arXiv subjects

Bingchen Miao

Publications and source records attributed to Bingchen Miao.

9 recordsLinked to original sources

arXiv2026

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

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.

cs.CV↗

arXiv2026

Learning to Adapt: Self-Improving Web Agent via Cognitive-Aware Exploration

Recent advances in Multimodal Large Language Models (MLLMs) have led to promising progress in web agents. However, existing web agents often rely on handcrafted execution pipelines or expensive expert trajectories, limiting their adaptability to complex, dynamic environments. To address these challenges, we propose SCALE (Self-Cognitive-Aware Learning and Exploration), which leverages three adversarial roles, Selector, Predictor, and Judger to autonomously discover the agent's limitations and expand its cognitive boundaries through environmental exploration. Moreover, we propose SCALE-Hop, a graph exploration strategy that facilitates global planning and helps agents avoid local exploration traps. To further support learning, we construct SCALE-20k, a large-scale dataset collected from 19 real-world websites, containing diverse task types and structured demonstrations generated from SCALE's exploration traces. Experimental results show that our approach significantly improves the performance and generalization of multiple MLLMs in various web environments. Our framework offers a scalable and generalizable solution for building truly autonomous and adaptive web agents.

cs.AI↗

arXiv2026

CORE: Code-based Inverse Self-Training Framework with Graph Expansion for Virtual Agents

The development of Multimodal Virtual Agents has made significant progress through the integration of Multimodal Large Language Models. However, mainstream training paradigms face key challenges: Behavior Cloning is simple and effective through imitation but suffers from low behavioral diversity, while Reinforcement Learning is capable of discovering novel strategies through exploration but heavily relies on manually designed reward functions. To address the conflict between these two methods, we present CORE, a Code-based Inverse Self-Training Framework with Graph Expansion that bridges imitation and exploration, offering a novel training framework that promotes behavioral diversity while eliminating the reliance on manually reward design. Specifically, we introduce Semantic Code Abstraction to automatically infers reward functions from expert demonstrations without manual design. The inferred reward function, referred to as the Label Function, is executable code that verifies one key step within a task. Building on this, we propose Strategy Graph Expansion to enhance in-domain behavioral diversity, which constructs a multi-path graph called Strategy Graph that captures diverse valid solutions beyond expert demonstrations. Furthermore, we introduce Trajectory-Guided Extrapolation, which enriches out-of-domain behavioral diversity by utilizing both successful and failed trajectories to expand the task space. Experiments on Web and Android platforms demonstrate that CORE significantly improves both overall performance and generalization, highlighting its potential as a robust and generalizable training paradigm for building powerful virtual agents.

cs.LG↗

arXiv2025

RoboTidy : A 3D Gaussian Splatting Household Tidying Benchmark for Embodied Navigation and Action

Household tidying is an important application area, yet current benchmarks neither model user preferences nor support mobility, and they generalize poorly, making it hard to comprehensively assess integrated language-to-action capabilities. To address this, we propose RoboTidy, a unified benchmark for language-guided household tidying that supports Vision-Language-Action (VLA) and Vision-Language-Navigation (VLN) training and evaluation. RoboTidy provides 500 photorealistic 3D Gaussian Splatting (3DGS) household scenes (covering 500 objects and containers) with collisions, formulates tidying as an "Action (Object, Container)" list, and supplies 6.4k high-quality manipulation demonstration trajectories and 1.5k naviagtion trajectories to support both few-shot and large-scale training. We also deploy RoboTidy in the real world for object tidying, establishing an end-to-end benchmark for household tidying. RoboTidy offers a scalable platform and bridges a key gap in embodied AI by enabling holistic and realistic evaluation of language-guided robots.

cs.RO↗

arXiv2025

Towards Physically Executable 3D Gaussian for Embodied Navigation

3D Gaussian Splatting (3DGS), a 3D representation method with photorealistic real-time rendering capabilities, is regarded as an effective tool for narrowing the sim-to-real gap. However, it lacks fine-grained semantics and physical executability for Visual-Language Navigation (VLN). To address this, we propose SAGE-3D (Semantically and Physically Aligned Gaussian Environments for 3D Navigation), a new paradigm that upgrades 3DGS into an executable, semantically and physically aligned environment. It comprises two components: (1) Object-Centric Semantic Grounding, which adds object-level fine-grained annotations to 3DGS; and (2) Physics-Aware Execution Jointing, which embeds collision objects into 3DGS and constructs rich physical interfaces. We release InteriorGS, containing 1K object-annotated 3DGS indoor scene data, and introduce SAGE-Bench, the first 3DGS-based VLN benchmark with 2M VLN data. Experiments show that 3DGS scene data is more difficult to converge, while exhibiting strong generalizability, improving baseline performance by 31% on the VLN-CE Unseen task. Our data and code are available at: https://sage-3d.github.io.

cs.CV↗

arXiv2025

What Limits Virtual Agent Application? OmniBench: A Scalable Multi-Dimensional Benchmark for Essential Virtual Agent Capabilities

As multimodal large language models (MLLMs) advance, MLLM-based virtual agents have demonstrated remarkable performance. However, existing benchmarks face significant limitations, including uncontrollable task complexity, extensive manual annotation with limited scenarios, and a lack of multidimensional evaluation. In response to these challenges, we introduce OmniBench, a self-generating, cross-platform, graph-based benchmark with an automated pipeline for synthesizing tasks of controllable complexity through subtask composition. To evaluate the diverse capabilities of virtual agents on the graph, we further present OmniEval, a multidimensional evaluation framework that includes subtask-level evaluation, graph-based metrics, and comprehensive tests across 10 capabilities. Our synthesized dataset contains 36k graph-structured tasks across 20 scenarios, achieving a 91\% human acceptance rate. Training on our graph-structured data shows that it can more efficiently guide agents compared to manually annotated data. We conduct multidimensional evaluations for various open-source and closed-source models, revealing their performance across various capabilities and paving the way for future advancements. Our project is available at https://omni-bench.github.io/.

cs.CV↗

arXiv2025

Boosting Virtual Agent Learning and Reasoning: A Step-Wise, Multi-Dimensional, and Generalist Reward Model with Benchmark

The development of Generalist Virtual Agents (GVAs) has shown significant promise in autonomous task execution. However, current training paradigms face critical limitations, including reliance on outcome supervision and labor-intensive human annotations. To address these challenges, we propose Similar, a Step-Wise Multi-Dimensional Generalist Reward Model, which offers fine-grained signals for agent training and can choose better action for inference-time scaling. Specifically, we begin by systematically defining five dimensions for evaluating agent actions. Building on this framework, we design an MCTS-P algorithm to automatically collect and annotate step-wise, five-dimensional agent execution data. Using this data, we train Similar with the Triple-M strategy. Furthermore, we introduce the first benchmark in the virtual agent domain for step-wise, multi-dimensional reward model training and evaluation, named SRM. This benchmark consists of two components: SRMTrain, which serves as the training set for Similar, and SRMEval, a manually selected test set for evaluating the reward model. Experimental results demonstrate that Similar, through its step-wise, multi-dimensional assessment and synergistic gain, provides GVAs with effective intermediate signals during both training and inference-time scaling. The project is available at https://github.com/antgroup/Similar.

cs.CV↗

arXiv2024

Generalist Virtual Agents: A Survey on Autonomous Agents Across Digital Platforms

In this paper, we introduce the Generalist Virtual Agent (GVA), an autonomous entity engineered to function across diverse digital platforms and environments, assisting users by executing a variety of tasks. This survey delves into the evolution of GVAs, tracing their progress from early intelligent assistants to contemporary implementations that incorporate large-scale models. We explore both the philosophical underpinnings and practical foundations of GVAs, addressing their developmental challenges and the methodologies currently employed in their design and operation. By presenting a detailed taxonomy of GVA environments, tasks, and capabilities, this paper aims to bridge the theoretical and practical aspects of GVAs, concluding those that operate in environments closely mirroring the real world are more likely to demonstrate human-like intelligence. We discuss potential future directions for GVA research, highlighting the necessity for realistic evaluation metrics and the enhancement of long-sequence decision-making capabilities to advance the field toward more systematic or embodied applications. This work not only synthesizes the existing body of literature but also proposes frameworks for future investigations, contributing significantly to the ongoing development of intelligent systems.

cs.MA↗

arXiv2024

Robust Modality-incomplete Anomaly Detection: A Modality-instructive Framework with Benchmark

Multimodal Industrial Anomaly Detection (MIAD), which utilizes 3D point clouds and 2D RGB images to identify abnormal regions in products, plays a crucial role in industrial quality inspection. However, traditional MIAD settings assume that all 2D and 3D modalities are paired, ignoring the fact that multimodal data collected from the real world is often imperfect due to missing modalities. Additionally, models trained on modality-incomplete data are prone to overfitting. Therefore, MIAD models that demonstrate robustness against modality-incomplete data are highly desirable in practice. To address this, we introduce a pioneering study that comprehensively investigates Modality-Incomplete Industrial Anomaly Detection (MIIAD), and under the guidance of experts, we construct the MIIAD Bench with rich modality-missing settings to account for imperfect learning environments with incomplete multimodal information. As expected, we find that most existing MIAD methods perform poorly on the MIIAD Bench, leading to significant performance degradation. To tackle this challenge, we propose a novel two-stage Robust modAlity-aware fusing and Detecting framewoRk, abbreviated as RADAR. Specifically: i) We propose Modality-incomplete Instruction to guide the multimodal Transformer to robustly adapt to various modality-incomplete scenarios, and implement adaptive parameter learning based on HyperNetwork. ii) Then, we construct a Double-Pseudo Hybrid Module to highlight the uniqueness of modality combinations, mitigating overfitting issues and further enhancing the robustness of the MIIAD model. Our experimental results demonstrate that the proposed RADAR significantly outperforms traditional MIAD methods on our newly created MIIAD dataset, proving its practical application value.

cs.CV↗