Boosting Self-Efficacy and Performance of Large Language Models via Verbal Efficacy Stimulations

2025-02-10

AI Summary

This work investigates how verbal efficacy stimuli (VES) modulate large language models’ (LLMs’) self-efficacy perceptions and task performance. Grounded in social cognitive theory, it introduces— for the first time—the concept of self-efficacy into LLM prompt design, proposing three VES types—encouraging, activating, and critical—that collectively operationalize six psychological dimensions (e.g., capability perception, effort attribution) within a structured zero-shot prompting template. Systematic evaluation across multi-difficulty tasks reveals that all three VES types consistently improve zero-shot performance across mainstream LLMs, though optimal VES type varies by model. Crucially, LLM response patterns align closely with classical psychological theory, demonstrating a regulatable, efficacy-like behavioral mechanism. The study establishes a novel paradigm for cognitive modeling of LLMs and controllable prompt engineering, bridging human motivation theory with foundation model behavior.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Natural Language Processing: Prompt Engineering / Prompting

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Significant improvements have been observed in the zero-shot capabilities of the Large Language Models (LLMs). Due to their high sensitivity to input, research has increasingly focused on enhancing LLMs' performance via direct and simple prompt engineering rather than intricate domain adaptation. Studies suggest that LLMs exhibit emotional intelligence, and both positive and negative emotions can potentially enhance task performances. However, prior interaction prompts have predominantly concentrated on a single stimulus type, neglecting to compare different stimulus effects, examine the influence of varying task difficulties, or explore underlying mechanisms. This paper, inspired by the positive correlation between self-efficacy and task performance within the social cognitive theory, introduces Verbal Efficacy Stimulations (VES). Our VES comprises three types of verbal prompts: encouraging, provocative, and critical, addressing six aspects such as helpfulness and competence. And we further categorize task difficulty, aiming to extensively investigate how distinct VES influence the self-efficacy and task achievements of language models at varied levels of difficulty. The experimental results show that the three types of VES improve the performance of LLMs on most tasks, and the most effective VES varies for different models. In extensive experiments, we have obtained some findings consistent with psychological theories, providing novel insights for future research.

Problem

• Enhance LLMs' zero-shot capabilities
• Compare verbal efficacy stimulation effects
• Explore task difficulty influence on LLMs

Innovation

• Verbal Efficacy Stimulations (VES)
• Three types of verbal prompts
• Task difficulty categorization