Causal-aware Large Language Models: Enhancing Decision-Making Through Learning, Adapting and Acting

2025-05-30

AI Summary

Existing large language models (LLMs) lack explicit causal reasoning capabilities and environmental adaptability, limiting their effectiveness in complex, dynamic decision-making tasks. To address this, we propose Causal-LLM, the first LLM framework that deeply integrates structural causal models (SCMs) into the decision-making pipeline via a three-stage “Learn–Adapt–Act” paradigm: (1) *Learn*: causal entities and relationships are extracted from domain knowledge; (2) *Adapt*: external feedback triggers causal interventions to dynamically update the SCM; (3) *Act*: policy decisions are optimized via reinforcement learning, grounded in interpretable causal reasoning. This framework enables explicit causal knowledge modeling, online causal updating, and transparent, explainable decision-making—capabilities previously unattained in LLMs. Evaluated on 22 diverse tasks in the open-world game Crafter, Causal-LLM achieves significant improvements in decision accuracy and cross-task generalization, empirically validating its efficacy in complex, non-stationary environments.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Knowledge Representation and Reasoning: Action, Change, and Causality

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Large language models (LLMs) have shown great potential in decision-making due to the vast amount of knowledge stored within the models. However, these pre-trained models are prone to lack reasoning abilities and are difficult to adapt to new environments, further hindering their application to complex real-world tasks. To address these challenges, inspired by the human cognitive process, we propose Causal-aware LLMs, which integrate the structural causal model (SCM) into the decision-making process to model, update, and utilize structured knowledge of the environment in a ``learning-adapting-acting"paradigm. Specifically, in the learning stage, we first utilize an LLM to extract the environment-specific causal entities and their causal relations to initialize a structured causal model of the environment. Subsequently,in the adapting stage, we update the structured causal model through external feedback about the environment, via an idea of causal intervention. Finally, in the acting stage, Causal-aware LLMs exploit structured causal knowledge for more efficient policy-making through the reinforcement learning agent. The above processes are performed iteratively to learn causal knowledge, ultimately enabling the causal-aware LLMs to achieve a more accurate understanding of the environment and make more efficient decisions. Experimental results across 22 diverse tasks within the open-world game ``Crafter"validate the effectiveness of our proposed method.

Problem

• Enhancing LLM decision-making with causal reasoning
• Adapting LLMs to new environments via causal models
• Improving policy-making through causal-aware reinforcement learning

Innovation

• Integrates SCM into LLMs for decision-making
• Updates causal model via feedback intervention
• Uses reinforcement learning for policy-making