Author: Criteo Tech
-
๐R&D Community Awards 2025
Our annual ceremony, where we celebrate achievements together. For the third consecutive year, we gathered to celebrate our achievements within the R&D community and recognize the contributions of all our R&D Ambassadors. At Criteo, expressing gratitude to our teammates is a consistent practice throughout the entire company. In our R&D department, our expert community embraces…
-
Off-Policy Learning in Large Action Spaces
Optimization Matters More Than Estimation This Research Card shows that in large-scale recommendation systems, like those used at Criteo, focusing on how we optimize algorithms is more impactful than trying to perfectly estimate outcomes. By using simpler, more stable training objectives, we can achieve better click-through rates, faster experimentation, and more scalable performance across millions…
-
From Socca to Software: RivieraDev 2025 in Review
A sun-soaked celebration of tech, talent, and taste in Sophia Antipolis Set against the sun-drenched backdrop of Sophia Antipolis, RivieraDev 2025 once again proved why itโs one of the most beloved developer conferences in the region. Held in July at the SKEMA Business School, the event brought together around 700 attendees for three days of…
-
Inside Devoxx France 2025: Engineering at Scale with Criteo
A look at our talks on Kubernetes, Aerospike, CI/CD, and frontend reactivity. One more year, we participated in a new Devoxx France edition in Paris last April. This conference is an important milestone every year, and we are always excited to support the tech community in Paris and across France. This year, as proud Gold…
-
Pareto-Optimality, Smoothness, and Stochasticity in Learning-Augmented One-Max-Search
A rule for choosing when to sell/buy an asset when facing a sequence ofย prices This Research Card explains a rule for choosing when to sell/buy an asset when facing a sequence of prices. The key feature is that the algorithm has access to โpredictionsโ on the optimal price. The objectives are to take โsafeโ decisions,…
-
Improving Consistency Models with Generator-Augmented Flows
Delivering more personalized ads in real-timeโโโboosting engagement and cutting infrastructure costs. This Research Card presents Generator-Augmented Flows (GC), a technique improving consistency models, which are state-of-the-art models for fast image and video generation. For Criteo, this could mean delivering more personalized ads in real timeโโโboosting engagement and cutting infrastructure costs. Title: Improving Consistency Models with…
-
Criteo R&D Hits the Stage at Voxxed Days Bucharest 2025
What we learned, shared, and loved at this yearโs conference. Authors: Benjamin Degerbaix Diana Carmen Cibu Alex Morus Alexandru Tudor Andreea Dragomirescu Catalin Stan George Patrascu Mihai Iacob Nicu Murgu Ovidiu Badea Rareศ Nicolaide Sergiu Petrescu Voxxed Days Bucharest 2025 brought together an exceptional mix of industry professionals, passionate developers, and thought leaders in a…
-
Bringing NeRFs to the Latent Space: Inverse Graphics Autoencoder
A novel IG-AE that creates a 3D-aware latent space compatible with Neural Radianceย Fields This Research Card introduces a novel Inverse Graphics Autoencoder (IG-AE) that creates a 3D-aware latent space compatible with Neural Radiance Fields (NeRFs), enabling efficient 3D object reconstruction from 2D images and facilitating high-quality 3D ads. Title: Bringing NeRFs to the Latent Space:…
-
Differentially Private Gradient Flow based on the Sliced Wasserstein Distance
Novel differentially private model using gradient flows defined on an optimal transport metric. This Research Card introduces a novel, theoretically grounded method for differentially private generative modeling by leveraging a smooth mathematical process, achieving high-fidelity data generation with strong privacy guarantees and lower computational costs compared to traditional approaches. Title: Differentially Private Gradient Flow based…
-
All our contributions at NeurIPS 2024
This year, our research team got a record of 15 papers accepted at the major ML conference. The 38th annual edition of the prestigious international conference NeurIPS 2024 (Neural Information Processing Systems) took place in Vancouver. ๐ฅ๐ฒ๐๐ฒ๐ฎ๐ฟ๐ฐ๐ต๐ฒ๐ฟ๐ ๐ณ๐ฟ๐ผ๐บ ๐๐ต๐ฒ ๐๐ฟ๐ถ๐๐ฒ๐ผ ๐๐ ๐๐ฎ๐ฏ were there to present their latest ML papers (๐ญ๐ฏ ๐ฝ๐ฎ๐ฝ๐ฒ๐ฟ๐โโโ๐ฎ๐บ๐ผ๐ป๐ด ๐๐ต๐ถ๐ฐ๐ต ๐ผ๐ป๐ฒ ๐๐ฝ๐ผ๐๐น๐ถ๐ด๐ต๐โโโ,…