Self-controller:利用多轮逐步自我意识控制 LLM
Self-controller: Controlling LLMs with Multi-round Step-by-step Self-awareness
AI总结:
针对 LLM 可控性不足的问题,本文提出 Self-controller 框架,通过多轮状态感知与逐步推理实现文本长度等约束控制,并结合二分搜索和上下文缓存降低开销。
摘要:
大语言模型(LLM)的应用已广泛遍及各个领域。然而,LLM 的可控性等基础能力仍然有限。为解决这一问题,我们提出了“Self-controller”,这是一种将自我意识引入 LLM 推理逻辑的新型智能体框架。本工作的核心思想是基于 LLM 的响应维护状态,使 LLM 能够意识到当前状态,并在多轮思维链范式中逐步思考。我们在文本长度状态上的实验展示了 Self-controller 的可控性和有效性。基于文本长度状态的线性和单调性,我们进一步实现了二分搜索算法以加速生成过程。Self-controller 的另一个优势来自 DeepSeek 的 Context Caching 技术,当一组对话共享相同的上下文前缀时,该技术可显著节省计算 token 消耗。理论上,我们证明了在此场景下额外时间复杂度为 O(c log n)。粗略估算结果表明,我们方法的 token 消耗不超过简单单轮生成的两倍。此外,我们针对词数约束开展的消融研究证明,Self-controller 在所有基础模型上均具有一致的可控性。
英文摘要:
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.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。