ECI: Effective Contrastive Information to Evaluate Hard-Negatives

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Sinha, Aarush, Seetharaman, Rahul, Bansal, Aman
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914413444857856
author Sinha, Aarush
Seetharaman, Rahul
Bansal, Aman
author_facet Sinha, Aarush
Seetharaman, Rahul
Bansal, Aman
contents Hard negatives play a critical role in training and fine-tuning dense retrieval models, as they are semantically similar to positive documents yet non-relevant, and correctly distinguishing them is essential for improving retrieval accuracy. However, identifying effective hard negatives typically requires extensive ablation studies involving repeated fine-tuning with different negative sampling strategies and hyperparameters, resulting in substantial computational cost. In this paper, we introduce ECI: Effective Contrastive Information , a theoretically grounded metric grounded in Information Theory and Information Retrieval principles that enables practitioners to assess the quality of hard negatives prior to model fine-tuning. ECI evaluates negatives by optimizing the trade-off between Information Capacity the logarithmic bound on mutual information determined by set size and Discriminative Efficiency, a harmonic balance of Signal Magnitude (Hardness) and Safety (Max-Margin). Unlike heuristic approaches, ECI strictly penalizes unsafe, false-positive negatives prevalent in generative methods. We evaluate ECI across hard-negative sets mined or generated using BM25, cross-encoders, and large language models. Our results demonstrate that ECI accurately predicts downstream retrieval performance, identifying that hybrid strategies (BM25+Cross-Encoder) offer the optimal balance of volume and reliability, significantly reducing the need for costly end-to-end ablation studies.
format Preprint
id arxiv_https___arxiv_org_abs_2603_20990
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ECI: Effective Contrastive Information to Evaluate Hard-Negatives
Sinha, Aarush
Seetharaman, Rahul
Bansal, Aman
Information Retrieval
Artificial Intelligence
Hard negatives play a critical role in training and fine-tuning dense retrieval models, as they are semantically similar to positive documents yet non-relevant, and correctly distinguishing them is essential for improving retrieval accuracy. However, identifying effective hard negatives typically requires extensive ablation studies involving repeated fine-tuning with different negative sampling strategies and hyperparameters, resulting in substantial computational cost. In this paper, we introduce ECI: Effective Contrastive Information , a theoretically grounded metric grounded in Information Theory and Information Retrieval principles that enables practitioners to assess the quality of hard negatives prior to model fine-tuning. ECI evaluates negatives by optimizing the trade-off between Information Capacity the logarithmic bound on mutual information determined by set size and Discriminative Efficiency, a harmonic balance of Signal Magnitude (Hardness) and Safety (Max-Margin). Unlike heuristic approaches, ECI strictly penalizes unsafe, false-positive negatives prevalent in generative methods. We evaluate ECI across hard-negative sets mined or generated using BM25, cross-encoders, and large language models. Our results demonstrate that ECI accurately predicts downstream retrieval performance, identifying that hybrid strategies (BM25+Cross-Encoder) offer the optimal balance of volume and reliability, significantly reducing the need for costly end-to-end ablation studies.
title ECI: Effective Contrastive Information to Evaluate Hard-Negatives
topic Information Retrieval
Artificial Intelligence
url https://arxiv.org/abs/2603.20990