Our work "Distilled Neural Networks for Efficient Learning to Rank" has been accepted at IEEE TKDE. 🥳🎉🥳 Joint work with Cosimo Rulli, Salvatore Trani, Divby! to appear!

SPIRE 2023 will be held in Pisa (Italy) on September 26-28, 2023. Submissions due: May 19 (AoE). More info: spire2023.isti.cnr.it

Today, Cosimo Rulli successfully defended his Ph.D. thesis: "Efficiency-Effectiveness Trade-offs in Neural Network Compression." Congrats Dr. cosimorulli !!!

We are happy to announce that our paper "Efficient Multi-Vector Dense Retrieval with Bit Vectors" was accepted as full-paper at #ecir2024!! 🏴󠁧󠁢󠁳󠁣󠁴󠁿 Franco Maria Nardini Rossano Venturini

From 2.5x (SPLADE) to 11.3x (ESPLADE) faster than the winning methods of the Big-ANN '23 Sparse Track. "Efficient Inverted Indexes for Approximate Retrieval over Learned Sparse Representations". LP @ #SIGIR2024. Joint work with @snbruch, cosimorulli, Rossano Venturini

Interested in speeding up PLAID by up to 3x while reducing the memory by up to 2x with no loss in retrieval accuracy? Come see our "Efficient Multi-vector Dense Retrieval with Bit Vectors" presented by cosimorulli. 13:00, Neural IR. Joint work with Rossano Venturini. #ECIR2024

Omar Khattab @RossanoVent thanks Omar Khattab! Code and data of EMVB available on GitHub: github.com/CosimoRulli/em… cosimorulli

Interested in speeding up approximate retrieval over learnt sparse embeddings of up to 12x w.r.t. the winning methods of the Big ANN 2023 Sparse Track challenge? Check this out: arxiv.org/abs/2404.18812. Long paper @ #SIGIR2024 with @snbruch, cosimorulli, Rossano Venturini

The Rust code (plus an easy-to-use search Python API) of "Efficient Inverted Indexes for Approximate Retrieval over Learned Sparse Representations" (ACM SIGIR 2024) is now released on GitHub. @snbruch, cosimorulli, Rossano Venturini github.com/TusKANNy/seism…

All EasyChair tracks of #ECIR2025 are now open and awaiting your contributions! Lucca is waiting for you! The submission link is in the calls. For more details see: ecir2025.eu