Work

Unveiling DIME: Reproducibility, Generalizability, and Formal Analysis of Dimension Importance Estimation for Dense Retrieval

Year: 2025

Type: conference-paper

Abstract: Dimension IMportance Estimation (DIME) is a recently proposed technique to enhance ranking effectiveness of dense retrieval models by pruning irrelevant embedding dimensions through Pseudo Relevance Feedback (PRF DIME) or exploiting dense representations of Large Language Model-generated answers (LLM DIME). Despite strong empirical performance, its theoretical foundations and generalizability remain open questions.

Source: International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR)

DOI: 10.1145/3726302.3730318

Authors: Cesare Campagnano, Antonio Mallia, Fabrizio Silvestri

Institution: Sapienza University of Rome

Language: English

Topic: Information Retrieval and Search Behavior

Subfield: Information Systems

Field: Computer Science

Domain: Physical Sciences

Open Access status: bronze

Funders: European Commission, NextGenerationEU

Awards: G2993132335, G507880695