Hierarchical Skip Decoding for Efficient Autoregressive Text Generation
This addresses the problem of slow inference in text generation for users of pre-trained language models, offering an incremental improvement over existing early-exiting methods.
The paper tackles the computational inefficiency of autoregressive text generation by proposing Hierarchical Skip Decoding (HSD), a plug-and-play method that skips layers based on sequence length, reducing workload while maintaining 90% of text quality with half the layers skipped.
Autoregressive decoding strategy is a commonly used method for text generation tasks with pre-trained language models, while early-exiting is an effective approach to speedup the inference stage. In this work, we propose a novel decoding strategy named Hierarchical Skip Decoding (HSD) for efficient autoregressive text generation. Different from existing methods that require additional trainable components, HSD is a plug-and-play method applicable to autoregressive text generation models, it adaptively skips decoding layers in a hierarchical manner based on the current sequence length, thereby reducing computational workload and allocating computation resources. Comprehensive experiments on five text generation datasets with pre-trained language models demonstrate HSD's advantages in balancing efficiency and text quality. With almost half of the layers skipped, HSD can sustain 90% of the text quality compared to vanilla autoregressive decoding, outperforming the competitive approaches.