no code implementations • 22 Apr 2024 • Thibault Formal, Stéphane Clinchant, Hervé Déjean, Carlos Lassance
The late interaction paradigm introduced with ColBERT stands out in the neural Information Retrieval space, offering a compelling effectiveness-efficiency trade-off across many benchmarks.
no code implementations • 15 Mar 2024 • Hervé Déjean, Stéphane Clinchant, Thibault Formal
We present a comparative study between cross-encoder and LLMs rerankers in the context of re-ranking effective SPLADE retrievers.
no code implementations • 11 Mar 2024 • Carlos Lassance, Hervé Déjean, Thibault Formal, Stéphane Clinchant
A companion to the release of the latest version of the SPLADE library.
1 code implementation • 5 Jun 2023 • Hervé Déjean, Stéphane Clinchant, Carlos Lassance, Simon Lupart, Thibault Formal
We compare both dense and sparse approaches under various finetuning protocols and middle training on different collections (MS MARCO, Wikipedia or Tripclick).
1 code implementation • 20 Feb 2023 • Guglielmo Faggioli, Thibault Formal, Stefano Marchesin, Stéphane Clinchant, Nicola Ferro, Benjamin Piwowarski
On top of that, in lexical-oriented scenarios, QPPs fail to predict performance for neural IR systems on those queries where they differ from traditional approaches the most.
no code implementations • 11 Jan 2023 • Nam Le Hai, Thomas Gerald, Thibault Formal, Jian-Yun Nie, Benjamin Piwowarski, Laure Soulier
Conversational search is a difficult task as it aims at retrieving documents based not only on the current user query but also on the full conversation history.
1 code implementation • 10 May 2022 • Thibault Formal, Carlos Lassance, Benjamin Piwowarski, Stéphane Clinchant
Neural retrievers based on dense representations combined with Approximate Nearest Neighbors search have recently received a lot of attention, owing their success to distillation and/or better sampling of examples for training -- while still relying on the same backbone architecture.
1 code implementation • 5 May 2022 • Simon Lupart, Thibault Formal, Stéphane Clinchant
To this end, we build three query-based distribution shifts within MS MARCO (query-semantic, query-intent, query-length), which are used to evaluate the three main families of neural retrievers based on BERT: sparse, dense, and late-interaction -- as well as a monoBERT re-ranker.
no code implementations • 14 Apr 2022 • Carlos Lassance, Thibault Formal, Stephane Clinchant
Second, CCSA can be used as a binary quantization method and we propose to combine it with the recent graph based ANN techniques.
no code implementations • 10 Dec 2021 • Thibault Formal, Benjamin Piwowarski, Stéphane Clinchant
Neural Information Retrieval models hold the promise to replace lexical matching models, e. g. BM25, in modern search engines.
1 code implementation • 21 Sep 2021 • Thibault Formal, Carlos Lassance, Benjamin Piwowarski, Stéphane Clinchant
Meanwhile, there has been a growing interest in learning \emph{sparse} representations for documents and queries, that could inherit from the desirable properties of bag-of-words models such as the exact matching of terms and the efficiency of inverted indexes.
Ranked #5 on Zero-shot Text Search on BEIR
1 code implementation • 12 Jul 2021 • Thibault Formal, Benjamin Piwowarski, Stéphane Clinchant
In neural Information Retrieval, ongoing research is directed towards improving the first retriever in ranking pipelines.
no code implementations • 17 Dec 2020 • Thibault Formal, Benjamin Piwowarski, Stéphane Clinchant
Transformer-based models are nowadays state-of-the-art in ad-hoc Information Retrieval, but their behavior is far from being understood.