Saved in:
Bibliographic Details
Main Authors: Peng, Xiao, Geng, Xufan
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.00359
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916417514766336
author Peng, Xiao
Geng, Xufan
author_facet Peng, Xiao
Geng, Xufan
contents The applications of large language models (LLMs) have been widely spread across all domains. However, the basic abilities such as the controllability of LLMs are still limited. To address this, we propose "Self-controller", a novel agentic framework bringing self-awareness into LLMs' reasoning logic. The core idea of this work is to maintain states based on the LLM's response, letting the LLM become self-aware of current status and think step by step in a multi-round chain-of-thought paradigm. Our experiment on the state of textual length has shown the controllability and effectiveness of the Self-controller. We further implement a binary search algorithm to accelerate the generation process based on the linearity and monotonicity of the textual length state. Another advantage of the Self-controller comes with DeepSeek's Context Caching technology, which significantly saves computational token consumption when a cluster of conversations shares the same prefix of context. Theoretically, we prove that in this scenario the extra time complexity is $O(c \log n)$. Results of the back-of-the-envelope estimation suggest that the token consumption of our method is no more than twice as much as that of the trivial single-round generation. Furthermore, our ablation study on word constraints demonstrates the Self-controller's consistent controllability across all foundation models.
format Preprint
id arxiv_https___arxiv_org_abs_2410_00359
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Self-controller: Controlling LLMs with Multi-round Step-by-step Self-awareness
Peng, Xiao
Geng, Xufan
Computation and Language
Artificial Intelligence
The applications of large language models (LLMs) have been widely spread across all domains. However, the basic abilities such as the controllability of LLMs are still limited. To address this, we propose "Self-controller", a novel agentic framework bringing self-awareness into LLMs' reasoning logic. The core idea of this work is to maintain states based on the LLM's response, letting the LLM become self-aware of current status and think step by step in a multi-round chain-of-thought paradigm. Our experiment on the state of textual length has shown the controllability and effectiveness of the Self-controller. We further implement a binary search algorithm to accelerate the generation process based on the linearity and monotonicity of the textual length state. Another advantage of the Self-controller comes with DeepSeek's Context Caching technology, which significantly saves computational token consumption when a cluster of conversations shares the same prefix of context. Theoretically, we prove that in this scenario the extra time complexity is $O(c \log n)$. Results of the back-of-the-envelope estimation suggest that the token consumption of our method is no more than twice as much as that of the trivial single-round generation. Furthermore, our ablation study on word constraints demonstrates the Self-controller's consistent controllability across all foundation models.
title Self-controller: Controlling LLMs with Multi-round Step-by-step Self-awareness
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2410.00359