This paper proposes the Causal CLIP Adapter (CCA), which applies ICA to causally disentangle CLIP visual features, and enhances cross-modal alignment via unidirectional text classifier fine-tuning and ...
description [ICCV 2025][Multimodal VLM][Benchmark evaluation] This paper proposes an "information density" principle to evaluate MLLM benchmark quality along four dimensions — Fallacy, Difficulty, ...
Abstract: Zero-shot image captioning can harness the knowledge of pre-trained visual language models (VLMs) and language models (LMs) to generate captions for target domain images without paired ...