8 papers
Shieldstral
Antonia Calvi, Avinash Sooriyarachchi, Giada Pistilli +273
We introduce Shieldstral, a 3B-parameter policy-adaptive multimodal safety classifier that matches or outperforms models nearly 7 its size on text safety benchmarks and set…
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…
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…
DiffLoRA: Differential Low-Rank Adapters for Large Language Models
Alexandre Misrahi, Nadezhda Chirkova, Maxime Louis +1
Differential Transformer has recently been proposed to improve performance in Transformer models by canceling out noise through a denoiser attention mechanism. In this work, we int…
Adapting Large Language Models for Multi-Domain Retrieval-Augmented-Generation
Alexandre Misrahi, Nadezhda Chirkova, Maxime Louis +1
Retrieval-Augmented Generation (RAG) enhances LLM factuality, but multi-domain applications face challenges like lack of diverse benchmarks and poor out-of-domain generalization. T…
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…
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…