mlsanityData to 7 Oct 2026

CIFLEX: Contextual Instruction Flow for Sub-task Execution in Multi-Turn Interactions with a Single On-Device LLM

arXiv 2510.01239 ↗First posted 24 Sep 2025cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 16 of its 23 references that are in the map. First posted on arXiv 24 Sep 2025.

Builds on

16 of 23 references.

References · 16

  1. 2024The Llama 3 Herd of Modelscited by 14,192
  2. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  3. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  4. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  5. 2023Gorilla: Large Language Model Connected with Massive APIscited by 1,302
  6. 2025LLMs Get Lost In Multi-Turn Conversationcited by 400
  7. 2024A Survey on Recent Advances in LLM-Based Multi-turn Dialogue Systemscited by 170
  8. 2020Open-Domain Question Answering Goes Conversational via Question Rewritingcited by 147
  9. 2021TopiOCQA: Open-domain Conversational Question Answering with Topic Switchingcited by 95
  10. 2025KVLink: Accelerating Large Language Models via Efficient KV Cache Reusecited by 60
  11. 2025Evaluating LLM-based Agents for Multi-Turn Conversations: A Surveycited by 56
  12. 2024KVSharer: Efficient Inference via Layer-Wise Dissimilar KV Cache Sharingcited by 38
4 more
  1. 2024DroidSpeak: KV Cache Sharing for Cross-LLM Communication and Multi-LLM Servingcited by 30
  2. 2024Adaptive Query Rewriting: Aligning Rewriters through Marginal Probability of Conversational Answerscited by 15
  3. 2025Efficient LLM Inference with Activation Checkpointing and Hybrid Cachingcited by 3
  4. 2025Streamlining the Collaborative Chain of Models into A Single Forward Pass in Generation-Based Taskscited by 2

Built on by

Not cited yet.