arXiv subjects

Marcel Gregoriadis

Publications and source records attributed to Marcel Gregoriadis.

6 recordsLinked to original sources

arXiv2026

Efficient and Robust Online Learning to Rank in Decentralized Systems

In Online Learning to Rank (OLTR), ranking models are trained directly from live user interactions, but existing systems rely on a trusted central server to collect and process these interactions. This leaves operators free to introduce biases that conflict with user interests. Decentralized learning offers an attractive alternative, allowing users to collaboratively train a shared ranking model by exchanging model updates directly with one another, without any central authority. In such settings, however, malicious nodes can send poisoned model updates that degrade the ranking quality of honest nodes. We introduce RankGuard, a decentralized OLTR framework in which users collaboratively train ranking models and exchange model updates directly with other nodes. RankGuard defends against poisoning attacks by carefully evaluating incoming models against the user's own private click history, corrected for position bias. An incoming model is only aggregated if it better explains the user's past interactions than the current local model, making it fundamentally hard for malicious nodes to craft updates that pass this test without also genuinely helping the user. We derive a theoretical convergence guarantee of RankGuard. To the best of our knowledge, this is the first formal convergence analysis of a decentralized OLTR algorithm. We evaluate RankGuard against four poisoning attacks, including a powerful adaptive attack, using four standard benchmarks and three click models. RankGuard outperforms all baselines in most settings while being up to 62x more efficient than its closest competitors.

cs.DC↗

arXiv2025

SwarmSearch: Decentralized Search Engine with Self-Funding Economy

Centralized search engines control what we see, read, believe, and vote. Consequently, they raise concerns over information control, censorship, and bias. Decentralized search engines offer a remedy to this problem, but their adoption has been hindered by their inferior quality and lack of a self-sustaining economic framework. We present SwarmSearch, a fully decentralized, AI-powered search engine with a self-funding architecture. Our system is designed for deployment within the decentralized file-sharing software Tribler. SwarmSearch integrates volunteer-based with profit-driven mechanisms to foster an implicit marketplace for resources. Employing the state-of-the-art of AI-based retrieval and relevance ranking, we also aim to close the quality gap between decentralized search and centralized alternatives. Our system demonstrates high retrieval accuracy while showing robustness in the presence of 50% adversarial nodes.

cs.DC↗

arXiv2025

A Large-Scale Web Search Dataset for Federated Online Learning to Rank

The centralized collection of search interaction logs for training ranking models raises significant privacy concerns. Federated Online Learning to Rank (FOLTR) offers a privacy-preserving alternative by enabling collaborative model training without sharing raw user data. However, benchmarks in FOLTR are largely based on random partitioning of classical learning-to-rank datasets, simulated user clicks, and the assumption of synchronous client participation. This oversimplifies real-world dynamics and undermines the realism of experimental results. We present AOL4FOLTR, a large-scale web search dataset with 2.6 million queries from 10,000 users. Our dataset addresses key limitations of existing benchmarks by including user identifiers, real click data, and query timestamps, enabling realistic user partitioning, behavior modeling, and asynchronous federated learning scenarios.

cs.IR↗

arXiv2024

A Thorough Investigation of Content-Defined Chunking Algorithms for Data Deduplication

Data deduplication emerged as a powerful solution for reducing storage and bandwidth costs in cloud settings by eliminating redundancies at the level of chunks. This has spurred the development of numerous Content-Defined Chunking (CDC) algorithms over the past two decades. Despite advancements, the current state-of-the-art remains obscure, as a thorough and impartial analysis and comparison is lacking. We conduct a rigorous theoretical analysis and impartial experimental comparison of several leading CDC algorithms. Using four realistic datasets, we evaluate these algorithms against four key metrics: throughput, deduplication ratio, average chunk size, and chunk-size variance. Our analyses, in many instances, extend the findings of their original publications by reporting new results and putting existing ones into context. Moreover, we highlight limitations that have previously gone unnoticed. Our findings provide valuable insights that inform the selection and optimization of CDC algorithms for practical applications in data deduplication.

cs.DC↗

arXiv2024

De-DSI: Decentralised Differentiable Search Index

This study introduces De-DSI, a novel framework that fuses large language models (LLMs) with genuine decentralization for information retrieval, particularly employing the differentiable search index (DSI) concept in a decentralized setting. Focused on efficiently connecting novel user queries with document identifiers without direct document access, De-DSI operates solely on query-docid pairs. To enhance scalability, an ensemble of DSI models is introduced, where the dataset is partitioned into smaller shards for individual model training. This approach not only maintains accuracy by reducing the number of data each model needs to handle but also facilitates scalability by aggregating outcomes from multiple models. This aggregation uses a beam search to identify top docids and applies a softmax function for score normalization, selecting documents with the highest scores for retrieval. The decentralized implementation demonstrates that retrieval success is comparable to centralized methods, with the added benefit of the possibility of distributing computational complexity across the network. This setup also allows for the retrieval of multimedia items through magnet links, eliminating the need for platforms or intermediaries.

cs.IR↗

arXiv2022

Analysis of Arbitrary Content on Blockchain-Based Systems using BigQuery

Blockchain-based systems have gained immense popularity as enablers of independent asset transfers and smart contract functionality. They have also, since as early as the first Bitcoin blocks, been used for storing arbitrary contents such as texts and images. On-chain data storage functionality is useful for a variety of legitimate use cases. It does, however, also pose a systematic risk. If abused, for example by posting illegal contents on a public blockchain, data storage functionality can lead to legal consequences for operators and users that need to store and distribute the blockchain, thereby threatening the operational availability of entire blockchain ecosystems. In this paper, we develop and apply a cloud-based approach for quickly discovering and classifying content on public blockchains. Our method can be adapted to different blockchain systems and offers insights into content-related usage patterns and potential cases of abuse. We apply our method on the two most prominent public blockchain systems - Bitcoin and Ethereum - and discuss our results. To the best of our knowledge, the presented study is the first to systematically analyze non-financial content stored on the Ethereum blockchain and the first to present a side-by-side comparison between different blockchains in terms of the quality and quantity of stored data.

cs.CR↗