Hauff / Macdonald / Jannach Advances in Information Retrieval

47th European Conference on Information Retrieval, ECIR 2025, Lucca, Italy, April 6–10, 2025, Proceedings, Part I
Erscheinungsjahr 2025
ISBN: 978-3-031-88708-6
Verlag: Springer International Publishing
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 474 Seiten

Reihe: Lecture Notes in Computer Science

The five-volume set LNCS 15572, 15573, 15574, 15575 and 15576 constitutes the refereed conference proceedings of the 47th European Conference on Information Retrieval, ECIR 2025, held in Lucca, Italy, during April 6–10, 2025.

The 52 full papers, 11 findings, 42 short papers and 76 papers of other types presented in these proceedings were carefully reviewed and selected from 530 submissions. The accepted papers cover the state-of-the-art in information retrieval and recommender systems: user aspects, system and foundational aspects, artificial intelligence and machine learning, applications, evaluation, new social and technical challenges, and other topics of direct or indirect relevance to search and recommendation.

Zielgruppe

Research

Autoren/Hrsg.

Hauff, Claudia

Macdonald, Craig

Jannach, Dietmar

Kazai, Gabriella

Nardini, Franco Maria

Pinelli, Fabio

Silvestri, Fabrizio

Tonellotto, Nicola

Inhaltsverzeichnis

.- Crossing the Structure Chasm – Querying Data Without Limits.

.- Understanding the Interplay between LLMs’ Utilisation of Parametric and Contextual Knowledge.

.- Knowledge Graphs Are Dead, Long Live Knowledge Graphs.

.- LIBRA: Measuring Bias of Large Language Model from a Local Context.

.- Embedding Cultural Diversity in Prototype-based Recommender Systems.

.- Is Relevance Propagated from Retriever to Generator in RAG?.

.- Measuring Actual Privacy of Obfuscated Queries in Information Retrieval.

.- One size doesn’t fit all: Predicting the Number of Examples for In-Context Learning.

.- MURR: Model Updating with Regularized Replay for Searching a Document Stream.

.- Token Pruning Optimization for Efficient Dense Retrieval with Multi-Vector Representations.

.- Advancing Math Formula Search Using Diverse Structural and Symbolic Representations.

.- Ragnar¨ok: A Reusable RAG Framework and Baselines for TREC 2024 Retrieval-Augmented Generation Track.

.- Retrieve, Annotate, Evaluate, Repeat: Leveraging Multimodal LLMs for Large-Scale Product Retrieval Evaluation.

.- Graph Representation of Tables+Text and Compact Subgraph Retrieval for QA Tasks.

.- Higher Order Knowledge Graph Embeddings.

.- Improving the Re-Usability of Conversational Search Test Collections.

.- Repeat-bias-aware Optimization of Beyond-accuracy Metrics for Next Basket Recommendation.

.- Guiding Retrieval using LLM-based Listwise Rankers.

.- Lost but Not Only in the Middle: Positional Bias in Retrieval Augmented Generation.

.- Biased PromptORE: Enhancing Relation Extraction in Gendered Languages and Complex Texts – The Case of Spanish Documents from the XVI Century.

.- LSTM-based Selective Dense Text Retrieval Guided by Sparse Lexical Retrieval.

.- Context Example Selection For LLM Generated Relevance Assessments.

.- Enhancing FEVER-Style Claim Fact-Checking Against Wikipedia: A Diagnostic Taxonomy and Generative Framework.

.- Evaluating Auto-complete Ranking for Diversity and Relevance.

.- Semantically Proportioned nDCG for Explaining ColBERT’s Learning Process.

.- Opt-in Transparent Fairness for Recommender Systems.

.- Malevolence Attacks Against Pretrained Dialogue Models.

.- Zero-Shot and Efficient Clarification Need Prediction in Conversational Search.

.- Decoding the Hierarchy: A Hybrid Approach to Hierarchical Multi-Label Text Classification.

.- A Multi-modal Recipe for Improved Multi-domain Recommendation.

.- Towards Identity-Aware Cross-Modal Retrieval: a Dataset and a Baseline.

.- Corpus Subsampling: Estimating the Effectiveness of Neural Retrieval Models on Large Corpora.