We present FlagEvalMM, an open-source evaluation framework designed to comprehensively assess multimodal models across a diverse range of vision-language understanding and generation tasks, such as visual question answering, text-to-image/video generation, and image-text retrieval. We decouple model inference from evaluation through an independent evaluation service, thus enabling flexible resource allocation and seamless integration of new tasks and models. Moreover, FlagEvalMM utilizes advanced inference acceleration tools (e.g., vLLM, SGLang) and asynchronous data loading to significantly enhance evaluation efficiency. Extensive experiments show that FlagEvalMM offers accurate and efficient insights into model strengths and limitations, making it a valuable tool for advancing multimodal research. The framework is publicly accessible athttps://github.com/flageval-baai/FlagEvalMM.
Adapting text-to-image (T2I) latent diffusion models for video editing has shown strong visual fidelity and controllability, but challenges remain in maintaining causal relationships in video content. Edits affecting causally dependent attributes risk generating unrealistic or misleading outcomes if these relationships are ignored. In this work, we propose a causally faithful framework for counterfactual video generation, guided by a vision-language model (VLM). Our method is agnostic to the underlying video editing system and does not require access to its internal mechanisms or finetuning. Instead, we guide the generation by optimizing text prompts based on an assumed causal graph, addressing the challenge of latent space control in LDMs. We evaluate our approach using standard video quality metrics and counterfactual-specific criteria, such as causal effectiveness and minimality. Our results demonstrate that causally faithful video counterfactuals can be effectively generated within the learned distribution of LDMs through prompt-based causal steering. With its compatibility with any black-box video editing system, our method holds significant potential for generating realistic "what-if" video scenarios in diverse areas such as healthcare and digital media.
Digital human video generation is gaining traction in fields like education and e-commerce, driven by advancements in head-body animation and lip-syncing technologies. However, realistic Hand-Object Interaction (HOI) - the complex dynamics between human hands and objects - continues to pose challenges. Generating natural and believable HOI reenactments is difficult due to issues such as occlusion between hands and objects, variations in object shapes and orientations, and the necessity for precise physical interactions, and importantly, the ability to generalize to unseen humans and objects. This paper presents a novel framework iDiT-HOI that enables in-the-wild HOI reenactment generation. Specifically, we propose a unified inpainting-based token process method, called Inp-TPU, with a two-stage video diffusion transformer (DiT) model. The first stage generates a key frame by inserting the designated object into the hand region, providing a reference for subsequent frames. The second stage ensures temporal coherence and fluidity in hand-object interactions. The key contribution of our method is to reuse the pretrained model's context perception capabilities without introducing additional parameters, enabling strong generalization to unseen objects and scenarios, and our proposed paradigm naturally supports long video generation. Comprehensive evaluations demonstrate that our approach outperforms existing methods, particularly in challenging real-world scenes, offering enhanced realism and more seamless hand-object interactions.
Masked-based autoregressive models have demonstrated promising image generation capability in continuous space. However, their potential for video generation remains under-explored. In this paper, we propose \textbf{VideoMAR}, a concise and efficient decoder-only autoregressive image-to-video model with continuous tokens, composing temporal frame-by-frame and spatial masked generation. We first identify temporal causality and spatial bi-directionality as the first principle of video AR models, and propose the next-frame diffusion loss for the integration of mask and video generation. Besides, the huge cost and difficulty of long sequence autoregressive modeling is a basic but crucial issue. To this end, we propose the temporal short-to-long curriculum learning and spatial progressive resolution training, and employ progressive temperature strategy at inference time to mitigate the accumulation error. Furthermore, VideoMAR replicates several unique capacities of language models to video generation. It inherently bears high efficiency due to simultaneous temporal-wise KV cache and spatial-wise parallel generation, and presents the capacity of spatial and temporal extrapolation via 3D rotary embeddings. On the VBench-I2V benchmark, VideoMAR surpasses the previous state-of-the-art (Cosmos I2V) while requiring significantly fewer parameters ($9.3\%$), training data ($0.5\%$), and GPU resources ($0.2\%$).
We introduce PlayerOne, the first egocentric realistic world simulator, facilitating immersive and unrestricted exploration within vividly dynamic environments. Given an egocentric scene image from the user, PlayerOne can accurately construct the corresponding world and generate egocentric videos that are strictly aligned with the real scene human motion of the user captured by an exocentric camera. PlayerOne is trained in a coarse-to-fine pipeline that first performs pretraining on large-scale egocentric text-video pairs for coarse-level egocentric understanding, followed by finetuning on synchronous motion-video data extracted from egocentric-exocentric video datasets with our automatic construction pipeline. Besides, considering the varying importance of different components, we design a part-disentangled motion injection scheme, enabling precise control of part-level movements. In addition, we devise a joint reconstruction framework that progressively models both the 4D scene and video frames, ensuring scene consistency in the long-form video generation. Experimental results demonstrate its great generalization ability in precise control of varying human movements and worldconsistent modeling of diverse scenarios. It marks the first endeavor into egocentric real-world simulation and can pave the way for the community to delve into fresh frontiers of world modeling and its diverse applications.
Despite rapid advancements in video generation models, generating coherent storytelling videos that span multiple scenes and characters remains challenging. Current methods often rigidly convert pre-generated keyframes into fixed-length clips, resulting in disjointed narratives and pacing issues. Furthermore, the inherent instability of video generation models means that even a single low-quality clip can significantly degrade the entire output animation's logical coherence and visual continuity. To overcome these obstacles, we introduce AniMaker, a multi-agent framework enabling efficient multi-candidate clip generation and storytelling-aware clip selection, thus creating globally consistent and story-coherent animation solely from text input. The framework is structured around specialized agents, including the Director Agent for storyboard generation, the Photography Agent for video clip generation, the Reviewer Agent for evaluation, and the Post-Production Agent for editing and voiceover. Central to AniMaker's approach are two key technical components: MCTS-Gen in Photography Agent, an efficient Monte Carlo Tree Search (MCTS)-inspired strategy that intelligently navigates the candidate space to generate high-potential clips while optimizing resource usage; and AniEval in Reviewer Agent, the first framework specifically designed for multi-shot animation evaluation, which assesses critical aspects such as story-level consistency, action completion, and animation-specific features by considering each clip in the context of its preceding and succeeding clips. Experiments demonstrate that AniMaker achieves superior quality as measured by popular metrics including VBench and our proposed AniEval framework, while significantly improving the efficiency of multi-candidate generation, pushing AI-generated storytelling animation closer to production standards.
The flourishing of video generation technologies has endangered the credibility of real-world information and intensified the demand for AI-generated video detectors. Despite some progress, the lack of high-quality real-world datasets hinders the development of trustworthy detectors. In this paper, we propose GenWorld, a large-scale, high-quality, and real-world simulation dataset for AI-generated video detection. GenWorld features the following characteristics: (1) Real-world Simulation: GenWorld focuses on videos that replicate real-world scenarios, which have a significant impact due to their realism and potential influence; (2) High Quality: GenWorld employs multiple state-of-the-art video generation models to provide realistic and high-quality forged videos; (3) Cross-prompt Diversity: GenWorld includes videos generated from diverse generators and various prompt modalities (e.g., text, image, video), offering the potential to learn more generalizable forensic features. We analyze existing methods and find they fail to detect high-quality videos generated by world models (i.e., Cosmos), revealing potential drawbacks of ignoring real-world clues. To address this, we propose a simple yet effective model, SpannDetector, to leverage multi-view consistency as a strong criterion for real-world AI-generated video detection. Experiments show that our method achieves superior results, highlighting a promising direction for explainable AI-generated video detection based on physical plausibility. We believe that GenWorld will advance the field of AI-generated video detection. Project Page: https://chen-wl20.github.io/GenWorld
Recent progress in diffusion models has greatly enhanced video generation quality, yet these models still require fine-tuning to improve specific dimensions like instance preservation, motion rationality, composition, and physical plausibility. Existing fine-tuning approaches often rely on human annotations and large-scale computational resources, limiting their practicality. In this work, we propose GigaVideo-1, an efficient fine-tuning framework that advances video generation without additional human supervision. Rather than injecting large volumes of high-quality data from external sources, GigaVideo-1 unlocks the latent potential of pre-trained video diffusion models through automatic feedback. Specifically, we focus on two key aspects of the fine-tuning process: data and optimization. To improve fine-tuning data, we design a prompt-driven data engine that constructs diverse, weakness-oriented training samples. On the optimization side, we introduce a reward-guided training strategy, which adaptively weights samples using feedback from pre-trained vision-language models with a realism constraint. We evaluate GigaVideo-1 on the VBench-2.0 benchmark using Wan2.1 as the baseline across 17 evaluation dimensions. Experiments show that GigaVideo-1 consistently improves performance on almost all the dimensions with an average gain of about 4% using only 4 GPU-hours. Requiring no manual annotations and minimal real data, GigaVideo-1 demonstrates both effectiveness and efficiency. Code, model, and data will be publicly available.
In e-commerce and digital marketing, generating high-fidelity human-product demonstration videos is important for effective product presentation. However, most existing frameworks either fail to preserve the identities of both humans and products or lack an understanding of human-product spatial relationships, leading to unrealistic representations and unnatural interactions. To address these challenges, we propose a Diffusion Transformer (DiT)-based framework. Our method simultaneously preserves human identities and product-specific details, such as logos and textures, by injecting paired human-product reference information and utilizing an additional masked cross-attention mechanism. We employ a 3D body mesh template and product bounding boxes to provide precise motion guidance, enabling intuitive alignment of hand gestures with product placements. Additionally, structured text encoding is used to incorporate category-level semantics, enhancing 3D consistency during small rotational changes across frames. Trained on a hybrid dataset with extensive data augmentation strategies, our approach outperforms state-of-the-art techniques in maintaining the identity integrity of both humans and products and generating realistic demonstration motions. Project page: https://submit2025-dream.github.io/DreamActor-H1/.
Traffic Accident Anticipation (TAA) in traffic scenes is a challenging problem for achieving zero fatalities in the future. Current approaches typically treat TAA as a supervised learning task needing the laborious annotation of accident occurrence duration. However, the inherent long-tailed, uncertain, and fast-evolving nature of traffic scenes has the problem that real causal parts of accidents are difficult to identify and are easily dominated by data bias, resulting in a background confounding issue. Thus, we propose an Attentive Video Diffusion (AVD) model that synthesizes additional accident video clips by generating the causal part in dashcam videos, i.e., from normal clips to accident clips. AVD aims to generate causal video frames based on accident or accident-free text prompts while preserving the style and content of frames for TAA after video generation. This approach can be trained using datasets collected from various driving scenes without any extra annotations. Additionally, AVD facilitates an Equivariant TAA (EQ-TAA) with an equivariant triple loss for an anchor accident-free video clip, along with the generated pair of contrastive pseudo-normal and pseudo-accident clips. Extensive experiments have been conducted to evaluate the performance of AVD and EQ-TAA, and competitive performance compared to state-of-the-art methods has been obtained.
Text-to-video generation has significantly enriched content creation and holds the potential to evolve into powerful world simulators. However, modeling the vast spatiotemporal space remains computationally demanding, particularly when employing Transformers, which incur quadratic complexity in sequence processing and thus limit practical applications. Recent advancements in linear-time sequence modeling, particularly the Mamba architecture, offer a more efficient alternative. Nevertheless, its plain design limits its direct applicability to multi-modal and spatiotemporal video generation tasks. To address these challenges, we introduce M4V, a Multi-Modal Mamba framework for text-to-video generation. Specifically, we propose a multi-modal diffusion Mamba (MM-DiM) block that enables seamless integration of multi-modal information and spatiotemporal modeling through a multi-modal token re-composition design. As a result, the Mamba blocks in M4V reduce FLOPs by 45% compared to the attention-based alternative when generating videos at 768$\times$1280 resolution. Additionally, to mitigate the visual quality degradation in long-context autoregressive generation processes, we introduce a reward learning strategy that further enhances per-frame visual realism. Extensive experiments on text-to-video benchmarks demonstrate M4V's ability to produce high-quality videos while significantly lowering computational costs. Code and models will be publicly available at https://huangjch526.github.io/M4V_project.
We introduce Self Forcing, a novel training paradigm for autoregressive video diffusion models. It addresses the longstanding issue of exposure bias, where models trained on ground-truth context must generate sequences conditioned on their own imperfect outputs during inference. Unlike prior methods that denoise future frames based on ground-truth context frames, Self Forcing conditions each frame's generation on previously self-generated outputs by performing autoregressive rollout with key-value (KV) caching during training. This strategy enables supervision through a holistic loss at the video level that directly evaluates the quality of the entire generated sequence, rather than relying solely on traditional frame-wise objectives. To ensure training efficiency, we employ a few-step diffusion model along with a stochastic gradient truncation strategy, effectively balancing computational cost and performance. We further introduce a rolling KV cache mechanism that enables efficient autoregressive video extrapolation. Extensive experiments demonstrate that our approach achieves real-time streaming video generation with sub-second latency on a single GPU, while matching or even surpassing the generation quality of significantly slower and non-causal diffusion models. Project website: http://self-forcing.github.io/
Audio is inherently temporal and closely synchronized with the visual world, making it a naturally aligned and expressive control signal for controllable video generation (e.g., movies). Beyond control, directly translating audio into video is essential for understanding and visualizing rich audio narratives (e.g., Podcasts or historical recordings). However, existing approaches fall short in generating high-quality videos with precise audio-visual synchronization, especially across diverse and complex audio types. In this work, we introduce MTV, a versatile framework for audio-sync video generation. MTV explicitly separates audios into speech, effects, and music tracks, enabling disentangled control over lip motion, event timing, and visual mood, respectively -- resulting in fine-grained and semantically aligned video generation. To support the framework, we additionally present DEMIX, a dataset comprising high-quality cinematic videos and demixed audio tracks. DEMIX is structured into five overlapped subsets, enabling scalable multi-stage training for diverse generation scenarios. Extensive experiments demonstrate that MTV achieves state-of-the-art performance across six standard metrics spanning video quality, text-video consistency, and audio-video alignment. Project page: https://hjzheng.net/projects/MTV/.
Generative models, with their success in image and video generation, have recently been explored for synthesizing effective neural network weights. These approaches take trained neural network checkpoints as training data, and aim to generate high-performing neural network weights during inference. In this work, we examine four representative methods on their ability to generate novel model weights, i.e., weights that are different from the checkpoints seen during training. Surprisingly, we find that these methods synthesize weights largely by memorization: they produce either replicas, or at best simple interpolations, of the training checkpoints. Current methods fail to outperform simple baselines, such as adding noise to the weights or taking a simple weight ensemble, in obtaining different and simultaneously high-performing models. We further show that this memorization cannot be effectively mitigated by modifying modeling factors commonly associated with memorization in image diffusion models, or applying data augmentations. Our findings provide a realistic assessment of what types of data current generative models can model, and highlight the need for more careful evaluation of generative models in new domains. Our code is available at https://github.com/boyazeng/weight_memorization.
Despite recent advances in video generation, existing models still lack fine-grained controllability, especially for multi-subject customization with consistent identity and interaction. In this paper, we propose PolyVivid, a multi-subject video customization framework that enables flexible and identity-consistent generation. To establish accurate correspondences between subject images and textual entities, we design a VLLM-based text-image fusion module that embeds visual identities into the textual space for precise grounding. To further enhance identity preservation and subject interaction, we propose a 3D-RoPE-based enhancement module that enables structured bidirectional fusion between text and image embeddings. Moreover, we develop an attention-inherited identity injection module to effectively inject fused identity features into the video generation process, mitigating identity drift. Finally, we construct an MLLM-based data pipeline that combines MLLM-based grounding, segmentation, and a clique-based subject consolidation strategy to produce high-quality multi-subject data, effectively enhancing subject distinction and reducing ambiguity in downstream video generation. Extensive experiments demonstrate that PolyVivid achieves superior performance in identity fidelity, video realism, and subject alignment, outperforming existing open-source and commercial baselines.
Existing acceleration techniques for video diffusion models often rely on uniform heuristics or time-embedding variants to skip timesteps and reuse cached features. These approaches typically require extensive calibration with curated prompts and risk inconsistent outputs due to prompt-specific overfitting. In this paper, we introduce a novel and robust discovery: a unified magnitude law observed across different models and prompts. Specifically, the magnitude ratio of successive residual outputs decreases monotonically and steadily in most timesteps while rapidly in the last several steps. Leveraging this insight, we introduce a Magnitude-aware Cache (MagCache) that adaptively skips unimportant timesteps using an error modeling mechanism and adaptive caching strategy. Unlike existing methods requiring dozens of curated samples for calibration, MagCache only requires a single sample for calibration. Experimental results show that MagCache achieves 2.1x and 2.68x speedups on Open-Sora and Wan 2.1, respectively, while preserving superior visual fidelity. It significantly outperforms existing methods in LPIPS, SSIM, and PSNR, under comparable computational budgets.
Collecting and annotating real-world data for safety-critical physical AI systems, such as Autonomous Vehicle (AV), is time-consuming and costly. It is especially challenging to capture rare edge cases, which play a critical role in training and testing of an AV system. To address this challenge, we introduce the Cosmos-Drive-Dreams - a synthetic data generation (SDG) pipeline that aims to generate challenging scenarios to facilitate downstream tasks such as perception and driving policy training. Powering this pipeline is Cosmos-Drive, a suite of models specialized from NVIDIA Cosmos world foundation model for the driving domain and are capable of controllable, high-fidelity, multi-view, and spatiotemporally consistent driving video generation. We showcase the utility of these models by applying Cosmos-Drive-Dreams to scale the quantity and diversity of driving datasets with high-fidelity and challenging scenarios. Experimentally, we demonstrate that our generated data helps in mitigating long-tail distribution problems and enhances generalization in downstream tasks such as 3D lane detection, 3D object detection and driving policy learning. We open source our pipeline toolkit, dataset and model weights through the NVIDIA's Cosmos platform. Project page: https://research.nvidia.com/labs/toronto-ai/cosmos_drive_dreams
To address key limitations in human-object interaction (HOI) video generation -- specifically the reliance on curated motion data, limited generalization to novel objects/scenarios, and restricted accessibility -- we introduce HunyuanVideo-HOMA, a weakly conditioned multimodal-driven framework. HunyuanVideo-HOMA enhances controllability and reduces dependency on precise inputs through sparse, decoupled motion guidance. It encodes appearance and motion signals into the dual input space of a multimodal diffusion transformer (MMDiT), fusing them within a shared context space to synthesize temporally consistent and physically plausible interactions. To optimize training, we integrate a parameter-space HOI adapter initialized from pretrained MMDiT weights, preserving prior knowledge while enabling efficient adaptation, and a facial cross-attention adapter for anatomically accurate audio-driven lip synchronization. Extensive experiments confirm state-of-the-art performance in interaction naturalness and generalization under weak supervision. Finally, HunyuanVideo-HOMA demonstrates versatility in text-conditioned generation and interactive object manipulation, supported by a user-friendly demo interface. The project page is at https://anonymous.4open.science/w/homa-page-0FBE/.
With the rapid development of text-to-vision generation diffusion models, classifier-free guidance has emerged as the most prevalent method for conditioning. However, this approach inherently requires twice as many steps for model forwarding compared to unconditional generation, resulting in significantly higher costs. While previous study has introduced the concept of adaptive guidance, it lacks solid analysis and empirical results, making previous method unable to be applied to general diffusion models. In this work, we present another perspective of applying adaptive guidance and propose Step AG, which is a simple, universally applicable adaptive guidance strategy. Our evaluations focus on both image quality and image-text alignment. whose results indicate that restricting classifier-free guidance to the first several denoising steps is sufficient for generating high-quality, well-conditioned images, achieving an average speedup of 20% to 30%. Such improvement is consistent across different settings such as inference steps, and various models including video generation models, highlighting the superiority of our method.
From professional filmmaking to user-generated content, creators and consumers have long recognized that the power of video depends on the harmonious integration of what we hear (the video's audio track) with what we see (the video's image sequence). Current approaches to video generation either ignore sound to focus on general-purpose but silent image sequence generation or address both visual and audio elements but focus on restricted application domains such as re-dubbing. We introduce Mirage, an audio-to-video foundation model that excels at generating realistic, expressive output imagery from scratch given an audio input. When integrated with existing methods for speech synthesis (text-to-speech, or TTS), Mirage results in compelling multimodal video. When trained on audio-video footage of people talking (A-roll) and conditioned on audio containing speech, Mirage generates video of people delivering a believable interpretation of the performance implicit in input audio. Our central technical contribution is a unified method for training self-attention-based audio-to-video generation models, either from scratch or given existing weights. This methodology allows Mirage to retain generality as an approach to audio-to-video generation while producing outputs of superior subjective quality to methods that incorporate audio-specific architectures or loss components specific to people, speech, or details of how images or audio are captured. We encourage readers to watch and listen to the results of Mirage for themselves (see paper and comments for links).