qCLEF

Go to Lab page (external)

Welcome to QuantumCLEF (qCLEF), an innovative evaluation lab at the intersection of Quantum Computing (QC), Information Retrieval (IR), and Recommender Systems (RS).

In today’s data-driven world, IR and RS systems are challenged by the explosive growth of data and the need for computationally intensive algorithms. Quantum Computing offers new possibilities to address these demands, and while Quantum Computing has already seen applications in various domains, its potential remains largely untapped within IR and RS. The emerging field of Quantum IR explores quantum mechanics principles to model IR problems but has yet to explore the practical implementation of IR and RS systems using quantum technologies.

At QuantumCLEF, we focus on Quantum Annealing (QA), a Quantum Computing approach that uses specialized devices to efficiently solve complex optimization problems by leveraging quantum-mechanical effects. Our mission is to explore whether Quantum Annealing can improve the efficiency and effectiveness of IR and RS systems.

QuantumCLEF aims to:

  • Benchmark the performance of Quantum Annealing against traditional approaches in IR and RS;

  • Identify novel formulations for IR and RS algorithms that can leverage Quantum Annealing;

  • Foster a research community dedicated to applying Quantum Computing technologies in IR and RS.

  • Quantum Annealing is accessible to researchers with or without a background in quantum physics, thanks to user-friendly tools and libraries designed for this paradigm. We invite you to join us in advancing the field and unlocking new capabilities in IR and RS through Quantum Computing."

Tasks

The qCLEF Lab features three different tasks:

Feature Selection
Apply quantum annealers to find the most relevant subset of features to train a learning model, e.g., for ranking. This problem is very impactful, since many IR and RS systems involve the optimization of learning models, and reducing the dimensionality of the input data can improve their performance.
Instance Selection
Apply quantum annealers to identify the most representative subset of instances to train a learning model, specifically for text classification tasks. This problem is critical as it addresses the high computational costs of fine-tuning large language models (LLMs) like Llama3.1, while ensuring that their effectiveness remains comparable to training on the entire dataset.
Clustering
Use quantum annealers to cluster different documents in the form of embeddings to ease the browsing process of large collections. Clustering can be helpful for organizing large collections, helping users to explore a collection and providing similar search results to a given query.

Organizers

  • Andrea Pasin (University of Padua, Italy)
  • Maurizio Ferrari Dacrema (Politecnico di Milano, Italy)
  • Paolo Cremonesi (Politecnico di Milano, Italy)
  • Washington Cunha (Federal University of Minas Gerais, Brazil)
  • Marcos André Goncalves (Federal University of Minas Gerais, Brazil)
  • Nicola Ferro (University of Padua, Italy)