HyperLLaVA: Dynamic Visual and Language Expert Tuning for Multimodal Large Language Models
This work addresses the problem of constrained performance across different multimodal tasks for researchers and practitioners in AI, representing an incremental advancement over existing MLLM methods.
The paper tackles the limitation of static tuning in Multimodal Large Language Models (MLLMs) by introducing HyperLLaVA, which uses dynamic visual and language experts for adaptive parameter tuning, resulting in significant performance improvements over LLaVA on benchmarks like MME, MMBench, SEED-Bench, and LLaVA-Bench.
Recent advancements indicate that scaling up Multimodal Large Language Models (MLLMs) effectively enhances performance on downstream multimodal tasks. The prevailing MLLM paradigm, \emph{e.g.}, LLaVA, transforms visual features into text-like tokens using a \emph{static} vision-language mapper, thereby enabling \emph{static} LLMs to develop the capability to comprehend visual information through visual instruction tuning. Although promising, the \emph{static} tuning strategy~\footnote{The static tuning refers to the trained model with static parameters.} that shares the same parameters may constrain performance across different downstream multimodal tasks. In light of this, we introduce HyperLLaVA, which involves adaptive tuning of the projector and LLM parameters, in conjunction with a dynamic visual expert and language expert, respectively. These experts are derived from HyperNetworks, which generates adaptive parameter shifts through visual and language guidance, enabling dynamic projector and LLM modeling in two-stage training. Our experiments demonstrate that our solution significantly surpasses LLaVA on existing MLLM benchmarks, including MME, MMBench, SEED-Bench, and LLaVA-Bench. ~\footnote{Our project is available on the link https://github.com/DCDmllm/HyperLLaVA}.