Lune

EMNLP Findings2024

Devil's Advocate: Anticipatory Reflection for LLM Agents

Haoyu Wang, Tao Li, Zhiwei Deng, Dan Roth, Yang Li

Abstract

In this work, we introduce a novel approach that equips LLM agents with introspection, enhancing consistency and adaptability in solving complex tasks.Our approach prompts LLM agents to decompose a given task into manageable subtasks (i.e., to make a plan), and to continuously introspect upon the suitability and results of their actions.We implement a three-fold introspective intervention: 1) anticipatory reflection on potential failures and alternative remedy before action execution, 2) post-action alignment with subtask objectives and backtracking with remedy to ensure utmost effort in plan execution, and 3) comprehensive review upon plan completion for future strategy refinement.By deploying and experimenting with this methodologya zero-shot approach-within WebArena for practical tasks in web environments, our agent demonstrates superior performance with a success rate of 23.5% over existing zero-shot methods by 3.5%.The experimental results suggest that our introspection-driven approach not only enhances the agent's ability to navigate unanticipated challenges through a robust mechanism of plan execution, but also improves efficiency by reducing the number of trials and plan revisions by 45% needed to achieve a task.

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.

5 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Tree of Thoughts: Deliberate Problem Solving with Large Language Models cites Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents
  • Look Before You Leap: A GUI-Critic-R1 Model for Pre-Operative Error Diagnosis in GUI Automation cites Aguvis: Unified Pure Vision Agents for Autonomous GUI Interaction
  • Look Before You Leap: A GUI-Critic-R1 Model for Pre-Operative Error Diagnosis in GUI Automation cites Reflexion: language agents with verbal reinforcement learning
  • PaLM-E: An Embodied Multimodal Language Model cites Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents
  • Reflexion: language agents with verbal reinforcement learning cites WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents
  • Richelieu: Self-Evolving LLM-Based Agents for AI Diplomacy cites Tree of Thoughts: Deliberate Problem Solving with Large Language Models
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers5

  1. Richelieu: Self-Evolving LLM-Based Agents for AI DiplomacyZhenyu Guan, Xiangyu Kong, Fangwei Zhong, Yizhou WangNeurIPS 2024 · 48 citations
  2. Look Before You Leap: A GUI-Critic-R1 Model for Pre-Operative Error Diagnosis in GUI AutomationYuyang Wanyan, Xi Zhang, Haiyang Xu, Haowei Liu et al.NeurIPS 2025 · 26 citations
  3. Attention-Driven GUI Grounding: Leveraging Pretrained Multimodal Large Language Models Without Fine-TuningHai-Ming Xu, Qi Chen, Lei Wang, Lingqiao LiuAAAI 2025 · 13 citations
  4. Aguvis: Unified Pure Vision Agents for Autonomous GUI InteractionYiheng Xu, Zekun Wang, Junli Wang, Dunjie Lu et al.ICML 2025
  5. Learning to Reason Over Time: Timeline Self-Reflection for Improved Temporal Reasoning in Language ModelsAdrián Bazaga, Rexhina Blloshmi, Bill Byrne, Adrià de GispertACL 2025

Builds on16

  1. Reflexion: language agents with verbal reinforcement learningNoah Shinn, Federico Cassano, Ashwin Gopinath, Karthik Narasimhan et al.NeurIPS 2023 · 5,828 citations
  2. Tree of Thoughts: Deliberate Problem Solving with Large Language ModelsShunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran et al.NeurIPS 2023 · 5,068 citations
  3. PaLM-E: An Embodied Multimodal Language ModelDanny Driess, Fei Xia, Mehdi S. M. Sajjadi, Corey Lynch et al.ICML 2023 · 2,601 citations
  4. Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied AgentsWenlong Huang, Pieter Abbeel, Deepak Pathak, Igor MordatchICML 2022 · 1,539 citations
  5. WebShop: Towards Scalable Real-World Web Interaction with Grounded Language AgentsShunyu Yao, Howard Chen, John Yang, Karthik NarasimhanNeurIPS 2022 · 1,477 citations

Related papers

  1. A Zero-Shot Language Agent for Computer Control with Structured ReflectionTao Li, Gang Li, Zhiwei Deng, Bryan Wang et al.EMNLP Findings 2023 · 6 citations
  2. Branch-and-Browse: Efficient and Controllable Web Exploration with Tree-Structured Reasoning and Action MemoryShiqi He, Yue Cui, Xinyu Ma, Yaliang Li et al.ACL 2026 · 5 citations
  3. Prospector: Improving LLM Agents with Self-Asking and Trajectory RankingByoungjip Kim, Youngsoo Jang, Lajanugen Logeswaran, Geon-Hyeong Kim et al.EMNLP Findings 2024 · 3 citations
  4. WebPilot: A Versatile and Autonomous Multi-Agent System for Web Task Execution with Strategic ExplorationYao Zhang, Zijian Ma, Yunpu Ma, Zhen Han et al.AAAI 2025 · 101 citations
  5. SAMULE: Self-Learning Agents Enhanced by Multi-level ReflectionYubin Ge, Salvatore Romeo, Jason Cai, Monica Sunkara et al.EMNLP 2025
Dusk over the sea between two cliffs drawn in fine vertical lines