相关论文: Interpreting Attention Heads for Image-to-Text Inf…

Recent advancements in dialogue systems have highlighted the significance of integrating multimodal responses, which enable conveying ideas through diverse modalities rather than solely relying on text-based interactions. This enrichment…

While large text-to-image models are able to synthesize "novel" images, these images are necessarily a reflection of the training data. The problem of data attribution in such models -- which of the images in the training set are most…

Vision-language models (VLMs) excel at extracting and reasoning about information from images. Yet, their capacity to leverage internal knowledge about specific entities remains underexplored. This work investigates the disparity in model…

This paper presents VideoStreaming, an advanced vision-language large model (VLLM) for video understanding, that capably understands arbitrary-length video with a constant number of video tokens streamingly encoded and adaptively selected.…

Induction head mechanism is a part of the computational circuits for in-context learning (ICL) that enable large language models (LLMs) to adapt to new tasks without fine-tuning. Most existing work explains the training dynamics behind…

Multi-modal Large Language Models (MLLMs) have demonstrated remarkable capabilities in understanding and generating content across various modalities, such as images and text. However, their interpretability remains a challenge, hindering…

Large Language Models (LLMs) tend to attend heavily to the first token in the sequence -- creating a so-called attention sink. Many works have studied this phenomenon in detail, proposing various ways to either leverage or alleviate it.…

Visual attention plays an important role to understand images and demonstrates its effectiveness in generating natural language descriptions of images. On the other hand, recent studies show that language associated with an image can steer…

Vision Large Language Models (VLLMs) usually take input as a concatenation of image token embeddings and text token embeddings and conduct causal modeling. However, their internal behaviors remain underexplored, raising the question of…

Despite the impressive capabilities of Multimodal Large Language Models (MLLMs) in integrating text and image modalities, challenges remain in accurately interpreting detailed visual elements. Vision detection models excel at recognizing…

Large-scale diffusion models have made significant advances in image generation, particularly through cross-attention mechanisms. While cross-attention has been well-studied in text-to-image tasks, their interpretability in image-to-image…

In the Transformer model, "self-attention" combines information from attended embeddings into the representation of the focal embedding in the next layer. Thus, across layers of the Transformer, information originating from different tokens…

Vision Language Models (VLMs) excel at identifying and describing objects but often fail at spatial reasoning. We study why VLMs, such as LLaVA, underutilize spatial cues despite having positional encodings and spatially rich vision encoder…

Multimodal Large Language Models (MLLMs) have achieved remarkable progress in multimodal reasoning. However, they often excessively rely on textual information during the later stages of inference, neglecting the crucial integration of…

Multimodal Large Language Models (MLLMs) have significantly improved performance across various image-language applications. Recently, there has been a growing interest in adapting image pre-trained MLLMs for video-related tasks. However,…

Flowcharts are typically presented as images, driving the trend of using vision-language models (VLMs) for end-to-end flowchart understanding. However, two key challenges arise: (i) Limited controllability--users have minimal influence over…

Large language models (LLMs) exhibit strong in-context learning capabilities, but how they track and retrieve information from context remains underexplored. Drawing on the free recall paradigm in cognitive science (where participants…

The current large language models are mainly based on decode-only structure transformers, which have great in-context learning (ICL) capabilities. It is generally believed that the important foundation of its ICL capability is the induction…

Vision-Language Models (VLMs) have been shown to be blind, often underutilizing their visual inputs even on tasks that require visual reasoning. In this work, we demonstrate that VLMs are selectively blind. They modulate the amount of…

Large Language Models (LLMs) are becoming increasingly popular in pervasive computing due to their versatility and strong performance. However, despite their ubiquitous use, the exact mechanisms underlying their outstanding performance…