Lune

ACL Findings2025

Direct Behavior Optimization: Unlocking the Potential of Lightweight LLMs

Hongming Yang, Shi Lin, Jun Shao, Changting Lin, Donghai Zhu, Meng Han, Qinglei Kong

Abstract

Lightweight Large Language Models (LwLLMs) are reduced-parameter, optimized models designed to run efficiently on consumergrade hardware, offering significant advantages in resource efficiency, cost-effectiveness, and data privacy. However, these models often struggle with limited inference and reasoning capabilities, which restrict their performance on complex tasks and limit their practical applicability. Moreover, existing prompt optimization methods typically rely on extensive manual effort or the meta-cognitive abilities of state-of-the-art LLMs, making them less effective for LwLLMs.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Self-Refine: Iterative Refinement with Self-Feedback cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  • Least-to-Most Prompting Enables Complex Reasoning in Large Language Models cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  • Teaching Large Language Models to Self-Debug cites Self-Refine: Iterative Refinement with Self-Feedback
  • Teaching Large Language Models to Self-Debug cites Least-to-Most Prompting Enables Complex Reasoning in Large Language Models
  • Teaching Large Language Models to Self-Debug cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  • Graph of Thoughts: Solving Elaborate Problems with Large Language Models cites Teaching Large Language Models to Self-Debug
  • Graph of Thoughts: Solving Elaborate Problems with Large Language Models cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on18

  1. Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
  2. Self-Refine: Iterative Refinement with Self-FeedbackAman Madaan, Niket Tandon, Prakhar Gupta, Skyler Hallinan et al.NeurIPS 2023 · 4,972 citations
  3. Graph of Thoughts: Solving Elaborate Problems with Large Language ModelsMaciej Besta, Nils Blach, Ales Kubicek, Robert Gerstenberger et al.AAAI 2024 · 1,292 citations
  4. Teaching Large Language Models to Self-DebugXinyun Chen, Maxwell Lin, Nathanael Schärli, Denny ZhouICLR 2024 · 1,085 citations
  5. Least-to-Most Prompting Enables Complex Reasoning in Large Language ModelsDenny Zhou, Nathanael Schärli, Le Hou, Jason Wei et al.ICLR 2023 · 318 citations

Related papers

  1. GReaTer: Gradients Over Reasoning Makes Smaller Language Models Strong Prompt OptimizersSarkar Snigdha Sarathi Das, Ryo Kamoi, Bo Pang, Yusen Zhang et al.ICLR 2025
  2. Unlocking Black-Box Prompt Tuning Efficiency via Zeroth-Order OptimizationHeshen Zhan, Congliang Chen, Tian Ding, Ziniu Li et al.EMNLP Findings 2024 · 3 citations
  3. DLPO: Towards a Robust, Efficient, and Generalizable Prompt Optimization Framework from a Deep-Learning PerspectiveDengyun Peng, Yuhang Zhou, Qiguang Chen, Jinhao Liu et al.EMNLP Findings 2025 · 2 citations
  4. Dual-Phase Accelerated Prompt OptimizationMuchen Yang, Moxin Li, Yongle Li, Zijun Chen et al.EMNLP Findings 2024
  5. Revisiting OPRO: The Limitations of Small-Scale LLMs as OptimizersTuo Zhang, Jinyue Yuan, Salman AvestimehrACL Findings 2024 · 5 citations
Dusk over the sea between two cliffs drawn in fine vertical lines