Iadh Ounis
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Iadh Ounis has a total of 204 co-authors. Here are the top co-authors by number of shared publications:
Craig Macdonald
University of Glasgow, Glasgow, UK
112 shared publications- Craig Macdonald
University of Glasgow, Glasgow, United Kingdom
44 shared publications - Graham McDonald
University of Glasgow, Glasgow, United Kingdom
24 shared publications Ben He
University of Chinese Academy of Sciences
23 shared publications- Rodrygo L. T. Santos
Department of Computer Science, Universidade Federal de Minas Gerais
19 shared publications Richard McCreadie
University of Glasgow
18 shared publications
Latest Publications
- Open Access
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 3031-3042
Transforming User Actions into Audit Signals: Participatory Auditing to Improve Search Results
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 3031-3042
Open AccessTransforming User Actions into Audit Signals: Participatory Auditing to Improve Search Results
- Open Access
- 22 August 2026
- Published by Springer Nature in Journal of Intelligent Information Systems
- Vol. 64 (6) , 3209-3230
URecJPQ: memory-efficient multimodal recommendation models through RecJPQ in large-scale scenarios
Training state-of-the-art recommendation models on large-scale industrial datasets can be a challenging task due to the high number of users and items which are typically represented through ID embeddings. Such embeddings typically require a large amount of memory resources, which are not always available. This problem is further exacerbated in multimodal recommendation, in which multimodal item features generally improve recommendation performance, but require more resources to encode. In this paper, we introduce URecJPQ , a Joint Product Quantization method specifically designed for large-scale and multimodal top- k recommendation tasks, in which the vast number of users and items, combined with the available modalities, further increases the memory demands for the computation. The core idea is to represent both users and items not as a fully learned, unique embedding, but rather as a concatenation of shared learned sub-embeddings, thereby significantly reducing the total number of trainable parameters. Our experiments on three widely-used datasets across different domains (movies, baby and sports products) show that URecJPQ can be effectively applied to multimodal recommendation settings. In large scale scenarios, we observe a substantial reduction in checkpoint sizes and the number of trainable parameters (ranging from 86% to 98%, and 98% to 99%, respectively), with only a marginal decrease in accuracy (8.5% on recall and 16% on NDCG, on average), and, in some cases, even performance improvements (up to 85%), as in the baby products domain. Our codebase is available at https://github.com/giuspillo/urecjpq .
- 22 August 2026
- Published by Springer Nature in Journal of Intelligent Information Systems
- Vol. 64 (6) , 3209-3230
Open AccessURecJPQ: memory-efficient multimodal recommendation models through RecJPQ in large-scale scenarios
Training state-of-the-art recommendation models on large-scale industrial datasets can be a challenging task due to the high number of users and items which are typically represented through ID embeddings. Such embeddings typically require a large amount of memory resources, which are not always available. This problem is further exacerbated in multimodal recommendation, in which multimodal item features generally improve recommendation performance, but require more resources to encode. In this paper, we introduce URecJPQ , a Joint Product Quantization method specifically designed for large-scale and multimodal top- k recommendation tasks, in which the vast number of users and items, combined with the available modalities, further increases the memory demands for the computation. The core idea is to represent both users and items not as a fully learned, unique embedding, but rather as a concatenation of shared learned sub-embeddings, thereby significantly reducing the total number of trainable parameters. Our experiments on three widely-used datasets across different domains (movies, baby and sports products) show that URecJPQ can be effectively applied to multimodal recommendation settings. In large scale scenarios, we observe a substantial reduction in checkpoint sizes and the number of trainable parameters (ranging from 86% to 98%, and 98% to 99%, respectively), with only a marginal decrease in accuracy (8.5% on recall and 16% on NDCG, on average), and, in some cases, even performance improvements (up to 85%), as in the baby products domain. Our codebase is available at https://github.com/giuspillo/urecjpq .
- Open Access
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 3694-3699
APR: Adaptive Personalised Reranking For Conversational Search
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 3694-3699
Open AccessAPR: Adaptive Personalised Reranking For Conversational Search
- Open Access
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 3952-3957
Auditing Query Drift: Do Users Actually Benefit from Pseudo-Relevance Feedback?
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 3952-3957
Open AccessAuditing Query Drift: Do Users Actually Benefit from Pseudo-Relevance Feedback?
- Open Access
- 26 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Information Systems
- Vol. 44 (6) , 1-9
Introduction to the Special Issue on Query Performance Prediction
Query Performance Prediction (QPP) aims to estimate the effectiveness of a retrieval system for a given query without requiring relevance judgments. While traditionally studied for sparse retrieval, recent advances in neural ranking, dense retrieval, and large language models (LLMs) have prompted a shift towards QPP methods that better reflect modern IR systems. This Special Issue on Query Performance Prediction Towards Novel Information Retrieval Paradigms presents recent advances along two complementary directions. The first focuses on LLM- and representation-based approaches that leverage query variants, semantic interactions, and comparative assessment to improve prediction for neural retrieval models. The second explores new formulations and application settings, including inverse learning, prediction of reliability in retrieval-augmented generation, and the incorporation of multimodal and neurophysiological signals. These works highlight emerging challenges in evaluating QPP in neural and generative settings by demonstrating that QPP is evolving into a broader framework for estimating uncertainty and reliability across complex IR pipelines. We believe that this Special Issue will foster further research towards robust and generalizable QPP methods for next-generation information access systems.
- 26 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Information Systems
- Vol. 44 (6) , 1-9
Open AccessIntroduction to the Special Issue on Query Performance Prediction
Query Performance Prediction (QPP) aims to estimate the effectiveness of a retrieval system for a given query without requiring relevance judgments. While traditionally studied for sparse retrieval, recent advances in neural ranking, dense retrieval, and large language models (LLMs) have prompted a shift towards QPP methods that better reflect modern IR systems. This Special Issue on Query Performance Prediction Towards Novel Information Retrieval Paradigms presents recent advances along two complementary directions. The first focuses on LLM- and representation-based approaches that leverage query variants, semantic interactions, and comparative assessment to improve prediction for neural retrieval models. The second explores new formulations and application settings, including inverse learning, prediction of reliability in retrieval-augmented generation, and the incorporation of multimodal and neurophysiological signals. These works highlight emerging challenges in evaluating QPP in neural and generative settings by demonstrating that QPP is evolving into a broader framework for estimating uncertainty and reliability across complex IR pipelines. We believe that this Special Issue will foster further research towards robust and generalizable QPP methods for next-generation information access systems.
- Open Access
- 25 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 6180-6202
All Eyes on the Ranker: Participatory Auditing to Surface Blind Spots in Ranked Search Results
- 25 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 6180-6202
Open AccessAll Eyes on the Ranker: Participatory Auditing to Surface Blind Spots in Ranked Search Results
- 25 March 2026
- Published by Springer Nature
- p. 566-576
Knowledge-Enhanced Multi-agent for LLM-Based Recommendation
- 25 March 2026
- Published by Springer Nature
- p. 566-576
Knowledge-Enhanced Multi-agent for LLM-Based Recommendation
- 25 March 2026
- Published by Springer Nature
- p. 565-581
LURE-RAG: Lightweight Utility-Driven Reranking for Efficient RAG
- 25 March 2026
- Published by Springer Nature
- p. 565-581
LURE-RAG: Lightweight Utility-Driven Reranking for Efficient RAG
- 24 March 2026
- Published by Springer Nature
- p. 115-130
Temporal Conflicts in LLMs: Reproducibility Insights from Unifying DYNAMICQA and MULAN
- 24 March 2026
- Published by Springer Nature
- p. 115-130
Temporal Conflicts in LLMs: Reproducibility Insights from Unifying DYNAMICQA and MULAN
- 24 March 2026
- Published by Springer Nature
- p. 98-114
RecRankerEval: A Reproducible Framework for Deploying and Evaluating LLM-Based Top-k Recommenders
- 24 March 2026
- Published by Springer Nature
- p. 98-114
RecRankerEval: A Reproducible Framework for Deploying and Evaluating LLM-Based Top-k Recommenders