7 papers
CrossCult-KIBench: A Benchmark for Cross-Cultural Knowledge Insertion in MLLMs
Zhen Zeng, Leijiang Gu, Feng Li +2
Multimodal Large Language Models (MLLMs), trained primarily on English-centric data, frequently generate culturally inappropriate or misaligned responses in cross-cultural settings…
ProAPO: Progressively Automatic Prompt Optimization for Visual Classification
Xiangyan Qu, Gaopeng Gou, Jiamin Zhuang +5
Vision-language models (VLMs) have made significant progress in image classification by training with large-scale paired image-text data. Their performances largely depend on the p…
Respond to Change with Constancy: Instruction-tuning with LLM for Non-I.I.D. Network Traffic Classification
Xinjie Lin, Gang Xiong, Gaopeng Gou +4
Encrypted traffic classification is highly challenging in network security due to the need for extracting robust features from content-agnostic traffic data. Existing approaches fa…
Missing Target-Relevant Information Prediction with World Model for Accurate Zero-Shot Composed Image Retrieval
Yuanmin Tang, Jing Yu, Keke Gai +4
Zero-Shot Composed Image Retrieval (ZS-CIR) involves diverse tasks with a broad range of visual content manipulation intent across domain, scene, object, and attribute. The key cha…
MADS: Multi-Attribute Document Supervision for Zero-Shot Image Classification
Xiangyan Qu, Jing Yu, Jiamin Zhuang +3
Zero-shot learning (ZSL) aims to train a model on seen classes and recognize unseen classes by knowledge transfer through shared auxiliary information. Recent studies reveal that d…
Reason-before-Retrieve: One-Stage Reflective Chain-of-Thoughts for Training-Free Zero-Shot Composed Image Retrieval
Yuanmin Tang, Xiaoting Qin, Jue Zhang +7
Composed Image Retrieval (CIR) aims to retrieve target images that closely resemble a reference image while integrating user-specified textual modifications, thereby capturing user…