Dense bi-encoders and late-interaction retrievers are usually treated as alternatives, yet they fail on different queries: one vector captures what a text is about, token-level matching finds its exact entities and phrases. Using both today means two models, two indexes and a score-fusion step. We present UIBK-Embeddin...
Abdelrahman Abdallah, Mohammed Ali, Adam Jatowt· 0 citations
Reasoning-intensive retrieval remains difficult for small models. Compact public ColBERTs are usually trained on general-purpose corpora and underperform reasoning-tuned 150M+ baselines on BRIGHT~\cite{bright} by several nDCG@10 points. However, no public reasoning-tuned ColBERT exists at edge scale. We introduce \text...
Abdelrahman Abdallah, Mohammed Ali, Adam Jatowt· 0 citations
Across six collections and three backbones, EXCISE is the strongest system in all eighteen backbone-collection cells against that backbone's own frozen and fine-tuned baselines, and outperforms every fine-tuned cross-encoder, each of which loses no-harm nDCG@10.
Mohammed Ali, Abdelrahman Abdallah, Adam Jatowt· 0 citations
A lightweight inference-time correction that estimates and removes a popularity prior from model predictions is proposed, which requires no fine-tuning, is label-free at test time, and adds negligible computational cost.
Abdelrahman Abdallah, Mohammed Ali, Bhawna Piryani et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
