Abstract:Long video understanding is a complex task that requires both spatial detail and temporal awareness. While Vision-Language Models (VLMs) obtain frame-level understanding capabilities through multi-frame input, they suffer from information loss due to the sparse sampling strategy. In contrast, Video Large Language Models (Video-LLMs) capture temporal relationships within visual features but are limited by the scarcity of high-quality video-text datasets. To transfer long video understanding capabilities to VLMs with minimal data and computational cost, we propose Lightweight Video Compression (LVC), a novel method featuring the Query-Attention Video Compression mechanism, which effectively tackles the sparse sampling problem in VLMs. By training only the alignment layer with 10k short video-text pairs, LVC significantly enhances the temporal reasoning abilities of VLMs. Extensive experiments show that LVC provides consistent performance improvements across various models, including the InternVL2 series and Phi-3.5-Vision. Notably, the InternVL2-40B-LVC achieves scores of 68.2 and 65.9 on the long video understanding benchmarks MLVU and Video-MME, respectively, with relative improvements of 14.6% and 7.7%. The enhanced models and code will be publicly available soon.
Abstract:Multi-timestep simulation of brain-inspired Spiking Neural Networks (SNNs) boost memory requirements during training and increase inference energy cost. Current training methods cannot simultaneously solve both training and inference dilemmas. This work proposes a novel Temporal Reversible architecture for SNNs (T-RevSNN) to jointly address the training and inference challenges by altering the forward propagation of SNNs. We turn off the temporal dynamics of most spiking neurons and design multi-level temporal reversible interactions at temporal turn-on spiking neurons, resulting in a $O(L)$ training memory. Combined with the temporal reversible nature, we redesign the input encoding and network organization of SNNs to achieve $O(1)$ inference energy cost. Then, we finely adjust the internal units and residual connections of the basic SNN block to ensure the effectiveness of sparse temporal information interaction. T-RevSNN achieves excellent accuracy on ImageNet, while the memory efficiency, training time acceleration, and inference energy efficiency can be significantly improved by $8.6 \times$, $2.0 \times$, and $1.6 \times$, respectively. This work is expected to break the technical bottleneck of significantly increasing memory cost and training time for large-scale SNNs while maintaining high performance and low inference energy cost. Source code and models are available at: https://github.com/BICLab/T-RevSNN.