QUITO: Accelerating Long-Context Reasoning through Query-Guided Context Compression
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913454949924864 |
|---|---|
| author | Wang, Wenshan Wang, Yihang Fan, Yixing Liao, Huaming Guo, Jiafeng |
| author_facet | Wang, Wenshan Wang, Yihang Fan, Yixing Liao, Huaming Guo, Jiafeng |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_00274 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | QUITO: Accelerating Long-Context Reasoning through Query-Guided Context Compression Wang, Wenshan Wang, Yihang Fan, Yixing Liao, Huaming Guo, Jiafeng Computation and Language Artificial Intelligence 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. |
| title | QUITO: Accelerating Long-Context Reasoning through Query-Guided Context Compression |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2408.00274 |