TimeViper: A Hybrid Mamba-Transformer Vision-Language Model for Efficient Long Video Understanding
This addresses the problem of processing long videos for applications requiring efficient and effective temporal context handling, representing an incremental advancement in hybrid model architectures.
The authors tackled the challenge of efficient long video understanding by introducing TimeViper, a hybrid Mamba-Transformer vision-language model that processes hour-long videos exceeding 10,000 frames and competes with state-of-the-art models on benchmarks.
We introduce TimeViper, a hybrid vision-language model designed to tackle challenges of long video understanding. Processing long videos demands both an efficient model architecture and an effective mechanism for handling extended temporal contexts. To this end, TimeViper adopts a hybrid Mamba-Transformer backbone that combines the efficiency of state-space models with the expressivity of attention mechanisms. Through this hybrid design, we reveal the vision-to-text information aggregation phenomenon, where information progressively flows from vision tokens to text tokens across increasing LLM depth, resulting in severe vision token redundancy. Motivated by this observation, we propose TransV, a token information transfer module that transfers and compresses vision tokens into instruction tokens while maintaining multimodal understanding capabilities. This design enables TimeViper to process hour-long videos exceeding 10,000 frames. Extensive experiments across multiple benchmarks demonstrate that TimeViper competes with state-of-the-art models while extending frame numbers. We further analyze attention behaviors of both Mamba and Transformer layers, offering new insights into hybrid model interpretability. This work represents an initial step towards developing, interpreting, and compressing hybrid Mamba-Transformer architectures.