MIRAGE: Mobile Agents with Implicit Reasoning and Generative World Models

• URL: http://arxiv.org/abs/2606.04627v2
• Date: Sat, 06 Jun 2026 08:34:56 GMT
• Title: MIRAGE: Mobile Agents with Implicit Reasoning and Generative World Models
• Abstract summary: We introduce MIRAGE, a framework that learns continuous latent reasoning representations from visible textual reasoning traces.<n>On AndroidWorld, MIRAGE matches explicit chain-of-thought supervised fine-tuning in the 4B ablation with a 3-5x lower decoded-token budget.<n>On AndroidControl, it improves action grounding while generating over 75% fewer tokens.
• Score: 53.09772037247959
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Mobile agents are increasingly expected to operate everyday applications from screenshots and language goals, where reliable control requires reasoning over screen affordances, multi-step navigation, and future state changes. However, many agents externalize this computation as long textual chains of thought, which slows interaction, increases supervision cost, and complicates deployment. We introduce MIRAGE, a framework that learns continuous latent reasoning representations from visible textual reasoning traces. MIRAGE transfers explicit reasoning into compact hidden states, enabling the agent to reason internally without decoding long rationales. It also incorporates a generative world-model objective: latent reasoning vectors are aligned with future screenshots, encouraging the agent to anticipate upcoming interface states before acting. This turns hidden computation into both a compressed thought representation and a forward-looking model of environment dynamics. At inference time, MIRAGE reasons in continuous latent space, reducing token generation while improving execution efficiency. On AndroidWorld, MIRAGE matches explicit chain-of-thought supervised fine-tuning in the 4B ablation with a 3-5x lower decoded-token budget and improves a comparable instruction-tuned baseline by 10.2 points; on AndroidControl, it improves action grounding while generating over 75% fewer tokens.

Related papers

• Enfold: Folding World Model Imagination into Predictive Representations for Ultra-Efficient Embodied Control [41.15180918658398]
  We argue that world generative models are most reusable when their computation that constructs a future is reusable.<n>We present Enfold, which transfers this computation into a representation predicted from the current visual context and language instruction.<n>Across LIBERO, RoboTwin2.0, and real-robot tasks, Enfold supports strong control while reducing action latency by $3.7times$ relative to Fast--WAM, Enfold-Flash reaches $10.1times$.
  arXiv  Detail & Related papers  (2026-07-29T09:17:33Z)
• A Task-State Representation for Long-Horizon Mobile GUI Agents [53.359524744953575]
  Task-State Representation (TSR) is a training-free framework that explicitly decouples task state from sensory input.<n> Experiments across four mobile GUI benchmarks validate TSR's effectiveness, yielding up to a 12 absolute point increase in success rate.
  arXiv  Detail & Related papers  (2026-07-01T06:37:21Z)
• ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both [55.182037225013836]
  ATLAS is a framework in which a single discrete 'word', termed as a functional token, serves both as an agentic operation and a latent visual reasoning unit.<n> ATLAS achieves superior performance on challenging benchmarks while maintaining clear interpretability.
  arXiv  Detail & Related papers  (2026-05-14T17:59:55Z)
• M2A: Synergizing Mathematical and Agentic Reasoning in Large Language Models [9.712139528293946]
  We propose M2A, a novel paradigm that synergizes mathematical and agentic reasoning via model merging.<n>M2A operates directly in parameter space, identifies the feature subspace critical for agent behavior, and merges the mathematical reasoning task vector only along its null space.<n>Unlike SFT or RL, M2A requires no additional gradient-update and exposes the coefficient merging as a simple knob for controlling reasoning length.
  arXiv  Detail & Related papers  (2026-05-11T02:05:30Z)
• Latent Agents: A Post-Training Procedure for Internalized Multi-Agent Debate [16.68699018207298]
  Multi-agent debate has been shown to improve reasoning in large language models (LLMs)<n>We develop a framework that distills multi-agent debate into a single LLM through a two-stage fine-tuning pipeline.<n>Our internalized models match or exceed explicit multi-agent debate performance using up to 93% fewer tokens.
  arXiv  Detail & Related papers  (2026-04-27T18:06:03Z)
• LatentPilot: Scene-Aware Vision-and-Language Navigation by Dreaming Ahead with Latent Visual Reasoning [51.969318585152116]
  LatentPilot exploits future observations during training as a valuable data source to learn action-conditioned visual dynamics.<n>We propose a flywheel-style training mechanism that iteratively collects on-policy trajectories and retrains the model to better match the agent's behavior distribution.<n> Experiments on R2R-CE, RxR-CE, and R2R-PE benchmarks achieve new SOTA results, and real-robot tests across diverse environments demonstrate LatentPilot's superior understanding of environment-action dynamics in scene.
  arXiv  Detail & Related papers  (2026-03-31T02:21:59Z)
• Show, Don't Tell: Morphing Latent Reasoning into Image Generation [60.743575139661154]
  Text-to-image (T2I) generation has achieved remarkable progress, yet existing methods often lack the ability to dynamically reason and refine during generation.<n>We introduce LatentMorph, a framework that seamlessly integrates implicit latent reasoning into the T2I generation process.<n>By performing reasoning entirely in continuous latent spaces, LatentMorph avoids the bottlenecks of explicit reasoning and enables more adaptive self-refinement.
  arXiv  Detail & Related papers  (2026-02-02T15:29:48Z)
• MGA: Memory-Driven GUI Agent for Observation-Centric Interaction [30.45490249299358]
  We introduce the Memory-Driven GUI Agent (MGA), which reframes GUI interaction around the principle of observe first, then decide.<n>MGA achieves substantial gains in robustness, generalization, and efficiency compared to state-of-the-art baselines.
  arXiv  Detail & Related papers  (2025-10-28T08:19:58Z)
• MAPLE: A Mobile Agent with Persistent Finite State Machines for Structured Task Reasoning [46.18718721121415]
  We present MAPLE, a state-aware multi-agent framework that abstracts app interactions as a Finite State Machine (FSM)<n>We computationally model each UI screen as a discrete state and user actions as transitions, allowing the FSM to provide a structured representation of the app execution.<n> MAPLE consists of specialized agents responsible for four phases of task execution: planning, execution, verification, error recovery, and knowledge retention.
  arXiv  Detail & Related papers  (2025-05-29T16:08:51Z)
• Mobile-Agent-E: Self-Evolving Mobile Assistant for Complex Tasks [85.48034185086169]
  Mobile-Agent-E is a hierarchical multi-agent framework capable of self-evolution through past experience.<n>Mobile-Agent-E achieves a 22% absolute improvement over previous state-of-the-art approaches.
  arXiv  Detail & Related papers  (2025-01-20T20:35:46Z)
• MageBench: Bridging Large Multimodal Models to Agents [90.59091431806793]
  LMMs have shown impressive visual understanding capabilities, with the potential to be applied in agents.<n>Existing benchmarks mostly assess their reasoning abilities in language part.<n>MageBench is a reasoning capability oriented multimodal agent benchmark.
  arXiv  Detail & Related papers  (2024-12-05T17:08:19Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.