QUITO: Accelerating Long-Context Reasoning through Query-Guided Context Compression
This work addresses efficiency issues in long-context reasoning for users of large language models, representing an incremental improvement in context compression techniques.
The paper tackles the problem of reducing reasoning complexities and computation costs in large language models by introducing QUITO, a query-guided attention compression method that filters context based on question attention, and it significantly outperforms established baselines on datasets like NaturalQuestions and ASQA.
In-context learning (ICL) capabilities are foundational to the success of large language models (LLMs). Recently, context compression has attracted growing interest since it can largely reduce reasoning complexities and computation costs of LLMs. In this paper, we introduce a novel Query-gUIded aTtention cOmpression (QUITO) method, which leverages attention of the question over the contexts to filter useless information. Specifically, we take a trigger token to calculate the attention distribution of the context in response to the question. Based on the distribution, we propose three different filtering methods to satisfy the budget constraints of the context length. We evaluate the QUITO using two widely-used datasets, namely, NaturalQuestions and ASQA. Experimental results demonstrate that QUITO significantly outperforms established baselines across various datasets and downstream LLMs, underscoring its effectiveness. Our code is available at https://github.com/Wenshansilvia/attention_compressor.