Order Matters: Exploring Order Sensitivity in Multimodal Large Language Models
원문 초록
Multimodal Large Language Models (MLLMs) utilize multimodal contexts consisting of text, images, or videos to solve various multimodal tasks. However, we find that changing the order of multimodal input can cause the model's performance to fluctuate between advanced performance and random guessing. This phenomenon exists in both single-modality (text-only or image-only) and mixed-modality (image-text-pair) contexts. Furthermore, we demonstrate that popular MLLMs pay special attention to certain multimodal context positions, particularly the beginning and end. Leveraging this special attention, we place key video frames and important image/text content in special positions within the context and submit them to the MLLM for inference. This method results in average performance gains of 14.7% for video-caption matching and 17.8% for visual question answering tasks. Additionally, we propose a new metric, Position-Invariant Accuracy (PIA), to address order bias in MLLM evaluation. Our research findings contribute to a better understanding of Multi-Modal In-Context Learning (MMICL) and provide practical strategies for enhancing MLLM performance without increasing computational costs.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
SMMILE: An Expert-Driven Benchmark for Multimodal Medical In-Context Learning
Provoking Multi-modal Few-Shot LVLM via Exploration-Exploitation In-Context Learning
Meta-Adaptive Prompt Distillation for Few-Shot Visual Question Answering
Describe Anything Model for Visual Question Answering on Text-rich Images
Revisiting CroPA: A Reproducibility Study and Enhancements for Cross-Prompt Adversarial Transferability in Vision-Language Models
DrishtiKon: Multi-Granular Visual Grounding for Text-Rich Document Images
토론
이 논문과 연결된 포럼 글이 아직 없습니다.