On Path to Multimodal Generalist: Levels and Benchmarks

[📖 Project] [🏆 Leaderboard] [📄 Paper] [🤗 Dataset-HF] [📝 Dataset-Github]

---

Does higher performance across tasks indicate a stronger capability of MLLM, and closer to AGI?
NO! But synergy does.

Most current MLLMs predominantly build on the language intelligence of LLMs to simulate the indirect intelligence of multimodality, which is merely extending language intelligence to aid multimodal understanding. While LLMs (e.g., ChatGPT) have already demonstrated such synergy in NLP, reflecting language intelligence, unfortunately, the vast majority of MLLMs do not really achieve it across modalities and tasks.

We argue that the key to advancing towards AGI lies in the synergy effect—a capability that enables knowledge learned in one modality or task to generalize and enhance mastery in other modalities or tasks, fostering mutual improvement across different modalities and tasks through interconnected learning.

🏆🏆🏆 Overall Leaderboard

This project introduces General-Level and General-Bench.

🚀🚀🚀 General-Level: a 5-scale level evaluation system with a new norm for assessing the multimodal generalists (multimodal LLMs/agents). The core is the use of Synergy as the evaluative criterion, categorizing capabilities based on whether MLLMs preserve synergy across comprehension and generation, as well as across multimodal interactions.

🌐🌐🌐 General-Bench, a companion massive multimodal benchmark dataset, encompasses a broader spectrum of skills, modalities, formats, and capabilities, including over 700 tasks and 325K instances.