LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model

by   Peng Gao, et al.

How to efficiently transform large language models (LLMs) into instruction followers is recently a popular research direction, while training LLM for multi-modal reasoning remains less explored. Although the recent LLaMA-Adapter demonstrates the potential to handle visual inputs with LLMs, it still cannot generalize well to open-ended visual instructions and lags behind GPT-4. In this paper, we present LLaMA-Adapter V2, a parameter-efficient visual instruction model. Specifically, we first augment LLaMA-Adapter by unlocking more learnable parameters (e.g., norm, bias and scale), which distribute the instruction-following ability across the entire LLaMA model besides adapters. Secondly, we propose an early fusion strategy to feed visual tokens only into the early LLM layers, contributing to better visual knowledge incorporation. Thirdly, a joint training paradigm of image-text pairs and instruction-following data is introduced by optimizing disjoint groups of learnable parameters. This strategy effectively alleviates the interference between the two tasks of image-text alignment and instruction following and achieves strong multi-modal reasoning with only a small-scale image-text and instruction dataset. During inference, we incorporate additional expert models (e.g. captioning/OCR systems) into LLaMA-Adapter to further enhance its image understanding capability without incurring training costs. Compared to the original LLaMA-Adapter, our LLaMA-Adapter V2 can perform open-ended multi-modal instructions by merely introducing 14M parameters over LLaMA. The newly designed framework also exhibits stronger language-only instruction-following capabilities and even excels in chat interactions. Our code and models are available at https://github.com/ZrrSkywalker/LLaMA-Adapter.


page 5

page 8

page 9

page 10


LLaSM: Large Language and Speech Model

Multi-modal large language models have garnered significant interest rec...

LLaVAR: Enhanced Visual Instruction Tuning for Text-Rich Image Understanding

Instruction tuning unlocks the superior capability of Large Language Mod...

Mitigating Hallucination in Large Multi-Modal Models via Robust Instruction Tuning

Despite the promising progress in multi-modal tasks, current large multi...

What Matters in Training a GPT4-Style Language Model with Multimodal Inputs?

Recent advancements in Large Language Models (LLMs) such as GPT4 have di...

ImageBind-LLM: Multi-modality Instruction Tuning

We present ImageBind-LLM, a multi-modality instruction tuning method of ...

LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention

We present LLaMA-Adapter, a lightweight adaption method to efficiently f...

A Preliminary Study of the Intrinsic Relationship between Complexity and Alignment

Training large language models (LLMs) with open-domain instruction data ...

Please sign up or login with your details

Forgot password? Click here to reset