Characterizing Prompt Compression Methods for Long Context Inference
This addresses a standardization gap for researchers and practitioners in AI/ML working with long-context models, though it is incremental as it compares existing methods rather than introducing new ones.
The paper tackles the problem of conflicting results in prompt compression methods for long context inference by performing a comprehensive characterization and evaluation, finding that extractive compression often outperforms other approaches with up to 10x compression and minimal accuracy degradation.
Long context inference presents challenges at the system level with increased compute and memory requirements, as well as from an accuracy perspective in being able to reason over long contexts. Recently, several methods have been proposed to compress the prompt to reduce the context length. However, there has been little work on comparing the different proposed methods across different tasks through a standardized analysis. This has led to conflicting results. To address this, here we perform a comprehensive characterization and evaluation of different prompt compression methods. In particular, we analyze extractive compression, summarization-based abstractive compression, and token pruning methods. Surprisingly, we find that extractive compression often outperforms all the other approaches, and enables up to 10x compression with minimal accuracy degradation. Interestingly, we also find that despite several recent claims, token pruning methods often lag behind extractive compression. We only found marginal improvements on summarization tasks.