This repository is a curated list of papers on multimodal and large language models (LLMs), primarily for personal tracking of daily arXiv publications. It covers AI, computation and language, computer vision, and machine learning, focusing on significant contributions since June 2024.

How It Works

The project functions as a comprehensive, categorized bibliography. It organizes papers by topic, including surveys, representation learning, LLM analysis, safety, evaluation, reasoning, applications, and specific multimodal areas like vision-language models and diffusion models. This structured approach allows for efficient browsing and identification of relevant research within the rapidly evolving fields of LLMs and multimodality.

Quick Start & Requirements

This is a static list of papers and does not require installation or execution. It serves as a reference guide.

Highlighted Details

• Extensive categorization of papers across numerous sub-fields of LLMs and multimodal AI.
• Includes foundational surveys and recent trends, offering a broad overview of the landscape.
• Covers a wide range of specific tasks and techniques, from image generation and document understanding to reasoning and agent development.
• Features papers on theoretical aspects like scaling laws, interpretability, and cognitive neuroscience connections.

Maintenance & Community

The repository is maintained by Yangyi-Chen. Updates focus on papers offering unique insights and substantial contributions. There are no explicit community links or forums mentioned.

Licensing & Compatibility

The repository itself, as a list of links and titles, is not subject to software licensing. The linked papers are subject to their respective copyright and licensing terms.

Limitations & Caveats

This is a personal, non-exhaustive list and may not cover every relevant paper. The focus has shifted to "unique insights and substantial contributions" since June 2024, potentially excluding other valuable work.