Stéphane Clinchant

[Eiso Kant @eisokant]

https://t.co/13VRNrqv45

[Sumit @_reachsumit]

Inference-Free Multimodal Learned Sparse Retrieval for Production-Scale Visual Document Search

Naver introduces a sparse retriever that directly indexes visual documents and serves text queries without neural query encoding.

📝 https://t.co/BNXWR5GH7F
👨🏽‍💻 https://t.co/DBC82Fwv2Q

[Amélie Chatelain @AmelieTabatta]

I told you it would be a stacked line-up!
@N1colAIs @thibault_formal and @antoine_chaffin talking tomorrow at the Search Meetup™️ (along with others whose twitter @ I don't have)
Join us for nice presentations and chats ❤️

[Stéphane Clinchant @sclincha]

@_reachsumit @santoshradha FYI, I use the q-log on the term frequencies in 2012 : https://t.co/kmAmjsuZSz

[Databricks @databricks]

Databricks is proud to be a Founding Gold Sponsor of @TheOfficialACM Conference on AI and Agentic Systems—the first ACM conference dedicated to compound AI and agentic systems, with our co-founder @matei_zaharia on the organizing committee.

Join us May 26–29 in San Jose for the premier event for rigorous, reproducible research in compound AI architectures, optimization, and deployment.

Register today: https://t.co/Y0b2NhQjWv

[Simon Lupart @simon_lupart]

Most code search systems rely on dense embeddings. In this work, we release SPLADE-Code, learned sparse retrieval models for code retrieval, with strong generalization, high interpretability, compatibility with inverted indexes, and working across 20+ programming languages.

[Thibault Formal @thibault_formal]

New sparse retrieval model: introducing SPLARE, which extends SPLADE by replacing the vocabulary head with pretrained SAEs!
paper: https://t.co/Un2zhX14KR (ICLR'26)
also how we won the WSDM'26 Cup on multilingual retrieval: https://t.co/77QlgZsnls (model weights coming soon!)

[Xin Eric Wang @xwang_lk]

𝘏𝘶𝘮𝘢𝘯𝘴 𝘵𝘩𝘪𝘯𝘬 𝘧𝘭𝘶𝘪𝘥𝘭𝘺—𝘯𝘢𝘷𝘪𝘨𝘢𝘵𝘪𝘯𝘨 𝘢𝘣𝘴𝘵𝘳𝘢𝘤𝘵 𝘤𝘰𝘯𝘤𝘦𝘱𝘵𝘴 𝘦𝘧𝘧𝘰𝘳𝘵𝘭𝘦𝘴𝘴𝘭𝘺, 𝘧𝘳𝘦𝘦 𝘧𝘳𝘰𝘮 𝘳𝘪𝘨𝘪𝘥 𝘭𝘪𝘯𝘨𝘶𝘪𝘴𝘵𝘪𝘤 𝘣𝘰𝘶𝘯𝘥𝘢𝘳𝘪𝘦𝘴. But current reasoning models remain constrained by discrete tokens, limiting their full potential.

Introducing 𝐒𝐨𝐟𝐭 𝐓𝐡𝐢𝐧𝐤𝐢𝐧𝐠: a training-free method that mimics human-like “soft” reasoning by generating continuous, abstract concept tokens. These tokens smoothly blend multiple meanings through probability-weighted mixtures of embeddings, enabling richer representations and seamless exploration of diverse reasoning paths.

𝐓𝐡𝐞 𝐢𝐦𝐩𝐚𝐜𝐭?
✅ Improved accuracy on math & code benchmarks by up to 2.48% (pass@1).
✅ Reduced token usage by up to 22.4%, making reasoning models both smarter and more efficient.

[Nadia Chirkova @nadiinchi]

Arrived in Singapore for #ICLR2025 and will be presenting PROVENCE on Friday, Poster session 3 at 10am, poster #255!

Blogpost: https://t.co/Q7TRYP04ZV

Will be happy to meet & chat about #LLMs, #RAG, #InformationRetrieval and #MultilingualNLP :)

#NLProc @naverlabseurope

[Vaibhav (VB) Srivastav @reach_vb]

AllenAI COOKED, Llama 3.1 Tulu 405B beats DeepSeek V3 - all whilst being 40% SMALLER! 🔥

Fully open model weights, data and training pipeline 🤗

[Stéphane Clinchant @sclincha]

We welcome contributors to add datasets, metrics, and other tasks to BERGEN. Join us! 🤝
#RAG #LLMs

[Stéphane Clinchant @sclincha]

What’s a good baseline for RAG? 🤔

The literature shows consistent differences in experimental setups, retrievers, datasets, and metrics. So, we built the BERGEN library https://t.co/9srOoFQNQ5 to enhance reproducibility and identify strong baselines : 🧵

@naverlabseurope

[Stéphane Clinchant @sclincha]

Our recommendations are detailed in our first Arxiv paper, with additional findings on multilingual RAG in our second paper.
https://t.co/3MkkSraIcd
https://t.co/qk6jp9GQvF

[Stéphane Clinchant @sclincha]

😀We're looking for a talented researcher to join our team at Naver Labs Europe (@naverlabseurope) , working on LLMs and Retrieval!😃
Please apply here: https://t.co/0lea7ABHld !

[Stéphane Clinchant @sclincha]

@jerryjliu0 @rpradeep42 If efficiency matters, a simpler solution is to actually use a state of the art reranker (cf our study comparing LLMs and cross-encoders) https://t.co/SOpZOgQIFy

[Stéphane Clinchant @sclincha]

... especially when reviewers said ‘dense retrieval on its own has shown to surpass sparse retrieval considerably ‘ and that our ‘approach is quite incremental’ 2/2

[Stéphane Clinchant @sclincha]

It feels good when someone from a big company shares that they saw ‘ pretty promising results in terms of quality and space savings [for SPLADE] compared to dense embedding models’ ... 1/2

[Stéphane Clinchant @sclincha]

@andysingal @thibault_formal @naverlabseurope https://t.co/woBcjVkDQn

[Laure Soulier @LaureSoulier]

What a great pleasure and honor to share this session about generative AI, ethics, bias, and politics with 3 passionate speakers @plimantour, Andrew Wyckoff, and Juha Heikkilä.

Thanks @AI2S2Symposium for the invitation. See you in Geneva on Monday!