Abstract
Despite the success of Vision-Language Models (VLMs) like CLIP in aligning vision and language, their proficiency in detailed, fine-grained visual comprehension remains a key challenge. We present CLIP-IN, a novel framework that bolsters CLIP’s fine-grained perception through two core innovations. Firstly, we leverage instruction-editing datasets, originally designed for image manipulation, as a unique source of hard negative image-text pairs. Coupled with a symmetric hard negative contrastive loss, this enables the model to effectively distinguish subtle visual-semantic differences. Secondly, CLIP-IN incorporates long descriptive captions, utilizing rotary positional encodings to capture rich semantic context often missed by standard CLIP. Our experiments demonstrate that CLIP-IN achieves substantial gains on the MMVP benchmark and various fine-grained visual recognition tasks, without compromising robust zero-shot performance on broader classification and retrieval tasks. Critically, integrating CLIP-IN’s visual representations into Multimodal Large Language Models significantly reduces visual hallucinations and enhances reasoning abilities. This work underscores the considerable potential of synergizing targeted, instruction-based contrastive learning with comprehensive descriptive information to elevate the fine-grained understanding of VLMs.
Overview of CLIP-IN
Symmetric Hard Negative Loss
Zero-Shot classification and retrieval Performance
MMVP Benchmark
Compositional Benchmarks
Performance of MLLM
Performance on zero-shot classification variants
Ablation studies on synthetic data source and data scale.
Ablation studies on different components-CLIP benchmarks.
Ablation studies on different components-MLLM benchmarks
Ablation studies on the impact of Rotary Positional Embeddings (RoPE).
Ablation studies on Symmetric Hard Negative Loss (L_HN).
Ablation studies on training data.
Ablation studies on long caption data scale.
Samples of instruction editing hard negative data
Feature visualization
BibTeX
@article{Wang2025CLIPINEF,
title={CLIP-IN: Enhancing Fine-Grained Visual Understanding in CLIP via Instruction Editing Data and Long Captions},
author={Ziteng Wang and Siqi Yang and Limeng Qiao and Lin Ma},
journal={NeurIPS},
year={2025}
}