Yuanzhi Liang

Yu Lu, Yuanzhi Liang, Linchao Zhu et al.

Video diffusion models have made substantial progress in various video generation applications. However, training models for long video generation tasks require significant computational and data resources, posing a challenge to developing long video diffusion models. This paper investigates a straightforward and training-free approach to extend an existing short video diffusion model (e.g. pre-trained on 16-frame videos) for consistent long video generation (e.g. 128 frames). Our preliminary observation has found that directly applying the short video diffusion model to generate long videos can lead to severe video quality degradation. Further investigation reveals that this degradation is primarily due to the distortion of high-frequency components in long videos, characterized by a decrease in spatial high-frequency components and an increase in temporal high-frequency components. Motivated by this, we propose a novel solution named FreeLong to balance the frequency distribution of long video features during the denoising process. FreeLong blends the low-frequency components of global video features, which encapsulate the entire video sequence, with the high-frequency components of local video features that focus on shorter subsequences of frames. This approach maintains global consistency while incorporating diverse and high-quality spatiotemporal details from local videos, enhancing both the consistency and fidelity of long video generation. We evaluated FreeLong on multiple base video diffusion models and observed significant improvements. Additionally, our method supports coherent multi-prompt generation, ensuring both visual coherence and seamless transitions between scenes.

Yuanzhi Liang, Linchao Zhu, Yi Yang

Recent advancements in natural language and Large Language Models (LLMs) have enabled AI agents to simulate human-like interactions within virtual worlds. However, these interactions still face limitations in complexity and flexibility, particularly in scenarios involving multiple characters and novel objects. Pre-defining all interactable objects in the agent's world model presents challenges, and conveying implicit intentions to multiple characters through complex interactions remains difficult. To address these issues, we propose integrating virtual Game Masters (GMs) into the agent's world model, drawing inspiration from Tabletop Role-Playing Games (TRPGs). GMs play a crucial role in overseeing information, estimating players' intentions, providing environment descriptions, and offering feedback, compensating for current world model deficiencies. To facilitate future explorations for complex interactions, we introduce a benchmark named Tachikuma, comprising a Multiple character and novel Object based interaction Estimation (MOE) task and a supporting dataset. MOE challenges models to understand characters' intentions and accurately determine their actions within intricate contexts involving multi-character and novel object interactions. Besides, the dataset captures log data from real-time communications during gameplay, providing diverse, grounded, and complex interactions for further explorations. Finally, we present a simple prompting baseline and evaluate its performance, demonstrating its effectiveness in enhancing interaction understanding. We hope that our dataset and task will inspire further research in complex interactions with natural language, fostering the development of more advanced AI agents.

Dianbing Xi, Jiepeng Wang, Yuanzhi Liang et al.

We tackle the dual challenges of video understanding and controllable video generation within a unified diffusion framework. Our key insights are two-fold: geometry-only cues (e.g., depth, edges) are insufficient: they specify layout but under-constrain appearance, materials, and illumination, limiting physically meaningful edits such as relighting or material swaps and often causing temporal drift. Enriching the model with additional graphics-based modalities (intrinsics and semantics) provides complementary constraints that both disambiguate understanding and enable precise, predictable control during generation. However, building a single model that uses many heterogeneous cues introduces two core difficulties. Architecturally, the model must accept any subset of modalities, remain robust to missing inputs, and inject control signals without sacrificing temporal consistency. Data-wise, training demands large-scale, temporally aligned supervision that ties real videos to per-pixel multimodal annotations. We then propose CtrlVDiff, a unified diffusion model trained with a Hybrid Modality Control Strategy (HMCS) that routes and fuses features from depth, normals, segmentation, edges, and graphics-based intrinsics (albedo, roughness, metallic), and re-renders videos from any chosen subset with strong temporal coherence. To enable this, we build MMVideo, a hybrid real-and-synthetic dataset aligned across modalities and captions. Across understanding and generation benchmarks, CtrlVDiff delivers superior controllability and fidelity, enabling layer-wise edits (relighting, material adjustment, object insertion) and surpassing state-of-the-art baselines while remaining robust when some modalities are unavailable.

Rui Li, Bingyu Li, Yuanzhi Liang et al.

Achieving high-fidelity generation in extremely few sampling steps has long been a central goal of generative modeling. Existing approaches largely rely on distillation-based frameworks to compress the original multi-step denoising process into a few-step generator. However, such methods inherently constrain the student to imitate a stronger multi-step teacher, imposing the teacher as an upper bound on student performance. We argue that introducing \textbf{preference alignment awareness} enables the student to optimize toward reward-preferred generation quality, potentially surpassing the teacher instead of being restricted to rigid teacher imitation. To this end, we propose \textbf{Reward-Aware Trajectory Shaping (RATS)}, a lightweight framework for preference-aligned few-step generation. Specifically, teacher and student latent trajectories are aligned at key denoising stages through horizon matching, while a \textbf{reward-aware gate} is introduced to adaptively regulate teacher guidance based on their relative reward performance. Trajectory shaping is strengthened when the teacher achieves higher rewards, and relaxed when the student matches or surpasses the teacher, thereby enabling continued reward-driven improvement. By seamlessly integrating trajectory distillation, reward-aware gating, and preference alignment, RATS effectively transfers preference-relevant knowledge from high-step generators without incurring additional test-time computational overhead. Experimental results demonstrate that RATS substantially improves the efficiency--quality trade-off in few-step visual generation, significantly narrowing the gap between few-step students and stronger multi-step generators.

Jialun Liu, Yukuo Ma, Xiao Cao et al.

Recent advances in diffusion-based video generation have substantially improved visual fidelity and temporal coherence. However, most existing approaches remain task-specific and rely primarily on textual instructions, limiting their ability to handle multimodal inputs, contextual references, and diverse video generation and editing scenarios within a unified framework. Moreover, many video editing methods depend on carefully engineered pipelines tailored to individual operations, which hinders scalability and composability. In this paper, we propose Tele-Omni, a unified multimodal framework for video generation and editing that follows multimodal instructions, including text, images, and reference videos, within a single model. Tele-Omni leverages pretrained multimodal large language models to parse heterogeneous instructions and infer structured generation or editing intents, while diffusion-based generators perform high-quality video synthesis conditioned on these structured signals. To enable joint training across heterogeneous video tasks, we introduce a task-aware data processing pipeline that unifies multimodal inputs into a structured instruction format while preserving task-specific constraints. Tele-Omni supports a wide range of video-centric tasks, including text-to-video generation, image-to-video generation, first-last-frame video generation, in-context video generation, and in-context video editing. By decoupling instruction parsing from video synthesis and combining it with task-aware data design, Tele-Omni achieves flexible multimodal control while maintaining strong temporal coherence and visual consistency. Experimental results demonstrate that Tele-Omni achieves competitive performance across multiple tasks.

Yiyi Ma, Yuanzhi Liang, Xiu Li et al.

We present Interleaved Learning for Motion Synthesis (InterSyn), a novel framework that targets the generation of realistic interaction motions by learning from integrated motions that consider both solo and multi-person dynamics. Unlike previous methods that treat these components separately, InterSyn employs an interleaved learning strategy to capture the natural, dynamic interactions and nuanced coordination inherent in real-world scenarios. Our framework comprises two key modules: the Interleaved Interaction Synthesis (INS) module, which jointly models solo and interactive behaviors in a unified paradigm from a first-person perspective to support multiple character interactions, and the Relative Coordination Refinement (REC) module, which refines mutual dynamics and ensures synchronized motions among characters. Experimental results show that the motion sequences generated by InterSyn exhibit higher text-to-motion alignment and improved diversity compared with recent methods, setting a new benchmark for robust and natural motion synthesis. Additionally, our code will be open-sourced in the future to promote further research and development in this area.

Yabo Chen, Yuanzhi Liang, Jiepeng Wang et al.

World models aim to endow AI systems with the ability to represent, generate, and interact with dynamic environments in a coherent and temporally consistent manner. While recent video generation models have demonstrated impressive visual quality, they remain limited in real-time interaction, long-horizon consistency, and persistent memory of dynamic scenes, hindering their evolution into practical world models. In this report, we present TeleWorld, a real-time multimodal 4D world modeling framework that unifies video generation, dynamic scene reconstruction, and long-term world memory within a closed-loop system. TeleWorld introduces a novel generation-reconstruction-guidance paradigm, where generated video streams are continuously reconstructed into a dynamic 4D spatio-temporal representation, which in turn guides subsequent generation to maintain spatial, temporal, and physical consistency. To support long-horizon generation with low latency, we employ an autoregressive diffusion-based video model enhanced with Macro-from-Micro Planning (MMPL)--a hierarchical planning method that reduces error accumulation from frame-level to segment-level-alongside efficient Distribution Matching Distillation (DMD), enabling real-time synthesis under practical computational budgets. Our approach achieves seamless integration of dynamic object modeling and static scene representation within a unified 4D framework, advancing world models toward practical, interactive, and computationally accessible systems. Extensive experiments demonstrate that TeleWorld achieves strong performance in both static and dynamic world understanding, long-term consistency, and real-time generation efficiency, positioning it as a practical step toward interactive, memory-enabled world models for multimodal generation and embodied intelligence.

Sheng Liu, Yuanzhi Liang, Sidan Du

Recent 3D human motion generation models demonstrate remarkable reconstruction accuracy yet struggle to generalize beyond training distributions. This limitation arises partly from the use of precise 3D supervision, which encourages models to fit fixed coordinate patterns instead of learning the essential 3D structure and motion semantic cues required for robust generalization.To overcome this limitation, we propose Free3D, a framework that synthesizes realistic 3D motions without any 3D motion annotations. Free3D introduces a Motion-Lifting Residual Quantized VAE (ML-RQ) that maps 2D motion sequences into 3D-consistent latent spaces, and a suite of 3D-free regularization objectives enforcing view consistency, orientation coherence, and physical plausibility. Trained entirely on 2D motion data, Free3D generates diverse, temporally coherent, and semantically aligned 3D motions, achieving performance comparable to or even surpassing fully 3D-supervised counterparts. These results suggest that relaxing explicit 3D supervision encourages stronger structural reasoning and generalization, offering a scalable and data-efficient paradigm for 3D motion generation.

Hongjun An, Wenhan Hu, Sida Huang et al.

Pioneered by the foundational information theory by Claude Shannon and the visionary framework of machine intelligence by Alan Turing, the convergent evolution of information and communication technologies (IT/CT) has created an unbroken wave of connectivity and computation. This synergy has sparked a technological revolution, now reaching its peak with large artificial intelligence (AI) models that are reshaping industries and redefining human-machine collaboration. However, the realization of ubiquitous intelligence faces considerable challenges due to substantial resource consumption in large models and high communication bandwidth demands. To address these challenges, AI Flow has been introduced as a multidisciplinary framework that integrates cutting-edge IT and CT advancements, with a particular emphasis on the following three key points. First, device-edge-cloud framework serves as the foundation, which integrates end devices, edge servers, and cloud clusters to optimize scalability and efficiency for low-latency model inference. Second, we introduce the concept of familial models, which refers to a series of different-sized models with aligned hidden features, enabling effective collaboration and the flexibility to adapt to varying resource constraints and dynamic scenarios. Third, connectivity- and interaction-based intelligence emergence is a novel paradigm of AI Flow. By leveraging communication networks to enhance connectivity, the collaboration among AI models across heterogeneous nodes achieves emergent intelligence that surpasses the capability of any single model. The innovations of AI Flow provide enhanced intelligence, timely responsiveness, and ubiquitous accessibility to AI services, paving the way for the tighter fusion of AI techniques and communication systems.

Chi Zhang, Yuanzhi Liang, Xi Qiu et al.

Generating high-quality videos from textual descriptions poses challenges in maintaining temporal coherence and control over subject motion. We propose VAST (Video As Storyboard from Text), a two-stage framework to address these challenges and enable high-quality video generation. In the first stage, StoryForge transforms textual descriptions into detailed storyboards, capturing human poses and object layouts to represent the structural essence of the scene. In the second stage, VisionForge generates videos from these storyboards, producing high-quality videos with smooth motion, temporal consistency, and spatial coherence. By decoupling text understanding from video generation, VAST enables precise control over subject dynamics and scene composition. Experiments on the VBench benchmark demonstrate that VAST outperforms existing methods in both visual quality and semantic expression, setting a new standard for dynamic and coherent video generation.

Rui Li, Ke Hao, Yuanzhi Liang et al.

Reinforcement learning, particularly Group Relative Policy Optimization (GRPO), has emerged as an effective framework for post-training visual generative models with human preference signals. However, its effectiveness is fundamentally limited by coarse reward credit assignment. In modern visual generation, multiple reward models are often used to capture heterogeneous objectives, such as visual quality, motion consistency, and text alignment. Existing GRPO pipelines typically collapse these rewards into a single static scalar and propagate it uniformly across the entire diffusion trajectory. This design ignores the stage-specific roles of different denoising steps and produces mistimed or incompatible optimization signals. To address this issue, we propose Objective-aware Trajectory Credit Assignment (OTCA), a structured framework for fine-grained GRPO training. OTCA consists of two key components. Trajectory-Level Credit Decomposition estimates the relative importance of different denoising steps. Multi-Objective Credit Allocation adaptively weights and combines multiple reward signals throughout the denoising process. By jointly modeling temporal credit and objective-level credit, OTCA converts coarse reward supervision into a structured, timestep-aware training signal that better matches the iterative nature of diffusion-based generation. Extensive experiments show that OTCA consistently improves both image and video generation quality across evaluation metrics.

Dianbing Xi, Jiepeng Wang, Yuanzhi Liang et al.

In this paper, we propose a novel framework for controllable video diffusion, OmniVDiff , aiming to synthesize and comprehend multiple video visual content in a single diffusion model. To achieve this, OmniVDiff treats all video visual modalities in the color space to learn a joint distribution, while employing an adaptive control strategy that dynamically adjusts the role of each visual modality during the diffusion process, either as a generation modality or a conditioning modality. Our framework supports three key capabilities: (1) Text-conditioned video generation, where all modalities are jointly synthesized from a textual prompt; (2) Video understanding, where structural modalities are predicted from rgb inputs in a coherent manner; and (3) X-conditioned video generation, where video synthesis is guided by finegrained inputs such as depth, canny and segmentation. Extensive experiments demonstrate that OmniVDiff achieves state-of-the-art performance in video generation tasks and competitive results in video understanding. Its flexibility and scalability make it well-suited for downstream applications such as video-to-video translation, modality adaptation for visual tasks, and scene reconstruction.

Sheng Liu, Yuanzhi Liang, Jiepeng Wang et al.

We present Uni-Inter, a unified framework for human motion generation that supports a wide range of interaction scenarios: including human-human, human-object, and human-scene-within a single, task-agnostic architecture. In contrast to existing methods that rely on task-specific designs and exhibit limited generalization, Uni-Inter introduces the Unified Interactive Volume (UIV), a volumetric representation that encodes heterogeneous interactive entities into a shared spatial field. This enables consistent relational reasoning and compound interaction modeling. Motion generation is formulated as joint-wise probabilistic prediction over the UIV, allowing the model to capture fine-grained spatial dependencies and produce coherent, context-aware behaviors. Experiments across three representative interaction tasks demonstrate that Uni-Inter achieves competitive performance and generalizes well to novel combinations of entities. These results suggest that unified modeling of compound interactions offers a promising direction for scalable motion synthesis in complex environments.

Rui Li, Yuanzhi Liang, Ziqi Ni et al.

Group Relative Policy Optimization (GRPO) has emerged as a powerful reinforcement learning paradigm for post-training video generation models. However, existing GRPO pipelines rely on static, fixed-capacity reward models whose evaluation behavior is frozen during training. Such rigid rewards introduce distributional bias, saturate quickly as the generator improves, and ultimately limit the stability and effectiveness of reinforcement-based alignment. We propose Self-Paced GRPO, a competence-aware GRPO framework in which reward feedback co-evolves with the generator. Our method introduces a progressive reward mechanism that automatically shifts its emphasis from coarse visual fidelity to temporal coherence and fine-grained text-video semantic alignment as generation quality increases. This self-paced curriculum alleviates reward-policy mismatch, mitigates reward exploitation, and yields more stable optimization. Experiments on VBench across multiple video generation backbones demonstrate consistent improvements in both visual quality and semantic alignment over GRPO baselines with static rewards, validating the effectiveness and generality of Self-Paced GRPO.

Ruiying Liu, Yuanzhi Liang, Haibin Huang et al.

Group Relative Policy Optimization (GRPO) has emerged as an effective and lightweight framework for post-training visual generative models. However, its performance is fundamentally limited by the ambiguity of textual visual correspondence: a single prompt may validly describe diverse visual outputs, and a single image or video may support multiple equally correct interpretations. This many to many relationship leads reward models to generate uncertain and weakly discriminative signals, causing GRPO to underutilize reliable feedback and overfit noisy ones. We introduce Bayesian Prior-Guided Optimization (BPGO), a novel extension of GRPO that explicitly models reward uncertainty through a semantic prior anchor. BPGO adaptively modulates optimization trust at two levels: inter-group Bayesian trust allocation emphasizes updates from groups consistent with the prior while down-weighting ambiguous ones, and intra-group prior-anchored renormalization sharpens sample distinctions by expanding confident deviations and compressing uncertain scores. Across both image and video generation tasks, BPGO delivers consistently stronger semantic alignment, enhanced perceptual fidelity, and faster convergence than standard GRPO and recent variants.

Ziqi Ni, Yuanzhi Liang, Rui Li et al.

Reinforcement learning (RL) has become a powerful tool for post-training visual generative models, with Group Relative Policy Optimization (GRPO) increasingly used to align generators with human preferences. However, existing GRPO pipelines rely on a single scalar reward per sample, treating each image or video as a holistic entity and ignoring the rich spatial and temporal structure of visual content. This coarse supervision hinders the correction of localized artifacts and the modeling of fine-grained perceptual cues. We introduce Visual Preference Policy Optimization (ViPO), a GRPO variant that lifts scalar feedback into structured, pixel-level advantages. ViPO employs a Perceptual Structuring Module that uses pretrained vision backbones to construct spatially and temporally aware advantage maps, redistributing optimization pressure toward perceptually important regions while preserving the stability of standard GRPO. Across both image and video benchmarks, ViPO consistently outperforms vanilla GRPO, improving in-domain alignment with human-preference rewards and enhancing generalization on out-of-domain evaluations. The method is architecture-agnostic, lightweight, and fully compatible with existing GRPO training pipelines, providing a more expressive and informative learning signal for visual generation.

Biao Liu, Yuanzhi Liang

Text-to-Image (T2I) synthesis has made significant advancements in recent years, driving applications such as generating datasets automatically. However, precise control over object localization in generated images remains a challenge. Existing methods fail to fully utilize positional information, leading to an inadequate understanding of object spatial layouts. To address this issue, we propose SpatialLock, a novel framework that leverages perception signals and grounding information to jointly control the generation of spatial locations. SpatialLock incorporates two components: Position-Engaged Injection (PoI) and Position-Guided Learning (PoG). PoI directly integrates spatial information through an attention layer, encouraging the model to learn the grounding information effectively. PoG employs perception-based supervision to further refine object localization. Together, these components enable the model to generate objects with precise spatial arrangements and improve the visual quality of the generated images. Experiments show that SpatialLock sets a new state-of-the-art for precise object positioning, achieving IOU scores above 0.9 across multiple datasets.

Chi Zhang, Jiepeng Wang, Youming Wang et al.

We present UniModel, a unified generative model that jointly supports visual understanding and visual generation within a single pixel-to-pixel diffusion framework. Our goal is to achieve unification along three axes: the model, the tasks, and the representations. At the representation level, we eliminate modality discrepancies by mapping both text and images into a shared visual space: textual prompts are rendered as painted text images on a clean canvas, and all inputs and outputs are treated purely as RGB pixels. This yields a fully vision-native formulation of multimodal learning. At the task level, a broad range of vision-language problems are cast as pixel-to-pixel transformations in this visual space. For understanding tasks, the model takes an RGB image and produces a painted text image that visually encodes the semantic prediction. For generation tasks, painted text images serve as visual conditions that guide realistic and semantically aligned image synthesis. Captioning and text-to-image generation thus become different directions of the same underlying visual translation process. At the model level, we instantiate a single Unified Diffusion Transformer trained with rectified flow in pixel space. A shared backbone jointly learns bidirectional mappings between natural images and painted text images, with lightweight task embeddings to specify the desired direction. Experiments on text-to-image synthesis and image-to-text understanding demonstrate strong cross-modal alignment and emergent controllability such as cycle-consistent image-caption-image loops. Our initial exploration suggests that unifying model, tasks, and representations in a single visual space is a promising paradigm for general-purpose multimodal intelligence.

Yuanzhi Liang, Yijie Fang, Rui Li et al.

Generative models have made significant progress in synthesizing visual content, including images, videos, and 3D/4D structures. However, they are typically trained with surrogate objectives such as likelihood or reconstruction loss, which often misalign with perceptual quality, semantic accuracy, or physical realism. Reinforcement learning (RL) offers a principled framework for optimizing non-differentiable, preference-driven, and temporally structured objectives. Recent advances demonstrate its effectiveness in enhancing controllability, consistency, and human alignment across generative tasks. This survey provides a systematic overview of RL-based methods for visual content generation. We review the evolution of RL from classical control to its role as a general-purpose optimization tool, and examine its integration into image, video, and 3D/4D generation. Across these domains, RL serves not only as a fine-tuning mechanism but also as a structural component for aligning generation with complex, high-level goals. We conclude with open challenges and future research directions at the intersection of RL and generative modeling.

Yuanzhi Liang, Linchao Zhu, Yi Yang

Large Language Models (LLMs) have demonstrated their ability to replicate human behaviors across a wide range of scenarios. However, their capability in handling complex, multi-character social interactions has yet to be fully explored, primarily due to the absence of robust, quantitative evaluation methods. This gap has slowed the development of agents proficient in more nuanced interactions beyond simple exchanges, for example, small talk. To address this challenge, we introduce the Multi-Agent Interaction Evaluation Framework (AntEval), encompassing a novel interaction framework and evaluation methods. The interaction framework aims to foster an complex interaction environment that bolsters information exchange and intention expression within social interactions. Furthermore, we introduce evaluation methods, including two metrics: Information Exchanging Precision (IEP) and Interaction Expressiveness Gap (IEG), designed for the quantitative and objective assessment of agents' interaction competencies. Our findings highlight the utility of these evaluative methods and show significant potential for improving LLMs' ability to construct agents that interact in a more natural manner with human-like intricacy.

Yuanzhi Liang, Yalong Bai, Wei Zhang et al.

Relationships encode the interactions among individual instances, and play a critical role in deep visual scene understanding. Suffering from the high predictability with non-visual information, existing methods tend to fit the statistical bias rather than ``learning'' to ``infer'' the relationships from images. To encourage further development in visual relationships, we propose a novel method to automatically mine more valuable relationships by pruning visually-irrelevant ones. We construct a new scene-graph dataset named Visually-Relevant Relationships Dataset (VrR-VG) based on Visual Genome. Compared with existing datasets, the performance gap between learnable and statistical method is more significant in VrR-VG, and frequency-based analysis does not work anymore. Moreover, we propose to learn a relationship-aware representation by jointly considering instances, attributes and relationships. By applying the representation-aware feature learned on VrR-VG, the performances of image captioning and visual question answering are systematically improved with a large margin, which demonstrates the gain of our dataset and the features embedding schema. VrR-VG is available via http://vrr-vg.com/.