Multimodal Transformer with Variable-length Memory for Vision-and-Language Navigation
This addresses a bottleneck in VLN for agents needing to follow instructions over extended trajectories, representing an incremental improvement over existing Transformer methods.
The paper tackles the problem of capturing long-term temporal context in Vision-and-Language Navigation by introducing a Multimodal Transformer with Variable-length Memory, which improves Success Rate by 2% on R2R unseen sets and reduces Goal Process by 1.6m on CVDN.
Vision-and-Language Navigation (VLN) is a task that an agent is required to follow a language instruction to navigate to the goal position, which relies on the ongoing interactions with the environment during moving. Recent Transformer-based VLN methods have made great progress benefiting from the direct connections between visual observations and the language instruction via the multimodal cross-attention mechanism. However, these methods usually represent temporal context as a fixed-length vector by using an LSTM decoder or using manually designed hidden states to build a recurrent Transformer. Considering a single fixed-length vector is often insufficient to capture long-term temporal context, in this paper, we introduce Multimodal Transformer with Variable-length Memory (MTVM) for visually-grounded natural language navigation by modelling the temporal context explicitly. Specifically, MTVM enables the agent to keep track of the navigation trajectory by directly storing previous activations in a memory bank. To further boost the performance, we propose a memory-aware consistency loss to help learn a better joint representation of temporal context with random masked instructions. We evaluate MTVM on popular R2R and CVDN datasets, and our model improves Success Rate on R2R unseen validation and test set by 2% each, and reduce Goal Process by 1.6m on CVDN test set.