Tongtong Feng

Tongtong Feng, Xin Wang, Zekai Zhou et al.

Completing Long-Horizon (LH) tasks in open-ended worlds is an important yet difficult problem for embodied agents. Existing approaches suffer from two key challenges: (1) they heavily rely on experiences obtained from human-created data or curricula, failing to autonomously update and select multimodal experiences, and (2) they may encounter catastrophic forgetting issues when faced with new tasks, failing to autonomously update world knowledge. To solve these challenges, this paper presents {\bf EvolvingAgent}, a curriculum self-evolving agent with a continual World Model (WM), which can autonomously complete various LH tasks across environments through self-planning, self-control, and self-reflection, without human intervention. Specifically, EvolvingAgent contains three modules, i.e., i) the experience-driven task planner, which uses an LLM along with multimodal experiences to convert LH tasks into executable sub-tasks; ii) the WM-guided action controller, which leverages WM to generate low-level actions and incorporates a self-verification mechanism to update multimodal experiences; iii) the Curriculum Learning (CL) -based reflector, which implements a two-stage CL algorithm to select multimodal experiences for task-adaptive WM updates. By building a planner-controller-reflector closed-loop dynamic, the continual WM for EvolvingAgent can autonomously update multimodal experiences and world knowledge. We conducted extensive experiments on Minecraft, compared with existing methods, EvolvingAgent can improve 111.74{\%} average success rate, reduce more than 6x ineffective actions, and generalize to the Atari environment with human-level performance.

Tongtong Feng, Qing Li, Xin Wang et al.

Cross-view geo-localization in GNSS-denied environments aims to determine an unknown location by matching drone-view images with the correct geo-tagged satellite-view images from a large gallery. Recent research shows that learning discriminative image representations under specific weather conditions can significantly enhance performance. However, the frequent occurrence of unseen extreme weather conditions hinders progress. This paper introduces MCGF, a Multi-weather Cross-view Geo-localization Framework designed to dynamically adapt to unseen weather conditions. MCGF establishes a joint optimization between image restoration and geo-localization using denoising diffusion models. For image restoration, MCGF incorporates a shared encoder and a lightweight restoration module to help the backbone eliminate weather-specific information. For geo-localization, MCGF uses EVA-02 as a backbone for feature extraction, with cross-entropy loss for training and cosine distance for testing. Extensive experiments on University160k-WX demonstrate that MCGF achieves competitive results for geo-localization in varying weather conditions.

Yu-Wei Zhan, Xin Wang, Hong Chen et al.

Video Large Language Models (Video LLMs) have shown impressive performance across a wide range of video-language tasks. However, they often fail in scenarios requiring a deeper understanding of physical dynamics. This limitation primarily arises from their reliance on appearance-based matching. Incorporating physical motion modeling is crucial for deeper video understanding, but presents three key challenges: (1) motion signals are often entangled with appearance variations, making it difficult to extract clean physical cues; (2) effective motion modeling requires not only continuous-time motion representations but also capturing physical dynamics; and (3) collecting accurate annotations for physical attributes is costly and often impractical. To address these issues, we propose PhyVLLM, a physical-guided video-language framework that explicitly incorporates physical motion into Video LLMs. Specifically, PhyVLLM disentangles visual appearance and object motion through a dual-branch encoder. To model physical dynamics over time, we incorporate a Neural Ordinary Differential Equation (Neural ODE) module, which generates differentiable physical dynamic representations. The resulting motion-aware representations are projected into the token space of a pretrained LLM, enabling physics reasoning without compromising the model's original multimodal capabilities. To circumvent the need for explicit physical labels, PhyVLLM employs a self-supervised manner to model the continuous evolution of object motion. Experimental results demonstrate that PhyVLLM significantly outperforms state-of-the-art Video LLMs on both physical reasoning and general video understanding tasks, highlighting the advantages of incorporating explicit physical modeling.

Yu-Wei Zhan, Xin Wang, Pengzhe Mao et al.

Building generalist embodied agents requires a unified system that can interpret multimodal goals, model environment dynamics, and execute reliable actions across diverse real-world tasks. Multimodal large language models (MLLMs) offer strong semantic priors and cross-modal generalization, while world models (WMs) provide actionable latent dynamics for prediction and control. Their combination holds promise for open-ended embodied intelligence, yet introduces two key challenges: (1) establishing a tight coupling between the semantic intent from MLLMs and the dynamic state representations within the WM's latent space, and (2) achieving task-aware adaptability that supports multi-task learning and cross-environment generalization. To address these limitations, we propose BiTAgent, a task-aware dynamic joint framework that enables bidirectional coupling between MLLMs and WMs. BiTAgent establishes two complementary pathways: a forward path that injects MLLM representations into the WM's latent space for semantically guided imagination, and a backward path where WM-generated feedback refines the MLLM's semantic space via dense text-conditioned rewards. This bidirectional interaction is realized through three synergistic components: Task-Aware Dynamic Joint Learning, Task-Aware Behavior Learning, and MLLM-WM Joint Optimization, which together harmonize semantic reasoning and dynamic prediction. Extensive experiments across multi-task and cross-environment settings demonstrate superior stability and generalization over state-of-the-art baselines, marking a step toward open-ended embodied learning.

Tongtong Feng, Xin Wang, Wenwu Zhu

Embodied Artificial Intelligence (AI) is an intelligent system formed by agents and their environment through active perception, embodied cognition, and action interaction. Existing embodied AI remains confined to human-crafted setting, in which agents are trained on given memory and construct models for given tasks, enabling fixed embodiments to interact with relatively static environments. Such methods fail in in-the-wild setting characterized by variable embodiments and dynamic open environments. This paper introduces self-evolving embodied AI, a new paradigm in which agents operate based on their changing state and environment with memory self-updating, task self-switching, environment self-prediction, embodiment self-adaptation, and model self-evolution, aiming to achieve continually adaptive intelligence with autonomous evolution. Specifically, we present the definition, framework, components, and mechanisms of self-evolving embodied AI, systematically review state-of-the-art works for realized components, discuss practical applications, and point out future research directions. We believe that self-evolving embodied AI enables agents to autonomously learn and interact with environments in a human-like manner and provide a new perspective toward general artificial intelligence.

Ziyu Shang, Haoran Liu, Rongchao Zhang et al.

Generating consistent human images with controllable pose and appearance is essential for applications in virtual try on, image editing, and digital human creation. Current methods often suffer from occlusions, garment style drift, and pose misalignment. We propose Pose-guided Multi-view Multimodal Diffusion (PMMD), a diffusion framework that synthesizes photorealistic person images conditioned on multi-view references, pose maps, and text prompts. A multimodal encoder jointly models visual views, pose features, and semantic descriptions, which reduces cross modal discrepancy and improves identity fidelity. We further design a ResCVA module to enhance local detail while preserving global structure, and a cross modal fusion module that integrates image semantics with text throughout the denoising pipeline. Experiments on the DeepFashion MultiModal dataset show that PMMD outperforms representative baselines in consistency, detail preservation, and controllability. Project page and code are available at https://github.com/ZANMANGLOOPYE/PMMD.

Hui Liu, Jiacheng Gu, Xiyuan Huang et al.

Accurate sports prediction is a crucial skill for professional coaches, which can assist in developing effective training strategies and scientific competition tactics. Traditional methods often use complex mathematical statistical techniques to boost predictability, but this often is limited by dataset scale and has difficulty handling long-term predictions with variable distributions, notably underperforming when predicting point-set-game multi-level matches. To deal with this challenge, this paper proposes TM2, a TCDformer-based Momentum Transfer Model for long-term sports prediction, which encompasses a momentum encoding module and a prediction module based on momentum transfer. TM2 initially encodes momentum in large-scale unstructured time series using the local linear scaling approximation (LLSA) module. Then it decomposes the reconstructed time series with momentum transfer into trend and seasonal components. The final prediction results are derived from the additive combination of a multilayer perceptron (MLP) for predicting trend components and wavelet attention mechanisms for seasonal components. Comprehensive experimental results show that on the 2023 Wimbledon men's tournament datasets, TM2 significantly surpasses existing sports prediction models in terms of performance, reducing MSE by 61.64% and MAE by 63.64%.

Yutong Shen, Hangxu Liu, Lei Zhang et al.

Long-Horizon (LH) tasks in Human-Scene Interaction (HSI) are complex multi-step tasks that require continuous planning, sequential decision-making, and extended execution across domains to achieve the final goal. However, existing methods heavily rely on skill chaining by concatenating pre-trained subtasks, with environment observations and self-state tightly coupled, lacking the ability to generalize to new combinations of environments and skills, failing to complete various LH tasks across domains. To solve this problem, this paper presents ALAS, a cross-domain learning framework for LH tasks via biologically inspired dual-stream disentanglement. Inspired by the brain's "where-what" dual pathway mechanism, ALAS comprises two core modules: i) an environment learning module for spatial understanding, which captures object functions, spatial relationships, and scene semantics, achieving cross-domain transfer through complete environment-self disentanglement; ii) a skill learning module for task execution, which processes self-state information including joint degrees of freedom and motor patterns, enabling cross-skill transfer through independent motor pattern encoding. We conducted extensive experiments on various LH tasks in HSI scenes. Compared with existing methods, ALAS can achieve an average subtasks success rate improvement of 23\% and average execution efficiency improvement of 29\%.

Tongtong Feng, Xin Wang, Yu-Gang Jiang et al.

Embodied Artificial Intelligence (AI) is an intelligent system paradigm for achieving Artificial General Intelligence (AGI), serving as the cornerstone for various applications and driving the evolution from cyberspace to physical systems. Recent breakthroughs in Large Language Models (LLMs) and World Models (WMs) have drawn significant attention for embodied AI. On the one hand, LLMs empower embodied AI via semantic reasoning and task decomposition, bringing high-level natural language instructions and low-level natural language actions into embodied cognition. On the other hand, WMs empower embodied AI by building internal representations and future predictions of the external world, facilitating physical law-compliant embodied interactions. As such, this paper comprehensively explores the literature in embodied AI from basics to advances, covering both LLM driven and WM driven works. In particular, we first present the history, key technologies, key components, and hardware systems of embodied AI, as well as discuss its development via looking from unimodal to multimodal angle. We then scrutinize the two burgeoning fields of embodied AI, i.e., embodied AI with LLMs/multimodal LLMs (MLLMs) and embodied AI with WMs, meticulously delineating their indispensable roles in end-to-end embodied cognition and physical laws-driven embodied interactions. Building upon the above advances, we further share our insights on the necessity of the joint MLLM-WM driven embodied AI architecture, shedding light on its profound significance in enabling complex tasks within physical worlds. In addition, we examine representative applications of embodied AI, demonstrating its wide applicability in real-world scenarios. Last but not least, we point out future research directions of embodied AI that deserve further investigation.

Yutong Shen, Hangxu Liu, Lei Zhang et al.

Long-Horizon (LH) tasks in Human-Scene Interaction (HSI) are complex multi-step tasks that require continuous planning, sequential decision-making, and extended execution across domains to achieve the final goal. However, existing methods heavily rely on skill chaining by concatenating pre-trained subtasks, with environment observations and self-state tightly coupled, lacking the ability to generalize to new combinations of environments and skills, failing to complete various LH tasks across domains. To solve this problem, this paper presents DETACH, a cross-domain learning framework for LH tasks via biologically inspired dual-stream disentanglement. Inspired by the brain's "where-what" dual pathway mechanism, DETACH comprises two core modules: i) an environment learning module for spatial understanding, which captures object functions, spatial relationships, and scene semantics, achieving cross-domain transfer through complete environment-self disentanglement; ii) a skill learning module for task execution, which processes self-state information including joint degrees of freedom and motor patterns, enabling cross-skill transfer through independent motor pattern encoding. We conducted extensive experiments on various LH tasks in HSI scenes. Compared with existing methods, DETACH can achieve an average subtasks success rate improvement of 23% and average execution efficiency improvement of 29%.

Mingzi Wang, Yuan Meng, Chen Tang et al.

The co-design of neural network architectures, quantization precisions, and hardware accelerators offers a promising approach to achieving an optimal balance between performance and efficiency, particularly for model deployment on resource-constrained edge devices. In this work, we propose the JAQ Framework, which jointly optimizes the three critical dimensions. However, effectively automating the design process across the vast search space of those three dimensions poses significant challenges, especially when pursuing extremely low-bit quantization. Specifical, the primary challenges include: (1) Memory overhead in software-side: Low-precision quantization-aware training can lead to significant memory usage due to storing large intermediate features and latent weights for back-propagation, potentially causing memory exhaustion. (2) Search time-consuming in hardware-side: The discrete nature of hardware parameters and the complex interplay between compiler optimizations and individual operators make the accelerator search time-consuming. To address these issues, JAQ mitigates the memory overhead through a channel-wise sparse quantization (CSQ) scheme, selectively applying quantization to the most sensitive components of the model during optimization. Additionally, JAQ designs BatchTile, which employs a hardware generation network to encode all possible tiling modes, thereby speeding up the search for the optimal compiler mapping strategy. Extensive experiments demonstrate the effectiveness of JAQ, achieving approximately 7% higher Top-1 accuracy on ImageNet compared to previous methods and reducing the hardware search time per iteration to 0.15 seconds.

Yanlong Wang, Hang Yu, Jian Xu et al.

Large time series foundation models often adopt channel-independent architectures to handle varying data dimensions, but this design ignores crucial cross-channel dependencies. Concurrently, existing multimodal approaches have not fully exploited the power of large vision models (LVMs) to interpret spatiotemporal data. Additionally, there remains significant unexplored potential in leveraging the advantages of information extraction from different modalities to enhance time series forecasting performance. To address these gaps, we propose the VIFO, a cross-modal forecasting model. VIFO uniquely renders multivariate time series into image, enabling pre-trained LVM to extract complex cross-channel patterns that are invisible to channel-independent models. These visual features are then aligned and fused with representations from the time series modality. By freezing the LVM and training only 7.45% of its parameters, VIFO achieves competitive performance on multiple benchmarks, offering an efficient and effective solution for capturing cross-variable relationships in

Tongtong Feng, Xin Wang, Feilin Han et al.

Swarm UAV autonomous flight for Embodied Long-Horizon (ELH) tasks is crucial for advancing the low-altitude economy. However, existing methods focus only on specific basic tasks due to dataset limitations, failing in real-world deployment for ELH tasks. ELH tasks are not mere concatenations of basic tasks, requiring handling long-term dependencies, maintaining embodied persistent states, and adapting to dynamic goal shifts. This paper presents U2UData+, the first large-scale swarm UAV autonomous flight dataset for ELH tasks and the first scalable swarm UAV data online collection and algorithm closed-loop verification platform. The dataset is captured by 15 UAVs in autonomous collaborative flights for ELH tasks, comprising 12 scenes, 720 traces, 120 hours, 600 seconds per trajectory, 4.32M LiDAR frames, and 12.96M RGB frames. This dataset also includes brightness, temperature, humidity, smoke, and airflow values covering all flight routes. The platform supports the customization of simulators, UAVs, sensors, flight algorithms, formation modes, and ELH tasks. Through a visual control window, this platform allows users to collect customized datasets through one-click deployment online and to verify algorithms by closed-loop simulation. U2UData+ also introduces an ELH task for wildlife conservation and provides comprehensive benchmarks with 9 SOTA models. U2UData+ can be found at https://fengtt42.github.io/U2UData-2/.