# Iron: Intent-Aligned and Retrospective Dual Learning Framework for Enhancing Generalist Virtual Agents

arXiv: https://arxiv.org/abs/2608.27866 · HTML: https://mlsanity.com/paper/2608.27866/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Multimodal and embodied agents, one of 16 directions in LLM agents and tool use.
- Builds on 29 of its 40 references that are in the map.
- First posted on arXiv 28 Aug 2026.

## Placed in

- LLM agents and tool use / [Multimodal and embodied agents](https://mlsanity.com/agents/multimodal-and-embodied-agents/)

## Builds on (most cited first, 29 of 29)

- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [WebArena: A Realistic Web Environment for Building Autonomous Agents](https://mlsanity.com/paper/2307.13854/) (2023)
- [Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling](https://mlsanity.com/paper/2412.05271/) (2024)
- [Mind2Web: Towards a Generalist Agent for the Web](https://mlsanity.com/paper/2306.06070/) (2023)
- [Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations](https://mlsanity.com/paper/2312.08935/) (2023)
- [Reasoning with Language Model is Planning with World Model](https://mlsanity.com/paper/2305.14992/) (2023)
- [CogAgent: A Visual Language Model for GUI Agents](https://mlsanity.com/paper/2312.08914/) (2023)
- [The Dawn of LMMs: Preliminary Explorations with GPT-4V(ision)](https://mlsanity.com/paper/2309.17421/) (2023)
- [Scaling Relationship on Learning Mathematical Reasoning with Large Language Models](https://mlsanity.com/paper/2308.01825/) (2023)
- [OS-ATLAS: A Foundation Action Model for Generalist GUI Agents](https://mlsanity.com/paper/2410.23218/) (2024)
- [Step-DPO: Step-wise Preference Optimization for Long-chain Reasoning of LLMs](https://mlsanity.com/paper/2406.18629/) (2024)
- [ShowUI: One Vision-Language-Action Model for GUI Visual Agent](https://mlsanity.com/paper/2411.17465/) (2024)
- [Skywork-Reward-V2: Scaling Preference Data Curation via Human-AI Synergy](https://mlsanity.com/paper/2507.01352/) (2025)
- [Dual Learning for Machine Translation](https://mlsanity.com/paper/1611.00179/) (2016)
- [SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvement](https://mlsanity.com/paper/2504.07934/) (2025)
- [GUI Agents with Foundation Models: A Comprehensive Survey](https://mlsanity.com/paper/2411.04890/) (2024)
- [Agent-R: Training Language Model Agents to Reflect via Iterative Self-Training](https://mlsanity.com/paper/2501.11425/) (2025)
- [Process Reward Model with Q-Value Rankings](https://mlsanity.com/paper/2410.11287/) (2024)
- [ExACT: Teaching AI Agents to Explore with Reflective-MCTS and Exploratory Learning](https://mlsanity.com/paper/2410.02052/) (2024)
- [OpenWebVoyager: Building Multimodal Web Agents via Iterative Real-World Exploration, Feedback and Optimization](https://mlsanity.com/paper/2410.19609/) (2024)
- [GUI-Xplore: Empowering Generalizable GUI Agents with One Exploration](https://mlsanity.com/paper/2503.17709/) (2025)
- [ProgRM: Build Better GUI Agents with Progress Rewards](https://mlsanity.com/paper/2505.18121/) (2025)
- [HandMeThat: Human-Robot Communication in Physical and Social Environments](https://mlsanity.com/paper/2310.03779/) (2023)
- [DUAL-REFLECT: Enhancing Large Language Models for Reflective Translation through Dual Learning Feedback Mechanisms](https://mlsanity.com/paper/2406.07232/) (2024)
- [Infer Human's Intentions Before Following Natural Language Instructions](https://mlsanity.com/paper/2409.18073/) (2024)
- [MobileA3gent: Training Mobile GUI Agents Using Decentralized Self-Sourced Data from Diverse Users](https://mlsanity.com/paper/2502.02982/) (2025)
- [GROOT-2: Weakly Supervised Multi-Modal Instruction Following Agents](https://mlsanity.com/paper/2412.10410/) (2024)
- [Boosting Virtual Agent Learning and Reasoning: A Step-Wise, Multi-Dimensional, and Generalist Reward Model with Benchmark](https://mlsanity.com/paper/2503.18665/) (2025)

## Abstract (arXiv)

Achieving virtual agents capable of automating tasks across diverse digital environments remains a pivotal challenge in Embodied AI. While Multimodal Large Language Models (MLLMs) offer enhanced visual perception and reasoning, their agentic deployment faces three challenges: costly data annotation, imprecise action-intent alignment, and inefficient exploration from discarded failed trajectories. To address these, we introduce Iron, an intent-aligned, self-improved, and annotation-efficient framework for training GUI agents. Iron employs a novel dual learning strategy that utilizes a stepwise cycle-consistent (SCC) reward to achieve fine-grained alignment between low-level actions and high-level intents, thereby improving instruction grounding and intent understanding. Concurrently, Iron introduces a hindsight reproduction mechanism to repurpose failed trajectories for training, improving both learning efficiency and task diversity. Extensive experiments demonstrate that Iron-trained generalist agents consistently improve performance on cross-environment and cross-device tasks, outperforming models trained with three times more data. Iron also achieves a substantial 25.06% relative improvement on unseen web tasks, with further gains observed on inherently complex tasks, demonstrating the feasibility of building more capable virtual agents.