Baoxiong Jia

Ran Gong, Jiangyong Huang, Yizhou Zhao et al. · berkeley

Understanding the continuous states of objects is essential for task learning and planning in the real world. However, most existing task learning benchmarks assume discrete (e.g., binary) object goal states, which poses challenges for the learning of complex tasks and transferring learned policy from simulated environments to the real world. Furthermore, state discretization limits a robot's ability to follow human instructions based on the grounding of actions and states. To tackle these challenges, we present ARNOLD, a benchmark that evaluates language-grounded task learning with continuous states in realistic 3D scenes. ARNOLD is comprised of 8 language-conditioned tasks that involve understanding object states and learning policies for continuous goals. To promote language-instructed learning, we provide expert demonstrations with template-generated language descriptions. We assess task performance by utilizing the latest language-conditioned policy learning models. Our results indicate that current models for language-conditioned manipulations continue to experience significant challenges in novel goal-state generalizations, scene generalizations, and object generalizations. These findings highlight the need to develop new algorithms that address this gap and underscore the potential for further research in this area. Project website: https://arnold-benchmark.github.io.

Siyuan Huang, Zan Wang, Puhao Li et al. · pku

We introduce SceneDiffuser, a conditional generative model for 3D scene understanding. SceneDiffuser provides a unified model for solving scene-conditioned generation, optimization, and planning. In contrast to prior works, SceneDiffuser is intrinsically scene-aware, physics-based, and goal-oriented. With an iterative sampling strategy, SceneDiffuser jointly formulates the scene-aware generation, physics-based optimization, and goal-oriented planning via a diffusion-based denoising process in a fully differentiable fashion. Such a design alleviates the discrepancies among different modules and the posterior collapse of previous scene-conditioned generative models. We evaluate SceneDiffuser with various 3D scene understanding tasks, including human pose and motion generation, dexterous grasp generation, path planning for 3D navigation, and motion planning for robot arms. The results show significant improvements compared with previous models, demonstrating the tremendous potential of SceneDiffuser for the broad community of 3D scene understanding.

Baoxiong Jia, Ting Lei, Song-Chun Zhu et al.

Understanding human tasks through video observations is an essential capability of intelligent agents. The challenges of such capability lie in the difficulty of generating a detailed understanding of situated actions, their effects on object states (i.e., state changes), and their causal dependencies. These challenges are further aggravated by the natural parallelism from multi-tasking and partial observations in multi-agent collaboration. Most prior works leverage action localization or future prediction as an indirect metric for evaluating such task understanding from videos. To make a direct evaluation, we introduce the EgoTaskQA benchmark that provides a single home for the crucial dimensions of task understanding through question-answering on real-world egocentric videos. We meticulously design questions that target the understanding of (1) action dependencies and effects, (2) intents and goals, and (3) agents' beliefs about others. These questions are divided into four types, including descriptive (what status?), predictive (what will?), explanatory (what caused?), and counterfactual (what if?) to provide diagnostic analyses on spatial, temporal, and causal understandings of goal-oriented tasks. We evaluate state-of-the-art video reasoning models on our benchmark and show their significant gaps between humans in understanding complex goal-oriented egocentric videos. We hope this effort will drive the vision community to move onward with goal-oriented video understanding and reasoning.

Peiyu Yu, Sirui Xie, Xiaojian Ma et al.

Latent space Energy-Based Models (EBMs), also known as energy-based priors, have drawn growing interests in generative modeling. Fueled by its flexibility in the formulation and strong modeling power of the latent space, recent works built upon it have made interesting attempts aiming at the interpretability of text modeling. However, latent space EBMs also inherit some flaws from EBMs in data space; the degenerate MCMC sampling quality in practice can lead to poor generation quality and instability in training, especially on data with complex latent structures. Inspired by the recent efforts that leverage diffusion recovery likelihood learning as a cure for the sampling issue, we introduce a novel symbiosis between the diffusion models and latent space EBMs in a variational learning framework, coined as the latent diffusion energy-based model. We develop a geometric clustering-based regularization jointly with the information bottleneck to further improve the quality of the learned latent space. Experiments on several challenging tasks demonstrate the superior performance of our model on interpretable text modeling over strong counterparts.

Rundong Luo, Haoran Geng, Congyue Deng et al. · berkeley

Interactable objects are ubiquitous in our daily lives. Recent advances in 3D generative models make it possible to automate the modeling of these objects, benefiting a range of applications from 3D printing to the creation of robot simulation environments. However, while significant progress has been made in modeling 3D shapes and appearances, modeling object physics, particularly for interactable objects, remains challenging due to the physical constraints imposed by inter-part motions. In this paper, we tackle the problem of physically plausible part completion for interactable objects, aiming to generate 3D parts that not only fit precisely into the object but also allow smooth part motions. To this end, we propose a diffusion-based part generation model that utilizes geometric conditioning through classifier-free guidance and formulates physical constraints as a set of stability and mobility losses to guide the sampling process. Additionally, we demonstrate the generation of dependent parts, paving the way toward sequential part generation for objects with complex part-whole hierarchies. Experimentally, we introduce a new metric for measuring physical plausibility based on motion success rates. Our model outperforms existing baselines over shape and physical metrics, especially those that do not adequately model physical constraints. We also demonstrate our applications in 3D printing, robot manipulation, and sequential part generation, showing our strength in realistic tasks with the demand for high physical plausibility.

Jiangyong Huang, Silong Yong, Xiaojian Ma et al.

Leveraging massive knowledge from large language models (LLMs), recent machine learning models show notable successes in general-purpose task solving in diverse domains such as computer vision and robotics. However, several significant challenges remain: (i) most of these models rely on 2D images yet exhibit a limited capacity for 3D input; (ii) these models rarely explore the tasks inherently defined in 3D world, e.g., 3D grounding, embodied reasoning and acting. We argue these limitations significantly hinder current models from performing real-world tasks and approaching general intelligence. To this end, we introduce LEO, an embodied multi-modal generalist agent that excels in perceiving, grounding, reasoning, planning, and acting in the 3D world. LEO is trained with a unified task interface, model architecture, and objective in two stages: (i) 3D vision-language (VL) alignment and (ii) 3D vision-language-action (VLA) instruction tuning. We collect large-scale datasets comprising diverse object-level and scene-level tasks, which require considerable understanding of and interaction with the 3D world. Moreover, we meticulously design an LLM-assisted pipeline to produce high-quality 3D VL data. Through extensive experiments, we demonstrate LEO's remarkable proficiency across a wide spectrum of tasks, including 3D captioning, question answering, embodied reasoning, navigation and manipulation. Our ablative studies and scaling analyses further provide valuable insights for developing future embodied generalist agents. Code and data are available on project page.

Baoxiong Jia, Yu Liu, Siyuan Huang

The ability to decompose complex natural scenes into meaningful object-centric abstractions lies at the core of human perception and reasoning. In the recent culmination of unsupervised object-centric learning, the Slot-Attention module has played an important role with its simple yet effective design and fostered many powerful variants. These methods, however, have been exceedingly difficult to train without supervision and are ambiguous in the notion of object, especially for complex natural scenes. In this paper, we propose to address these issues by investigating the potential of learnable queries as initializations for Slot-Attention learning, uniting it with efforts from existing attempts on improving Slot-Attention learning with bi-level optimization. With simple code adjustments on Slot-Attention, our model, Bi-level Optimized Query Slot Attention, achieves state-of-the-art results on 3 challenging synthetic and 7 complex real-world datasets in unsupervised image segmentation and reconstruction, outperforming previous baselines by a large margin. We provide thorough ablative studies to validate the necessity and effectiveness of our design. Additionally, our model exhibits great potential for concept binding and zero-shot learning. Our work is made publicly available at https://bo-qsa.github.io

Xiongkun Linghu, Jiangyong Huang, Xuesong Niu et al.

Situation awareness is essential for understanding and reasoning about 3D scenes in embodied AI agents. However, existing datasets and benchmarks for situated understanding are limited in data modality, diversity, scale, and task scope. To address these limitations, we propose Multi-modal Situated Question Answering (MSQA), a large-scale multi-modal situated reasoning dataset, scalably collected leveraging 3D scene graphs and vision-language models (VLMs) across a diverse range of real-world 3D scenes. MSQA includes 251K situated question-answering pairs across 9 distinct question categories, covering complex scenarios within 3D scenes. We introduce a novel interleaved multi-modal input setting in our benchmark to provide text, image, and point cloud for situation and question description, resolving ambiguity in previous single-modality convention (e.g., text). Additionally, we devise the Multi-modal Situated Next-step Navigation (MSNN) benchmark to evaluate models' situated reasoning for navigation. Comprehensive evaluations on MSQA and MSNN highlight the limitations of existing vision-language models and underscore the importance of handling multi-modal interleaved inputs and situation modeling. Experiments on data scaling and cross-domain transfer further demonstrate the efficacy of leveraging MSQA as a pre-training dataset for developing more powerful situated reasoning models.

Yixuan Li, Le Ma, Yutang Lin et al. · pku

Whole-body humanoid teleoperation enables humans to remotely control humanoid robots, serving as both a real-time operational tool and a scalable engine for collecting demonstrations for autonomous learning. Despite recent advances, existing systems are validated using aggregate metrics that conflate distinct motion regimes, masking critical failure modes. This lack of diagnostic granularity, compounded by tightly coupled and labor-intensive system configurations, hinders robust real-world deployment. A key open challenge is building a teleoperation system that is simultaneously robust, versatile, and affordable for practical use. Here we present OmniClone, a whole-body humanoid teleoperation system that achieves high-fidelity, multi-skill control on a single consumer GPU with modest data requirements. Central to our approach is OmniBench, a diagnostic benchmark that evaluates policies across stratified motion categories and difficulty levels on unseen motions, exposing the narrow specialization of prior systems. Guided by these diagnostics, we identify an optimized training data recipe and integrate system-level improvements: subject-agnostic retargeting and robust communication, that collectively reduce Mean Per-Joint Position Error (MPJPE) by over 66% while requiring orders-of-magnitude fewer computational resources than comparable methods. Crucially, OmniClone is control-source-agnostic: a single unified policy supports real-time teleoperation, generated motion playback, and Vision-Language-Action (VLA) models, while generalizing across operators of vastly different body proportions. By uniting diagnostic evaluation with practical engineering, OmniClone provides an accessible foundation for scalable humanoid teleoperation and autonomous learning.

Bo Dai, Linge Wang, Baoxiong Jia et al.

Intuitive physics is pivotal for human understanding of the physical world, enabling prediction and interpretation of events even in infancy. Nonetheless, replicating this level of intuitive physics in artificial intelligence (AI) remains a formidable challenge. This study introduces X-VoE, a comprehensive benchmark dataset, to assess AI agents' grasp of intuitive physics. Built on the developmental psychology-rooted Violation of Expectation (VoE) paradigm, X-VoE establishes a higher bar for the explanatory capacities of intuitive physics models. Each VoE scenario within X-VoE encompasses three distinct settings, probing models' comprehension of events and their underlying explanations. Beyond model evaluation, we present an explanation-based learning system that captures physics dynamics and infers occluded object states solely from visual sequences, without explicit occlusion labels. Experimental outcomes highlight our model's alignment with human commonsense when tested against X-VoE. A remarkable feature is our model's ability to visually expound VoE events by reconstructing concealed scenes. Concluding, we discuss the findings' implications and outline future research directions. Through X-VoE, we catalyze the advancement of AI endowed with human-like intuitive physics capabilities.

Jiangyong Huang, William Yicheng Zhu, Baoxiong Jia et al.

Current computer vision models, unlike the human visual system, cannot yet achieve general-purpose visual understanding. Existing efforts to create a general vision model are limited in the scope of assessed tasks and offer no overarching framework to perform them holistically. We present a new comprehensive benchmark, General-purpose Visual Understanding Evaluation (G-VUE), covering the full spectrum of visual cognitive abilities with four functional domains $\unicode{x2014}$ Perceive, Ground, Reason, and Act. The four domains are embodied in 11 carefully curated tasks, from 3D reconstruction to visual reasoning and manipulation. Along with the benchmark, we provide a general encoder-decoder framework to allow for the evaluation of arbitrary visual representation on all 11 tasks. We evaluate various pre-trained visual representations with our framework and observe that (1) Transformer-based visual backbone generally outperforms CNN-based backbone on G-VUE, (2) visual representations from vision-language pre-training are superior to those with vision-only pre-training across visual tasks. With G-VUE, we provide a holistic evaluation standard to motivate research toward building general-purpose visual systems via obtaining more general-purpose visual representations.

Yu Liu, Baoxiong Jia, Yixin Chen et al.

The ability to distill object-centric abstractions from intricate visual scenes underpins human-level generalization. Despite the significant progress in object-centric learning methods, learning object-centric representations in the 3D physical world remains a crucial challenge. In this work, we propose SlotLifter, a novel object-centric radiance model addressing scene reconstruction and decomposition jointly via slot-guided feature lifting. Such a design unites object-centric learning representations and image-based rendering methods, offering state-of-the-art performance in scene decomposition and novel-view synthesis on four challenging synthetic and four complex real-world datasets, outperforming existing 3D object-centric learning methods by a large margin. Through extensive ablative studies, we showcase the efficacy of designs in SlotLifter, revealing key insights for potential future directions.

Zhuofan Zhang, Ziyu Zhu, Junhao Li et al.

Grounding natural language in 3D environments is a critical step toward achieving robust 3D vision-language alignment. Current datasets and models for 3D visual grounding predominantly focus on identifying and localizing objects from static, object-centric descriptions. These approaches do not adequately address the dynamic and sequential nature of task-oriented scenarios. In this work, we introduce a novel task: Task-oriented Sequential Grounding and Navigation in 3D Scenes, where models must interpret step-by-step instructions for daily activities by either localizing a sequence of target objects in indoor scenes or navigating toward them within a 3D simulator. To facilitate this task, we present SG3D, a large-scale dataset comprising 22,346 tasks with 112,236 steps across 4,895 real-world 3D scenes. The dataset is constructed by combining RGB-D scans from various 3D scene datasets with an automated task generation pipeline, followed by human verification for quality assurance. We benchmark contemporary methods on SG3D, revealing the significant challenges in understanding task-oriented context across multiple steps. Furthermore, we propose SG-LLM, a state-of-the-art approach leveraging a stepwise grounding paradigm to tackle the sequential grounding task. Our findings underscore the need for further research to advance the development of more capable and context-aware embodied agents.

Jieming Cui, Ziren Gong, Baoxiong Jia et al.

The challenge of replicating research results has posed a significant impediment to the field of molecular biology. The advent of modern intelligent systems has led to notable progress in various domains. Consequently, we embarked on an investigation of intelligent monitoring systems as a means of tackling the issue of the reproducibility crisis. Specifically, we first curate a comprehensive multimodal dataset, named ProBio, as an initial step towards this objective. This dataset comprises fine-grained hierarchical annotations intended for the purpose of studying activity understanding in BioLab. Next, we devise two challenging benchmarks, transparent solution tracking and multimodal action recognition, to emphasize the unique characteristics and difficulties associated with activity understanding in BioLab settings. Finally, we provide a thorough experimental evaluation of contemporary video understanding models and highlight their limitations in this specialized domain to identify potential avenues for future research. We hope ProBio with associated benchmarks may garner increased focus on modern AI techniques in the realm of molecular biology.

Yanpeng Zhao, Wentao Ding, Hongtao Li et al.

A recent trend in vision-language models (VLMs) has been to enhance their spatial cognition for embodied domains. Despite progress, existing evaluations have been limited both in paradigm and in coverage, hindering rapid, iterative model development. To address these limitations, we propose ESPIRE, a diagnostic benchmark for embodied spatial reasoning. ESPIRE offers a simulated world that physically grounds VLMs and evaluates them on spatial-reasoning-centric robotic tasks, thus narrowing the gap between evaluation and real-world deployment. To adapt VLMs to robotic tasks, we decompose each task into localization and execution, and frame both as generative problems, in stark contrast to predominant discriminative evaluations (e.g., via visual-question answering) that rely on distractors and discard execution. This decomposition further enables a fine-grained analysis beyond passive spatial reasoning toward reasoning to act. We systematically design ESPIRE both at the instruction level and at the environment level, ensuring broad coverage of spatial reasoning scenarios. We use ESPIRE to diagnose a range of frontier VLMs and provide in-depth analysis of their spatial reasoning behaviors.

Yixin Chen, Yaowei Zhang, Huangyue Yu et al.

Annotated 3D scene data is scarce and expensive to acquire, while abundant unlabeled videos are readily available on the internet. In this paper, we demonstrate that carefully designed data engines can leverage web-curated, unlabeled videos to automatically generate training data, to facilitate end-to-end models in 3D scene understanding alongside human-annotated datasets. We identify and analyze bottlenecks in automated data generation, revealing critical factors that determine the efficiency and effectiveness of learning from unlabeled data. To validate our approach across different perception granularities, we evaluate on three tasks spanning low-level perception, i.e., 3D object detection and instance segmentation, to high-evel reasoning, i.e., 3D spatial Visual Question Answering (VQA) and Vision-Lanugage Navigation (VLN). Models trained on our generated data demonstrate strong zero-shot performance and show further improvement after finetuning. This demonstrates the viability of leveraging readily available web data as a path toward more capable scene understanding systems.

Baoxiong Jia, Yixin Chen, Huangyue Yu et al.

3D vision-language grounding, which focuses on aligning language with the 3D physical environment, stands as a cornerstone in the development of embodied agents. In comparison to recent advancements in the 2D domain, grounding language in 3D scenes faces several significant challenges: (i) the inherent complexity of 3D scenes due to the diverse object configurations, their rich attributes, and intricate relationships; (ii) the scarcity of paired 3D vision-language data to support grounded learning; and (iii) the absence of a unified learning framework to distill knowledge from grounded 3D data. In this work, we aim to address these three major challenges in 3D vision-language by examining the potential of systematically upscaling 3D vision-language learning in indoor environments. We introduce the first million-scale 3D vision-language dataset, SceneVerse, encompassing about 68K 3D indoor scenes and comprising 2.5M vision-language pairs derived from both human annotations and our scalable scene-graph-based generation approach. We demonstrate that this scaling allows for a unified pre-training framework, Grounded Pre-training for Scenes (GPS), for 3D vision-language learning. Through extensive experiments, we showcase the effectiveness of GPS by achieving state-of-the-art performance on all existing 3D visual grounding benchmarks. The vast potential of SceneVerse and GPS is unveiled through zero-shot transfer experiments in the challenging 3D vision-language tasks. Project website: https://scene-verse.github.io.

Siyuan Qi, Wenguan Wang, Baoxiong Jia et al.

This paper addresses the task of detecting and recognizing human-object interactions (HOI) in images and videos. We introduce the Graph Parsing Neural Network (GPNN), a framework that incorporates structural knowledge while being differentiable end-to-end. For a given scene, GPNN infers a parse graph that includes i) the HOI graph structure represented by an adjacency matrix, and ii) the node labels. Within a message passing inference framework, GPNN iteratively computes the adjacency matrices and node labels. We extensively evaluate our model on three HOI detection benchmarks on images and videos: HICO-DET, V-COCO, and CAD-120 datasets. Our approach significantly outperforms state-of-art methods, verifying that GPNN is scalable to large datasets and applies to spatial-temporal settings. The code is available at https://github.com/SiyuanQi/gpnn.

Zan Wang, Yixin Chen, Baoxiong Jia et al.

Despite significant advancements in text-to-motion synthesis, generating language-guided human motion within 3D environments poses substantial challenges. These challenges stem primarily from (i) the absence of powerful generative models capable of jointly modeling natural language, 3D scenes, and human motion, and (ii) the generative models' intensive data requirements contrasted with the scarcity of comprehensive, high-quality, language-scene-motion datasets. To tackle these issues, we introduce a novel two-stage framework that employs scene affordance as an intermediate representation, effectively linking 3D scene grounding and conditional motion generation. Our framework comprises an Affordance Diffusion Model (ADM) for predicting explicit affordance map and an Affordance-to-Motion Diffusion Model (AMDM) for generating plausible human motions. By leveraging scene affordance maps, our method overcomes the difficulty in generating human motion under multimodal condition signals, especially when training with limited data lacking extensive language-scene-motion pairs. Our extensive experiments demonstrate that our approach consistently outperforms all baselines on established benchmarks, including HumanML3D and HUMANISE. Additionally, we validate our model's exceptional generalization capabilities on a specially curated evaluation set featuring previously unseen descriptions and scenes.

Yandan Yang, Baoxiong Jia, Peiyuan Zhi et al.

With recent developments in Embodied Artificial Intelligence (EAI) research, there has been a growing demand for high-quality, large-scale interactive scene generation. While prior methods in scene synthesis have prioritized the naturalness and realism of the generated scenes, the physical plausibility and interactivity of scenes have been largely left unexplored. To address this disparity, we introduce PhyScene, a novel method dedicated to generating interactive 3D scenes characterized by realistic layouts, articulated objects, and rich physical interactivity tailored for embodied agents. Based on a conditional diffusion model for capturing scene layouts, we devise novel physics- and interactivity-based guidance mechanisms that integrate constraints from object collision, room layout, and object reachability. Through extensive experiments, we demonstrate that PhyScene effectively leverages these guidance functions for physically interactable scene synthesis, outperforming existing state-of-the-art scene synthesis methods by a large margin. Our findings suggest that the scenes generated by PhyScene hold considerable potential for facilitating diverse skill acquisition among agents within interactive environments, thereby catalyzing further advancements in embodied AI research. Project website: http://physcene.github.io.

Ziyu Zhu, Zhuofan Zhang, Xiaojian Ma et al.

A unified model for 3D vision-language (3D-VL) understanding is expected to take various scene representations and perform a wide range of tasks in a 3D scene. However, a considerable gap exists between existing methods and such a unified model, due to the independent application of representation and insufficient exploration of 3D multi-task training. In this paper, we introduce PQ3D, a unified model capable of using Promptable Queries to tackle a wide range of 3D-VL tasks, from low-level instance segmentation to high-level reasoning and planning. This is achieved through three key innovations: (1) unifying various 3D scene representations (i.e., voxels, point clouds, multi-view images) into a shared 3D coordinate space by segment-level grouping, (2) an attention-based query decoder for task-specific information retrieval guided by prompts, and (3) universal output heads for different tasks to support multi-task training. Tested across ten diverse 3D-VL datasets, PQ3D demonstrates impressive performance on these tasks, setting new records on most benchmarks. Particularly, PQ3D improves the state-of-the-art on ScanNet200 by 4.9% (AP25), ScanRefer by 5.4% (acc@0.5), Multi3DRefer by 11.7% (F1@0.5), and Scan2Cap by 13.4% (CIDEr@0.5). Moreover, PQ3D supports flexible inference with individual or combined forms of available 3D representations, e.g., solely voxel input.

Peiyuan Zhi, Zhiyuan Zhang, Yu Zhao et al.

Autonomous robot navigation and manipulation in open environments require reasoning and replanning with closed-loop feedback. In this work, we present COME-robot, the first closed-loop robotic system utilizing the GPT-4V vision-language foundation model for open-ended reasoning and adaptive planning in real-world scenarios.COME-robot incorporates two key innovative modules: (i) a multi-level open-vocabulary perception and situated reasoning module that enables effective exploration of the 3D environment and target object identification using commonsense knowledge and situated information, and (ii) an iterative closed-loop feedback and restoration mechanism that verifies task feasibility, monitors execution success, and traces failure causes across different modules for robust failure recovery. Through comprehensive experiments involving 8 challenging real-world mobile and tabletop manipulation tasks, COME-robot demonstrates a significant improvement in task success rate (~35%) compared to state-of-the-art methods. We further conduct comprehensive analyses to elucidate how COME-robot's design facilitates failure recovery, free-form instruction following, and long-horizon task planning.

Yu Liu, Baoxiong Jia, Ruijie Lu et al.

Building articulated objects is a key challenge in computer vision. Existing methods often fail to effectively integrate information across different object states, limiting the accuracy of part-mesh reconstruction and part dynamics modeling, particularly for complex multi-part articulated objects. We introduce ArtGS, a novel approach that leverages 3D Gaussians as a flexible and efficient representation to address these issues. Our method incorporates canonical Gaussians with coarse-to-fine initialization and updates for aligning articulated part information across different object states, and employs a skinning-inspired part dynamics modeling module to improve both part-mesh reconstruction and articulation learning. Extensive experiments on both synthetic and real-world datasets, including a new benchmark for complex multi-part objects, demonstrate that ArtGS achieves state-of-the-art performance in joint parameter estimation and part mesh reconstruction. Our approach significantly improves reconstruction quality and efficiency, especially for multi-part articulated objects. Additionally, we provide comprehensive analyses of our design choices, validating the effectiveness of each component to highlight potential areas for future improvement. Our work is made publicly available at: https://articulate-gs.github.io.

Jiangyong Huang, Baoxiong Jia, Yan Wang et al.

Existing 3D vision-language (3D-VL) benchmarks fall short in evaluating 3D-VL models, creating a "mist" that obscures rigorous insights into model capabilities and 3D-VL tasks. This mist persists due to three key limitations. First, flawed test data, like ambiguous referential text in the grounding task, can yield incorrect and unreliable test results. Second, oversimplified metrics such as simply averaging accuracy per question answering (QA) pair, cannot reveal true model capability due to their vulnerability to language variations. Third, existing benchmarks isolate the grounding and QA tasks, disregarding the underlying coherence that QA should be based on solid grounding capabilities. To unveil the "mist", we propose Beacon3D, a benchmark for 3D-VL grounding and QA tasks, delivering a perspective shift in the evaluation of 3D-VL understanding. Beacon3D features (i) high-quality test data with precise and natural language, (ii) object-centric evaluation with multiple tests per object to ensure robustness, and (iii) a novel chain-of-analysis paradigm to address language robustness and model performance coherence across grounding and QA. Our evaluation of state-of-the-art 3D-VL models on Beacon3D reveals that (i) object-centric evaluation elicits true model performance and particularly weak generalization in QA; (ii) grounding-QA coherence remains fragile in current 3D-VL models, and (iii) incorporating large language models (LLMs) to 3D-VL models, though as a prevalent practice, hinders grounding capabilities and has yet to elevate QA capabilities. We hope Beacon3D and our comprehensive analysis could benefit the 3D-VL community towards faithful developments.

Huangyue Yu, Baoxiong Jia, Yixin Chen et al.

Embodied AI (EAI) research requires high-quality, diverse 3D scenes to effectively support skill acquisition, sim-to-real transfer, and generalization. Achieving these quality standards, however, necessitates the precise replication of real-world object diversity. Existing datasets demonstrate that this process heavily relies on artist-driven designs, which demand substantial human effort and present significant scalability challenges. To scalably produce realistic and interactive 3D scenes, we first present MetaScenes, a large-scale, simulatable 3D scene dataset constructed from real-world scans, which includes 15366 objects spanning 831 fine-grained categories. Then, we introduce Scan2Sim, a robust multi-modal alignment model, which enables the automated, high-quality replacement of assets, thereby eliminating the reliance on artist-driven designs for scaling 3D scenes. We further propose two benchmarks to evaluate MetaScenes: a detailed scene synthesis task focused on small item layouts for robotic manipulation and a domain transfer task in vision-and-language navigation (VLN) to validate cross-domain transfer. Results confirm MetaScene's potential to enhance EAI by supporting more generalizable agent learning and sim-to-real applications, introducing new possibilities for EAI research. Project website: https://meta-scenes.github.io/.

Ziyu Zhu, Xilin Wang, Yixuan Li et al.

Embodied scene understanding requires not only comprehending visual-spatial information that has been observed but also determining where to explore next in the 3D physical world. Existing 3D Vision-Language (3D-VL) models primarily focus on grounding objects in static observations from 3D reconstruction, such as meshes and point clouds, but lack the ability to actively perceive and explore their environment. To address this limitation, we introduce \underline{\textbf{M}}ove \underline{\textbf{t}}o \underline{\textbf{U}}nderstand (\textbf{\model}), a unified framework that integrates active perception with \underline{\textbf{3D}} vision-language learning, enabling embodied agents to effectively explore and understand their environment. This is achieved by three key innovations: 1) Online query-based representation learning, enabling direct spatial memory construction from RGB-D frames, eliminating the need for explicit 3D reconstruction. 2) A unified objective for grounding and exploring, which represents unexplored locations as frontier queries and jointly optimizes object grounding and frontier selection. 3) End-to-end trajectory learning that combines \textbf{V}ision-\textbf{L}anguage-\textbf{E}xploration pre-training over a million diverse trajectories collected from both simulated and real-world RGB-D sequences. Extensive evaluations across various embodied navigation and question-answering benchmarks show that MTU3D outperforms state-of-the-art reinforcement learning and modular navigation approaches by 14\%, 23\%, 9\%, and 2\% in success rate on HM3D-OVON, GOAT-Bench, SG3D, and A-EQA, respectively. \model's versatility enables navigation using diverse input modalities, including categories, language descriptions, and reference images. These findings highlight the importance of bridging visual grounding and exploration for embodied intelligence.

Ruijie Lu, Yixin Chen, Junfeng Ni et al.

Repurposing pre-trained diffusion models has been proven to be effective for NVS. However, these methods are mostly limited to a single object; directly applying such methods to compositional multi-object scenarios yields inferior results, especially incorrect object placement and inconsistent shape and appearance under novel views. How to enhance and systematically evaluate the cross-view consistency of such models remains under-explored. To address this issue, we propose MOVIS to enhance the structural awareness of the view-conditioned diffusion model for multi-object NVS in terms of model inputs, auxiliary tasks, and training strategy. First, we inject structure-aware features, including depth and object mask, into the denoising U-Net to enhance the model's comprehension of object instances and their spatial relationships. Second, we introduce an auxiliary task requiring the model to simultaneously predict novel view object masks, further improving the model's capability in differentiating and placing objects. Finally, we conduct an in-depth analysis of the diffusion sampling process and carefully devise a structure-guided timestep sampling scheduler during training, which balances the learning of global object placement and fine-grained detail recovery. To systematically evaluate the plausibility of synthesized images, we propose to assess cross-view consistency and novel view object placement alongside existing image-level NVS metrics. Extensive experiments on challenging synthetic and realistic datasets demonstrate that our method exhibits strong generalization capabilities and produces consistent novel view synthesis, highlighting its potential to guide future 3D-aware multi-object NVS tasks. Our project page is available at https://jason-aplp.github.io/MOVIS/.

Guanxing Lu, Baoxiong Jia, Puhao Li et al.

Training robot policies within a learned world model is trending due to the inefficiency of real-world interactions. The established image-based world models and policies have shown prior success, but lack robust geometric information that requires consistent spatial and physical understanding of the three-dimensional world, even pre-trained on internet-scale video sources. To this end, we propose a novel branch of world model named Gaussian World Model (GWM) for robotic manipulation, which reconstructs the future state by inferring the propagation of Gaussian primitives under the effect of robot actions. At its core is a latent Diffusion Transformer (DiT) combined with a 3D variational autoencoder, enabling fine-grained scene-level future state reconstruction with Gaussian Splatting. GWM can not only enhance the visual representation for imitation learning agent by self-supervised future prediction training, but can serve as a neural simulator that supports model-based reinforcement learning. Both simulated and real-world experiments depict that GWM can precisely predict future scenes conditioned on diverse robot actions, and can be further utilized to train policies that outperform the state-of-the-art by impressive margins, showcasing the initial data scaling potential of 3D world model.

Yan Wang, Baoxiong Jia, Ziyu Zhu et al.

Open-vocabulary 3D scene understanding is pivotal for enhancing physical intelligence, as it enables embodied agents to interpret and interact dynamically within real-world environments. This paper introduces MPEC, a novel Masked Point-Entity Contrastive learning method for open-vocabulary 3D semantic segmentation that leverages both 3D entity-language alignment and point-entity consistency across different point cloud views to foster entity-specific feature representations. Our method improves semantic discrimination and enhances the differentiation of unique instances, achieving state-of-the-art results on ScanNet for open-vocabulary 3D semantic segmentation and demonstrating superior zero-shot scene understanding capabilities. Extensive fine-tuning experiments on 8 datasets, spanning from low-level perception to high-level reasoning tasks, showcase the potential of learned 3D features, driving consistent performance gains across varied 3D scene understanding tasks. Project website: https://mpec-3d.github.io/

Yandan Yang, Baoxiong Jia, Shujie Zhang et al.

Indoor scene synthesis has become increasingly important with the rise of Embodied AI, which requires 3D environments that are not only visually realistic but also physically plausible and functionally diverse. While recent approaches have advanced visual fidelity, they often remain constrained to fixed scene categories, lack sufficient object-level detail and physical consistency, and struggle to align with complex user instructions. In this work, we present SceneWeaver, a reflective agentic framework that unifies diverse scene synthesis paradigms through tool-based iterative refinement. At its core, SceneWeaver employs a language model-based planner to select from a suite of extensible scene generation tools, ranging from data-driven generative models to visual- and LLM-based methods, guided by self-evaluation of physical plausibility, visual realism, and semantic alignment with user input. This closed-loop reason-act-reflect design enables the agent to identify semantic inconsistencies, invoke targeted tools, and update the environment over successive iterations. Extensive experiments on both common and open-vocabulary room types demonstrate that SceneWeaver not only outperforms prior methods on physical, visual, and semantic metrics, but also generalizes effectively to complex scenes with diverse instructions, marking a step toward general-purpose 3D environment generation. Project website: https://scene-weaver.github.io/.

Jiangyong Huang, Xiaojian Ma, Xiongkun Linghu et al.

Developing vision-language models (VLMs) capable of understanding 3D scenes has been a longstanding goal in the 3D-VL community. Despite recent progress, 3D VLMs still fall short of their 2D counterparts in capability and robustness. A key bottleneck is that current scene representations struggle to balance performance and efficiency: competitive performance comes at the cost of heavy token overhead, which in turn hampers the scalability of 3D-VL learning. To address this, we propose the condensed feature grid (CFG), an efficient scene representation featuring significantly reduced token overhead and strong perception capability. Building on CFG, we introduce LEO-VL, a 3D VLM trained on 700k 3D-VL data spanning four real-world indoor domains and five tasks such as captioning and dialogue. To enhance the robustness of 3D VLM, we further propose SceneDPO for post-training, which involves contrasts across answers and scenes. LEO-VL achieves state-of-the-art performance on various 3D QA benchmarks, including SQA3D, MSQA, and Beacon3D. Our extensive experiments highlight the efficiency of our representation, the benefit of task and scene diversity, consistent scaling effects, and the advantages of SceneDPO compared to SFT and GRPO. We hope our findings advance the efficiency, scalability, and robustness of future 3D VLMs.

Xiongkun Linghu, Jiangyong Huang, Baoxiong Jia et al.

Reinforcement Learning with Verifiable Rewards ( RLVR ) has emerged as a transformative paradigm for enhancing the reasoning capabilities of Large Language Models ( LLMs), yet its potential in 3D scene understanding remains under-explored. Existing approaches largely rely on Supervised Fine-Tuning ( SFT), where the token-level cross-entropy loss acts as an indirect proxy for optimization, leading to a misalignment between training objectives and task performances. To bridge this gap, we present Reinforcement Fine-Tuning for Video-based 3D Scene Understanding (3D-RFT ), the first framework to extend RLVR to video-based 3D perception and reasoning. 3D-RFT shifts the paradigm by directly optimizing the model towards evaluation metrics. 3D-RFT first activates 3D-aware Multi-modal Large Language Models ( MLLM s) via SFT, followed by reinforcement fine-tuning using Group Relative Policy Optimization ( GRPO) with strictly verifiable reward functions. We design task-specific reward functions directly from metrics like 3D IoU and F1-Score to provide more effective signals to guide model training. Extensive experiments demonstrate that 3D-RFT-4B achieves state-of-the-art performance on various video-based 3D scene understanding tasks. Notably, 3D-RFT-4B significantly outperforms larger models (e.g., VG LLM-8B) on 3D video detection, 3D visual grounding, and spatial reasoning benchmarks. We further reveal good properties of 3D-RFT such as robust efficacy, and valuable insights into training strategies and data impact. We hope 3D-RFT can serve as a robust and promising paradigm for future development of 3D scene understanding.

Zan Wang, Jingze Zhang, Yixin Chen et al.

Despite significant advancements in human motion generation, current motion representations, typically formulated as discrete frame sequences, still face two critical limitations: (i) they fail to capture motion from a multi-scale perspective, limiting the capability in complex patterns modeling; (ii) they lack compositional flexibility, which is crucial for model's generalization in diverse generation tasks. To address these challenges, we introduce MSQ, a novel quantization method that compresses the motion sequence into multi-scale discrete tokens across spatial and temporal dimensions. MSQ employs distinct encoders to capture body parts at varying spatial granularities and temporally interpolates the encoded features into multiple scales before quantizing them into discrete tokens. Building on this representation, we establish a generative mask modeling model to effectively support motion editing, motion control, and conditional motion generation. Through quantitative and qualitative analysis, we show that our quantization method enables the seamless composition of motion tokens without requiring specialized design or re-training. Furthermore, extensive evaluations demonstrate that our approach outperforms existing baseline methods on various benchmarks.

Wentao Jiang, Jingya Wang, Kaiyang Ji et al.

Human action-reaction synthesis, a fundamental challenge in modeling causal human interactions, plays a critical role in applications ranging from virtual reality to social robotics. While diffusion-based models have demonstrated promising performance, they exhibit two key limitations for interaction synthesis: reliance on complex noise-to-reaction generators with intricate conditional mechanisms, and frequent physical violations in generated motions. To address these issues, we propose Action-Reaction Flow Matching (ARFlow), a novel framework that establishes direct action-to-reaction mappings, eliminating the need for complex conditional mechanisms. Our approach introduces a physical guidance mechanism specifically designed for Flow Matching (FM) that effectively prevents body penetration artifacts during sampling. Moreover, we discover the bias of traditional flow matching sampling algorithm and employ a reprojection method to revise the sampling direction of FM. To further enhance the reaction diversity, we incorporate randomness into the sampling process. Extensive experiments on NTU120, Chi3D and InterHuman datasets demonstrate that ARFlow not only outperforms existing methods in terms of Fréchet Inception Distance and motion diversity but also significantly reduces body collisions, as measured by our new Intersection Volume and Intersection Frequency metrics.

Xiongkun Linghu, Jiangyong Huang, Ziyu Zhu et al.

Existing research on 3D Large Language Models (LLMs) still struggles to achieve grounded question-answering, primarily due to the under-exploration of the mechanism of human-like scene-object grounded reasoning. This paper bridges the gap by presenting a novel framework. We first introduce a grounded Chain-of-Thought reasoning method in 3D scenes (SCENECOT), decoupling a complex reasoning task into simpler and manageable problems, and building corresponding visual clues based on multimodal expert modules. To enable such a method, we develop SCENECOT-185K, the first large-scale grounded CoT reasoning dataset, consisting of 185K high-quality instances. Extensive experiments across various complex 3D scene reasoning benchmarks demonstrate that our new framework achieves strong performance with high grounding-QA coherence. To the best of our knowledge, this is the first successful application of CoT reasoning to 3D scene understanding, enabling step-by-step human-like reasoning and showing potential for extension to broader 3D scene understanding scenarios.

Yushi Du, Yixuan Li, Baoxiong Jia et al. · pku

Human-humanoid collaboration shows significant promise for applications in healthcare, domestic assistance, and manufacturing. While compliant robot-human collaboration has been extensively developed for robotic arms, enabling compliant human-humanoid collaboration remains largely unexplored due to humanoids' complex whole-body dynamics. In this paper, we propose a proprioception-only reinforcement learning approach, COLA, that combines leader and follower behaviors within a single policy. The model is trained in a closed-loop environment with dynamic object interactions to predict object motion patterns and human intentions implicitly, enabling compliant collaboration to maintain load balance through coordinated trajectory planning. We evaluate our approach through comprehensive simulator and real-world experiments on collaborative carrying tasks, demonstrating the effectiveness, generalization, and robustness of our model across various terrains and objects. Simulation experiments demonstrate that our model reduces human effort by 24.7%. compared to baseline approaches while maintaining object stability. Real-world experiments validate robust collaborative carrying across different object types (boxes, desks, stretchers, etc.) and movement patterns (straight-line, turning, slope climbing). Human user studies with 23 participants confirm an average improvement of 27.4% compared to baseline models. Our method enables compliant human-humanoid collaborative carrying without requiring external sensors or complex interaction models, offering a practical solution for real-world deployment.

Yu Liu, Baoxiong Jia, Ruijie Lu et al.

Building digital twins of articulated objects from monocular video presents an essential challenge in computer vision, which requires simultaneous reconstruction of object geometry, part segmentation, and articulation parameters from limited viewpoint inputs. Monocular video offers an attractive input format due to its simplicity and scalability; however, it's challenging to disentangle the object geometry and part dynamics with visual supervision alone, as the joint movement of the camera and parts leads to ill-posed estimation. While motion priors from pre-trained tracking models can alleviate the issue, how to effectively integrate them for articulation learning remains largely unexplored. To address this problem, we introduce VideoArtGS, a novel approach that reconstructs high-fidelity digital twins of articulated objects from monocular video. We propose a motion prior guidance pipeline that analyzes 3D tracks, filters noise, and provides reliable initialization of articulation parameters. We also design a hybrid center-grid part assignment module for articulation-based deformation fields that captures accurate part motion. VideoArtGS demonstrates state-of-the-art performance in articulation and mesh reconstruction, reducing the reconstruction error by about two orders of magnitude compared to existing methods. VideoArtGS enables practical digital twin creation from monocular video, establishing a new benchmark for video-based articulated object reconstruction. Our work is made publicly available at: https://videoartgs.github.io.

Chi Zhang, Sirui Xie, Baoxiong Jia et al.

Is intelligence realized by connectionist or classicist? While connectionist approaches have achieved superhuman performance, there has been growing evidence that such task-specific superiority is particularly fragile in systematic generalization. This observation lies in the central debate between connectionist and classicist, wherein the latter continually advocates an algebraic treatment in cognitive architectures. In this work, we follow the classicist's call and propose a hybrid approach to improve systematic generalization in reasoning. Specifically, we showcase a prototype with algebraic representation for the abstract spatial-temporal reasoning task of Raven's Progressive Matrices (RPM) and present the ALgebra-Aware Neuro-Semi-Symbolic (ALANS) learner. The ALANS learner is motivated by abstract algebra and the representation theory. It consists of a neural visual perception frontend and an algebraic abstract reasoning backend: the frontend summarizes the visual information from object-based representation, while the backend transforms it into an algebraic structure and induces the hidden operator on the fly. The induced operator is later executed to predict the answer's representation, and the choice most similar to the prediction is selected as the solution. Extensive experiments show that by incorporating an algebraic treatment, the ALANS learner outperforms various pure connectionist models in domains requiring systematic generalization. We further show the generative nature of the learned algebraic representation; it can be decoded by isomorphism to generate an answer.

Chi Zhang, Baoxiong Jia, Mark Edmonds et al.

Causal induction, i.e., identifying unobservable mechanisms that lead to the observable relations among variables, has played a pivotal role in modern scientific discovery, especially in scenarios with only sparse and limited data. Humans, even young toddlers, can induce causal relationships surprisingly well in various settings despite its notorious difficulty. However, in contrast to the commonplace trait of human cognition is the lack of a diagnostic benchmark to measure causal induction for modern Artificial Intelligence (AI) systems. Therefore, in this work, we introduce the Abstract Causal REasoning (ACRE) dataset for systematic evaluation of current vision systems in causal induction. Motivated by the stream of research on causal discovery in Blicket experiments, we query a visual reasoning system with the following four types of questions in either an independent scenario or an interventional scenario: direct, indirect, screening-off, and backward-blocking, intentionally going beyond the simple strategy of inducing causal relationships by covariation. By analyzing visual reasoning architectures on this testbed, we notice that pure neural models tend towards an associative strategy under their chance-level performance, whereas neuro-symbolic combinations struggle in backward-blocking reasoning. These deficiencies call for future research in models with a more comprehensive capability of causal induction.

Chi Zhang, Baoxiong Jia, Song-Chun Zhu et al.

Spatial-temporal reasoning is a challenging task in Artificial Intelligence (AI) due to its demanding but unique nature: a theoretic requirement on representing and reasoning based on spatial-temporal knowledge in mind, and an applied requirement on a high-level cognitive system capable of navigating and acting in space and time. Recent works have focused on an abstract reasoning task of this kind -- Raven's Progressive Matrices (RPM). Despite the encouraging progress on RPM that achieves human-level performance in terms of accuracy, modern approaches have neither a treatment of human-like reasoning on generalization, nor a potential to generate answers. To fill in this gap, we propose a neuro-symbolic Probabilistic Abduction and Execution (PrAE) learner; central to the PrAE learner is the process of probabilistic abduction and execution on a probabilistic scene representation, akin to the mental manipulation of objects. Specifically, we disentangle perception and reasoning from a monolithic model. The neural visual perception frontend predicts objects' attributes, later aggregated by a scene inference engine to produce a probabilistic scene representation. In the symbolic logical reasoning backend, the PrAE learner uses the representation to abduce the hidden rules. An answer is predicted by executing the rules on the probabilistic representation. The entire system is trained end-to-end in an analysis-by-synthesis manner without any visual attribute annotations. Extensive experiments demonstrate that the PrAE learner improves cross-configuration generalization and is capable of rendering an answer, in contrast to prior works that merely make a categorical choice from candidates.

Baoxiong Jia, Yixin Chen, Siyuan Huang et al.

Understanding and interpreting human actions is a long-standing challenge and a critical indicator of perception in artificial intelligence. However, a few imperative components of daily human activities are largely missed in prior literature, including the goal-directed actions, concurrent multi-tasks, and collaborations among multi-agents. We introduce the LEMMA dataset to provide a single home to address these missing dimensions with meticulously designed settings, wherein the number of tasks and agents varies to highlight different learning objectives. We densely annotate the atomic-actions with human-object interactions to provide ground-truths of the compositionality, scheduling, and assignment of daily activities. We further devise challenging compositional action recognition and action/task anticipation benchmarks with baseline models to measure the capability of compositional action understanding and temporal reasoning. We hope this effort would drive the machine vision community to examine goal-directed human activities and further study the task scheduling and assignment in the real world.

Chi Zhang, Baoxiong Jia, Feng Gao et al.

"Thinking in pictures," [1] i.e., spatial-temporal reasoning, effortless and instantaneous for humans, is believed to be a significant ability to perform logical induction and a crucial factor in the intellectual history of technology development. Modern Artificial Intelligence (AI), fueled by massive datasets, deeper models, and mighty computation, has come to a stage where (super-)human-level performances are observed in certain specific tasks. However, current AI's ability in "thinking in pictures" is still far lacking behind. In this work, we study how to improve machines' reasoning ability on one challenging task of this kind: Raven's Progressive Matrices (RPM). Specifically, we borrow the very idea of "contrast effects" from the field of psychology, cognition, and education to design and train a permutation-invariant model. Inspired by cognitive studies, we equip our model with a simple inference module that is jointly trained with the perception backbone. Combining all the elements, we propose the Contrastive Perceptual Inference network (CoPINet) and empirically demonstrate that CoPINet sets the new state-of-the-art for permutation-invariant models on two major datasets. We conclude that spatial-temporal reasoning depends on envisaging the possibilities consistent with the relations between objects and can be solved from pixel-level inputs.

Chi Zhang, Feng Gao, Baoxiong Jia et al.

Dramatic progress has been witnessed in basic vision tasks involving low-level perception, such as object recognition, detection, and tracking. Unfortunately, there is still an enormous performance gap between artificial vision systems and human intelligence in terms of higher-level vision problems, especially ones involving reasoning. Earlier attempts in equipping machines with high-level reasoning have hovered around Visual Question Answering (VQA), one typical task associating vision and language understanding. In this work, we propose a new dataset, built in the context of Raven's Progressive Matrices (RPM) and aimed at lifting machine intelligence by associating vision with structural, relational, and analogical reasoning in a hierarchical representation. Unlike previous works in measuring abstract reasoning using RPM, we establish a semantic link between vision and reasoning by providing structure representation. This addition enables a new type of abstract reasoning by jointly operating on the structure representation. Machine reasoning ability using modern computer vision is evaluated in this newly proposed dataset. Additionally, we also provide human performance as a reference. Finally, we show consistent improvement across all models by incorporating a simple neural module that combines visual understanding and structure reasoning.

Siyuan Qi, Baoxiong Jia, Song-Chun Zhu

Future predictions on sequence data (e.g., videos or audios) require the algorithms to capture non-Markovian and compositional properties of high-level semantics. Context-free grammars are natural choices to capture such properties, but traditional grammar parsers (e.g., Earley parser) only take symbolic sentences as inputs. In this paper, we generalize the Earley parser to parse sequence data which is neither segmented nor labeled. This generalized Earley parser integrates a grammar parser with a classifier to find the optimal segmentation and labels, and makes top-down future predictions. Experiments show that our method significantly outperforms other approaches for future human activity prediction.