CIFLEX: Contextual Instruction Flow for Sub-task Execution in Multi-Turn Interactions with a Single On-Device LLM
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908572799991808 |
|---|---|
| author | Lee, Juntae Bang, Jihwan Yang, Seunghan Chang, Simyung |
| author_facet | Lee, Juntae Bang, Jihwan Yang, Seunghan Chang, Simyung |
| contents | We present CIFLEX (Contextual Instruction Flow for Sub-task Execution), which is a novel execution system for efficient sub-task handling in multi-turn interactions with a single on-device large language model (LLM). As LLMs become increasingly capable, a single model is expected to handle diverse sub-tasks that more effectively and comprehensively support answering user requests. Naive approach reprocesses the entire conversation context when switching between main and sub-tasks (e.g., query rewriting, summarization), incurring significant computational overhead. CIFLEX mitigates this overhead by reusing the key-value (KV) cache from the main task and injecting only task-specific instructions into isolated side paths. After sub-task execution, the model rolls back to the main path via cached context, thereby avoiding redundant prefill computation. To support sub-task selection, we also develop a hierarchical classification strategy tailored for small-scale models, decomposing multi-choice decisions into binary ones. Experiments show that CIFLEX significantly reduces computational costs without degrading task performance, enabling scalable and efficient multi-task dialogue on-device. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_01239 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CIFLEX: Contextual Instruction Flow for Sub-task Execution in Multi-Turn Interactions with a Single On-Device LLM Lee, Juntae Bang, Jihwan Yang, Seunghan Chang, Simyung Computation and Language We present CIFLEX (Contextual Instruction Flow for Sub-task Execution), which is a novel execution system for efficient sub-task handling in multi-turn interactions with a single on-device large language model (LLM). As LLMs become increasingly capable, a single model is expected to handle diverse sub-tasks that more effectively and comprehensively support answering user requests. Naive approach reprocesses the entire conversation context when switching between main and sub-tasks (e.g., query rewriting, summarization), incurring significant computational overhead. CIFLEX mitigates this overhead by reusing the key-value (KV) cache from the main task and injecting only task-specific instructions into isolated side paths. After sub-task execution, the model rolls back to the main path via cached context, thereby avoiding redundant prefill computation. To support sub-task selection, we also develop a hierarchical classification strategy tailored for small-scale models, decomposing multi-choice decisions into binary ones. Experiments show that CIFLEX significantly reduces computational costs without degrading task performance, enabling scalable and efficient multi-task dialogue on-device. |
| title | CIFLEX: Contextual Instruction Flow for Sub-task Execution in Multi-Turn Interactions with a Single On-Device LLM |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2510.01239 |