Seeing is Understanding: Unlocking Causal Attention into Modality-Mutual Attention for Multimodal LLMs
This addresses a critical challenge in MLLMs for improving multimodal understanding, though it is incremental as it builds on existing decoder-only LLM architectures.
The paper tackles vision-language misalignment in Multimodal Large Language Models (MLLMs) by proposing a novel architecture that unlocks causal attention into modality-mutual attention, enabling image tokens to attend to text tokens, which achieves an average performance improvement of +7.2% across 12 multimodal benchmarks without extra parameters or training time.
Recent Multimodal Large Language Models (MLLMs) have demonstrated significant progress in perceiving and reasoning over multimodal inquiries, ushering in a new research era for foundation models. However, vision-language misalignment in MLLMs has emerged as a critical challenge, where the textual responses generated by these models are not factually aligned with the given text-image inputs. Existing efforts to address vision-language misalignment have focused on developing specialized vision-language connectors or leveraging visual instruction tuning from diverse domains. In this paper, we tackle this issue from a fundamental yet unexplored perspective by revisiting the core architecture of MLLMs. Most MLLMs are typically built on decoder-only LLMs consisting of a causal attention mechanism, which limits the ability of the earlier modalities (e.g., images) to incorporate information from the latter modalities (e.g., text). To address this problem, we propose \MapleLeaf AKI, a novel MLLM that unlocks causal attention into modality-mutual attention (MMA) to enable image tokens to attend to text tokens. This simple yet effective design allows AKI to achieve superior performance in 12 multimodal understanding benchmarks (+7.2% on average) without introducing additional parameters and increasing training time. Our MMA design is intended to be generic, allowing for application across various modalities, and scalable to accommodate diverse multimodal scenarios. The code and model are publicly available at https://github.com/sony/aki to encourage further advancements in MLLMs across various directions.