72 citations · 148 across the 43 of their papers we have counts for
45 papers
Inference-Free Multimodal Learned Sparse Retrieval for Production-Scale Visual Document Search
Gyu-Hwung Cho, Youngjune Lee, Kiyoon Jeong +5
As large-scale visual-document corpora such as arXiv papers and enterprise PDFs continue to grow, visual-document retrieval has gained increasing attention; yet it still lacks a de…
Efficient Listwise Reranking with Compressed Document Representations
Hervé Déjean, Stéphane Clinchant
Reranking, the process of refining the output from a first-stage retriever, is often considered computationally expensive, especially when using Large Language Models (LLMs). A com…
On the Challenges and Opportunities of Learned Sparse Retrieval for Code
Simon Lupart, Maxime Louis, Thibault Formal +2
Retrieval over large codebases is a key component of modern LLM-based software engineering systems. Existing approaches predominantly rely on dense embedding models, while learned…
Retrieval-Augmented LLM Agents: Learning to Learn from Experience
Thomas Palmeira Ferraz, Romain Deffayet, Vassilina Nikoulina +2
While large language models (LLMs) have advanced the development of general-purpose agents, robust generalization to unseen tasks remains challenging. Two common approaches are sup…
Learning Retrieval Models with Sparse Autoencoders
Thibault Formal, Maxime Louis, Hervé Dejean +1
Sparse autoencoders (SAEs) provide a powerful mechanism for decomposing the dense representations produced by Large Language Models (LLMs) into interpretable latent features. We po…
Naver Labs Europe @ WSDM CUP | Multilingual Retrieval
Thibault Formal, Maxime Louis, Hervé Déjean +1
This report presents our participation to the WSDM Cup 2026 shared task on multilingual document retrieval from English queries. The task provides a challenging benchmark for cross…
Reranking with Compressed Document Representation
Hervé Déjean, Stéphane Clinchant
Reranking, the process of refining the output of a first-stage retriever, is often considered computationally expensive, especially with Large Language Models. Borrowing from recen…
OSCAR: Online Soft Compression And Reranking
Maxime Louis, Thibault Formal, Hervé Dejean +1
Retrieval-Augmented Generation (RAG) enhances Large Language Models (LLMs) by integrating external knowledge, leading to improved accuracy and relevance. However, scaling RAG pipel…
RANa: Retrieval-Augmented Navigation
Gianluca Monaci, Rafael S. Rezende, Romain Deffayet +5
Methods for navigation based on large-scale learning typically treat each episode as a new problem, where the agent is spawned with a clean memory in an unknown environment. While…
Provence: efficient and robust context pruning for retrieval-augmented generation
Nadezhda Chirkova, Thibault Formal, Vassilina Nikoulina +1
Retrieval-augmented generation improves various aspects of large language models (LLMs) generation, but suffers from computational overhead caused by long contexts as well as the p…
PISCO: Pretty Simple Compression for Retrieval-Augmented Generation
Maxime Louis, Hervé Déjean, Stéphane Clinchant
Retrieval-Augmented Generation (RAG) pipelines enhance Large Language Models (LLMs) by retrieving relevant documents, but they face scalability issues due to high inference costs a…
Context Embeddings for Efficient Answer Generation in RAG
David Rau, Shuai Wang, Hervé Déjean +1
Retrieval-Augmented Generation (RAG) allows overcoming the limited knowledge of LLMs by extending the input with external information. As a consequence, the contextual inputs to th…
Retrieval-augmented generation in multilingual settings
Nadezhda Chirkova, David Rau, Hervé Déjean +3
Retrieval-augmented generation (RAG) has recently emerged as a promising solution for incorporating up-to-date or domain-specific knowledge into large language models (LLMs) and im…
BERGEN: A Benchmarking Library for Retrieval-Augmented Generation
David Rau, Hervé Déjean, Nadezhda Chirkova +4
Retrieval-Augmented Generation allows to enhance Large Language Models with external knowledge. In response to the recent popularity of generative LLMs, many RAG approaches have be…
SPLATE: Sparse Late Interaction Retrieval
Thibault Formal, Stéphane Clinchant, Hervé Déjean +1
The late interaction paradigm introduced with ColBERT stands out in the neural Information Retrieval space, offering a compelling effectiveness-efficiency trade-off across many ben…
Two-Step SPLADE: Simple, Efficient and Effective Approximation of SPLADE
Carlos Lassance, Hervé Dejean, Stéphane Clinchant +1
Learned sparse models such as SPLADE have successfully shown how to incorporate the benefits of state-of-the-art neural information retrieval models into the classical inverted ind…
A Thorough Comparison of Cross-Encoders and LLMs for Reranking SPLADE
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. We conduct a large evaluation on TREC Deep Learnin…
SPLADE-v3: New baselines for SPLADE
Carlos Lassance, Hervé Déjean, Thibault Formal +1
A companion to the release of the latest version of the SPLADE library. We describe changes to the training structure and present our latest series of models -- SPLADE-v3. We compa…
Benchmarking Middle-Trained Language Models for Neural Search
Hervé Déjean, Stéphane Clinchant, Carlos Lassance +2
Middle training methods aim to bridge the gap between the Masked Language Model (MLM) pre-training and the final finetuning for retrieval. Recent models such as CoCondenser, RetroM…
The tale of two MS MARCO -- and their unfair comparisons
Carlos Lassance, Stéphane Clinchant
The MS MARCO-passage dataset has been the main large-scale dataset open to the IR community and it has fostered successfully the development of novel neural retrieval models over t…
A Static Pruning Study on Sparse Neural Retrievers
Carlos Lassance, Simon Lupart, Hervé Dejean +2
Sparse neural retrievers, such as DeepImpact, uniCOIL and SPLADE, have been introduced recently as an efficient and effective way to perform retrieval with inverted indexes. They a…
AToMiC: An Image/Text Retrieval Test Collection to Support Multimedia Content Creation
Jheng-Hong Yang, Carlos Lassance, Rafael Sampaio de Rezende +4
This paper presents the AToMiC (Authoring Tools for Multimedia Content) dataset, designed to advance research in image/text cross-modal retrieval. While vision-language pretrained…
Parameter-Efficient Sparse Retrievers and Rerankers using Adapters
Vaishali Pal, Carlos Lassance, Hervé Déjean +1
Parameter-Efficient transfer learning with Adapters have been studied in Natural Language Processing (NLP) as an alternative to full fine-tuning. Adapters are memory-efficient and…
Naver Labs Europe (SPLADE) @ TREC NeuCLIR 2022
Carlos Lassance, Stéphane Clinchant
This paper describes our participation in the 2022 TREC NeuCLIR challenge. We submitted runs to two out of the three languages (Farsi and Russian), with a focus on first-stage rank…