Wenchao Ding

Xu Zhao, Wenchao Ding, Yongqi An et al.

The recently proposed segment anything model (SAM) has made a significant influence in many computer vision tasks. It is becoming a foundation step for many high-level tasks, like image segmentation, image caption, and image editing. However, its huge computation costs prevent it from wider applications in industry scenarios. The computation mainly comes from the Transformer architecture at high-resolution inputs. In this paper, we propose a speed-up alternative method for this fundamental task with comparable performance. By reformulating the task as segments-generation and prompting, we find that a regular CNN detector with an instance segmentation branch can also accomplish this task well. Specifically, we convert this task to the well-studied instance segmentation task and directly train the existing instance segmentation method using only 1/50 of the SA-1B dataset published by SAM authors. With our method, we achieve a comparable performance with the SAM method at 50 times higher run-time speed. We give sufficient experimental results to demonstrate its effectiveness. The codes and demos will be released at https://github.com/CASIA-IVA-Lab/FastSAM.

Yupeng Zheng, Jichao Peng, Weize Li et al. · cmu, tsinghua

We introduce World In Your Hands (WIYH), a large-scale open-source ecosystem comprising over 1,000 hours of human manipulation data collected in-the-wild with millimeter-scale motion accuracy. Specifically, WIYH includes (1) the Oracle Suite, a wearable data collection kit with an auto-labeling pipeline for accurate motion capture; (2) the WIYH Dataset, featuring over 1,000 hours of multimodal manipulation data across hundreds of skills in diverse real-world scenarios; and (3) extensive annotations and benchmarks supporting tasks from perception to action. Furthermore, experiments based on the WIYH ecosystem show that integrating WIYH's human-centric data improves robotic manipulation success rates from 8% to 60% in cluttered scenes. World In Your Hands provides a foundation for advancing human-centric data collection and cross-embodiment policy learning. All data and hardware design will be open-source.

Mingzhe Gao, Jieru Zhao, Zhe Lin et al.

Recently, the use of large language models (LLMs) for software code generation, e.g., C/C++ and Python, has proven a great success. However, LLMs still suffer from low syntactic and functional correctness when it comes to the generation of register-transfer level (RTL) code, such as Verilog. To address this issue, in this paper, we develop AutoVCoder, a systematic open-source framework that significantly improves the LLMs' correctness of generating Verilog code and enhances the quality of its output at the same time. Our framework integrates three novel techniques, including a high-quality hardware dataset generation approach, a two-round LLM fine-tuning method and a domain-specific retrieval-augmented generation (RAG) mechanism. Experimental results demonstrate that AutoVCoder outperforms both industrial and academic LLMs in Verilog code generation. Specifically, AutoVCoder shows a 0.5% and 2.2% improvement in functional correctness on the EvalMachine and EvalHuman benchmarks compared with BetterV, and also achieves a 3.4% increase in syntax correctness and a 3.4% increase in functional correctness on the RTLLM benchmark compared with RTLCoder.

Kunyun Wang, Yuhang Zheng, Yupeng Zheng et al.

Modern robotic policies increasingly rely on action chunking to execute complex tasks in the physical world. While action chunking improves temporal consistency at moderate action frequencies, it becomes insufficient when the action frequency is further increased (e.g., to 60~Hz). At such high frequencies, policies often fail to generate actions that are both temporally smooth and spatially consistent. We address this challenge by shifting high-frequency action learning from the action space to a latent space with variational autoencoder (VAE). This formulation significantly improves both temporal and spatial consistency of high-frequency control. To enable smooth real-time execution, we further introduce Reuse-then-Refine, a chunk-level refine strategy that improves continuity between adjacent action chunks under asynchronous inference. As a result, robots controlled by our policy can execute complex contact-rich tasks continuously, with less pauses and jerky motions. Experiments on three real-world contact-rich robotic tasks show that our approach consistently completes tasks with smooth motions. Our code and data are available at https://github.com/tars-robotics/RTR.

Yijie Zhou, Likun Cai, Xianhui Cheng et al.

In the era of big data and large models, automatic annotating functions for multi-modal data are of great significance for real-world AI-driven applications, such as autonomous driving and embodied AI. Unlike traditional closed-set annotation, open-vocabulary annotation is essential to achieve human-level cognition capability. However, there are few open-vocabulary auto-labeling systems for multi-modal 3D data. In this paper, we introduce OpenAnnotate3D, an open-source open-vocabulary auto-labeling system that can automatically generate 2D masks, 3D masks, and 3D bounding box annotations for vision and point cloud data. Our system integrates the chain-of-thought capabilities of Large Language Models (LLMs) and the cross-modality capabilities of vision-language models (VLMs). To the best of our knowledge, OpenAnnotate3D is one of the pioneering works for open-vocabulary multi-modal 3D auto-labeling. We conduct comprehensive evaluations on both public and in-house real-world datasets, which demonstrate that the system significantly improves annotation efficiency compared to manual annotation while providing accurate open-vocabulary auto-annotating results.

Guan Shen, Jieru Zhao, Zeke Wang et al.

Along with the fast evolution of deep neural networks, the hardware system is also developing rapidly. As a promising solution achieving high scalability and low manufacturing cost, multi-accelerator systems widely exist in data centers, cloud platforms, and SoCs. Thus, a challenging problem arises in multi-accelerator systems: selecting a proper combination of accelerators from available designs and searching for efficient DNN mapping strategies. To this end, we propose MARS, a novel mapping framework that can perform computation-aware accelerator selection, and apply communication-aware sharding strategies to maximize parallelism. Experimental results show that MARS can achieve 32.2% latency reduction on average for typical DNN workloads compared to the baseline, and 59.4% latency reduction on heterogeneous models compared to the corresponding state-of-the-art method.

Julong Wei, Shanshuai Yuan, Pengfei Li et al.

The rise of multi-modal large language models(MLLMs) has spurred their applications in autonomous driving. Recent MLLM-based methods perform action by learning a direct mapping from perception to action, neglecting the dynamics of the world and the relations between action and world dynamics. In contrast, human beings possess world model that enables them to simulate the future states based on 3D internal visual representation and plan actions accordingly. To this end, we propose OccLLaMA, an occupancy-language-action generative world model, which uses semantic occupancy as a general visual representation and unifies vision-language-action(VLA) modalities through an autoregressive model. Specifically, we introduce a novel VQVAE-like scene tokenizer to efficiently discretize and reconstruct semantic occupancy scenes, considering its sparsity and classes imbalance. Then, we build a unified multi-modal vocabulary for vision, language and action. Furthermore, we enhance LLM, specifically LLaMA, to perform the next token/scene prediction on the unified vocabulary to complete multiple tasks in autonomous driving. Extensive experiments demonstrate that OccLLaMA achieves competitive performance across multiple tasks, including 4D occupancy forecasting, motion planning, and visual question answering, showcasing its potential as a foundation model in autonomous driving.

Hongjin Chen, Wei Zhang, Pengfei Li et al.

Realizing interactive whole-body control for multi-humanoid systems is critical for unlocking complex collaborative capabilities in shared environments. Although recent advancements have significantly enhanced the agility of individual robots, bridging the gap to physically coupled multi-humanoid interaction remains challenging, primarily due to severe kinematic mismatches and complex contact dynamics. To address this, we introduce Rhythm, the first unified framework enabling real-world deployment of dual-humanoid systems for complex, physically plausible interactions. Our framework integrates three core components: (1) an Interaction-Aware Motion Retargeting (IAMR) module that generates feasible humanoid interaction references from human data; (2) an Interaction-Guided Reinforcement Learning (IGRL) policy that masters coupled dynamics via graph-based rewards; and (3) a real-world deployment system that enables robust transfer of dual-humanoid interaction. Extensive experiments on physical Unitree G1 robots demonstrate that our framework achieves robust interactive whole-body control, successfully transferring diverse behaviors such as hugging and dancing from simulation to reality.

Yonghao Wang, Yang Yin, Hongqin Lyu et al.

LLMs can generate SystemVerilog assertions (SVAs) from natural language specs, but single-pass outputs often lack functional coverage due to limited IC design understanding. We propose CoverAssert, an iterative framework that clusters semantic and AST-based structural features of assertions, maps them to specifications, and uses functional coverage feedback to guide LLMs in prioritizing uncovered points. Experiments on four open-source designs show that integrating CoverAssert with AssertLLM and Spec2Assertion improves average improvements of 9.57 % in branch coverage, 9.64 % in statement coverage, and 15.69 % in toggle coverage.

Zhenyu Ning, Jieru Zhao, Qihao Jin et al.

Multimodal Large Language Models (MLLMs) are distinguished by their multimodal comprehensive ability and widely used in many real-world applications including GPT-4o, autonomous driving and robotics. Despite their impressive performance, the multimodal inputs always incur long context. The inference under long context requires caching massive Key and Value states (KV cache) of previous tokens, which introduces high latency and excessive memory consumption. Due to this reason, it is challenging to deploy streaming inference of MLLMs on edge devices, which largely constrains the power and usage of MLLMs in real-world applications. In this paper, we introduce Inf-MLLM, an efficient inference framework for MLLMs, which enable streaming inference of MLLM on a single GPU with infinite context. Inf-MLLM is based on our key observation of the attention pattern in both LLMs and MLLMs called "attention saddles". Thanks to the newly discovered attention pattern, Inf-MLLM maintains a size-constrained KV cache by dynamically caching recent tokens and relevant tokens. Furthermore, Inf-MLLM proposes attention bias, a novel approach to enable MLLMs to capture long-term dependency. We show that Inf-MLLM enables multiple LLMs and MLLMs to achieve stable performance over 4M-token long texts and multi-round conversations with 1-hour-long videos on a single GPU. In addition, Inf-MLLM exhibits superior streaming reasoning quality than existing methods such as StreamingLLM and 2x speedup than H2O.

Xiang Li, Yupeng Zheng, Pengfei Li et al.

Occupancy prediction provides critical geometric and semantic understanding for robotics but faces efficiency-accuracy trade-offs. Current dense methods suffer computational waste on empty voxels, while sparse query-based approaches lack robustness in diverse and complex indoor scenes. In this paper, we propose DiScene, a novel sparse query-based framework that leverages multi-level distillation to achieve efficient and robust occupancy prediction. In particular, our method incorporates two key innovations: (1) a Multi-level Consistent Knowledge Distillation strategy, which transfers hierarchical representations from large teacher models to lightweight students through coordinated alignment across four levels, including encoder-level feature alignment, query-level feature matching, prior-level spatial guidance, and anchor-level high-confidence knowledge transfer and (2) a Teacher-Guided Initialization policy, employing optimized parameter warm-up to accelerate model convergence. Validated on the Occ-Scannet benchmark, DiScene achieves 23.2 FPS without depth priors while outperforming our baseline method, OPUS, by 36.1% and even better than the depth-enhanced version, OPUS†. With depth integration, DiScene† attains new SOTA performance, surpassing EmbodiedOcc by 3.7% with 1.62$\times$ faster inference speed. Furthermore, experiments on the Occ3D-nuScenes benchmark and in-the-wild scenarios demonstrate the versatility of our approach in various environments. Code and models can be accessed at https://github.com/getterupper/DiScene.

Jie Jia, Yaofeng Su, Zeyu Bao et al.

Occlusion-aware prediction remains a critical challenge in autonomous driving due to the inherent uncertainty of unobserved regions. Existing approaches either overestimate risk based on reachable states or struggle to predict accurate trajectories under high occlusion uncertainty. To address these limitations, we propose a unified risk map modeling and learning framework for partially observable environments. Our method integrates traffic flow risk and collision risk through spatiotemporal modeling, enabling fine-grained assessment of occlusion-induced hazards. To address the scarcity of scenarios involving occluded interactions, we introduce a diffusion-based scenario generation framework that produces realistic yet adversarial scenarios. We integrate the modeling and learning of a unified risk map into a framework that supports risk-aware planning under partial observability. Experiments on the Waymo Open Motion Dataset show that our method significantly outperforms the state-of-the-art occlusion-aware baseline, improving minimum time-to-collision by 0.78 times and average time-to-collision by 1.67 times. The proposed framework offers a comprehensive and practical solution for risk-aware planning in partially observable environments.

Yuhang Zheng, Songen Gu, Weize Li et al.

Contact-rich manipulation tasks, such as wiping and assembly, require accurate perception of contact forces, friction changes, and state transitions that cannot be reliably inferred from vision alone. Despite growing interest in visuo-tactile manipulation, progress is constrained by two persistent limitations: existing datasets are small in scale and narrow in task coverage, and current methods treat tactile signals as passive observations rather than using them to model contact dynamics or enable closed-loop control explicitly. In this paper, we present \textbf{OmniViTac}, a large-scale visuo-tactile-action dataset comprising $21{,}000+$ trajectories across $86$ tasks and $100+$ objects, organized into six physics-grounded interaction patterns. Building on this dataset, we propose \textbf{OmniVTA}, a world-model-based visuo-tactile manipulation framework that integrates four tightly coupled modules: a self-supervised tactile encoder, a two-stream visuo-tactile world model for predicting short-horizon contact evolution, a contact-aware fusion policy for action generation, and a 60Hz reflexive controller that corrects deviations between predicted and observed tactile signals in a closed loop. Real-robot experiments across all six interaction categories show that OmniVTA outperforms existing methods and generalizes well to unseen objects and geometric configurations, confirming the value of combining predictive contact modeling with high-frequency tactile feedback for contact-rich manipulation. All data, models, and code will be made publicly available on the project website at https://mrsecant.github.io/OmniVTA.

Zhihua Hua, Junli Wang, Pengfei LI et al.

Global navigation information and local scene understanding are two crucial components of autonomous driving systems. However, our experimental results indicate that many end-to-end autonomous driving systems tend to over-rely on local scene understanding while failing to utilize global navigation information. These systems exhibit weak correlation between their planning capabilities and navigation input, and struggle to perform navigation-following in complex scenarios. To overcome this limitation, we propose the Sequential Navigation Guidance (SNG) framework, an efficient representation of global navigation information based on real-world navigation patterns. The SNG encompasses both navigation paths for constraining long-term trajectories and turn-by-turn (TBT) information for real-time decision-making logic. We constructed the SNG-QA dataset, a visual question answering (VQA) dataset based on SNG that aligns global and local planning. Additionally, we introduce an efficient model SNG-VLA that fuses local planning with global planning. The SNG-VLA achieves state-of-the-art performance through precise navigation information modeling without requiring auxiliary loss functions from perception tasks. Project page: SNG-VLA

Tian Niu, Kaizhao Zhang, Zhongxue Gan et al.

Planning safe trajectories for autonomous vehicles is essential for operational safety but remains extremely challenging due to the complex interactions among traffic participants. Recent autonomous driving frameworks have focused on improving prediction accuracy to explicitly model these interactions. However, some methods overlook the significant influence of the ego vehicle's planning on the possible trajectories of other agents, which can alter prediction accuracy and lead to unsafe planning decisions. In this paper, we propose a novel motion Planning approach by Simulation with learning-based parallel scenario prediction (PS). PS deduces predictions iteratively based on Monte Carlo Tree Search (MCTS), jointly inferring scenarios that cooperate with the ego vehicle's planning set. Our method simulates possible scenes and calculates their costs after the ego vehicle executes potential actions. To balance and prune unreasonable actions and scenarios, we adopt MCTS as the foundation to explore possible future interactions encoded within the prediction network. Moreover, the query-centric trajectory prediction streamlines our scene generation, enabling a sophisticated framework that captures the mutual influence between other agents' predictions and the ego vehicle's planning. We evaluate our framework on the Argoverse 2 dataset, and the results demonstrate that our approach effectively achieves parallel ego vehicle planning.

Yupeng Zheng, Xiang Li, Songen Gu et al.

Recent advances in Vision-Language-Action (VLA) models have opened new avenues for robot manipulation, yet existing methods exhibit limited efficiency and a lack of high-level knowledge and spatial awareness. To address these challenges, we propose PokeVLA, a lightweight yet powerful foundation model for embodied manipulation that effectively infuses vision-language understanding into action learning. Our framework introduces a two-stage training paradigm: first, we pre-train a compact vision-language model (PokeVLM) on a curated multimodal dataset of 2.4M samples encompassing spatial grounding, affordance, and embodied reasoning tasks; second, we inject manipulation-relevant representations into the action space through multi-view goal-aware semantics learning, geometry alignment, and a novel action expert. Extensive experiments demonstrate state-of-the-art performance on the LIBERO-Plus benchmark and in real-world deployment, outperforming comparable baselines in success rate and robustness under diverse perturbations. To foster reproducibility and community progress, we will open-source our code, model weights, and the scripts for the curated pre-training dataset. Project page: https://getterupper.github.io/PokeVLA

Zhijian Qiao, Zehuan Yu, Tong Li et al.

Crowdsourcing enables scalable autonomous driving map construction, but low-cost sensor noise hinders quality from improving with data volume. We propose CSMapping, a system that produces accurate semantic maps and topological road centerlines whose quality consistently increases with more crowdsourced data. For semantic mapping, we train a latent diffusion model on HD maps (optionally conditioned on SD maps) to learn a generative prior of real-world map structure, without requiring paired crowdsourced/HD-map supervision. This prior is incorporated via constrained MAP optimization in latent space, ensuring robustness to severe noise and plausible completion in unobserved areas. Initialization uses a robust vectorized mapping module followed by diffusion inversion; optimization employs efficient Gaussian-basis reparameterization, projected gradient descent zobracket multi-start, and latent-space factor-graph for global consistency. For topological mapping, we apply confidence-weighted k-medoids clustering and kinematic refinement to trajectories, yielding smooth, human-like centerlines robust to trajectory variation. Experiments on nuScenes, Argoverse 2, and a large proprietary dataset achieve state-of-the-art semantic and topological mapping performance, with thorough ablation and scalability studies.

Yonghao Wang, Hongqin Lyu, Boling Chen et al.

Functional verification has become the most time-consuming phase in IC development, and Assertion-Based Verification (ABV) is key to reducing debugging time. However, existing LLM-based assertion generation methods typically pursue indiscriminate verification, aiming for maximal coverage without considering signal criticality, whereas industrial practice demands maximizing coverage with minimal verification cost. Consequently, identifying signals that have the greatest impact on design functionality and error propagation-enabling a shift from indiscriminate to targeted verification-remains a key challenge. To address this, we propose AgileAssert, a key signal-driven assertion generation framework that constructs RTL semantic graphs and identifies the top-K critical signals via a hybrid scoring and selection mechanism, followed by structure-aware RTL slicing to provide the LLM with precise targets and contextual information, thereby guiding LLMs to generate tightly constrained targeted assertions for efficient verification. Evaluated on block- and CPU-level designs, with an average 66.68% reduction in assertions, our approach outperforms three existing SOTA methods, and significantly improving coverage metrics while reducing input token consumption by 64%. In mutation testing, when our approach surpasses existing methods in error detection rate, the average number of assertions used decreases by 72.74%.

Zhaoyang Wei, Wenchao Ding, Yanchao Hao et al.

Models capable of "thinking with images" by dynamically grounding their reasoning in visual evidence represent a major leap in multimodal AI. However, replicating and advancing this ability is non-trivial, with current methods often trapped between the instability of end-to-end reinforcement learning (RL) and the rigidity of supervised fine-tuning (SFT). This leads to models that either struggle to learn or lack the cognitive flexibility required for complex, real-world scenes. To navigate this dilemma, we introduce GRiP (Guided Reasoning and Perception), a novel two-stage training framework that cultivates robust and flexible visual grounded reasoning by explicitly guiding the model's perceptual focus and logical pathways. GRiP's core lies in its cognitive-enhanced RL stage, which features two key innovations: (1) a Salience-Weighted IoU Reward that incentivizes the model to prioritize the localization of mission-critical objects over trivial distractors, and (2) a Multi-Heuristic Reward that encourages cognitive flexibility by rewarding diverse yet logically valid reasoning pathways. Initialized from the Qwen2.5-VL-7B model, GRiP demonstrates significant performance gains across multiple challenging benchmarks. It achieves state-of-the-art results among open-source models on the highly challenging TreeBench and V* Bench, proving its effectiveness in complex visual reasoning. Our work demonstrates that moving beyond simplistic rewards and instead guiding models with cognitively-inspired signals for what to see and how to think is crucial for unlocking the next level of multimodal intelligence. The code will be made publicly available.

Jieru Zhao, Tingyuan Liang, Liang Feng et al.

Fast and accurate depth estimation, or stereo matching, is essential in embedded stereo vision systems, requiring substantial design effort to achieve an appropriate balance among accuracy, speed and hardware cost. To reduce the design effort and achieve the right balance, we propose FP-Stereo for building high-performance stereo matching pipelines on FPGAs automatically. FP-Stereo consists of an open-source hardware-efficient library, allowing designers to obtain the desired implementation instantly. Diverse methods are supported in our library for each stage of the stereo matching pipeline and a series of techniques are developed to exploit the parallelism and reduce the resource overhead. To improve the usability, FP-Stereo can generate synthesizable C code of the FPGA accelerator with our optimized HLS templates automatically. To guide users for the right design choice meeting specific application requirements, detailed comparisons are performed on various configurations of our library to investigate the accuracy/speed/cost trade-off. Experimental results also show that FP-Stereo outperforms the state-of-the-art FPGA design from all aspects, including 6.08% lower error, 2x faster speed, 30% less resource usage and 40% less energy consumption. Compared to GPU designs, FP-Stereo achieves the same accuracy at a competitive speed while consuming much less energy.

Ke Wu, Kaizhao Zhang, Zhiwei Zhang et al.

Online dense mapping of urban scenes forms a fundamental cornerstone for scene understanding and navigation of autonomous vehicles. Recent advancements in mapping methods are mainly based on NeRF, whose rendering speed is too slow to meet online requirements. 3D Gaussian Splatting (3DGS), with its rendering speed hundreds of times faster than NeRF, holds greater potential in online dense mapping. However, integrating 3DGS into a street-view dense mapping framework still faces two challenges, including incomplete reconstruction due to the absence of geometric information beyond the LiDAR coverage area and extensive computation for reconstruction in large urban scenes. To this end, we propose HGS-Mapping, an online dense mapping framework in unbounded large-scale scenes. To attain complete construction, our framework introduces Hybrid Gaussian Representation, which models different parts of the entire scene using Gaussians with distinct properties. Furthermore, we employ a hybrid Gaussian initialization mechanism and an adaptive update method to achieve high-fidelity and rapid reconstruction. To the best of our knowledge, we are the first to integrate Gaussian representation into online dense mapping of urban scenes. Our approach achieves SOTA reconstruction accuracy while only employing 66% number of Gaussians, leading to 20% faster reconstruction speed.

Muer Tie, Julong Wei, Zhengjun Wang et al.

Online construction of open-ended language scenes is crucial for robotic applications, where open-vocabulary interactive scene understanding is required. Recently, neural implicit representation has provided a promising direction for online interactive mapping. However, implementing open-vocabulary scene understanding capability into online neural implicit mapping still faces three challenges: lack of local scene updating ability, blurry spatial hierarchical semantic segmentation and difficulty in maintaining multi-view consistency. To this end, we proposed O2V-mapping, which utilizes voxel-based language and geometric features to create an open-vocabulary field, thus allowing for local updates during online training process. Additionally, we leverage a foundational model for image segmentation to extract language features on object-level entities, achieving clear segmentation boundaries and hierarchical semantic features. For the purpose of preserving consistency in 3D object properties across different viewpoints, we propose a spatial adaptive voxel adjustment mechanism and a multi-view weight selection method. Extensive experiments on open-vocabulary object localization and semantic segmentation demonstrate that O2V-mapping achieves online construction of language scenes while enhancing accuracy, outperforming the previous SOTA method.

Xiaze Zhang, Ziheng Ding, Qi Jing et al.

Point clouds have shown significant potential in various domains, including Simultaneous Localization and Mapping (SLAM). However, existing approaches either rely on dense point clouds to achieve high localization accuracy or use generalized descriptors to reduce map size. Unfortunately, these two aspects seem to conflict with each other. To address this limitation, we propose a unified architecture, DeepPointMap, achieving excellent preference on both aspects. We utilize neural network to extract highly representative and sparse neural descriptors from point clouds, enabling memory-efficient map representation and accurate multi-scale localization tasks (e.g., odometry and loop-closure). Moreover, we showcase the versatility of our framework by extending it to more challenging multi-agent collaborative SLAM. The promising results obtained in these scenarios further emphasize the effectiveness and potential of our approach.

Songen Gu, Yuhang Zheng, Weize Li et al.

Recently, end-to-end robotic manipulation models have gained significant attention for their generalizability and scalability. However, they often suffer from limited robustness to camera viewpoint changes when training with a fixed camera. In this paper, we propose VistaBot, a novel framework that integrates feed-forward geometric models with video diffusion models to achieve view-robust closed-loop manipulation without requiring camera calibration at test time. Our approach consists of three key components: 4D geometry estimation, view synthesis latent extraction, and latent action learning. VistaBot is integrated into both action-chunking (ACT) and diffusion-based ($π_0$) policies and evaluated across simulation and real-world tasks. We further introduce the View Generalization Score (VGS) as a new metric for comprehensive evaluation of cross-view generalization. Results show that VistaBot improves VGS by 2.79$\times$ and 2.63$\times$ over ACT and $π_0$, respectively, while also achieving high-quality novel view synthesis. Our contributions include a geometry-aware synthesis model, a latent action planner, a new benchmark metric, and extensive validation across diverse environments. The code and models will be made publicly available.

Ke Wu, Zicheng Zhang, Muer Tie et al.

VINGS-Mono is a monocular (inertial) Gaussian Splatting (GS) SLAM framework designed for large scenes. The framework comprises four main components: VIO Front End, 2D Gaussian Map, NVS Loop Closure, and Dynamic Eraser. In the VIO Front End, RGB frames are processed through dense bundle adjustment and uncertainty estimation to extract scene geometry and poses. Based on this output, the mapping module incrementally constructs and maintains a 2D Gaussian map. Key components of the 2D Gaussian Map include a Sample-based Rasterizer, Score Manager, and Pose Refinement, which collectively improve mapping speed and localization accuracy. This enables the SLAM system to handle large-scale urban environments with up to 50 million Gaussian ellipsoids. To ensure global consistency in large-scale scenes, we design a Loop Closure module, which innovatively leverages the Novel View Synthesis (NVS) capabilities of Gaussian Splatting for loop closure detection and correction of the Gaussian map. Additionally, we propose a Dynamic Eraser to address the inevitable presence of dynamic objects in real-world outdoor scenes. Extensive evaluations in indoor and outdoor environments demonstrate that our approach achieves localization performance on par with Visual-Inertial Odometry while surpassing recent GS/NeRF SLAM methods. It also significantly outperforms all existing methods in terms of mapping and rendering quality. Furthermore, we developed a mobile app and verified that our framework can generate high-quality Gaussian maps in real time using only a smartphone camera and a low-frequency IMU sensor. To the best of our knowledge, VINGS-Mono is the first monocular Gaussian SLAM method capable of operating in outdoor environments and supporting kilometer-scale large scenes.

Zhenyu Ning, Guangda Liu, Qihao Jin et al.

Recent developments in Video Large Language Models (Video LLMs) have enabled models to process long video sequences and demonstrate remarkable performance. Nonetheless, studies predominantly focus on offline video question answering, neglecting memory usage and response speed that are essential in various real-world applications, such as Deepseek services, autonomous driving, and robotics. To mitigate these challenges, we propose $\textbf{LiveVLM}$, a training-free framework specifically designed for streaming, online video understanding and real-time interaction. Unlike existing works that process videos only after one question is posed, LiveVLM constructs an innovative streaming-oriented KV cache to process video streams in real-time, retain long-term video details and eliminate redundant KVs, ensuring prompt responses to user queries. For continuous video streams, LiveVLM generates and compresses video key-value tensors (video KVs) to reserve visual information while improving memory efficiency. Furthermore, when a new question is proposed, LiveVLM incorporates an online question-answering process that efficiently fetches both short-term and long-term visual information, while minimizing interference from redundant context. Extensive experiments demonstrate that LiveVLM enables the foundation LLaVA-OneVision model to process 44$\times$ number of frames on the same device, and achieves up to 5$\times$ speedup in response speed compared with SoTA online methods at an input of 256 frames, while maintaining the same or better model performance.

Zicheng Zhang, Xiangting Meng, Ke Wu et al.

Recent progress in feed-forward 3D Gaussian Splatting (3DGS) has notably improved rendering quality. However, the spatially uniform and highly redundant 3DGS map generated by previous feed-forward 3DGS methods limits their integration into downstream reconstruction tasks. We propose SparseSplat, the first feed-forward 3DGS model that adaptively adjusts Gaussian density according to scene structure and information richness of local regions, yielding highly compact 3DGS maps. To achieve this, we propose entropy-based probabilistic sampling, generating large, sparse Gaussians in textureless areas and assigning small, dense Gaussians to regions with rich information. Additionally, we designed a specialized point cloud network that efficiently encodes local context and decodes it into 3DGS attributes, addressing the receptive field mismatch between the general 3DGS optimization pipeline and feed-forward models. Extensive experimental results demonstrate that SparseSplat can achieve state-of-the-art rendering quality with only 22% of the Gaussians and maintain reasonable rendering quality with only 1.5% of the Gaussians. Project page: https://victkk.github.io/SparseSplat-page/.

Zicheng Zhang, Ke Wu, Xiangting Meng et al.

Monocular 3D Gaussian Splatting SLAM suffers from critical limitations in time efficiency, geometric accuracy, and multi-view consistency. These issues stem from the time-consuming $\textit{Train-from-Scratch}$ optimization and the lack of inter-frame scale consistency from single-frame geometry priors. We contend that a feed-forward paradigm, leveraging multi-frame context to predict Gaussian attributes directly, is crucial for addressing these challenges. We present Flash-Mono, a system composed of three core modules: a feed-forward prediction frontend, a 2D Gaussian Splatting mapping backend, and an efficient hidden-state-based loop closure module. We trained a recurrent feed-forward frontend model that progressively aggregates multi-frame visual features into a hidden state via cross attention and jointly predicts camera poses and per-pixel Gaussian properties. By directly predicting Gaussian attributes, our method bypasses the burdensome per-frame optimization required in optimization-based GS-SLAM, achieving a $\textbf{10x}$ speedup while ensuring high-quality rendering. The power of our recurrent architecture extends beyond efficient prediction. The hidden states act as compact submap descriptors, facilitating efficient loop closure and global $\mathrm{Sim}(3)$ optimization to mitigate the long-standing challenge of drift. For enhanced geometric fidelity, we replace conventional 3D Gaussian ellipsoids with 2D Gaussian surfels. Extensive experiments demonstrate that Flash-Mono achieves state-of-the-art performance in both tracking and mapping quality, highlighting its potential for embodied perception and real-time reconstruction applications. Project page: https://victkk.github.io/flash-mono.

Chenqi Zhang, Yu Feng, Jieru Zhao et al.

3D Gaussian Splatting (3DGS) has gained popularity for its efficiency and sparse Gaussian-based representation. However, 3DGS struggles to meet the real-time requirement of 90 frames per second (FPS) on resource-constrained mobile devices, achieving only 2 to 9 FPS.Existing accelerators focus on compute efficiency but overlook memory efficiency, leading to redundant DRAM traffic. We introduce STREAMINGGS, a fully streaming 3DGS algorithm-architecture co-design that achieves fine-grained pipelining and reduces DRAM traffic by transforming from a tile-centric rendering to a memory-centric rendering. Results show that our design achieves up to 45.7 $\times$ speedup and 62.9 $\times$ energy savings over mobile Ampere GPUs.

Shanshuai Yuan, Julong Wei, Muer Tie et al.

Vision-based 3D semantic occupancy prediction is critical for autonomous driving, enabling unified modeling of static infrastructure and dynamic agents. In practice, autonomous vehicles may repeatedly traverse identical geographic locations under varying environmental conditions, such as weather fluctuations and illumination changes. Existing methods in 3D occupancy prediction predominantly integrate adjacent temporal contexts. However, these works neglect to leverage perceptual information, which is acquired from historical traversals of identical geographic locations. In this paper, we propose Longterm Memory Prior Occupancy (LMPOcc), the first 3D occupancy prediction methodology that exploits long-term memory priors derived from historical traversal perceptual outputs. We introduce a plug-and-play architecture that integrates long-term memory priors to enhance local perception while simultaneously constructing global occupancy representations. To adaptively aggregate prior features and current features, we develop an efficient lightweight Current-Prior Fusion module. Moreover, we propose a model-agnostic prior format to ensure compatibility across diverse occupancy prediction baselines. LMPOcc achieves state-of-the-art performance validated on the Occ3D-nuScenes benchmark, especially on static semantic categories. Additionally, experimental results demonstrate LMPOcc's ability to construct global occupancy through multi-vehicle crowdsourcing.

Xiangting Meng, Jiaqi Yang, Mingshu Chen et al.

In the realm of object pose estimation, scenarios involving both dynamic objects and moving cameras are prevalent. However, the scarcity of corresponding real-world datasets significantly hinders the development and evaluation of robust pose estimation models. This is largely attributed to the inherent challenges in accurately annotating object poses in dynamic scenes captured by moving cameras. To bridge this gap, this paper presents a novel dataset DynOPETs and a dedicated data acquisition and annotation pipeline tailored for object pose estimation and tracking in such unconstrained environments. Our efficient annotation method innovatively integrates pose estimation and pose tracking techniques to generate pseudo-labels, which are subsequently refined through pose graph optimization. The resulting dataset offers accurate pose annotations for dynamic objects observed from moving cameras. To validate the effectiveness and value of our dataset, we perform comprehensive evaluations using 18 state-of-the-art methods, demonstrating its potential to accelerate research in this challenging domain. The dataset will be made publicly available to facilitate further exploration and advancement in the field.

Kunyun Wang, Shuo Yang, Jieru Zhao et al.

Deep learning models have become pivotal in the field of video processing and is increasingly critical in practical applications such as autonomous driving and object detection. Although Vision Transformers (ViTs) have demonstrated their power, Convolutional Neural Networks (CNNs) remain a highly efficient and high-performance choice for feature extraction and encoding. However, the intensive computational demands of convolution operations hinder its broader adoption as a video encoder. Given the inherent temporal continuity in video frames, changes between consecutive frames are minimal, allowing for the skipping of redundant computations. This technique, which we term as Diff Computation, presents two primary challenges. First, Diff Computation requires to cache intermediate feature maps to ensure the correctness of non-linear computations, leading to significant memory consumption. Second, the imbalance of sparsity among layers, introduced by Diff Computation, incurs accuracy degradation. To address these issues, we propose a memory-efficient scheduling method to eliminate memory overhead and an online adjustment mechanism to minimize accuracy degradation. We integrate these techniques into our framework, SparseTem, to seamlessly support various CNN-based video encoders. SparseTem achieves speedup of 1.79x for EfficientDet and 4.72x for CRNN, with minimal accuracy drop and no additional memory overhead. Extensive experimental results demonstrate that SparseTem sets a new state-of-the-art by effectively utilizing temporal continuity to accelerate CNN-based video encoders.

Wenchao Ding, Jieru Zhao, Yubin Chu et al.

There is extensive literature on perceiving road structures by fusing various sensor inputs such as lidar point clouds and camera images using deep neural nets. Leveraging the latest advance of neural architects (such as transformers) and bird-eye-view (BEV) representation, the road cognition accuracy keeps improving. However, how to cognize the ``road'' for automated vehicles where there is no well-defined ``roads'' remains an open problem. For example, how to find paths inside intersections without HD maps is hard since there is neither an explicit definition for ``roads'' nor explicit features such as lane markings. The idea of this paper comes from a proverb: it becomes a way when people walk on it. Although there are no ``roads'' from sensor readings, there are ``roads'' from tracks of other vehicles. In this paper, we propose FlowMap, a path generation framework for automated vehicles based on traffic flows. FlowMap is built by extending our previous work RoadMap, a light-weight semantic map, with an additional traffic flow layer. A path generation algorithm on traffic flow fields (TFFs) is proposed to generate human-like paths. The proposed framework is validated using real-world driving data and is amenable to generating paths for super complicated intersections without using HD maps.

Wenchao Ding, Lu Zhang, Jing Chen et al.

In this paper, we present an Efficient Planning System for automated vehicles In highLy interactive envirONments (EPSILON). EPSILON is an efficient interaction-aware planning system for automated driving, and is extensively validated in both simulation and real-world dense city traffic. It follows a hierarchical structure with an interactive behavior planning layer and an optimization-based motion planning layer. The behavior planning is formulated from a partially observable Markov decision process (POMDP), but is much more efficient than naively applying a POMDP to the decision-making problem. The key to efficiency is guided branching in both the action space and observation space, which decomposes the original problem into a limited number of closed-loop policy evaluations. Moreover, we introduce a new driver model with a safety mechanism to overcome the risk induced by the potential imperfectness of prior knowledge. For motion planning, we employ a spatio-temporal semantic corridor (SSC) to model the constraints posed by complex driving environments in a unified way. Based on the SSC, a safe and smooth trajectory is optimized, complying with the decision provided by the behavior planner. We validate our planning system in both simulations and real-world dense traffic, and the experimental results show that our EPSILON achieves human-like driving behaviors in highly interactive traffic flow smoothly and safely without being over-conservative compared to the existing planning methods.

Haoran Song, Di Luan, Wenchao Ding et al.

Predicting the future trajectories of on-road vehicles is critical for autonomous driving. In this paper, we introduce a novel prediction framework called PRIME, which stands for Prediction with Model-based Planning. Unlike recent prediction works that utilize neural networks to model scene context and produce unconstrained trajectories, PRIME is designed to generate accurate and feasibility-guaranteed future trajectory predictions. PRIME guarantees the trajectory feasibility by exploiting a model-based generator to produce future trajectories under explicit constraints and enables accurate multimodal prediction by utilizing a learning-based evaluator to select future trajectories. We conduct experiments on the large-scale Argoverse Motion Forecasting Benchmark, where PRIME outperforms the state-of-the-art methods in prediction accuracy, feasibility, and robustness under imperfect tracking.

Haoran Song, Wenchao Ding, Yuxuan Chen et al.

It is critical to predict the motion of surrounding vehicles for self-driving planning, especially in a socially compliant and flexible way. However, future prediction is challenging due to the interaction and uncertainty in driving behaviors. We propose planning-informed trajectory prediction (PiP) to tackle the prediction problem in the multi-agent setting. Our approach is differentiated from the traditional manner of prediction, which is only based on historical information and decoupled with planning. By informing the prediction process with the planning of ego vehicle, our method achieves the state-of-the-art performance of multi-agent forecasting on highway datasets. Moreover, our approach enables a novel pipeline which couples the prediction and planning, by conditioning PiP on multiple candidate trajectories of the ego vehicle, which is highly beneficial for autonomous driving in interactive scenarios.

Lu Zhang, Wenchao Ding, Jing Chen et al.

Decision-making in dense traffic scenarios is challenging for automated vehicles (AVs) due to potentially stochastic behaviors of other traffic participants and perception uncertainties (e.g., tracking noise and prediction errors, etc.). Although the partially observable Markov decision process (POMDP) provides a systematic way to incorporate these uncertainties, it quickly becomes computationally intractable when scaled to the real-world large-size problem. In this paper, we present an efficient uncertainty-aware decision-making (EUDM) framework, which generates long-term lateral and longitudinal behaviors in complex driving environments in real-time. The computation complexity is controlled to an appropriate level by two novel techniques, namely, the domain-specific closed-loop policy tree (DCP-Tree) structure and conditional focused branching (CFB) mechanism. The key idea is utilizing domain-specific expert knowledge to guide the branching in both action and intention space. The proposed framework is validated using both onboard sensing data captured by a real vehicle and an interactive multi-agent simulation platform. We also release the code of our framework to accommodate benchmarking.

Wenchao Ding, Lu Zhang, Jing Chen et al.

Planning safe trajectories for autonomous vehicles in complex urban environments is challenging since there are numerous semantic elements (such as dynamic agents, traffic lights and speed limits) to consider. These semantic elements may have different mathematical descriptions such as obstacle, constraint and cost. It is non-trivial to tune the effects from different combinations of semantic elements for a stable and generalizable behavior. In this paper, we propose a novel unified spatio-temporal semantic corridor (SSC) structure, which provides a level of abstraction for different types of semantic elements. The SSC consists of a series of mutually connected collision-free cubes with dynamical constraints posed by the semantic elements in the spatio-temporal domain. The trajectory generation problem then boils down to a general quadratic programming (QP) formulation. Thanks to the unified SSC representation, our framework can generalize to any combination of semantic elements. Moreover, our formulation provides a theoretical guarantee that the entire trajectory is safe and constraint-satisfied, by using the convex hull and hodograph properties of piecewise Bezier curve parameterization. We also release the code of our method to accommodate benchmarking.

Wenchao Ding, Wenliang Gao, Kaixuan Wang et al.

Trajectory replanning for quadrotors is essential to enable fully autonomous flight in unknown environments. Hierarchical motion planning frameworks, which combine path planning with path parameterization, are popular due to their time efficiency. However, the path planning cannot properly deal with non-static initial states of the quadrotor, which may result in non-smooth or even dynamically infeasible trajectories. In this paper, we present an efficient kinodynamic replanning framework by exploiting the advantageous properties of the B-spline, which facilitates dealing with the non-static state and guarantees safety and dynamical feasibility. Our framework starts with an efficient B-spline-based kinodynamic (EBK) search algorithm which finds a feasible trajectory with minimum control effort and time. To compensate for the discretization induced by the EBK search, an elastic optimization (EO) approach is proposed to refine the control point placement to the optimal location. Systematic comparisons against the state-of-the-art are conducted to validate the performance. Comprehensive onboard experiments using two different vision-based quadrotors are carried out showing the general applicability of the framework.

Wenchao Ding, Wenliang Gao, Kaixuan Wang et al.

We focus on a replanning scenario for quadrotors where considering time efficiency, non-static initial state and dynamical feasibility is of great significance. We propose a real-time B-spline based kinodynamic (RBK) search algorithm, which transforms a position-only shortest path search (such as A* and Dijkstra) into an efficient kinodynamic search, by exploring the properties of B-spline parameterization. The RBK search is greedy and produces a dynamically feasible time-parameterized trajectory efficiently, which facilitates non-static initial state of the quadrotor. To cope with the limitation of the greedy search and the discretization induced by a grid structure, we adopt an elastic optimization (EO) approach as a post-optimization process, to refine the control point placement provided by the RBK search. The EO approach finds the optimal control point placement inside an expanded elastic tube which represents the free space, by solving a Quadratically Constrained Quadratic Programming (QCQP) problem. We design a receding horizon replanner based on the local control property of B-spline. A systematic comparison of our method against two state-of-the-art methods is provided. We integrate our replanning system with a monocular vision-based quadrotor and validate our performance onboard.

Wenchao Ding, Jing Chen, Shaojie Shen

Anticipating possible behaviors of traffic participants is an essential capability of autonomous vehicles. Many behavior detection and maneuver recognition methods only have a very limited prediction horizon that leaves inadequate time and space for planning. To avoid unsatisfactory reactive decisions, it is essential to count long-term future rewards in planning, which requires extending the prediction horizon. In this paper, we uncover that clues to vehicle behaviors over an extended horizon can be found in vehicle interaction, which makes it possible to anticipate the likelihood of a certain behavior, even in the absence of any clear maneuver pattern. We adopt a recurrent neural network (RNN) for observation encoding, and based on that, we propose a novel vehicle behavior interaction network (VBIN) to capture the vehicle interaction from the hidden states and connection feature of each interaction pair. The output of our method is a probabilistic likelihood of multiple behavior classes, which matches the multimodal and uncertain nature of the distant future. A systematic comparison of our method against two state-of-the-art methods and another two baseline methods on a publicly available real highway dataset is provided, showing that our method has superior accuracy and advanced capability for interaction modeling.

Wenchao Ding, Shaojie Shen

In this paper, we present an online two-level vehicle trajectory prediction framework for urban autonomous driving where there are complex contextual factors, such as lane geometries, road constructions, traffic regulations and moving agents. Our method combines high-level policy anticipation with low-level context reasoning. We leverage a long short-term memory (LSTM) network to anticipate the vehicle's driving policy (e.g., forward, yield, turn left, turn right, etc.) using its sequential history observations. The policy is then used to guide a low-level optimization-based context reasoning process. We show that it is essential to incorporate the prior policy anticipation due to the multimodal nature of the future trajectory. Moreover, contrary to existing regression-based trajectory prediction methods, our optimization-based reasoning process can cope with complex contextual factors. The final output of the two-level reasoning process is a continuous trajectory that automatically adapts to different traffic configurations and accurately predicts future vehicle motions. The performance of the proposed framework is analyzed and validated in an emerging autonomous driving simulation platform (CARLA).