Research

Paper

AI LLM March 25, 2026

Powerful Teachers Matter: Text-Guided Multi-view Knowledge Distillation with Visual Prior Enhancement

Authors

Xin Zhang, Jianyang Xu, Hao Peng, Dongjing Wang, Jingyuan Zheng, Yu Li, Yuyu Yin, Hongbo Wang

Abstract

Knowledge distillation transfers knowledge from large teacher models to smaller students for efficient inference. While existing methods primarily focus on distillation strategies, they often overlook the importance of enhancing teacher knowledge quality. In this paper, we propose Text-guided Multi-view Knowledge Distillation (TMKD), which leverages dual-modality teachers, a visual teacher and a text teacher (CLIP), to provide richer supervisory signals. Specifically, we enhance the visual teacher with multi-view inputs incorporating visual priors (edge and high-frequency features), while the text teacher generates semantic weights through prior-aware prompts to guide adaptive feature fusion. Additionally, we introduce vision-language contrastive regularization to strengthen semantic knowledge in the student model. Extensive experiments on five benchmarks demonstrate that TMKD consistently improves knowledge distillation performance by up to 4.49\%, validating the effectiveness of our dual-teacher multi-view enhancement strategy. Code is available at https://anonymous.4open.science/r/TMKD-main-44D1.

Metadata

arXiv ID: 2603.24208
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-03-25
Fetched: 2026-03-26 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.24208v1</id>\n    <title>Powerful Teachers Matter: Text-Guided Multi-view Knowledge Distillation with Visual Prior Enhancement</title>\n    <updated>2026-03-25T11:33:27Z</updated>\n    <link href='https://arxiv.org/abs/2603.24208v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.24208v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Knowledge distillation transfers knowledge from large teacher models to smaller students for efficient inference. While existing methods primarily focus on distillation strategies, they often overlook the importance of enhancing teacher knowledge quality. In this paper, we propose Text-guided Multi-view Knowledge Distillation (TMKD), which leverages dual-modality teachers, a visual teacher and a text teacher (CLIP), to provide richer supervisory signals. Specifically, we enhance the visual teacher with multi-view inputs incorporating visual priors (edge and high-frequency features), while the text teacher generates semantic weights through prior-aware prompts to guide adaptive feature fusion. Additionally, we introduce vision-language contrastive regularization to strengthen semantic knowledge in the student model. Extensive experiments on five benchmarks demonstrate that TMKD consistently improves knowledge distillation performance by up to 4.49\\%, validating the effectiveness of our dual-teacher multi-view enhancement strategy. Code is available at https://anonymous.4open.science/r/TMKD-main-44D1.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-03-25T11:33:27Z</published>\n    <arxiv:comment>9 pages, 6 figures</arxiv:comment>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Xin Zhang</name>\n    </author>\n    <author>\n      <name>Jianyang Xu</name>\n    </author>\n    <author>\n      <name>Hao Peng</name>\n    </author>\n    <author>\n      <name>Dongjing Wang</name>\n    </author>\n    <author>\n      <name>Jingyuan Zheng</name>\n    </author>\n    <author>\n      <name>Yu Li</name>\n    </author>\n    <author>\n      <name>Yuyu Yin</name>\n    </author>\n    <author>\n      <name>Hongbo Wang</name>\n    </author>\n  </entry>"
}