance score, we averaged the results that we obtained when evaluating on the datasets of the BEIR benchmark. Note that all these datasets are in English.

Dataset	NDCG@10
Average	0.466
Arguana	0.567
CLIMATE-FEVER	0.162
DBPedia Entity	0.363
FEVER	0.721
FiQA-2018	0.304
HotpotQA	0.680
MS MARCO	0.342
NFCorpus	0.346
NQ	0.487
Quora	0.779
SCIDOCS	0.150
SciFact	0.649
TREC-COVID	0.683
Webis-Touche-2020	0.287