Marco Favier, Noah Daniëls, Jens Leysen et al.
Item-item similarity is one of the oldest and most foundational signals in recommender systems: two items are assumed to be similar when the same users consume them.Item-item similarity is one of the oldest and most foundational signals in recommender systems: two items are assumed to be similar when the same users consume them.
Guangwei Xu, Wenkai Zhang, Yawen Liu et al.
Navigational search aims to find a single target page among hundreds of millions of web pages. Click tables work well for queries seen before but cannot generalise to new…Navigational search aims to find a single target page among hundreds of millions of web pages. Click tables work well for queries seen before but cannot generalise to new formulations, while embedding retrievers generalise across…
Sébastien Gracia
Official texts are corrected after publication: EU corrigenda, verified RFC errata, retractions. We measured whether these corrections are present in four public datasets…Official texts are corrected after publication: EU corrigenda, verified RFC errata, retractions. We measured whether these corrections are present in four public datasets used to train language models, and what models answer about…
Qinsi Wang, Wei Wen, Chong Zhou et al.
People have long dreamed of an assistant that is always there to guide them and remind what they forget, from new dishes to lost keys.People have long dreamed of an assistant that is always there to guide them and remind what they forget, from new dishes to lost keys.
Lychee was 9.7 to 117 times faster than the baseline IVF-Flat at equal recall with 200 probed lists on datasets of up to 100 million vectors.Lychee was 9.7 to 117 times faster than the baseline IVF-Flat at equal recall with 200 probed lists on datasets of up to 100 million vectors.
Aaditya Chauhan, Arun Agarwal
Hybrid retrieval, which fuses a sparse lexical retriever with a dense neural one, is widely recommended in the software documentation surveyed in September 2026, usually…Hybrid retrieval, which fuses a sparse lexical retriever with a dense neural one, is widely recommended in the software documentation surveyed in September 2026, usually with reciprocal rank fusion (RRF), on the strength of…
Vihan Lakshman, Franco Maria Nardini, Cosimo Rulli et al.
Learned sparse retrieval (LSR) has recently gained traction in the information retrieval community for the ability to combine the generalization capabilities of neural…Learned sparse retrieval (LSR) has recently gained traction in the information retrieval community for the ability to combine the generalization capabilities of neural representations with the interpretability of keyword-based…
Ju-Chun Ko
"Abliterated" ("uncensored") builds of open-weight language models remove the activation direction that mediates refusal, and are popular partly on the assumption that a…"Abliterated" ("uncensored") builds of open-weight language models remove the activation direction that mediates refusal, and are popular partly on the assumption that a model that no longer refuses is free of political…
Saber Zerhoudi, Jelena Mitrovic, Michael Granitzer
User simulators promise to evaluate conversational search systems at a scale human studies cannot reach, but only if their behavior is realistic enough to trust.User simulators promise to evaluate conversational search systems at a scale human studies cannot reach, but only if their behavior is realistic enough to trust.
Liang Yue, Eric A. Vance
What the Web community believes about toxic conversations, such as that toxicity rises in longer and deeper threads, was measured mostly with Google's Perspective API…What the Web community believes about toxic conversations, such as that toxicity rises in longer and deeper threads, was measured mostly with Google's Perspective API, which is shutting down; large language models (LLMs) will make…
Daniel Domínguez Figaredo, Rafael Fernández De la Cruz
It is assumed that a natural way to improve the pedagogical quality of large language model tutors is to define a metric of instructional performance and fine-tune the…It is assumed that a natural way to improve the pedagogical quality of large language model tutors is to define a metric of instructional performance and fine-tune the model against it.
Yujiao Chen
Language-model agents favor their own group because they have watched their members favor each other. The group label alone does little once the decision has a cost; what…Language-model agents favor their own group because they have watched their members favor each other. The group label alone does little once the decision has a cost; what drives favoritism is observed behavior, and an individual's…
Zhuchenyang Liu, Yao Zhang, Yu Xiao
Prevailing multi-vector visual document retrievers store each page as about a thousand patch vectors, often in vector databases run by a third party.Prevailing multi-vector visual document retrievers store each page as about a thousand patch vectors, often in vector databases run by a third party.
Erica Chiang, Kenny Peng, Rebecca Lichtenstein et al.
Algorithmic recommendations can help participants navigate large matching markets. For example, recommendations for school and college choices may reduce information…Algorithmic recommendations can help participants navigate large matching markets. For example, recommendations for school and college choices may reduce information frictions and disparities in access to high-performing programs.
Wikum A, Mishra N, Saberi A et al.
Efficient vector retrieval requires both a corpus geometry that supports retrieving the right documents through vector similarity, and a query encoder that can embed…Efficient vector retrieval requires both a corpus geometry that supports retrieving the right documents through vector similarity, and a query encoder that can embed queries near their desired documents in the embedding space.
Ye R, Fan J, Bernuy AZ et al.
Large language models can generate plausible solutions to programming assignments, making it tempting to detect their use by matching student code against a reference…Large language models can generate plausible solutions to programming assignments, making it tempting to detect their use by matching student code against a reference bank of generated solutions.