Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Jayaraman, Srideepika, Fokoue, Achille, Patel, Dhaval, Kalagnanam, Jayant
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2603.22294
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914414733557760
author Jayaraman, Srideepika
Fokoue, Achille
Patel, Dhaval
Kalagnanam, Jayant
author_facet Jayaraman, Srideepika
Fokoue, Achille
Patel, Dhaval
Kalagnanam, Jayant
contents Synthetic Data Generation (SDG), leveraging Large Language Models (LLMs), has recently been recognized and broadly adopted as an effective approach to improve the performance of smaller but more resource and compute efficient LLMs through fine-tuning. A key challenge in SDG is ensuring the quality and diversity of the generated data. In this paper, we analyze the diversity and distribution of generated data in the embedding space, and demonstrate a strong correlation between the density of examples within a specific neighborhood and the accuracy of predictions on examples drawn from that region. Building on this insight, we present a targeted pipeline for embedding-based sampling that enhances data diversity and consistently improves performance across several benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2603_22294
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Efficient Embedding-based Synthetic Data Generation for Complex Reasoning Tasks
Jayaraman, Srideepika
Fokoue, Achille
Patel, Dhaval
Kalagnanam, Jayant
Machine Learning
Artificial Intelligence
Synthetic Data Generation (SDG), leveraging Large Language Models (LLMs), has recently been recognized and broadly adopted as an effective approach to improve the performance of smaller but more resource and compute efficient LLMs through fine-tuning. A key challenge in SDG is ensuring the quality and diversity of the generated data. In this paper, we analyze the diversity and distribution of generated data in the embedding space, and demonstrate a strong correlation between the density of examples within a specific neighborhood and the accuracy of predictions on examples drawn from that region. Building on this insight, we present a targeted pipeline for embedding-based sampling that enhances data diversity and consistently improves performance across several benchmarks.
title Efficient Embedding-based Synthetic Data Generation for Complex Reasoning Tasks
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2603.22294