SEARCH AND RECOMMENDATION
Human-centric, multimodal strategies and modules built on accountable algorithms for transparency and trust
Highlights
- 3 short papers & 1 resource paper accepted at SIGIR 2023
- 4 long papers & 1 short accepted at ECIR 2023
- An Intro to Neural Models IR by Stephane Clinchant given in May 22 at the Univ. of Padova.
- 4 papers at SIGIR 2022
- New SPLADE model released!
- 2 papers at ECIR 2022
- 1 paper at WSDM 2022
- 2 papers at SIGIR 2021
- Best Short paper Award at ECIR 2021 for our paper ‘A white box analysis of ColBERT‘
Related Content
The ubiquity of sensors and the apparition of new user interfaces has profoundly changed the nature of search and recommendation engines and how we interact with them.
Our research is carried out in collaboration with the NAVER search group, responsible for the world's 5th biggest search engine.
This provides research opportunities that go far beyond the traditional information retrieval and recommendation frameworks.
We have become more reluctant to express our needs or preferences in an explicit manner as previously done through primarily text-based interfaces. The way we give feedback to search and recommendation (S&R) systems is noisier, more biased and more heterogeneous. Users, service providers and regulators have greater requirements whereby they want more accountable, transparent and ethical technologies that respect the privacy-utility trade-off, whilst also offering some kind of guarantee on the fairness of the results they provide.
This change brings new exciting research opportunities that our research group enjoys initiating and experimenting with in the real world.
We firmly believe that search and recommendation processes must be considered as multi-stakeholder, multi-round games (continuous learning loops), where the engine and the user interplay collaboratively with the different types of feedback they each provide. Facilitating communication between the system and the user naturally leads to the design of multimodal strategies where text in any language, images, sound and speech can be consistently handled together. With the same communication objective in mind, and in particular to establish mutual confidence, we focus on developing unbiased ranking algorithms that can explain the advice they provide in a transparent and understandable form, while preserving fairness and diversity. The technology we use to pursue our research relies on Deep-IR/Deep-matching architectures, context-aware sequential recommendation models, causal and counter-factual learning-to-rank algorithms, conversational S&R and reinforcement learning.
Some Recent Publications
- An efficiency study for SPLADE models, Carlos Lassance, Stephane Clinchant, SIGIR, 10-15 July, 2022
- Pareto-optimal fairness-utility amortizations in rankings with a DBN exposure model, Till Kletti, Jean-Michel Renders and Patrick Loiseau, SIGIR, 10-15 July, 2022
- Learned token pruning in contextualized late interaction over BERT (ColBERT), Carlos Lassance, Maroua Maachou, Joohee Park and Stephane Clinchant, SIGIR, 10-15 July, 2022
- From distillation to hard negative sampling: making sparse neural IR models more effective, Thibault Formal, Carlos Lassance, Benjamin Piwowarski and Stephane Clinchant, SIGIR, 10-15 July, 2022
- Match your words! A study of lexical matching in neural information retrieval, Thibault Formal, Benjamin Piwowarski, Stéphane Clinchant, ECIR, 10-14 April, 2022
- Joint personalized search and recommendation with hypergraph convolutional networks, Thibaut Thonet, Jean-Michel Renders, Mario Choi, Jinho Kim, ECIR, 10-14 April, 2022
- Introducing the Expohedron for efficient pareto-optimal-fairness utility amortizations in repeated rankings, Till Kletti, Jean-Michel Renders, Patrick Loiseau, WSDM 21-25 February, 2022.
- SPLADE: Spare Lexical and Expansion Model for First Stage Ranking, Thibault Formal, Stéphane Clinchant and Benjamin Piwowarski, SIGIR, 11-15 July, 2021
- A white box analysis of ColBERT, Thibault Formal, Benjamin Piwowarski, Stéphane Clinchant, ECIR, 28 Mar – 1 April, 2021. Best short paper award.
- Adaptive pointwise-pairwise learning-to-rank for content-based personalized recommendation, Yagmur Gizem Cinar, Jean-Michel Renders | ACM Conference on Recommender Systems (RecSys), 22-26 September, 2020
- Multi-grouping robust fair ranking, Thibaut Thonet, Jean-Michel Renders | International ACM SIGIR Conference on Research and Development in Information Retrieval, 25-30 July, 2020
- Designing ambient wanderer: mobile recommendations for urban exploration, Sruthi Viswanathan, Bernard Omidvar-Tehrani, Adrien Bruyat, Frédéric Roulland, Antonietta Grasso | ACM conference on Designing Interactive Systems (DIS), 6-11 July, 2020
- Learning to rank images with cross-modal graph convolutions, Thibault Formal, Stéphane Clinchant, Jean-Michel Renders, Sooyeol Lee, Geun Hee Cho | European Conference on Information Retrieval (ECIR), 14-17 April, 2020
- Seed-guided deep document clustering, Maziar Moradi Fard, Thibaut Thonet, Eric Gaussier | European Conference on Information Retrieval (ECIR), 14-17 April, 2020
- SAGE: interactive state-aware point-of-interest recommendation, Bernard Omidvar-Tehrani, Sruthi Viswanathan, Frédéric Roulland, Jean-Michel Renders | Workshop of State-based User Modelling, co-located with ACM International Conference on Web Search and Data Mining (WSDM) Conference, 4-7 February, 2020