Saved in:
Bibliographic Details
Main Authors: Lu, Rong, Liu, Hao, Hou, Song
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.04997
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • This work presents a comparative analysis of embedding-based and generative models for classifying geoscience technical documents. Using a multi-disciplinary benchmark dataset, we evaluated the trade-offs between model accuracy, stability, and computational cost. We find that generative Vision-Language Models (VLMs) like Qwen2.5-VL, enhanced with Chain-of-Thought (CoT) prompting, achieve superior zero-shot accuracy (82%) compared to state-of-the-art multimodal embedding models like QQMM (63%). We also demonstrate that while supervised fine-tuning (SFT) can improve VLM performance, it is sensitive to training data imbalance.