mlsanityData to 7 Oct 2026

Evaluation of Embedding-Based and Generative Methods for LLM-Driven Document Classification: Opportunities and Challenges

arXiv 2604.04997 ↗First posted 5 Apr 2026cs.IR cs.AI cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 8 of its 10 references that are in the map. First posted on arXiv 5 Apr 2026.

Builds on

8 of 10 references.

References · 8

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  3. 2025Qwen2.5-VL Technical Reportcited by 5,418
  4. 2025Gemma 3 Technical Reportcited by 1,686
  5. 2024GME: Improving Universal Multimodal Retrieval by Multimodal LLMscited by 166
  6. 2017Semantic Instance Segmentation with a Discriminative Loss Functioncited by 118
  7. 2025mmE5: Improving Multimodal Multilingual Embeddings via High-quality Synthetic Datacited by 49
  8. 2025Improve Multi-Modal Embedding Learning via Explicit Hard Negative Gradient Amplifyingcited by 21

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2023LLMs as Visual Explainers: Advancing Image Classification with Evolving Visual Descriptions
  2. 2024VLM2Vec: Training Vision-Language Models for Massive Multimodal Embedding Tasks
  3. 2025Vector embedding of multi-modal texts: a tool for discovery?
  4. 2024On the Feasibility of Vision-Language Models for Time-Series Classification
  5. 2024Rethinking VLMs and LLMs for Image Classification
  6. 2024Towards Vision-Language Geo-Foundation Model: A Survey
  7. 2026Analyzing Diffusion and Autoregressive Vision Language Models in Multimodal Embedding Space
  8. 2023CLAMP: Contrastive LAnguage Model Prompt-tuning