8 papers
Toward Collective-Centric Evaluation of Preference Inference for Participatory Democracy
Pierre-Antoine Lequeu, Salim Hafid, Paul Lerner +6
To scale up collective decision-making, participatory democracy platforms such as Polis and Remesh enable online deliberation among thousands of participants. However, at this scal…
From Tokens to Concepts: Leveraging SAE for SPLADE
Yuxuan Zong, Mathias Vast, Basile Van Cooten +2
Learned Sparse IR models, such as SPLADE, offer an excellent efficiency-effectiveness tradeoff. However, they rely on the underlying backbone vocabulary, which might hinder perform…
Building Better Encoder-only Cross-Encoders: A Controlled Study of Training Strategies for Neural Re-ranking
Victor Morand, Mathias Vast, Basile Van Cooten +3
Cross-encoders fine-tuned from Transformer backbones remain the standard for second-stage re-ranking, and recent knowledge-distillation strategies have closed much of the gap with…
MICE: Minimal Interaction Cross-Encoders for efficient Re-ranking
Mathias Vast, Victor Morand, Josiane Mothe +1
In Information Retrieval (IR), cross-encoders deliver state-of-the-art ranking effectiveness but have a high inference cost, limiting their use to second-stage re-rankers. Prior wo…
The GDN-CC Dataset: Automatic Corpus Clarification for AI-enhanced Democratic Citizen Consultations
Pierre-Antoine Lequeu, Léo Labat, Laurène Cave +3
LLMs are ubiquitous in modern NLP, and while their applicability extends to texts produced for democratic activities such as online deliberations or large-scale citizen consultatio…
On the Representations of Entities in Auto-regressive Large Language Models
Victor Morand, Josiane Mothe, Benjamin Piwowarski
Named entities are fundamental building blocks of knowledge in text, grounding factual information and structuring relationships within language. Despite their importance, it remai…
ToMMeR -- Efficient Entity Mention Detection from Large Language Models
Victor Morand, Nadi Tomeh, Josiane Mothe +1
Identifying which text spans refer to entities - mention detection - is both foundational for information extraction and a known performance bottleneck. We introduce ToMMeR, a ligh…
Understanding Matching Mechanisms in Cross-Encoders
Mathias Vast, Basile Van Cooten, Laure Soulier +1
Neural IR architectures, particularly cross-encoders, are highly effective models whose internal mechanisms are mostly unknown. Most works trying to explain their behavior focused…