Daily Papers

AI-generated videos (AIGVs) typically contain subtle temporal artifacts that arise from inter-frame inconsistencies rather than within individual frames. A detector that captures such artifacts should therefore benefit from video pretrained backbones over image only ones. In practice, however, video backbones with standard global readouts often fail to outperform strong image pretrained probes on AIGV benchmarks. We attribute this gap to excessive spatiotemporal aggregation in the readout. Video pretrained backbones tend to compress each frame into a single global descriptor. This compression suppresses local patch level temporal dynamics and discards inter patch relations, which are precisely the cues that AIGV detection most reliably depends on. Based on this, we propose Velocity Gated Patch Velocity Profiling (V-PVP), a lightweight readout that replaces only the aggregation layer with two parallel streams over the patch velocity field, adding only about 0.5M trainable parameters. V-PVP serves as a general plug-and-play module that consistently improves performance across diverse video backbones under both end-to-end fine-tuning and linear probing settings. Our method reaches 95.28 AUC on AIGVDBench while keeping the backbone fully frozen. The results show that simply replacing the aggregation layer reactivates the temporal potential of frozen video backbones, restoring their advantage on AIGV detection. Code is available at https://anonymous.4open.science/r/PVP-81B3/.

Recent world action models draw on video generation priors to jointly predict future observations and robot actions. Adapting these priors to control requires determining how future predictions should guide policy learning. RAP-WAM combines predictive Rehearsal and Action Perturbations (RAP) with future-supervised queries and Coupled Action Projection for policy learning. Predictive rehearsal trains on model-generated inputs while retaining recorded targets, and accumulates recurring errors into a spatial prior. We perturb actions and measure changes in video predictions. Action-response-based spatiotemporal credit assignment combines these responses, prediction error, and the rehearsal prior to allocate visual supervision; spatially aggregating the responses determines supervision intensity across future action steps. No separate scoring network or deployment-time probe is required. The base learns from a video-pretrained backbone without intermediate pretraining on external embodied data. We evaluate the fixed base and its agent-refined, task-conditioned system. On 42 RoboDojo tasks, their average success rates are 12.85% and 15.62%, with Scores of 17.86 and 20.59. Across 15 real-robot tasks, the complete system achieves 89.00% task-macro-average success, exceeding the strongest of five baselines by 37.67 percentage points. Across 315 fixed-output simulation windows, 45.21% of response-top20 positions fall outside error-top20, revealing information beyond prediction error. Videos and further details are available on the project website: https://rapwam.github.io/.

Driven by next-token prediction, NLP shifted from task-specific models into powerful generalist foundation models. What, then, is the equivalent catalyst needed to achieve a general-purpose model in computer vision? In this paper, we contend that large-scale text-to-video generation serves as a strong pre-training paradigm for computer vision, providing the necessary spatiotemporal priors, vision-language alignment, and scalability required for general visual intelligence. We introduce GenCeption, which leverages a pre-trained video generative diffusion backbone to define a feed-forward perception model, capable of performing various vision tasks steered by text instructions. Empirical results demonstrate that GenCeption achieves state-of-the-art performance across a diverse suite of tasks, including depth, surface normal, and camera pose estimation, expression-referring segmentation, and 3D keypoint prediction, often matching or surpassing specialized models (e.g. DepthAnything3, SAM3, D4RT, VGGT-Omega, Sapiens, David, Genmo, and Lotus-2). Furthermore, the video generative pretrained backbone outperforms alternative pretraining paradigms (e.g., V-JEPA, and Video MAE) under comparable settings. Importantly, GenCeption exhibits preliminary data and model scaling properties along with exceptional data efficiency, where it achieves comparable performance with leading models like D4RT and VGGT-Omega with 7 to 500 less training data. Finally, GenCeption also exhibits intriguing emergent behaviors: a model trained exclusively on synthetic human videos generalizes to real-world footage and out-of-distribution object categories (e.g., animals and robots). These findings suggest that video generation is not merely a synthesis tool, but a foundational path toward generalist vision intelligence for the physical world. Project page: https://genception.github.io

Instruction-based video editing is commonly built on video-pretrained generative backbones: a video diffusion transformer is adapted, at considerable cost, to condition on a source video and an editing instruction. In this report we explore a different route and show that a strong instruction-based image editing model can edit videos by operating directly on video-VAE latents. Starting from Qwen-Image-Edit, we arrange the latent frames of a Wan~2.1 video VAE as tiles of one large virtual image, reuse the editor's image positional encoding for every tile, and bridge the two latent spaces with a pair of lightweight input/output projections warm-started from the editor's own patchify and unpatchify layers, so that at initialization a (static) video is embedded exactly as an image the model already understands. The whole system is then fine-tuned on the public Ditto-1M editing triplets, and a few denoising steps of Wan~2.2 serve as an optional temporal enhancer. We motivate the design with a chain of zero-training observations: the stock image editor already edits a video presented as a contact sheet; it is indifferent to whether the sheet's tokens come from one joint encode or from per-frame encodes stitched in latent space; and it even edits genuine video latents zero-shot to a clearly recognizable degree, leaving fine-tuning only a fidelity gap to close. Our results suggest that, despite the large investment in training video latent spaces, per-frame video latents remain close enough to the image domain that mature image editing priors transfer with minimal adaptation. Project Page: https://yunpeng1998.github.io/Qwen-Video-Edit-Page; Code: https://github.com/yunpeng1998/Qwen-Video-Edit; Model: https://huggingface.co/yunpeng1998/Qwen-Video-Edit.

Unified models for robot manipulation aim to equip one policy with both the semantic priors of pretrained VLMs and the physical dynamics learned through future prediction. In practice, existing designs tend to erode the semantics of the pretrained backbone, suffer interference among heterogeneous objectives, and learn future prediction from scratch in pixel space, leaving the dynamics priors of pretrained video generators unexploited. We present InternVLA-A1.5, which builds the policy on a native VLM backbone that keeps training on VQA and subtask prediction, and attaches a lightweight unified expert for continuous action generation. Future prediction is recast as a latent-querying problem, where a small set of learnable foresight tokens condenses the task-relevant future into a compact latent code under the supervision of a frozen pretrained video generation model, so the policy inherits world-model dynamics priors without ever learning pixel-level generation. The video branch is discarded at inference, keeping real-time control. Pretrained on 1.2M robot episodes and 3M multimodal samples, InternVLA-A1.5 achieves the best overall results on all six simulation benchmarks. In the real world, the preserved semantics deliver the strongest compositional generalization on held-out instruction bindings, and the two designs together sustain long-horizon execution.

Temporal action localization (TAL) requires long-form reasoning to predict actions of various durations and complex content. Given limited GPU memory, training TAL end to end (i.e., from videos to predictions) on long videos is a significant challenge. Most methods can only train on pre-extracted features without optimizing them for the localization problem, consequently limiting localization performance. In this work, to extend the potential in TAL networks, we propose a novel end-to-end method Re2TAL, which rewires pretrained video backbones for reversible TAL. Re2TAL builds a backbone with reversible modules, where the input can be recovered from the output such that the bulky intermediate activations can be cleared from memory during training. Instead of designing one single type of reversible module, we propose a network rewiring mechanism, to transform any module with a residual connection to a reversible module without changing any parameters. This provides two benefits: (1) a large variety of reversible networks are easily obtained from existing and even future model designs, and (2) the reversible models require much less training effort as they reuse the pre-trained parameters of their original non-reversible versions. Re2TAL, only using the RGB modality, reaches 37.01% average mAP on ActivityNet-v1.3, a new state-of-the-art record, and mAP 64.9% at tIoU=0.5 on THUMOS-14, outperforming all other RGB-only methods.

Transitioning bidirectional video diffusion models into an autoregressive paradigm improves the interactivity of video world models, but existing causal pipelines need many stages (control fine-tuning, autoregressive training, causal initialization, few-step distillation) and still trail bidirectional models in quality due to error accumulation. Recent world models such as Yume-1.5 and Matrix-Game-3.0 instead adopt a bidirectional autoregressive approach, gaining fidelity and stable long-horizon rollout from self-correcting error propagation, yet open-source frameworks (e.g., minWM) support only causal models. We present BiWM, the first full-stack framework for interactive video world models under the bidirectional autoregressive paradigm, jointly optimizing generation quality and inference speed. From a pretrained video backbone, BiWM injects camera control by fine-tuning, then runs a few-step Distribution Matching Distillation (DMD) stage that turns the backbone into an action/camera-controllable world model: just two training stages instead of four in minWM, converging in a few hundred steps on 8xH200 GPUs. A single recipe spans Wan2.1-1.3B, Wan2.2-5B, HunyuanVideo-1.5-8B, and LTX-2.3-22B, and also supports secondary fine-tuning of existing bidirectional models. BiWM enables real-world camera control where minWM loses controllability, integrates pluggable history compression (FramePack-style and PackForcing-style) for long rollouts, and offers an optional NVFP4 4-bit training/inference pipeline. To counter DMD's mode-seeking degradation, we add GAN and mass-covering forward-KL objectives that preserve scene dynamics. We open-source BiWM for resource-constrained research and high-fidelity environment simulation.

Human video generation remains challenging due to the difficulty of jointly modeling human appearance, motion, and camera viewpoint under limited multi-view data. Existing methods often address these factors separately, resulting in limited controllability or reduced visual quality. We revisit this problem from an image-first perspective, where high-quality human appearance is learned via image generation and used as a prior for video synthesis, decoupling appearance modeling from temporal consistency. We propose a pose- and viewpoint-controllable pipeline that combines a pretrained image backbone with SMPL-X-based motion guidance, together with a training-free temporal refinement stage based on a pretrained video diffusion model. Our method produces high-quality, temporally consistent videos under diverse poses and viewpoints. We also release a canonical human dataset and an auxiliary model for compositional human image synthesis. Code and data are publicly available at https://github.com/Taited/ReImagine.

Interactive video generation systems for camera-controlled world exploration roll out growing sequences of latent video frames, entangling state transition with high-frequency observation synthesis. We propose Walking in the Implicit, a scene-centric paradigm that changes the rollout variable from frame latents to a fixed-length, renderable implicit state, termed Neural Implicit Scene (NIS). This factorizes interactive generation into stochastic transition of a compact scene state and deterministic pose-conditioned rendering given the sampled state. We instantiate this paradigm as NeuWorld: a transformer VAE learns locally anchored NIS from sparse posed frames, and a diffusion transformer evolves NIS conditioned on future camera trajectories and geometry-aware retrieved history. By reusing the VAE encoder as a unified conditioner, NeuWorld maps camera, reference-image, and history cues into the same NIS modality, avoiding external heterogeneous encoders. Trained from scratch on public posed-view data without pretrained video backbones or auxiliary 3D reconstructors, NeuWorld achieves strong long-horizon consistency with favorable inference efficiency.

We address the task of video style transfer with diffusion models, where the goal is to preserve the context of an input video while rendering it in a target style specified by a text prompt. A major challenge is the lack of paired video data for supervision. We propose PickStyle, a video-to-video style transfer framework that augments pretrained video diffusion backbones with style adapters and benefits from paired still image data with source-style correspondences for training. PickStyle inserts low-rank adapters into the self-attention layers of conditioning modules, enabling efficient specialization for motion-style transfer while maintaining strong alignment between video content and style. To bridge the gap between static image supervision and dynamic video, we construct synthetic training clips from paired images by applying shared augmentations that simulate camera motion, ensuring temporal priors are preserved. In addition, we introduce Context-Style Classifier-Free Guidance (CS-CFG), a novel factorization of classifier-free guidance into independent text (style) and video (context) directions. CS-CFG ensures that context is preserved in generated video while the style is effectively transferred. Experiments across benchmarks show that our approach achieves temporally coherent, style-faithful, and content-preserving video translations, outperforming existing baselines both qualitatively and quantitatively.

This study focuses on a challenging yet promising task, Text-to-Sounding-Video (T2SV) generation, which aims to generate a video with synchronized audio from text conditions, meanwhile ensuring both modalities are aligned with text. Despite progress in joint audio-video training, two critical challenges still remain unaddressed: (1) a single, shared text caption where the text for video is equal to the text for audio often creates modal interference, confusing the pretrained backbones, and (2) the optimal mechanism for cross-modal feature interaction remains unclear. To address these challenges, we first propose the Hierarchical Visual-Grounded Captioning (HVGC) framework that generates pairs of disentangled captions, a video caption, and an audio caption, eliminating interference at the conditioning stage. Based on HVGC, we further introduce BridgeDiT, a novel dual-tower diffusion transformer, which employs a Dual CrossAttention (DCA) mechanism that acts as a robust ``bridge" to enable a symmetric, bidirectional exchange of information, achieving both semantic and temporal synchronization. Extensive experiments on three benchmark datasets, supported by human evaluations, demonstrate that our method achieves state-of-the-art results on most metrics. Comprehensive ablation studies further validate the effectiveness of our contributions, offering key insights for the future T2SV task. All the codes and checkpoints will be publicly released.

While large-scale diffusion models have revolutionized video synthesis, achieving precise control over both multi-subject identity and multi-granularity motion remains a significant challenge. Recent attempts to bridge this gap often suffer from limited motion granularity, control ambiguity, and identity degradation, leading to suboptimal performance on identity preservation and motion control. In this work, we present DreamVideo-Omni, a unified framework enabling harmonious multi-subject customization with omni-motion control via a progressive two-stage training paradigm. In the first stage, we integrate comprehensive control signals for joint training, encompassing subject appearances, global motion, local dynamics, and camera movements. To ensure robust and precise controllability, we introduce a condition-aware 3D rotary positional embedding to coordinate heterogeneous inputs and a hierarchical motion injection strategy to enhance global motion guidance. Furthermore, to resolve multi-subject ambiguity, we introduce group and role embeddings to explicitly anchor motion signals to specific identities, effectively disentangling complex scenes into independent controllable instances. In the second stage, to mitigate identity degradation, we design a latent identity reward feedback learning paradigm by training a latent identity reward model upon a pretrained video diffusion backbone. This provides motion-aware identity rewards in the latent space, prioritizing identity preservation aligned with human preferences. Supported by our curated large-scale dataset and the comprehensive DreamOmni Bench for multi-subject and omni-motion control evaluation, DreamVideo-Omni demonstrates superior performance in generating high-quality videos with precise controllability.

We present XKD, a novel self-supervised framework to learn meaningful representations from unlabelled videos. XKD is trained with two pseudo objectives. First, masked data reconstruction is performed to learn modality-specific representations from audio and visual streams. Next, self-supervised cross-modal knowledge distillation is performed between the two modalities through a teacher-student setup to learn complementary information. We introduce a novel domain alignment strategy to tackle domain discrepancy between audio and visual modalities enabling effective cross-modal knowledge distillation. Additionally, to develop a general-purpose network capable of handling both audio and visual streams, modality-agnostic variants of XKD are introduced, which use the same pretrained backbone for different audio and visual tasks. Our proposed cross-modal knowledge distillation improves video action classification by 8% to 14% on UCF101, HMDB51, and Kinetics400. Additionally, XKD improves multimodal action classification by 5.5% on Kinetics-Sound. XKD shows state-of-the-art performance in sound classification on ESC50, achieving top-1 accuracy of 96.5%.

Interactive long video generation requires prompt switching to introduce new subjects or events, while maintaining perceptual fidelity and coherent motion over extended horizons. Recent distilled streaming video diffusion models reuse a rolling KV cache for long-range generation, enabling prompt-switch interaction through re-cache at each switch. However, existing streaming methods still exhibit progressive quality degradation and weakened motion dynamics. We identify two failure modes specific to interactive streaming generation: (i) at each prompt switch, current cache maintenance cannot simultaneously retain KV-based semantic context and recent latent cues, resulting in weak boundary conditioning and reduced perceptual quality; and (ii) during distillation, unbounded time indexing induces a positional distribution shift from the pretrained backbone's bounded RoPE regime, weakening pretrained motion priors and long-horizon motion retention. To address these issues, we propose Anchor Forcing, a cache-centric framework with two designs. First, an anchor-guided re-cache mechanism stores KV states in anchor caches and warm-starts re-cache from these anchors at each prompt switch, reducing post-switch evidence loss and stabilizing perceptual quality. Second, a tri-region RoPE with region-specific reference origins, together with RoPE re-alignment distillation, reconciles unbounded streaming indices with the pretrained RoPE regime to better retain motion priors. Experiments on long videos show that our method improves perceptual quality and motion metrics over prior streaming baselines in interactive settings. Project page: https://github.com/vivoCameraResearch/Anchor-Forcing

State-of-the-art Vision-Language-Action (VLA) models excel at semantic generalization but struggle to generalize to unseen physical motions in novel environments. We introduce DreamZero, a World Action Model (WAM) built upon a pretrained video diffusion backbone. Unlike VLAs, WAMs learn physical dynamics by predicting future world states and actions, using video as a dense representation of how the world evolves. By jointly modeling video and action, DreamZero learns diverse skills effectively from heterogeneous robot data without relying on repetitive demonstrations. This results in over 2x improvement in generalization to new tasks and environments compared to state-of-the-art VLAs in real robot experiments. Crucially, through model and system optimizations, we enable a 14B autoregressive video diffusion model to perform real-time closed-loop control at 7Hz. Finally, we demonstrate two forms of cross-embodiment transfer: video-only demonstrations from other robots or humans yield a relative improvement of over 42% on unseen task performance with just 10-20 minutes of data. More surprisingly, DreamZero enables few-shot embodiment adaptation, transferring to a new embodiment with only 30 minutes of play data while retaining zero-shot generalization.

Manipulation policies must know which objects matter and where they are, yet the pretrained backbones that current robot foundation models build on, from language in vision-language-action models (VLAs) to video generation in world-action models (WAMs), do not directly require this metric grounding, leaving it to be learned implicitly from robot demonstrations. We propose Grounded Action Models (GAMs), a new paradigm of robot foundation models built with 3D grounding. GAM can be conditioned using language, points, or box prompts, which are first transformed into a shared object-centric representation of the selected objects. This representation captures target-focused visual features and metric object geometry, which is mixed with robot state history through a multi-stream transformer to predict action chunks. Although GAMs can be run autonomously, they can also serve as a low-level controller that a high-level planner controls using its various input modalities, allowing for long-horizon and memory-dependent manipulation. On RoboTwin 2.0, GAM achieves an average success rate of 55.3% across 50 tasks (vs. 52.0% for Spatial Forcing), including 47.6% under scene randomization (vs. 30.4% for Abot-M0), with its action policy trained only on clean-scene demonstrations. On LIBERO-PRO, it achieves a state-of-the-art average success rate of 61% (vs. 53% for π_{0.5}) across 16 perturbation settings, with the largest gains when targets are relocated or newly designated. On two real robots, GAM retains 17/20 successes under visual shift on a bimanual YAM versus 4/20 for π_{0.5}, while its composition with a Molmo2 planner on a Franka achieves 64.7% ID and 49.8% OOD step completion on long-horizon and memory-dependent tasks.

Audio-video generation has often relied on complex multi-stage architectures or sequential synthesis of sound and visuals. We introduce Ovi, a unified paradigm for audio-video generation that models the two modalities as a single generative process. By using blockwise cross-modal fusion of twin-DiT modules, Ovi achieves natural synchronization and removes the need for separate pipelines or post hoc alignment. To facilitate fine-grained multimodal fusion modeling, we initialize an audio tower with an architecture identical to that of a strong pretrained video model. Trained from scratch on hundreds of thousands of hours of raw audio, the audio tower learns to generate realistic sound effects, as well as speech that conveys rich speaker identity and emotion. Fusion is obtained by jointly training the identical video and audio towers via blockwise exchange of timing (via scaled-RoPE embeddings) and semantics (through bidirectional cross-attention) on a vast video corpus. Our model enables cinematic storytelling with natural speech and accurate, context-matched sound effects, producing movie-grade video clips. All the demos, code and model weights are published at https://aaxwaz.github.io/Ovi

Modern interactive video world models have achieved impressive visual fidelity, yet lack fine-grained multi-entity control and cross-entity, cross-world generalization. We trace this gap to the action interface: standard control protocols (e.g. animation IDs, device inputs, scene-level captions) bind action semantics to specific entities or engines at design time. We propose natural language as the interface to unlock expressiveness that no prior interface can achieve, and we present Incantation, the first interactive video world model with per-latent-frame (0.25 s) natural-language conditioning that supports simultaneous multi-entity control and concept-level cross-entity transfer beyond any fixed rendering pipeline. We pair a pretrained bidirectional video backbone with frame-local text cross-attention, and enable real-time long-horizon streaming through ODE-initialized Self-Forcing distillation with a RoPE-decoupled sliding KV-cache. We surpass the Action-Index baseline on cross-entity transfer (89% vs. 43%) and out-of-vocabulary prompts (90% vs. 0%), and our 2-step student sustains 19.7 FPS at 480p with stable FVD over 2-hour rollouts. We further apply the same architecture and training recipe to The King of Fighters, changing only the per-entity action vocabulary slots. We have released a preview subset of the Incantation dataset at https://huggingface.co/datasets/zhush/incantation-elden-ring-scenes, containing manually collected Elden Ring player-boss combat clips with structured action-oriented metadata. Larger-scale Elden Ring and KOF data will be released with the full project.

Robotic manipulation requires understanding both the 3D spatial structure of the environment and its temporal evolution, yet most existing policies overlook one or both. They typically rely on 2D visual observations and backbones pretrained on static image--text pairs, resulting in high data requirements and limited understanding of environment dynamics. To address this, we introduce MV-VDP, a multi-view video diffusion policy that jointly models the 3D spatio-temporal state of the environment. The core idea is to simultaneously predict multi-view heatmap videos and RGB videos, which 1) align the representation format of video pretraining with action finetuning, and 2) specify not only what actions the robot should take, but also how the environment is expected to evolve in response to those actions. Extensive experiments show that MV-VDP enables data-efficient, robust, generalizable, and interpretable manipulation. With only ten demonstration trajectories and without additional pretraining, MV-VDP successfully performs complex real-world tasks, demonstrates strong robustness across a range of model hyperparameters, generalizes to out-of-distribution settings, and predicts realistic future videos. Experiments on Meta-World and real-world robotic platforms demonstrate that MV-VDP consistently outperforms video-prediction--based, 3D-based, and vision--language--action models, establishing a new state of the art in data-efficient multi-task manipulation.

Prevailing Vision-Language-Action Models (VLAs) for robotic manipulation are built upon vision-language backbones pretrained on large-scale, but disconnected static web data. As a result, despite improved semantic generalization, the policy must implicitly infer complex physical dynamics and temporal dependencies solely from robot trajectories. This reliance creates an unsustainable data burden, necessitating continuous, large-scale expert data collection to compensate for the lack of innate physical understanding. We contend that while vision-language pretraining effectively captures semantic priors, it remains blind to physical causality. A more effective paradigm leverages video to jointly capture semantics and visual dynamics during pretraining, thereby isolating the remaining task of low-level control. To this end, we introduce mimic-video, a novel Video-Action Model (VAM) that pairs a pretrained Internet-scale video model with a flow matching-based action decoder conditioned on its latent representations. The decoder serves as an Inverse Dynamics Model (IDM), generating low-level robot actions from the latent representation of video-space action plans. Our extensive evaluation shows that our approach achieves state-of-the-art performance on simulated and real-world robotic manipulation tasks, improving sample efficiency by 10x and convergence speed by 2x compared to traditional VLA architectures.

The current popular methods for video object segmentation (VOS) implement feature matching through several hand-crafted modules that separately perform feature extraction and matching. However, the above hand-crafted designs empirically cause insufficient target interaction, thus limiting the dynamic target-aware feature learning in VOS. To tackle these limitations, this paper presents a scalable Simplified VOS (SimVOS) framework to perform joint feature extraction and matching by leveraging a single transformer backbone. Specifically, SimVOS employs a scalable ViT backbone for simultaneous feature extraction and matching between query and reference features. This design enables SimVOS to learn better target-ware features for accurate mask prediction. More importantly, SimVOS could directly apply well-pretrained ViT backbones (e.g., MAE) for VOS, which bridges the gap between VOS and large-scale self-supervised pre-training. To achieve a better performance-speed trade-off, we further explore within-frame attention and propose a new token refinement module to improve the running speed and save computational cost. Experimentally, our SimVOS achieves state-of-the-art results on popular video object segmentation benchmarks, i.e., DAVIS-2017 (88.0% J&F), DAVIS-2016 (92.9% J&F) and YouTube-VOS 2019 (84.2% J&F), without applying any synthetic video or BL30K pre-training used in previous VOS approaches.

Reconstructing dynamic 3D scenes (i.e., 4D geometry) from monocular video is an important yet challenging problem. Conventional multiview geometry-based approaches often struggle with dynamic motion, whereas recent learning-based methods either require specialized 4D representation or sophisticated optimization. In this paper, we present Sora3R, a novel framework that taps into the rich spatiotemporal priors of large-scale video diffusion models to directly infer 4D pointmaps from casual videos. Sora3R follows a two-stage pipeline: (1) we adapt a pointmap VAE from a pretrained video VAE, ensuring compatibility between the geometry and video latent spaces; (2) we finetune a diffusion backbone in combined video and pointmap latent space to generate coherent 4D pointmaps for every frame. Sora3R operates in a fully feedforward manner, requiring no external modules (e.g., depth, optical flow, or segmentation) or iterative global alignment. Extensive experiments demonstrate that Sora3R reliably recovers both camera poses and detailed scene geometry, achieving performance on par with state-of-the-art methods for dynamic 4D reconstruction across diverse scenarios.

Long-horizon manipulation is partially observable: the information needed to choose the next action may appear only in observations from minutes earlier. Existing memory mechanisms: retrieval banks, learned compressors, recurrent states must decide what to keep from the past before knowing what a future decision will require. This was motivated by the assumption that minute-scale history is too large to process directly, which modern VLM backbones no longer make true. In this work, we introduce SimpleMemVLA, a VLA without a dedicated memory module. It keeps the sampled history intact and passes it to the backbone in the timestamped video format the backbone was pretrained to process; the hidden states of a generated sub-task then form the only channel from history to a standard flow-matching action head. Since consecutive decisions share most of their history, prefilling the shared prefix during action execution keeps latency close to a single-frame VLA. SimpleMemVLA sets a new state of the art on four memory benchmarks without cost on general-purpose control. Holding the backbone and training setup fixed, it outperforms retrieval, compression and recurrent-state mechanisms by a wide margin, and causal interventions confirm that the policy genuinely reads its history. Code available at https://github.com/wadeKeith/SimpleMemVLA

Recent Diffusion Transformers (DiTs) have shown impressive capabilities in generating high-quality single-modality content, including images, videos, and audio. However, it is still under-explored whether the transformer-based diffuser can efficiently denoise the Gaussian noises towards superb multimodal content creation. To bridge this gap, we introduce AV-DiT, a novel and efficient audio-visual diffusion transformer designed to generate high-quality, realistic videos with both visual and audio tracks. To minimize model complexity and computational costs, AV-DiT utilizes a shared DiT backbone pre-trained on image-only data, with only lightweight, newly inserted adapters being trainable. This shared backbone facilitates both audio and video generation. Specifically, the video branch incorporates a trainable temporal attention layer into a frozen pre-trained DiT block for temporal consistency. Additionally, a small number of trainable parameters adapt the image-based DiT block for audio generation. An extra shared DiT block, equipped with lightweight parameters, facilitates feature interaction between audio and visual modalities, ensuring alignment. Extensive experiments on the AIST++ and Landscape datasets demonstrate that AV-DiT achieves state-of-the-art performance in joint audio-visual generation with significantly fewer tunable parameters. Furthermore, our results highlight that a single shared image generative backbone with modality-specific adaptations is sufficient for constructing a joint audio-video generator. Our source code and pre-trained models will be released.

Pretrained video generative models are promising backbones for visuomotor control, but their imagined futures often drift from task intent and are not reliably action-conditional. As a result, these models can be difficult to use for planning or policy extraction. To address these limitations, we propose RoboTALES, a single-stage framework that learns task-aligned simulated futures and uses them to train robot policies. Our approach introduces two key innovations: (1) a hierarchical LLM-based planner that breaks complex tasks into a sequence of subgoals to guide the model's imagination; and (2) a VLM-based critic that evaluates these ``imagined'' futures and uses reward-based feedback to keep the model's internal representations focused on the goal. By anchoring the video generator in abstract reasoning, we produce temporally consistent rollouts and more coherent actions. We evaluate RoboTALES on diverse manipulation tasks from RoboCasa and LIBERO10, and show that our method consistently outperforms existing methods, especially in long-horizon tasks. Our code and models are publicly available at https://github.com/hananshafi/RoboTALES.

Vision-language-action (VLA) models are a widely adopted paradigm for embodied policies. They excel at efficient closed-loop control but do not explicitly model how physical scenes evolve as a task unfolds. Recently emerging world-action models (WAMs) leverage pretrained video world models to capture spatiotemporal evolution, yet retaining future generation or a large video backbone in the control loop substantially increases inference cost. We introduce World Tokens, an embodied policy architecture built around a World Adapter that bridges visual-language understanding, world-dynamics modeling, and action generation. It uses world modeling during training to enhance the action policy while preserving efficient deployment. Specifically, the World Adapter transforms VLM features into a fixed set of world tokens, which condition a jointly fine-tuned future-video denoiser and simultaneously serve as the action expert's sole visual-language context. This shared conditioning allows gradients from future-video denoising to directly shape the representation used for action prediction, while exclusive routing prevents the policy from bypassing that representation. At deployment, the world-model branch is removed, leaving only the VLM, World Adapter, and action expert, with no online video-model inference. With a 2B backbone and no embodied action pretraining, World Tokens is highly competitive on LIBERO, attains the best reported averages on SIMPLER, substantially improves real-world R1 Pro success over a matched action-only baseline, and generates each action chunk at VLA-level latency.

World Action Models (WAMs) leverage the capabilities of large-scale pretrained video diffusion models to jointly predict future observations and actions, inheriting rich visual and physical priors from internet-scale video. This has made them a promising paradigm for robot policy learning, yet the prevailing models operate exclusively on RGB observations and do not leverage 3D information. To bridge this gap, we introduce a Spatially Aware World Action Model (SA-WAM), which repurposes a pretrained video model for joint action, RGB, and depth prediction, enabling 3D-aware world modeling and action prediction within a single diffusion backbone. We use a nonlinear encoding that maps the unbounded depth signal into the bounded input domain expected by the frozen VAE tokenizer. This allows us to reuse the tokenizer without 3D-specific fine-tuning, incorporating geometric information without sacrificing the pretrained priors. SA-WAM achieves state-of-the-art results on the RoboCasa and LIBERO-Plus benchmarks, while simultaneously improving future-state predictions. Furthermore, SA-WAM outperforms strong baselines in real-world evaluation using a UR5 robotic arm, with strong gains in randomized environments. We analyze the correlation between world model prediction quality and rollout success, providing insights into WAM performance and avenues for its improvement.

In this paper, we explore the visual representations produced from a pre-trained text-to-video (T2V) diffusion model for video understanding tasks. We hypothesize that the latent representation learned from a pretrained generative T2V model encapsulates rich semantics and coherent temporal correspondences, thereby naturally facilitating video understanding. Our hypothesis is validated through the classic referring video object segmentation (R-VOS) task. We introduce a novel framework, termed "VD-IT", tailored with dedicatedly designed components built upon a fixed pretrained T2V model. Specifically, VD-IT uses textual information as a conditional input, ensuring semantic consistency across time for precise temporal instance matching. It further incorporates image tokens as supplementary textual inputs, enriching the feature set to generate detailed and nuanced masks. Besides, instead of using the standard Gaussian noise, we propose to predict the video-specific noise with an extra noise prediction module, which can help preserve the feature fidelity and elevates segmentation quality. Through extensive experiments, we surprisingly observe that fixed generative T2V diffusion models, unlike commonly used video backbones (e.g., Video Swin Transformer) pretrained with discriminative image/video pre-tasks, exhibit better potential to maintain semantic alignment and temporal consistency. On existing standard benchmarks, our VD-IT achieves highly competitive results, surpassing many existing state-of-the-art methods. The code is available at https://github.com/buxiangzhiren/VD-IT.

Consistent 3D geometry estimation from streaming RGB input is crucial for real-world applications such as autonomous driving, embodied AI, and large-scale reconstruction. While modern monocular geometry foundation models achieve strong single-image accuracy, they exhibit severe temporal inconsistency on continuous input, notably dominated by scale--shift drifting. Through targeted empirical analysis, we trace this instability to its root cause: fluctuations in latent feature statistics, whose mean and variance directly determine the predicted depth's scale and shift. Building on this insight, we introduce Dynamic Feature Normalization (DyFN), a lightweight, causal recurrent module that dynamically and robustly modulates feature statistics to maintain stable geometry over time. We adapt powerful pretrained monocular geometry models for streaming by finetuning only DyFN, a mere 2\% additional parameters, while keeping the backbone frozen, thereby achieving temporal consistency without compromising single-image accuracy. Extensive experiments across four benchmarks show that DyFN effectively eliminates temporal artifacts such as disjointed layering and positional jitter, and achieves state-of-the-art temporal stability, improving over prior streaming methods by up to 14\% and even outperforming heavier non-causal video baselines. Project Page: https://shawlyu.github.io/DyFN

We present TokenDial, a framework for continuous, slider-style attribute control in pretrained text-to-video generation models. While modern generators produce strong holistic videos, they offer limited control over how much an attribute changes (e.g., effect intensity or motion magnitude) without drifting identity, background, or temporal coherence. TokenDial is built on the observation: additive offsets in the intermediate spatiotemporal visual patch-token space form a semantic control direction, where adjusting the offset magnitude yields coherent, predictable edits for both appearance and motion dynamics. We learn attribute-specific token offsets without retraining the backbone, using pretrained understanding signals: semantic direction matching for appearance and motion-magnitude scaling for motion. We demonstrate TokenDial's effectiveness on diverse attributes and prompts, achieving stronger controllability and higher-quality edits than state-of-the-art baselines, supported by extensive quantitative evaluation and human studies.

Recent approaches on 3D camera control in video diffusion models (VDMs) often create anchor videos to guide diffusion models as a structured prior by rendering from estimated point clouds following annotated camera trajectories. However, errors inherent in point cloud estimation often lead to inaccurate anchor videos. Moreover, the requirement for extensive camera trajectory annotations further increases resource demands. To address these limitations, we introduce EPiC, an efficient and precise camera control learning framework that automatically constructs high-quality anchor videos without expensive camera trajectory annotations. Concretely, we create highly precise anchor videos for training by masking source videos based on first-frame visibility. This approach ensures high alignment, eliminates the need for camera trajectory annotations, and thus can be readily applied to any in-the-wild video to generate image-to-video (I2V) training pairs. Furthermore, we introduce Anchor-ControlNet, a lightweight conditioning module that integrates anchor video guidance in visible regions to pretrained VDMs, with less than 1% of backbone model parameters. By combining the proposed anchor video data and ControlNet module, EPiC achieves efficient training with substantially fewer parameters, training steps, and less data, without requiring modifications to the diffusion model backbone typically needed to mitigate rendering misalignments. Although being trained on masking-based anchor videos, our method generalizes robustly to anchor videos made with point clouds during inference, enabling precise 3D-informed camera control. EPiC achieves SOTA performance on RealEstate10K and MiraData for I2V camera control task, demonstrating precise and robust camera control ability both quantitatively and qualitatively. Notably, EPiC also exhibits strong zero-shot generalization to video-to-video scenarios.

Video world models can generate realistic futures from a single instruction, but they often fail to track the same physical points consistently across time. As a result, the generated videos appear plausible, yet lack the physical grounding required for reliable action execution, such as robot manipulation. We present GEM-4D, a geometry-grounded video world model that resolves this limitation by injecting dense 4D correspondence supervision distilled from a pretrained geometry foundation model into the video generative backbone during training. This supervision enables the model to jointly capture appearance and geometric structure while retaining a single-stream architecture with no additional inference cost. We further introduce an inverse dynamics module that converts correspondence-consistent video rollouts into executable robot trajectories, enabling direct deployment in both real-world and simulated manipulation. GEM-4D achieves state-of-the-art performance on both video prediction and geometric consistency across both simulation and realistic scenarios and improves real-world manipulation success from 61% to 81%. Additional results are available at https://gem-4d.github.io/.

Text-to-video (T2V) diffusion models have recently achieved impressive visual quality, yet most systems still generate silent clips and treat audio as a secondary concern. Existing audio-video generation pipelines typically decompose the task into cascaded stages, which accumulate errors across modalities and are trained under separate objectives. Recent joint audio-video generators alleviate this issue but often rely on dual-tower architectures with ad-hoc cross-modal bridges and static, single-shot text conditioning, making it difficult to both reuse T2V backbones and to reason about how audio, video and language interact over time. To address these challenges, we propose 3MDiT, a unified tri-modal diffusion transformer for text-driven synchronized audio-video generation. Our framework models video, audio and text as jointly evolving streams: an isomorphic audio branch mirrors a T2V backbone, tri-modal omni-blocks perform feature-level fusion across the three modalities, and an optional dynamic text conditioning mechanism updates the text representation as audio and video evidence co-evolve. The design supports two regimes: training from scratch on audio-video data, and orthogonally adapting a pretrained T2V model without modifying its backbone. Experiments show that our approach generates high-quality videos and realistic audio while consistently improving audio-video synchronization and tri-modal alignment across a range of quantitative metrics.

Generating a coherent multi-shot video requires structured cross-shot memory. Subject appearance, scene context, and speaker identity must persist across cuts. Existing approaches either train end-to-end over fixed-length sequences and cannot scale, generate shot-by-shot with memory banks that grow linearly, or orchestrate pretrained generators under an LLM planner without a multi-shot-aware backbone. We present UnityShots, a memory-driven multi-shot audio-video generation system built on LTX-2.3, trained on annotated cinematic and music-video shots. The video stream maintains two fixed-size slots, a long-term memory (LTM) slot anchored to the opening shot and a short-term memory (STM) slot holding the immediately preceding tail, both updated at every cut by a boundary-conditioned gate that fuses visual cut probability and beat-tracker signals. The audio stream injects a reference speaker token at every shot to preserve vocal timbre without a sliding audio bank. A discrete cut-type prior, learned through AdaLN, becomes an inference-time control knob over transition strength. We release a benchmark of 200 multi-cultural multi-shot sequences spanning six ethnic regions and ten or more languages, with per-shot reference identities, reference audio, and per-boundary transition labels. Evaluated across I2V, T2V, and R2V conditioning modes, UnityShots leads open-source baselines on every cross-shot coherence metric and matches the strongest closed-source system on the multi-shot axes.

We present BetterScene, an approach to enhance novel view synthesis (NVS) quality for diverse real-world scenes using extremely sparse, unconstrained photos. BetterScene leverages the production-ready Stable Video Diffusion (SVD) model pretrained on billions of frames as a strong backbone, aiming to mitigate artifacts and recover view-consistent details at inference time. Conventional methods have developed similar diffusion-based solutions to address these challenges of novel view synthesis. Despite significant improvements, these methods typically rely on off-the-shelf pretrained diffusion priors and fine-tune only the UNet module while keeping other components frozen, which still leads to inconsistent details and artifacts even when incorporating geometry-aware regularizations like depth or semantic conditions. To address this, we investigate the latent space of the diffusion model and introduce two components: (1) temporal equivariance regularization and (2) vision foundation model-aligned representation, both applied to the variational autoencoder (VAE) module within the SVD pipeline. BetterScene integrates a feed-forward 3D Gaussian Splatting (3DGS) model to render features as inputs for the SVD enhancer and generate continuous, artifact-free, consistent novel views. We evaluate on the challenging DL3DV-10K dataset and demonstrate superior performance compared to state-of-the-art methods.

We present FlowAct-R2, a framework for interactive humanoid video generation that combines continuous multimodal control with proactive agent planning. Our method consists of two coupled components. First, a Streaming Multimodal Reference Diffusion Transformer adapts the pretrained Seedance 2.0 Mini reference-to-video backbone to accept rolling action prompts, streaming audio, and dynamically updated image, audio, and video references. Video-driven rotary positional embeddings align reference chunks with the generation timeline, while reference-plus-image conditioning and partially noised historical motion frames preserve appearance and avoid accumulated drift. Second, a Proactive Interaction Agent separates pre-online planning from online scheduling and response: it prepares a persona, a long-horizon agenda, and reusable multimodal skills in advance, then autonomously schedules behaviors, responds to audience input, and handles interruptions during a live session. FlowAct-R2 supports real-time 720p generation and hour-scale streaming across entertainment streaming, live shopping, video chatting, and live vlogging.

In this report, we introduce Ovis-Embedding, a state-of-the-art omni-modal embedding family built on native integration of text, image, video, and audio. Instead of assembling separate modality towers, Ovis-Embedding uses a shared multimodal backbone to encode different modalities in a common representation space. Specifically, we make three key advances: (1) native omni-modal initialization: we adopt a pretrained Qwen-omni model as the embedding backbone and adapt it through contrastive training with low-rank initialization; (2) data-centric omni-modal training: we construct a broad, high-quality corpus spanning text, images, video, audio, and interleaved multimodal data. To improve data efficiency, we introduce homogeneous-source sampling to form task-consistent batches with informative in-batch negatives; and (3) embedding-specific training and inference optimization: we use focal loss to emphasize hard examples and similarity-based Embedding Distillation to transfer fine-grained similarity structure from complementary experts. At inference time, low-rank feature decomposition enables compact embeddings with flexible dimensionality and minimal performance loss. Empirical evaluations show that the Ovis-Embedding family achieves state-of-the-art performance on MMEB-v3, MMEB-v2, MVEB, MAEB, and RTEB, demonstrating its effectiveness across text, image, video, and audio modalities. These results highlight the potential of unified omni-modal training to overcome modality fragmentation and advance universal embedding models for any-to-any retrieval.

We present Video Pre-trained Transformer. VPT uses four SOTA encoder models from prior work to convert a video into a sequence of compact embeddings. Our backbone, based on a reference Flan-T5-11B architecture, learns a universal representation of the video that is a non-linear sum of the encoder models. It learns using an autoregressive causal language modeling loss by predicting the words spoken in YouTube videos. Finally, we evaluate on standard downstream benchmarks by training fully connected prediction heads for each task. To the best of our knowledge, this is the first use of multiple frozen SOTA models as encoders in an "embedding -> backbone -> prediction head" design pattern - all others have trained their own joint encoder models. Additionally, we include more modalities than the current SOTA, Merlot Reserve, by adding explicit Scene Graph information. For these two reasons, we believe it could combine the world's best open-source models to achieve SOTA performance. Initial experiments demonstrate the model is learning appropriately, but more experimentation and compute is necessary, and already in progress, to realize our loftier goals. Alongside this work, we build on the YT-20M dataset, reproducing it and adding 25,000 personally selected YouTube videos to its corpus. All code and model checkpoints are open sourced under a standard MIT license.

We propose Spatio-temporal Crop Aggregation for video representation LEarning (SCALE), a novel method that enjoys high scalability at both training and inference time. Our model builds long-range video features by learning from sets of video clip-level features extracted with a pre-trained backbone. To train the model, we propose a self-supervised objective consisting of masked clip feature prediction. We apply sparsity to both the input, by extracting a random set of video clips, and to the loss function, by only reconstructing the sparse inputs. Moreover, we use dimensionality reduction by working in the latent space of a pre-trained backbone applied to single video clips. These techniques make our method not only extremely efficient to train but also highly effective in transfer learning. We demonstrate that our video representation yields state-of-the-art performance with linear, non-linear, and KNN probing on common action classification and video understanding datasets.

Generating talking face videos from various conditions has recently become a highly popular research area within generative tasks. However, building a high-quality face video generation model requires a well-performing pre-trained backbone, a key obstacle that universal models fail to adequately address. Most existing works rely on universal video or image generation models and optimize control mechanisms, but they neglect the evident upper bound in video quality due to the limited capabilities of the backbones, which is a result of the lack of high-quality human face video datasets. In this work, we investigate the unsatisfactory results from related studies, gather and trim existing public talking face video datasets, and additionally collect and annotate a large-scale dataset, resulting in a comprehensive, high-quality multiracial face collection named FaceVid-1K. Using this dataset, we craft several effective pre-trained backbone models for face video generation. Specifically, we conduct experiments with several well-established video generation models, including text-to-video, image-to-video, and unconditional video generation, under various settings. We obtain the corresponding performance benchmarks and compared them with those trained on public datasets to demonstrate the superiority of our dataset. These experiments also allow us to investigate empirical strategies for crafting domain-specific video generation tasks with cost-effective settings. We will make our curated dataset, along with the pre-trained talking face video generation models, publicly available as a resource contribution to hopefully advance the research field.

We present the Decoupled VIdeo Segmentation (DVIS) framework, a novel approach for the challenging task of universal video segmentation, including video instance segmentation (VIS), video semantic segmentation (VSS), and video panoptic segmentation (VPS). Unlike previous methods that model video segmentation in an end-to-end manner, our approach decouples video segmentation into three cascaded sub-tasks: segmentation, tracking, and refinement. This decoupling design allows for simpler and more effective modeling of the spatio-temporal representations of objects, especially in complex scenes and long videos. Accordingly, we introduce two novel components: the referring tracker and the temporal refiner. These components track objects frame by frame and model spatio-temporal representations based on pre-aligned features. To improve the tracking capability of DVIS, we propose a denoising training strategy and introduce contrastive learning, resulting in a more robust framework named DVIS++. Furthermore, we evaluate DVIS++ in various settings, including open vocabulary and using a frozen pre-trained backbone. By integrating CLIP with DVIS++, we present OV-DVIS++, the first open-vocabulary universal video segmentation framework. We conduct extensive experiments on six mainstream benchmarks, including the VIS, VSS, and VPS datasets. Using a unified architecture, DVIS++ significantly outperforms state-of-the-art specialized methods on these benchmarks in both close- and open-vocabulary settings. Code:~https://github.com/zhang-tao-whu/DVIS_Plus.

Neural network based computer vision systems are typically built on a backbone, a pretrained or randomly initialized feature extractor. Several years ago, the default option was an ImageNet-trained convolutional neural network. However, the recent past has seen the emergence of countless backbones pretrained using various algorithms and datasets. While this abundance of choice has led to performance increases for a range of systems, it is difficult for practitioners to make informed decisions about which backbone to choose. Battle of the Backbones (BoB) makes this choice easier by benchmarking a diverse suite of pretrained models, including vision-language models, those trained via self-supervised learning, and the Stable Diffusion backbone, across a diverse set of computer vision tasks ranging from classification to object detection to OOD generalization and more. Furthermore, BoB sheds light on promising directions for the research community to advance computer vision by illuminating strengths and weakness of existing approaches through a comprehensive analysis conducted on more than 1500 training runs. While vision transformers (ViTs) and self-supervised learning (SSL) are increasingly popular, we find that convolutional neural networks pretrained in a supervised fashion on large training sets still perform best on most tasks among the models we consider. Moreover, in apples-to-apples comparisons on the same architectures and similarly sized pretraining datasets, we find that SSL backbones are highly competitive, indicating that future works should perform SSL pretraining with advanced architectures and larger pretraining datasets. We release the raw results of our experiments along with code that allows researchers to put their own backbones through the gauntlet here: https://github.com/hsouri/Battle-of-the-Backbones

Text-to-video and image-to-video generation have made rapid progress in visual quality, but they remain limited in controlling the precise timing of motion. In contrast, audio provides temporal cues aligned with video motion, making it a promising condition for temporally controlled video generation. However, existing audio-to-video (A2V) models struggle with fine-grained synchronization due to indirect conditioning mechanisms or limited temporal modeling capacity. We present Syncphony, which generates 380x640 resolution, 24fps videos synchronized with diverse audio inputs. Our approach builds upon a pre-trained video backbone and incorporates two key components to improve synchronization: (1) Motion-aware Loss, which emphasizes learning at high-motion regions; (2) Audio Sync Guidance, which guides the full model using a visually aligned off-sync model without audio layers to better exploit audio cues at inference while maintaining visual quality. To evaluate synchronization, we propose CycleSync, a video-to-audio-based metric that measures the amount of motion cues in the generated video to reconstruct the original audio. Experiments on AVSync15 and The Greatest Hits datasets demonstrate that Syncphony outperforms existing methods in both synchronization accuracy and visual quality. Project page is available at: https://jibin86.github.io/syncphony_project_page

Recent feed-forward 3D gaussian splatting methods have made dramatic progress on individual aspects of 3D scene reconstruction, but no existing method jointly addresses dynamic content, multi-view input, and unknown camera poses in a single feed-forward pass. Methods that handle dynamics either require accurate camera poses or accept only monocular input; pose-free multi-view methods address only static scenes; and per-scene optimization methods bridge some of these gaps but at minutes-to-hours cost per scene. We introduce NoPo4D, the first feed-forward system that addresses this empty quadrant. Building on a pretrained geometry backbone and recent 4D Gaussian frameworks, NoPo4D introduces a velocity decomposition that splits Gaussian motion into per-pixel image-plane shifts and depth changes, allowing direct supervision from pseudo ground-truth optical flow on the 2D component. This sidesteps both the differentiable rendering that couples prior posed methods to pose accuracy and the 3D motion ground truth that prior pose-free methods require. The system is rounded out by a bidirectional motion encoder for cross-view and cross-frame feature aggregation, and view-dependent opacity that mitigates cross-view and cross-timestep Gaussian misalignments. On four multi-view dynamic benchmarks, NoPo4D consistently outperforms prior feed-forward baselines, and with an optional post-optimization stage surpasses per-scene optimization methods, while running orders of magnitude faster.

Egocentric Referring Video Object Segmentation (Ego-RVOS) aims to segment the specific object actively involved in a human action, as described by a language query, within first-person videos. This task is critical for understanding egocentric human behavior. However, achieving such segmentation robustly is challenging due to ambiguities inherent in egocentric videos and biases present in training data. Consequently, existing methods often struggle, learning spurious correlations from skewed object-action pairings in datasets and fundamental visual confounding factors of the egocentric perspective, such as rapid motion and frequent occlusions. To address these limitations, we introduce Causal Ego-REferring Segmentation (CERES), a plug-in causal framework that adapts strong, pre-trained RVOS backbones to the egocentric domain. CERES implements dual-modal causal intervention: applying backdoor adjustment principles to counteract language representation biases learned from dataset statistics, and leveraging front-door adjustment concepts to address visual confounding by intelligently integrating semantic visual features with geometric depth information guided by causal principles, creating representations more robust to egocentric distortions. Extensive experiments demonstrate that CERES achieves state-of-the-art performance on Ego-RVOS benchmarks, highlighting the potential of applying causal reasoning to build more reliable models for broader egocentric video understanding.

Nutrition estimation of meals from visual data is an important problem for dietary monitoring and computational health, but existing approaches largely rely on single images of the finally completed dish. This setting is fundamentally limited because many nutritionally relevant ingredients and transformations, such as oils, sauces, and mixed components, become visually ambiguous after cooking, making accurate calorie and macronutrient estimation difficult. In this paper, we investigate whether the cooking process information from egocentric cooking videos can contribute to dish-level nutrition estimation. First, we further manually annotated the HD-EPIC dataset and established the first benchmark for video-based nutrition estimation. Most importantly, we propose V-Nutri, a staged framework that combines Nutrition5K-pretrained visual backbones with a lightweight fusion module that aggregates features from the final dish frame and cooking process keyframes extracted from the egocentric videos. V-Nutri also includes a cooking keyframes selection module, a VideoMamba-based event-detection model that targets ingredient-addition moments. Experiments on the HD-EPIC dataset show that process cues can provide complementary nutritional evidence, improving nutrition estimation under controlled conditions. Our results further indicate that the benefit of process keyframes depends strongly on backbone representation capacity and event detection quality. Our code and annotated dataset is available at https://github.com/K624-YCK/V-Nutri.

Video-language pre-training (VLP) has become increasingly important due to its ability to generalize to various vision and language tasks. However, existing egocentric VLP frameworks utilize separate video and language encoders and learn task-specific cross-modal information only during fine-tuning, limiting the development of a unified system. In this work, we introduce the second generation of egocentric video-language pre-training (EgoVLPv2), a significant improvement from the previous generation, by incorporating cross-modal fusion directly into the video and language backbones. EgoVLPv2 learns strong video-text representation during pre-training and reuses the cross-modal attention modules to support different downstream tasks in a flexible and efficient manner, reducing fine-tuning costs. Moreover, our proposed fusion in the backbone strategy is more lightweight and compute-efficient than stacking additional fusion-specific layers. Extensive experiments on a wide range of VL tasks demonstrate the effectiveness of EgoVLPv2 by achieving consistent state-of-the-art performance over strong baselines across all downstream. Our project page can be found at https://shramanpramanick.github.io/EgoVLPv2/.

In contrast to numerous NLP and 2D vision foundational models, learning a 3D foundational model poses considerably greater challenges. This is primarily due to the inherent data variability and diversity of downstream tasks. In this paper, we introduce a novel universal 3D pre-training framework designed to facilitate the acquisition of efficient 3D representation, thereby establishing a pathway to 3D foundational models. Considering that informative 3D features should encode rich geometry and appearance cues that can be utilized to render realistic images, we propose to learn 3D representations by differentiable neural rendering. We train a 3D backbone with a devised volumetric neural renderer by comparing the rendered with the real images. Notably, our approach seamlessly integrates the learned 3D encoder into various downstream tasks. These tasks encompass not only high-level challenges such as 3D detection and segmentation but also low-level objectives like 3D reconstruction and image synthesis, spanning both indoor and outdoor scenarios. Besides, we also illustrate the capability of pre-training a 2D backbone using the proposed methodology, surpassing conventional pre-training methods by a large margin. For the first time, PonderV2 achieves state-of-the-art performance on 11 indoor and outdoor benchmarks, implying its effectiveness. Code and models are available at https://github.com/OpenGVLab/PonderV2.

This paper proposes a novel pretext task to address the self-supervised video representation learning problem. Specifically, given an unlabeled video clip, we compute a series of spatio-temporal statistical summaries, such as the spatial location and dominant direction of the largest motion, the spatial location and dominant color of the largest color diversity along the temporal axis, etc. Then a neural network is built and trained to yield the statistical summaries given the video frames as inputs. In order to alleviate the learning difficulty, we employ several spatial partitioning patterns to encode rough spatial locations instead of exact spatial Cartesian coordinates. Our approach is inspired by the observation that human visual system is sensitive to rapidly changing contents in the visual field, and only needs impressions about rough spatial locations to understand the visual contents. To validate the effectiveness of the proposed approach, we conduct extensive experiments with four 3D backbone networks, i.e., C3D, 3D-ResNet, R(2+1)D and S3D-G. The results show that our approach outperforms the existing approaches across these backbone networks on four downstream video analysis tasks including action recognition, video retrieval, dynamic scene recognition, and action similarity labeling. The source code is publicly available at: https://github.com/laura-wang/video_repres_sts.

Mainstream Video-Language Pre-training models actbert,clipbert,violet consist of three parts, a video encoder, a text encoder, and a video-text fusion Transformer. They pursue better performance via utilizing heavier unimodal encoders or multimodal fusion Transformers, resulting in increased parameters with lower efficiency in downstream tasks. In this work, we for the first time introduce an end-to-end video-language model, namely all-in-one Transformer, that embeds raw video and textual signals into joint representations using a unified backbone architecture. We argue that the unique temporal information of video data turns out to be a key barrier hindering the design of a modality-agnostic Transformer. To overcome the challenge, we introduce a novel and effective token rolling operation to encode temporal representations from video clips in a non-parametric manner. The careful design enables the representation learning of both video-text multimodal inputs and unimodal inputs using a unified backbone model. Our pre-trained all-in-one Transformer is transferred to various downstream video-text tasks after fine-tuning, including text-video retrieval, video-question answering, multiple choice and visual commonsense reasoning. State-of-the-art performances with the minimal model FLOPs on nine datasets demonstrate the superiority of our method compared to the competitive counterparts. The code and pretrained model have been released in https://github.com/showlab/all-in-one.

Estimating full-hand grasp pressure from egocentric video is critical for immersive VR and robotic manipulation, yet dense tactile sensing often relies on intrusive hardware. Existing vision-based methods predominantly rely on planar surfaces or fingertip contacts, failing to generalize to complex 3D object interactions. Therefore, we introduce EgoTactile, a benchmark pairing egocentric video with full-hand pressure supervision for diverse everyday objects, incorporating a bare-hand transfer subset to enable generalization to natural scenarios. Leveraging this benchmark, we first establish EgoPressureFormer as a discriminative baseline. Beyond this, to explicitly address the uncertainty in partial observations, we propose EgoPressureDiff, a conditional diffusion framework that adapts a large-scale pre-trained video diffusion backbone. By combining rich world knowledge priors with a Physically-Informed Feature Rectification layer to inject semantic constraints, our approach effectively infers plausible contact patterns and resolves visual-physical ambiguities. Extensive experiments demonstrate that our method achieves superior performance on the benchmark and robust transferability to in-the-wild scenarios. Our project page is available at https://egotactile.github.io/.

Pre-trained video models learn powerful priors for generating high-quality, temporally coherent content. While these models excel at temporal coherence, their dynamics are often constrained by the continuous nature of their training data. We hypothesize that by injecting the rich and unconstrained content diversity from image data into this coherent temporal framework, we can generate image sets that feature both natural transitions and a far more expansive dynamic range. To this end, we introduce iMontage, a unified framework designed to repurpose a powerful video model into an all-in-one image generator. The framework consumes and produces variable-length image sets, unifying a wide array of image generation and editing tasks. To achieve this, we propose an elegant and minimally invasive adaptation strategy, complemented by a tailored data curation process and training paradigm. This approach allows the model to acquire broad image manipulation capabilities without corrupting its invaluable original motion priors. iMontage excels across several mainstream many-in-many-out tasks, not only maintaining strong cross-image contextual consistency but also generating scenes with extraordinary dynamics that surpass conventional scopes. Find our homepage at: https://kr1sjfu.github.io/iMontage-web/.

World-Action Models (WAM) initialized from pre-trained video generation backbones have demonstrated remarkable potential for robot policy learning. However, existing approaches face two critical bottlenecks that hinder performance and deployment. First, jointly reasoning over future visual dynamics and corresponding actions incurs substantial inference overhead. Second, joint modeling often entangles visual and motion representations, making motion prediction accuracy heavily dependent on the quality of future video forecasts. To address these issues, we introduce GigaWorld-Policy, an action-centered WAM that learns 2D pixel-action dynamics while enabling efficient action decoding, with optional video generation. Specifically, we formulate policy training into two coupled components: the model predicts future action sequences conditioned on the current observation, and simultaneously generates future videos conditioned on the predicted actions and the same observation. The policy is supervised by both action prediction and video generation, providing richer learning signals and encouraging physically plausible actions through visual-dynamics constraints. With a causal design that prevents future-video tokens from influencing action tokens, explicit future-video generation is optional at inference time, allowing faster action prediction during deployment. To support this paradigm, we curate a diverse, large-scale robot dataset to pre-train an action-centered video generation model, which is then adapted as the backbone for robot policy learning. Experimental results on real-world robotic platforms show that GigaWorld-Policy runs 9x faster than the leading WAM baseline, Motus, while improving task success rates by 7%. Moreover, compared with pi-0.5, GigaWorld-Policy improves performance by 95% on RoboTwin 2.0.

Contrastive language-image pretraining has shown great success in learning visual-textual joint representation from web-scale data, demonstrating remarkable "zero-shot" generalization ability for various image tasks. However, how to effectively expand such new language-image pretraining methods to video domains is still an open problem. In this work, we present a simple yet effective approach that adapts the pretrained language-image models to video recognition directly, instead of pretraining a new model from scratch. More concretely, to capture the long-range dependencies of frames along the temporal dimension, we propose a cross-frame attention mechanism that explicitly exchanges information across frames. Such module is lightweight and can be plugged into pretrained language-image models seamlessly. Moreover, we propose a video-specific prompting scheme, which leverages video content information for generating discriminative textual prompts. Extensive experiments demonstrate that our approach is effective and can be generalized to different video recognition scenarios. In particular, under fully-supervised settings, our approach achieves a top-1 accuracy of 87.1% on Kinectics-400, while using 12 times fewer FLOPs compared with Swin-L and ViViT-H. In zero-shot experiments, our approach surpasses the current state-of-the-art methods by +7.6% and +14.9% in terms of top-1 accuracy under two popular protocols. In few-shot scenarios, our approach outperforms previous best methods by +32.1% and +23.1% when the labeled data is extremely limited. Code and models are available at https://aka.ms/X-CLIP

Existing methodologies for animating portrait images face significant challenges, particularly in handling non-frontal perspectives, rendering dynamic objects around the portrait, and generating immersive, realistic backgrounds. In this paper, we introduce the first application of a pretrained transformer-based video generative model that demonstrates strong generalization capabilities and generates highly dynamic, realistic videos for portrait animation, effectively addressing these challenges. The adoption of a new video backbone model makes previous U-Net-based methods for identity maintenance, audio conditioning, and video extrapolation inapplicable. To address this limitation, we design an identity reference network consisting of a causal 3D VAE combined with a stacked series of transformer layers, ensuring consistent facial identity across video sequences. Additionally, we investigate various speech audio conditioning and motion frame mechanisms to enable the generation of continuous video driven by speech audio. Our method is validated through experiments on benchmark and newly proposed wild datasets, demonstrating substantial improvements over prior methods in generating realistic portraits characterized by diverse orientations within dynamic and immersive scenes. Further visualizations and the source code are available at: https://fudan-generative-vision.github.io/hallo3/.

Video Diffusion Models (VDMs) have emerged as powerful generative tools, capable of synthesizing high-quality spatiotemporal content. Yet, their potential goes far beyond mere video generation. We argue that the training dynamics of VDMs, driven by the need to model coherent sequences, naturally pushes them to internalize structured representations and an implicit understanding of the visual world. To probe the extent of this internal knowledge, we introduce a few-shot fine-tuning framework that repurposes VDMs for new tasks using only a handful of examples. Our method transforms each task into a visual transition, enabling the training of LoRA weights on short input-output sequences without altering the generative interface of a frozen VDM. Despite minimal supervision, the model exhibits strong generalization across diverse tasks, from low-level vision (for example, segmentation and pose estimation) to high-level reasoning (for example, on ARC-AGI). These results reframe VDMs as more than generative engines. They are adaptable visual learners with the potential to serve as the backbone for future foundation models in vision.

Current popular backbones in computer vision, such as Vision Transformers (ViT) and ResNets are trained to perceive the world from 2D images. However, to more effectively understand 3D structural priors in 2D backbones, we propose Mask3D to leverage existing large-scale RGB-D data in a self-supervised pre-training to embed these 3D priors into 2D learned feature representations. In contrast to traditional 3D contrastive learning paradigms requiring 3D reconstructions or multi-view correspondences, our approach is simple: we formulate a pre-text reconstruction task by masking RGB and depth patches in individual RGB-D frames. We demonstrate the Mask3D is particularly effective in embedding 3D priors into the powerful 2D ViT backbone, enabling improved representation learning for various scene understanding tasks, such as semantic segmentation, instance segmentation and object detection. Experiments show that Mask3D notably outperforms existing self-supervised 3D pre-training approaches on ScanNet, NYUv2, and Cityscapes image understanding tasks, with an improvement of +6.5% mIoU against the state-of-the-art Pri3D on ScanNet image semantic segmentation.

Building video-language foundation models is costly and difficult due to the redundant nature of video data and the lack of high-quality video-language datasets. In this paper, we propose an efficient framework to harvest video foundation models from image ones. Our method is intuitively simple by randomly dropping input video patches and masking out input text during the post-pretraining procedure. The patch dropping boosts the training efficiency significantly and text masking enforces the learning of cross-modal fusion. We conduct extensive experiments to validate the effectiveness of our method on a wide range of video-language downstream tasks including various zero-shot tasks, video question answering, and video-text retrieval. Despite its simplicity, our method achieves state-of-the-art performances, which are comparable to some heavily pretrained video foundation models. Our method is extremely efficient and can be trained in less than one day on 8 GPUs, requiring only WebVid-10M as pretraining data. We hope our method can serve as a simple yet strong counterpart for prevalent video foundation models, provide useful insights when building them, and make large pretrained models more accessible and sustainable. This is part of the InternVideo project https://github.com/OpenGVLab/InternVideo.

Most successful self-supervised learning methods are trained to align the representations of two independent views from the data. State-of-the-art methods in video are inspired by image techniques, where these two views are similarly extracted by cropping and augmenting the resulting crop. However, these methods miss a crucial element in the video domain: time. We introduce BraVe, a self-supervised learning framework for video. In BraVe, one of the views has access to a narrow temporal window of the video while the other view has a broad access to the video content. Our models learn to generalise from the narrow view to the general content of the video. Furthermore, BraVe processes the views with different backbones, enabling the use of alternative augmentations or modalities into the broad view such as optical flow, randomly convolved RGB frames, audio or their combinations. We demonstrate that BraVe achieves state-of-the-art results in self-supervised representation learning on standard video and audio classification benchmarks including UCF101, HMDB51, Kinetics, ESC-50 and AudioSet.

Training strong video generation models usually requires massive datasets, large parameter counts, and substantial compute. In this work, we ask whether strong text-to-video quality is possible at a much smaller budget: fewer than 10M clips and less than 100,000 H200 GPU hours. Our core claim is that part of the answer lies in how model capacity is organized, not only in how much of it is used. In video generation, prompt alignment, temporal consistency, and fine-detail recovery can interfere with one another when they are handled through the same pathway. Motif-Video 2B addresses this by separating these roles architecturally, rather than relying on scale alone. The model combines two key ideas. First, Shared Cross-Attention strengthens text control when video token sequences become long. Second, a three-part backbone separates early fusion, joint representation learning, and detail refinement. To make this design effective under a limited compute budget, we pair it with an efficient training recipe based on dynamic token routing and early-phase feature alignment to a frozen pretrained video encoder. Our analysis shows that later blocks develop clearer cross-frame attention structure than standard single-stream baselines. On VBench, Motif-Video~2B reaches 83.76\%, surpassing Wan2.1 14B while using 7times fewer parameters and substantially less training data. These results suggest that careful architectural specialization, combined with an efficiency-oriented training recipe, can narrow or exceed the quality gap typically associated with much larger video models.

Vision-language pre-training has significantly elevated performance across a wide range of image-language applications. Yet, the pre-training process for video-related tasks demands exceptionally large computational and data resources, which hinders the progress of video-language models. This paper investigates a straightforward, highly efficient, and resource-light approach to adapting an existing image-language pre-trained model for dense video understanding. Our preliminary experiments reveal that directly fine-tuning pre-trained image-language models with multiple frames as inputs on video datasets leads to performance saturation or even a drop. Our further investigation reveals that it is largely attributed to the bias of learned high-norm visual features. Motivated by this finding, we propose a simple but effective pooling strategy to smooth the feature distribution along the temporal dimension and thus reduce the dominant impacts from the extreme features. The new model is termed Pooling LLaVA, or in short. achieves new state-of-the-art performance on modern benchmark datasets for both video question-answer and captioning tasks. Notably, on the recent popular Video ChatGPT benchmark, PLLaVA achieves a score of 3.48 out of 5 on average of five evaluated dimensions, exceeding the previous SOTA results from GPT4V (IG-VLM) by 9\%. On the latest multi-choice benchmark MVBench, PLLaVA achieves 58.1\% accuracy on average across 20 sub-tasks, 14.5\% higher than GPT4V (IG-VLM). Code is available at https://github.com/magic-research/PLLaVA.

Learning discriminative spatiotemporal representation is the key problem of video understanding. Recently, Vision Transformers (ViTs) have shown their power in learning long-term video dependency with self-attention. Unfortunately, they exhibit limitations in tackling local video redundancy, due to the blind global comparison among tokens. UniFormer has successfully alleviated this issue, by unifying convolution and self-attention as a relation aggregator in the transformer format. However, this model has to require a tiresome and complicated image-pretraining phrase, before being finetuned on videos. This blocks its wide usage in practice. On the contrary, open-sourced ViTs are readily available and well-pretrained with rich image supervision. Based on these observations, we propose a generic paradigm to build a powerful family of video networks, by arming the pretrained ViTs with efficient UniFormer designs. We call this family UniFormerV2, since it inherits the concise style of the UniFormer block. But it contains brand-new local and global relation aggregators, which allow for preferable accuracy-computation balance by seamlessly integrating advantages from both ViTs and UniFormer. Without any bells and whistles, our UniFormerV2 gets the state-of-the-art recognition performance on 8 popular video benchmarks, including scene-related Kinetics-400/600/700 and Moments in Time, temporal-related Something-Something V1/V2, untrimmed ActivityNet and HACS. In particular, it is the first model to achieve 90% top-1 accuracy on Kinetics-400, to our best knowledge. Code will be available at https://github.com/OpenGVLab/UniFormerV2.

Modern top-performing object detectors depend heavily on backbone networks, whose advances bring consistent performance gains through exploring more effective network structures. In this paper, we propose a novel and flexible backbone framework, namely CBNetV2, to construct high-performance detectors using existing open-sourced pre-trained backbones under the pre-training fine-tuning paradigm. In particular, CBNetV2 architecture groups multiple identical backbones, which are connected through composite connections. Specifically, it integrates the high- and low-level features of multiple backbone networks and gradually expands the receptive field to more efficiently perform object detection. We also propose a better training strategy with assistant supervision for CBNet-based detectors. Without additional pre-training of the composite backbone, CBNetV2 can be adapted to various backbones (CNN-based vs. Transformer-based) and head designs of most mainstream detectors (one-stage vs. two-stage, anchor-based vs. anchor-free-based). Experiments provide strong evidence that, compared with simply increasing the depth and width of the network, CBNetV2 introduces a more efficient, effective, and resource-friendly way to build high-performance backbone networks. Particularly, our Dual-Swin-L achieves 59.4% box AP and 51.6% mask AP on COCO test-dev under the single-model and single-scale testing protocol, which is significantly better than the state-of-the-art result (57.7% box AP and 50.2% mask AP) achieved by Swin-L, while the training schedule is reduced by 6times. With multi-scale testing, we push the current best single model result to a new record of 60.1% box AP and 52.3% mask AP without using extra training data. Code is available at https://github.com/VDIGPKU/CBNetV2.

Recently, approaches that leverage human video datasets for robot policy training have become increasingly prevalent. However, most existing hand trackers regress pose from cropped frames with limited priors on hand motion and object interaction, resulting in inaccurate and physically inconsistent estimates. Moreover, the lack of physical cues, e.g., contact and force, limits the use of human videos for robot policy training. To this end, we propose RLHND, a video foundation model-based hand tracking model that jointly estimates hand pose and realistic tactile information from monocular egocentric videos. RLHND turns the pre-trained Cosmos 3 video diffusion backbone into a deterministic clip-level feature extractor via clean-latent conditioning, carrying its learned priors on hand motion and hand-object interaction into tracking. For pose estimation, RLHND (i) predicts hand poses with anatomically plausible joint angles and (ii) enables optional conditioning on the shape parameter to maintain consistent hand shape within the same video and even across videos recorded by the same actor. For tactile estimation, a separate tactile expert stream, trained with the pose stream frozen, predicts dense contact and force over the hand surface. We further adopt LBS-based feature spreading to enable vertex-wise feature extraction without costly per-vertex attention. RLHND achieves state-of-the-art performance across various benchmark datasets for pose estimation, while also achieving state-of-the-art performance in contact and force estimation. Moreover, we demonstrate the utility of RLHND for robot learning through retargeting results and real-world robot experiments. The code will be publicly available at https://seungjun-moon.github.io/rlhnd/.

With the overwhelming trend of mask image modeling led by MAE, generative pre-training has shown a remarkable potential to boost the performance of fundamental models in 2D vision. However, in 3D vision, the over-reliance on Transformer-based backbones and the unordered nature of point clouds have restricted the further development of generative pre-training. In this paper, we propose a novel 3D-to-2D generative pre-training method that is adaptable to any point cloud model. We propose to generate view images from different instructed poses via the cross-attention mechanism as the pre-training scheme. Generating view images has more precise supervision than its point cloud counterpart, thus assisting 3D backbones to have a finer comprehension of the geometrical structure and stereoscopic relations of the point cloud. Experimental results have proved the superiority of our proposed 3D-to-2D generative pre-training over previous pre-training methods. Our method is also effective in boosting the performance of architecture-oriented approaches, achieving state-of-the-art performance when fine-tuning on ScanObjectNN classification and ShapeNetPart segmentation tasks. Code is available at https://github.com/wangzy22/TAP.

Many recent studies leverage the pre-trained CLIP for text-video cross-modal retrieval by tuning the backbone with additional heavy modules, which not only brings huge computational burdens with much more parameters, but also leads to the knowledge forgetting from upstream models. In this work, we propose the VoP: Text-Video Co-operative Prompt Tuning for efficient tuning on the text-video retrieval task. The proposed VoP is an end-to-end framework with both video & text prompts introducing, which can be regarded as a powerful baseline with only 0.1% trainable parameters. Further, based on the spatio-temporal characteristics of videos, we develop three novel video prompt mechanisms to improve the performance with different scales of trainable parameters. The basic idea of the VoP enhancement is to model the frame position, frame context, and layer function with specific trainable prompts, respectively. Extensive experiments show that compared to full fine-tuning, the enhanced VoP achieves a 1.4% average R@1 gain across five text-video retrieval benchmarks with 6x less parameter overhead. The code will be available at https://github.com/bighuang624/VoP.

Early detection, accurate segmentation, classification and tracking of polyps during colonoscopy are critical for preventing colorectal cancer. Many existing deep-learning-based methods for analyzing colonoscopic videos either require task-specific fine-tuning, lack tracking capabilities, or rely on domain-specific pre-training. In this paper, we introduce PolypSegTrack, a novel foundation model that jointly addresses polyp detection, segmentation, classification and unsupervised tracking in colonoscopic videos. Our approach leverages a novel conditional mask loss, enabling flexible training across datasets with either pixel-level segmentation masks or bounding box annotations, allowing us to bypass task-specific fine-tuning. Our unsupervised tracking module reliably associates polyp instances across frames using object queries, without relying on any heuristics. We leverage a robust vision foundation model backbone that is pre-trained unsupervisedly on natural images, thereby removing the need for domain-specific pre-training. Extensive experiments on multiple polyp benchmarks demonstrate that our method significantly outperforms existing state-of-the-art approaches in detection, segmentation, classification, and tracking.

Most text-video retrieval methods utilize the text-image pre-trained models like CLIP as a backbone. These methods process each sampled frame independently by the image encoder, resulting in high computational overhead and limiting practical deployment. Addressing this, we focus on efficient text-video retrieval by tackling two key challenges: 1. From the perspective of trainable parameters, current parameter-efficient fine-tuning methods incur high inference costs; 2. From the perspective of model complexity, current token compression methods are mainly designed for images to reduce spatial redundancy but overlook temporal redundancy in consecutive frames of a video. To tackle these challenges, we propose Temporal Token Merging (TempMe), a parameter-efficient and training-inference efficient text-video retrieval architecture that minimizes trainable parameters and model complexity. Specifically, we introduce a progressive multi-granularity framework. By gradually combining neighboring clips, we reduce spatio-temporal redundancy and enhance temporal modeling across different frames, leading to improved efficiency and performance. Extensive experiments validate the superiority of our TempMe. Compared to previous parameter-efficient text-video retrieval methods, TempMe achieves superior performance with just 0.50M trainable parameters. It significantly reduces output tokens by 95% and GFLOPs by 51%, while achieving a 1.8X speedup and a 4.4% R-Sum improvement. With full fine-tuning, TempMe achieves a significant 7.9% R-Sum improvement, trains 1.57X faster, and utilizes 75.2% GPU memory usage. The code is available at https://github.com/LunarShen/TempMe.

Understanding continuous video streams plays a fundamental role in real-time applications including embodied AI and autonomous driving. Unlike offline video understanding, streaming video understanding requires the ability to process video streams frame by frame, preserve historical information, and make low-latency decisions. To address these challenges, our main contributions are three-fold. (i) We develop a novel streaming video backbone, termed as StreamFormer, by incorporating causal temporal attention into a pre-trained vision transformer. This enables efficient streaming video processing while maintaining image representation capability. (ii) To train StreamFormer, we propose to unify diverse spatial-temporal video understanding tasks within a multitask visual-language alignment framework. Hence, StreamFormer learns global semantics, temporal dynamics, and fine-grained spatial relationships simultaneously. (iii) We conduct extensive experiments on online action detection, online video instance segmentation, and video question answering. StreamFormer achieves competitive results while maintaining efficiency, demonstrating its potential for real-time applications.

With the rapid advancement in video generation, people can conveniently utilize video generation models to create videos tailored to their specific desires. Nevertheless, there are also growing concerns about their potential misuse in creating and disseminating false information. In this work, we introduce VGMShield: a set of three straightforward but pioneering mitigations through the lifecycle of fake video generation. We start from fake video detection trying to understand whether there is uniqueness in generated videos and whether we can differentiate them from real videos; then, we investigate the tracing problem, which maps a fake video back to a model that generates it. Towards these, we propose to leverage pre-trained models that focus on {\it spatial-temporal dynamics} as the backbone to identify inconsistencies in videos. Through experiments on seven state-of-the-art open-source models, we demonstrate that current models still cannot perfectly handle spatial-temporal relationships, and thus, we can accomplish detection and tracing with nearly perfect accuracy. Furthermore, anticipating future generative model improvements, we propose a {\it prevention} method that adds invisible perturbations to images to make the generated videos look unreal. Together with fake video detection and tracing, our multi-faceted set of solutions can effectively mitigate misuse of video generative models.

Speech understanding as an element of the more generic video understanding using audio-visual large language models (av-LLMs) is a crucial yet understudied aspect. This paper proposes video-SALMONN, a single end-to-end av-LLM for video processing, which can understand not only visual frame sequences, audio events and music, but speech as well. To obtain fine-grained temporal information required by speech understanding, while keeping efficient for other video elements, this paper proposes a novel multi-resolution causal Q-Former (MRC Q-Former) structure to connect pre-trained audio-visual encoders and the backbone large language model. Moreover, dedicated training approaches including the diversity loss and the unpaired audio-visual mixed training scheme are proposed to avoid frames or modality dominance. On the introduced speech-audio-visual evaluation benchmark, video-SALMONN achieves more than 25\% absolute accuracy improvements on the video-QA task and over 30\% absolute accuracy improvements on audio-visual QA tasks with human speech. In addition, video-SALMONN demonstrates remarkable video comprehension and reasoning abilities on tasks that are unprecedented by other av-LLMs. Our training code and model checkpoints are available at \url{https://github.com/bytedance/SALMONN/}.

Generating high fidelity human video with specified identities has attracted significant attention in the content generation community. However, existing techniques struggle to strike a balance between training efficiency and identity preservation, either requiring tedious case-by-case finetuning or usually missing the identity details in video generation process. In this study, we present ID-Animator, a zero-shot human-video generation approach that can perform personalized video generation given single reference facial image without further training. ID-Animator inherits existing diffusion-based video generation backbones with a face adapter to encode the ID-relevant embeddings from learnable facial latent queries. To facilitate the extraction of identity information in video generation, we introduce an ID-oriented dataset construction pipeline, which incorporates decoupled human attribute and action captioning technique from a constructed facial image pool. Based on this pipeline, a random face reference training method is further devised to precisely capture the ID-relevant embeddings from reference images, thus improving the fidelity and generalization capacity of our model for ID-specific video generation. Extensive experiments demonstrate the superiority of ID-Animator to generate personalized human videos over previous models. Moreover, our method is highly compatible with popular pre-trained T2V models like animatediff and various community backbone models, showing high extendability in real-world applications for video generation where identity preservation is highly desired. Our codes and checkpoints will be released at https://github.com/ID-Animator/ID-Animator.

Video inpainting, which aims to restore corrupted video content, has experienced substantial progress. Despite these advances, existing methods, whether propagating unmasked region pixels through optical flow and receptive field priors, or extending image-inpainting models temporally, face challenges in generating fully masked objects or balancing the competing objectives of background context preservation and foreground generation in one model, respectively. To address these limitations, we propose a novel dual-stream paradigm VideoPainter that incorporates an efficient context encoder (comprising only 6% of the backbone parameters) to process masked videos and inject backbone-aware background contextual cues to any pre-trained video DiT, producing semantically consistent content in a plug-and-play manner. This architectural separation significantly reduces the model's learning complexity while enabling nuanced integration of crucial background context. We also introduce a novel target region ID resampling technique that enables any-length video inpainting, greatly enhancing our practical applicability. Additionally, we establish a scalable dataset pipeline leveraging current vision understanding models, contributing VPData and VPBench to facilitate segmentation-based inpainting training and assessment, the largest video inpainting dataset and benchmark to date with over 390K diverse clips. Using inpainting as a pipeline basis, we also explore downstream applications including video editing and video editing pair data generation, demonstrating competitive performance and significant practical potential. Extensive experiments demonstrate VideoPainter's superior performance in both any-length video inpainting and editing, across eight key metrics, including video quality, mask region preservation, and textual coherence.

Controllable video generation aims to synthesize video content that aligns precisely with user-provided conditions, such as text descriptions and initial images. However, a significant challenge persists in this domain: existing models often struggle to maintain strong semantic consistency, frequently generating videos that deviate from the nuanced details specified in the prompts. To address this issue, we propose SSG-DiT (Spatial Signal Guided Diffusion Transformer), a novel and efficient framework for high-fidelity controllable video generation. Our approach introduces a decoupled two-stage process. The first stage, Spatial Signal Prompting, generates a spatially aware visual prompt by leveraging the rich internal representations of a pre-trained multi-modal model. This prompt, combined with the original text, forms a joint condition that is then injected into a frozen video DiT backbone via our lightweight and parameter-efficient SSG-Adapter. This unique design, featuring a dual-branch attention mechanism, allows the model to simultaneously harness its powerful generative priors while being precisely steered by external spatial signals. Extensive experiments demonstrate that SSG-DiT achieves state-of-the-art performance, outperforming existing models on multiple key metrics in the VBench benchmark, particularly in spatial relationship control and overall consistency.

The paradigm of large-scale pre-training followed by downstream fine-tuning has been widely employed in various object detection algorithms. In this paper, we reveal discrepancies in data, model, and task between the pre-training and fine-tuning procedure in existing practices, which implicitly limit the detector's performance, generalization ability, and convergence speed. To this end, we propose AlignDet, a unified pre-training framework that can be adapted to various existing detectors to alleviate the discrepancies. AlignDet decouples the pre-training process into two stages, i.e., image-domain and box-domain pre-training. The image-domain pre-training optimizes the detection backbone to capture holistic visual abstraction, and box-domain pre-training learns instance-level semantics and task-aware concepts to initialize the parts out of the backbone. By incorporating the self-supervised pre-trained backbones, we can pre-train all modules for various detectors in an unsupervised paradigm. As depicted in Figure 1, extensive experiments demonstrate that AlignDet can achieve significant improvements across diverse protocols, such as detection algorithm, model backbone, data setting, and training schedule. For example, AlignDet improves FCOS by 5.3 mAP, RetinaNet by 2.1 mAP, Faster R-CNN by 3.3 mAP, and DETR by 2.3 mAP under fewer epochs.

With the recent success of the pre-training technique for NLP and image-linguistic tasks, some video-linguistic pre-training works are gradually developed to improve video-text related downstream tasks. However, most of the existing multimodal models are pre-trained for understanding tasks, leading to a pretrain-finetune discrepancy for generation tasks. This paper proposes UniVL: a Unified Video and Language pre-training model for both multimodal understanding and generation. It comprises four components, including two single-modal encoders, a cross encoder, and a decoder with the Transformer backbone. Five objectives, including video-text joint, conditioned masked language model (CMLM), conditioned masked frame model (CMFM), video-text alignment, and language reconstruction, are designed to train each of the components. We further develop two pre-training strategies, stage by stage pre-training (StagedP) and enhanced video representation (EnhancedV), to make the training process of the UniVL more effective. The pre-train is carried out on a sizeable instructional video dataset HowTo100M. Experimental results demonstrate that the UniVL can learn strong video-text representation and achieves state-of-the-art results on five downstream tasks.

We present a self-supervised Contrastive Video Representation Learning (CVRL) method to learn spatiotemporal visual representations from unlabeled videos. Our representations are learned using a contrastive loss, where two augmented clips from the same short video are pulled together in the embedding space, while clips from different videos are pushed away. We study what makes for good data augmentations for video self-supervised learning and find that both spatial and temporal information are crucial. We carefully design data augmentations involving spatial and temporal cues. Concretely, we propose a temporally consistent spatial augmentation method to impose strong spatial augmentations on each frame of the video while maintaining the temporal consistency across frames. We also propose a sampling-based temporal augmentation method to avoid overly enforcing invariance on clips that are distant in time. On Kinetics-600, a linear classifier trained on the representations learned by CVRL achieves 70.4% top-1 accuracy with a 3D-ResNet-50 (R3D-50) backbone, outperforming ImageNet supervised pre-training by 15.7% and SimCLR unsupervised pre-training by 18.8% using the same inflated R3D-50. The performance of CVRL can be further improved to 72.9% with a larger R3D-152 (2x filters) backbone, significantly closing the gap between unsupervised and supervised video representation learning. Our code and models will be available at https://github.com/tensorflow/models/tree/master/official/.

In light of recent advances in multimodal Large Language Models (LLMs), there is increasing attention to scaling them from image-text data to more informative real-world videos. Compared to static images, video poses unique challenges for effective large-scale pre-training due to the modeling of its spatiotemporal dynamics. In this paper, we address such limitations in video-language pre-training with an efficient video decomposition that represents each video as keyframes and temporal motions. These are then adapted to an LLM using well-designed tokenizers that discretize visual and temporal information as a few tokens, thus enabling unified generative pre-training of videos, images, and text. At inference, the generated tokens from the LLM are carefully recovered to the original continuous pixel space to create various video content. Our proposed framework is both capable of comprehending and generating image and video content, as demonstrated by its competitive performance across 13 multimodal benchmarks in image and video understanding and generation. Our code and models will be available at https://video-lavit.github.io.

Fine-tuning a visual pre-trained model can leverage the semantic information from large-scale pre-training data and mitigate the over-fitting problem on downstream vision tasks with limited training examples. While the problem of catastrophic forgetting in pre-trained backbone has been extensively studied for fine-tuning, its potential bias from the corresponding pre-training task and data, attracts less attention. In this work, we investigate this problem by demonstrating that the obtained classifier after fine-tuning will be close to that induced by the pre-trained model. To reduce the bias in the classifier effectively, we introduce a reference distribution obtained from a fixed text classifier, which can help regularize the learned vision classifier. The proposed method, Text Supervised fine-tuning (TeS), is evaluated with diverse pre-trained vision models including ResNet and ViT, and text encoders including BERT and CLIP, on 11 downstream tasks. The consistent improvement with a clear margin over distinct scenarios confirms the effectiveness of our proposal. Code is available at https://github.com/idstcv/TeS.

Self-supervised learning has unlocked the potential of scaling up pretraining to billions of images, since annotation is unnecessary. But are we making the best use of data? How more economical can we be? In this work, we attempt to answer this question by making two contributions. First, we investigate first-person videos and introduce a "Walking Tours" dataset. These videos are high-resolution, hours-long, captured in a single uninterrupted take, depicting a large number of objects and actions with natural scene transitions. They are unlabeled and uncurated, thus realistic for self-supervision and comparable with human learning. Second, we introduce a novel self-supervised image pretraining method tailored for learning from continuous videos. Existing methods typically adapt image-based pretraining approaches to incorporate more frames. Instead, we advocate a "tracking to learn to recognize" approach. Our method called DoRA, leads to attention maps that Discover and tRAck objects over time in an end-to-end manner, using transformer cross-attention. We derive multiple views from the tracks and use them in a classical self-supervised distillation loss. Using our novel approach, a single Walking Tours video remarkably becomes a strong competitor to ImageNet for several image and video downstream tasks.

This paper presents a novel framework for converting 2D videos to immersive stereoscopic 3D, addressing the growing demand for 3D content in immersive experience. Leveraging foundation models as priors, our approach overcomes the limitations of traditional methods and boosts the performance to ensure the high-fidelity generation required by the display devices. The proposed system consists of two main steps: depth-based video splatting for warping and extracting occlusion mask, and stereo video inpainting. We utilize pre-trained stable video diffusion as the backbone and introduce a fine-tuning protocol for the stereo video inpainting task. To handle input video with varying lengths and resolutions, we explore auto-regressive strategies and tiled processing. Finally, a sophisticated data processing pipeline has been developed to reconstruct a large-scale and high-quality dataset to support our training. Our framework demonstrates significant improvements in 2D-to-3D video conversion, offering a practical solution for creating immersive content for 3D devices like Apple Vision Pro and 3D displays. In summary, this work contributes to the field by presenting an effective method for generating high-quality stereoscopic videos from monocular input, potentially transforming how we experience digital media.

This report presents TAP-JEPA, our runner-up submission to the EPIC-KITCHENS-100 (EK-100) Action Anticipation Challenge at EgoVis 2026. The task is to anticipate the next verb, noun, and verb-noun action from an egocentric clip that ends before the target action begins. Instead of fine-tuning a large video backbone, TAP-JEPA builds a compact anticipation model on frozen V-JEPA 2.1 features: a ViT-G/384 encoder extracts visible pre-action tokens, the pre-trained latent predictor estimates near-future tokens from the observed context, and both token groups are fused by attentive probes with task-specific queries for verbs, nouns, and action pairs. For the final submission, we expand supervised training with the official training split and most of the validation split, reserving a small subset for sanity checks and qualitative inspection, and adopt a two-stage score fusion that first averages eight independently initialized probe replicas within each epoch and then merges candidates from epochs 12-20 with field-dependent weights. On the official open-testing leaderboard, our sunshinesky entry achieves 27.91 percent overall action Mean Top-5 Recall (MT5R), ranking second and only 0.04 percentage points behind the top score.

Latent Action Models (LAMs) have emerged as an effective paradigm for handling heterogeneous datasets during Vision-Language-Action (VLA) model pretraining, offering a unified action space across embodiments. However, existing LAMs often rely on discrete quantization encode and decode pipelines, which can lead to trivial frame reconstruction behavior, limited representational capacity, and a lack of physically meaningful structure. We introduce RotVLA, a VLA framework built on a continuous rotational latent action representation. Latent actions are modeled as elements of SO(n), providing continuity, compositionality, and structured geometry aligned with real-world action dynamics. A triplet frame learning framework further enforces meaningful temporal dynamics while avoiding degeneration. RotVLA consists of a VLM backbone and a flow-matching action head, pretrained on large-scale cross-embodiment robotic datasets and human videos with latent-action supervision. For downstream robot control, the flow-matching head is extended into a unified action expert that jointly denoises latent and robot actions. Here, latent actions serve as a latent planner, providing high-level guidance that conditions action generation. With only 1.7B parameters and 1700+ hours of pretraining data, RotVLA achieves 98.2% on LIBERO and 89.6% / 88.5% on RoboTwin2.0 under clean and randomized settings, respectively. It also demonstrates strong real-world performance on manipulation tasks, consistently outperforming existing VLA models.

Concepts involved in long-form videos such as people, objects, and their interactions, can be viewed as following an implicit prior. They are notably complex and continue to pose challenges to be comprehensively learned. In recent years, generative pre-training (GPT) has exhibited versatile capacities in modeling any kind of text content even visual locations. Can this manner work for learning long-form video prior? Instead of operating on pixel space, it is efficient to employ visual locations like bounding boxes and keypoints to represent key information in videos, which can be simply discretized and then tokenized for consumption by GPT. Due to the scarcity of suitable data, we create a new dataset called Storyboard20K from movies to serve as a representative. It includes synopses, shot-by-shot keyframes, and fine-grained annotations of film sets and characters with consistent IDs, bounding boxes, and whole body keypoints. In this way, long-form videos can be represented by a set of tokens and be learned via generative pre-training. Experimental results validate that our approach has great potential for learning long-form video prior. Code and data will be released at https://github.com/showlab/Long-form-Video-Prior.

In the past decade, image foundation models (IFMs) have achieved unprecedented progress. However, the potential of directly using IFMs for video self-supervised representation learning has largely been overlooked. In this study, we propose an advancing video self-supervised learning (AdViSe) approach, aimed at significantly reducing the training overhead of video representation models using pre-trained IFMs. Specifically, we first introduce temporal modeling modules (ResNet3D) to IFMs, constructing a video representation model. We then employ a video self-supervised learning approach, playback rate perception, to train temporal modules while freezing the IFM components. Experiments on UCF101 demonstrate that AdViSe achieves performance comparable to state-of-the-art methods while reducing training time by 3.4times and GPU memory usage by 8.2times. This study offers fresh insights into low-cost video self-supervised learning based on pre-trained IFMs. Code is available at https://github.com/JingwWu/advise-video-ssl.

Modern computer vision is converging on a closed loop in which perception, reasoning and generation mutually reinforce each other. However, this loop remains incomplete: the top-down influence of high-level reasoning on the foundational learning of low-level perceptual features is not yet underexplored. This paper addresses this gap by proposing a new paradigm for pretraining foundation models in downstream domains. We introduce Visual insTruction Pretraining (ViTP), a novel approach that directly leverages reasoning to enhance perception. ViTP embeds a Vision Transformer (ViT) backbone within a Vision-Language Model and pretrains it end-to-end using a rich corpus of visual instruction data curated from target downstream domains. ViTP is powered by our proposed Visual Robustness Learning (VRL), which compels the ViT to learn robust and domain-relevant features from a sparse set of visual tokens. Extensive experiments on 16 challenging remote sensing and medical imaging benchmarks demonstrate that ViTP establishes new state-of-the-art performance across a diverse range of downstream tasks. The code is available at github.com/zcablii/ViTP.

To effectively exploit the potential of large-scale models, various pre-training strategies supported by massive data from different sources are proposed, including supervised pre-training, weakly-supervised pre-training, and self-supervised pre-training. It has been proved that combining multiple pre-training strategies and data from various modalities/sources can greatly boost the training of large-scale models. However, current works adopt a multi-stage pre-training system, where the complex pipeline may increase the uncertainty and instability of the pre-training. It is thus desirable that these strategies can be integrated in a single-stage manner. In this paper, we first propose a general multi-modal mutual information formula as a unified optimization target and demonstrate that all existing approaches are special cases of our framework. Under this unified perspective, we propose an all-in-one single-stage pre-training approach, named Maximizing Multi-modal Mutual Information Pre-training (M3I Pre-training). Our approach achieves better performance than previous pre-training methods on various vision benchmarks, including ImageNet classification, COCO object detection, LVIS long-tailed object detection, and ADE20k semantic segmentation. Notably, we successfully pre-train a billion-level parameter image backbone and achieve state-of-the-art performance on various benchmarks. Code shall be released at https://github.com/OpenGVLab/M3I-Pretraining.

Recent self-supervised video representation learning methods have found significant success by exploring essential properties of videos, e.g. speed, temporal order, etc. This work exploits an essential yet under-explored property of videos, the video continuity, to obtain supervision signals for self-supervised representation learning. Specifically, we formulate three novel continuity-related pretext tasks, i.e. continuity justification, discontinuity localization, and missing section approximation, that jointly supervise a shared backbone for video representation learning. This self-supervision approach, termed as Continuity Perception Network (CPNet), solves the three tasks altogether and encourages the backbone network to learn local and long-ranged motion and context representations. It outperforms prior arts on multiple downstream tasks, such as action recognition, video retrieval, and action localization. Additionally, the video continuity can be complementary to other coarse-grained video properties for representation learning, and integrating the proposed pretext task to prior arts can yield much performance gains.

In this paper, we show that useful video representations can be learned from synthetic videos and natural images, without incorporating natural videos in the training. We propose a progression of video datasets synthesized by simple generative processes, that model a growing set of natural video properties (e.g. motion, acceleration, and shape transformations). The downstream performance of video models pre-trained on these generated datasets gradually increases with the dataset progression. A VideoMAE model pre-trained on our synthetic videos closes 97.2% of the performance gap on UCF101 action classification between training from scratch and self-supervised pre-training from natural videos, and outperforms the pre-trained model on HMDB51. Introducing crops of static images to the pre-training stage results in similar performance to UCF101 pre-training and outperforms the UCF101 pre-trained model on 11 out of 14 out-of-distribution datasets of UCF101-P. Analyzing the low-level properties of the datasets, we identify correlations between frame diversity, frame similarity to natural data, and downstream performance. Our approach provides a more controllable and transparent alternative to video data curation processes for pre-training.

The use of pretrained backbones with fine-tuning has been successful for 2D vision and natural language processing tasks, showing advantages over task-specific networks. In this work, we introduce a pretrained 3D backbone, called {\SST}, for 3D indoor scene understanding. We design a 3D Swin transformer as our backbone network, which enables efficient self-attention on sparse voxels with linear memory complexity, making the backbone scalable to large models and datasets. We also introduce a generalized contextual relative positional embedding scheme to capture various irregularities of point signals for improved network performance. We pretrained a large {\SST} model on a synthetic Structured3D dataset, which is an order of magnitude larger than the ScanNet dataset. Our model pretrained on the synthetic dataset not only generalizes well to downstream segmentation and detection on real 3D point datasets, but also outperforms state-of-the-art methods on downstream tasks with +2.3 mIoU and +2.2 mIoU on S3DIS Area5 and 6-fold semantic segmentation, +1.8 mIoU on ScanNet segmentation (val), +1.9 mAP@0.5 on ScanNet detection, and +8.1 mAP@0.5 on S3DIS detection. A series of extensive ablation studies further validate the scalability, generality, and superior performance enabled by our approach. The code and models are available at https://github.com/microsoft/Swin3D .

Click-based interactive image segmentation aims at extracting objects with a limited user clicking. A hierarchical backbone is the de-facto architecture for current methods. Recently, the plain, non-hierarchical Vision Transformer (ViT) has emerged as a competitive backbone for dense prediction tasks. This design allows the original ViT to be a foundation model that can be finetuned for downstream tasks without redesigning a hierarchical backbone for pretraining. Although this design is simple and has been proven effective, it has not yet been explored for interactive image segmentation. To fill this gap, we propose SimpleClick, the first interactive segmentation method that leverages a plain backbone. Based on the plain backbone, we introduce a symmetric patch embedding layer that encodes clicks into the backbone with minor modifications to the backbone itself. With the plain backbone pretrained as a masked autoencoder (MAE), SimpleClick achieves state-of-the-art performance. Remarkably, our method achieves 4.15 NoC@90 on SBD, improving 21.8% over the previous best result. Extensive evaluation on medical images demonstrates the generalizability of our method. We further develop an extremely tiny ViT backbone for SimpleClick and provide a detailed computational analysis, highlighting its suitability as a practical annotation tool.

Our objective in this work is video-text retrieval - in particular a joint embedding that enables efficient text-to-video retrieval. The challenges in this area include the design of the visual architecture and the nature of the training data, in that the available large scale video-text training datasets, such as HowTo100M, are noisy and hence competitive performance is achieved only at scale through large amounts of compute. We address both these challenges in this paper. We propose an end-to-end trainable model that is designed to take advantage of both large-scale image and video captioning datasets. Our model is an adaptation and extension of the recent ViT and Timesformer architectures, and consists of attention in both space and time. The model is flexible and can be trained on both image and video text datasets, either independently or in conjunction. It is trained with a curriculum learning schedule that begins by treating images as 'frozen' snapshots of video, and then gradually learns to attend to increasing temporal context when trained on video datasets. We also provide a new video-text pretraining dataset WebVid-2M, comprised of over two million videos with weak captions scraped from the internet. Despite training on datasets that are an order of magnitude smaller, we show that this approach yields state-of-the-art results on standard downstream video-retrieval benchmarks including MSR-VTT, MSVD, DiDeMo and LSMDC.

Large scale Vision-Language (VL) models have shown tremendous success in aligning representations between visual and text modalities. This enables remarkable progress in zero-shot recognition, image generation & editing, and many other exciting tasks. However, VL models tend to over-represent objects while paying much less attention to verbs, and require additional tuning on video data for best zero-shot action recognition performance. While previous work relied on large-scale, fully-annotated data, in this work we propose an unsupervised approach. We adapt a VL model for zero-shot and few-shot action recognition using a collection of unlabeled videos and an unpaired action dictionary. Based on that, we leverage Large Language Models and VL models to build a text bag for each unlabeled video via matching, text expansion and captioning. We use those bags in a Multiple Instance Learning setup to adapt an image-text backbone to video data. Although finetuned on unlabeled video data, our resulting models demonstrate high transferability to numerous unseen zero-shot downstream tasks, improving the base VL model performance by up to 14\%, and even comparing favorably to fully-supervised baselines in both zero-shot and few-shot video recognition transfer. The code will be released later at https://github.com/wlin-at/MAXI.

In this paper, we propose SVEET, a framework that requires merely training on a pretrained bidirectional video diffusion model but supports high-quality streaming video editing in an auto-regressive fashion. To tackle this problem, we first systematically revisit existing video-to-video diffusion approaches and identify two key principles for such streaming adaptation: backbone feature disentanglement and conditional frame independence. Building on these insights, we develop a novel paradigm for controllable video generation. At its core, an auxiliary model branch encodes source video inputs with temporally independent self-attention, and the intermediate features are injected into the corresponding backbone blocks for streaming-compatible control. Moreover, to bridge the discrepancy between the feature spaces of bidirectional and streaming models, we propose a decoupled training scheme that explicitly enforces the orthogonality between the optimization directions of video controllability and model causality. Such disentanglement ensures compatibility between the two objectives at inference and facilitates smooth zero-shot knowledge transfer across heterogeneous backbone architectures. Extensive experiments demonstrate that SVEET achieves superior editing quality while maintaining real-time performance, attaining 15 FPS on a single H100 GPU 17 without any auxiliary acceleration techniques. Codes are available at https://github.com/YujiaHu1109/SVEET.

Video-language pre-trained models have shown remarkable success in guiding video question-answering (VideoQA) tasks. However, due to the length of video sequences, training large-scale video-based models incurs considerably higher costs than training image-based ones. This motivates us to leverage the knowledge from image-based pretraining, despite the obvious gaps between image and video domains. To bridge these gaps, in this paper, we propose Tem-Adapter, which enables the learning of temporal dynamics and complex semantics by a visual Temporal Aligner and a textual Semantic Aligner. Unlike conventional pretrained knowledge adaptation methods that only concentrate on the downstream task objective, the Temporal Aligner introduces an extra language-guided autoregressive task aimed at facilitating the learning of temporal dependencies, with the objective of predicting future states based on historical clues and language guidance that describes event progression. Besides, to reduce the semantic gap and adapt the textual representation for better event description, we introduce a Semantic Aligner that first designs a template to fuse question and answer pairs as event descriptions and then learns a Transformer decoder with the whole video sequence as guidance for refinement. We evaluate Tem-Adapter and different pre-train transferring methods on two VideoQA benchmarks, and the significant performance improvement demonstrates the effectiveness of our method.

We study joint video and language (VL) pre-training to enable cross-modality learning and benefit plentiful downstream VL tasks. Existing works either extract low-quality video features or learn limited text embedding, while neglecting that high-resolution videos and diversified semantics can significantly improve cross-modality learning. In this paper, we propose a novel High-resolution and Diversified VIdeo-LAnguage pre-training model (HD-VILA) for many visual tasks. In particular, we collect a large dataset with two distinct properties: 1) the first high-resolution dataset including 371.5k hours of 720p videos, and 2) the most diversified dataset covering 15 popular YouTube categories. To enable VL pre-training, we jointly optimize the HD-VILA model by a hybrid Transformer that learns rich spatiotemporal features, and a multimodal Transformer that enforces interactions of the learned video features with diversified texts. Our pre-training model achieves new state-of-the-art results in 10 VL understanding tasks and 2 more novel text-to-visual generation tasks. For example, we outperform SOTA models with relative increases of 40.4% R@1 in zero-shot MSR-VTT text-to-video retrieval task and 55.4% in high-resolution dataset LSMDC. The learned VL embedding is also effective in generating visually pleasing and semantically relevant results in text-to-visual editing and super-resolution tasks.

Video-language pre-training has advanced the performance of various downstream video-language tasks. However, most previous methods directly inherit or adapt typical image-language pre-training paradigms to video-language pre-training, thus not fully exploiting the unique characteristic of video, i.e., temporal. In this paper, we propose a Hierarchical Temporal-Aware video-language pre-training framework, HiTeA, with two novel pre-training tasks for modeling cross-modal alignment between moments and texts as well as the temporal relations of video-text pairs. Specifically, we propose a cross-modal moment exploration task to explore moments in videos, which results in detailed video moment representation. Besides, the inherent temporal relations are captured by aligning video-text pairs as a whole in different time resolutions with multi-modal temporal relation exploration task. Furthermore, we introduce the shuffling test to evaluate the temporal reliance of datasets and video-language pre-training models. We achieve state-of-the-art results on 15 well-established video-language understanding and generation tasks, especially on temporal-oriented datasets (e.g., SSv2-Template and SSv2-Label) with 8.6% and 11.1% improvement respectively. HiTeA also demonstrates strong generalization ability when directly transferred to downstream tasks in a zero-shot manner. Models and demo will be available on ModelScope.

Pre-trained video large language models (Video LLMs) exhibit remarkable reasoning capabilities, yet adapting these models to new tasks involving additional modalities or data types (e.g., audio or 3D information) remains challenging. In this paper, we present PAVE, a flexible framework for adapting pre-trained Video LLMs to downstream tasks with side-channel signals, such as audio, 3D cues, or multi-view videos. PAVE introduces lightweight adapters, referred to as "patches," which add a small number of parameters and operations to a base model without modifying its architecture or pre-trained weights. In doing so, PAVE can effectively adapt the pre-trained base model to support diverse downstream tasks, including audio-visual question answering, 3D reasoning, multi-view video recognition, and high frame rate video understanding. Across these tasks, PAVE significantly enhances the performance of the base model, surpassing state-of-the-art task-specific models while incurring a minor cost of ~0.1% additional FLOPs and parameters. Further, PAVE supports multi-task learning and generalizes well across different Video LLMs. Our code is available at https://github.com/dragonlzm/PAVE.

Multi-camera 3D object detection for autonomous driving is a challenging problem that has garnered notable attention from both academia and industry. An obstacle encountered in vision-based techniques involves the precise extraction of geometry-conscious features from RGB images. Recent approaches have utilized geometric-aware image backbones pretrained on depth-relevant tasks to acquire spatial information. However, these approaches overlook the critical aspect of view transformation, resulting in inadequate performance due to the misalignment of spatial knowledge between the image backbone and view transformation. To address this issue, we propose a novel geometric-aware pretraining framework called GAPretrain. Our approach incorporates spatial and structural cues to camera networks by employing the geometric-rich modality as guidance during the pretraining phase. The transference of modal-specific attributes across different modalities is non-trivial, but we bridge this gap by using a unified bird's-eye-view (BEV) representation and structural hints derived from LiDAR point clouds to facilitate the pretraining process. GAPretrain serves as a plug-and-play solution that can be flexibly applied to multiple state-of-the-art detectors. Our experiments demonstrate the effectiveness and generalization ability of the proposed method. We achieve 46.2 mAP and 55.5 NDS on the nuScenes val set using the BEVFormer method, with a gain of 2.7 and 2.1 points, respectively. We also conduct experiments on various image backbones and view transformations to validate the efficacy of our approach. Code will be released at https://github.com/OpenDriveLab/BEVPerception-Survey-Recipe.

Video virtual try-on (VVT) technology has garnered considerable academic interest owing to its promising applications in e-commerce advertising and entertainment. However, most existing end-to-end methods rely heavily on scarce paired garment-centric datasets and fail to effectively leverage priors of advanced visual models and test-time inputs, making it challenging to accurately preserve fine-grained garment details and maintain temporal consistency in unconstrained scenarios. To address these challenges, we propose DreamVVT, a carefully designed two-stage framework built upon Diffusion Transformers (DiTs), which is inherently capable of leveraging diverse unpaired human-centric data to enhance adaptability in real-world scenarios. To further leverage prior knowledge from pretrained models and test-time inputs, in the first stage, we sample representative frames from the input video and utilize a multi-frame try-on model integrated with a vision-language model (VLM), to synthesize high-fidelity and semantically consistent keyframe try-on images. These images serve as complementary appearance guidance for subsequent video generation. In the second stage, skeleton maps together with fine-grained motion and appearance descriptions are extracted from the input content, and these along with the keyframe try-on images are then fed into a pretrained video generation model enhanced with LoRA adapters. This ensures long-term temporal coherence for unseen regions and enables highly plausible dynamic motions. Extensive quantitative and qualitative experiments demonstrate that DreamVVT surpasses existing methods in preserving detailed garment content and temporal stability in real-world scenarios. Our project page https://virtu-lab.github.io/

Video-text pretraining has achieved remarkable progress through the scaling of models and datasets, yet the quality of language supervision remains underexplored. Existing web-scale datasets often provide only a single sparse caption per video that fails to capture rich spatiotemporal semantics, while directly using captioning models can generate noisy descriptions. We propose a large-scale multimodal large language model-based supervision generation framework that improves supervision diversity, fidelity, and semantic coverage. Starting from 10 million videos, our approach generates multi-view captions (MVC) through complementary summary and detailed captions, reasoning-based refinement, and semantic positive caption generation. To effectively exploit supervision at different granularities, we further introduce a granularity-aware text representation with separate CLS tokens for summary and detailed views. We pretrain video-text models using the resulting supervision corpus and evaluate them across standard, fine-grained and detailed text-to-video retrieval benchmarks. Our approach consistently improves both zero-shot and fine-tuned performance while using smaller pretraining corpora than existing methods, demonstrating the importance of rich and complementary textual supervision for video-text pretraining. Project page: https://rvandeghen.github.io/mvc/

This paper explores feature prediction as a stand-alone objective for unsupervised learning from video and introduces V-JEPA, a collection of vision models trained solely using a feature prediction objective, without the use of pretrained image encoders, text, negative examples, reconstruction, or other sources of supervision. The models are trained on 2 million videos collected from public datasets and are evaluated on downstream image and video tasks. Our results show that learning by predicting video features leads to versatile visual representations that perform well on both motion and appearance-based tasks, without adaption of the model's parameters; e.g., using a frozen backbone. Our largest model, a ViT-H/16 trained only on videos, obtains 81.9% on Kinetics-400, 72.2% on Something-Something-v2, and 77.9% on ImageNet1K.

The millions of movies produced in the human history are valuable resources for computer vision research. However, learning a vision model from movie data would meet with serious difficulties. A major obstacle is the computational cost -- the length of a movie is often over one hour, which is substantially longer than the short video clips that previous study mostly focuses on. In this paper, we explore an alternative approach to learning vision models from movies. Specifically, we consider a framework comprised of a visual module and a temporal analysis module. Unlike conventional learning methods, the proposed approach learns these modules from different sets of data -- the former from trailers while the latter from movies. This allows distinctive visual features to be learned within a reasonable budget while still preserving long-term temporal structures across an entire movie. We construct a large-scale dataset for this study and define a series of tasks on top. Experiments on this dataset showed that the proposed method can substantially reduce the training time while obtaining highly effective features and coherent temporal structures.

Image-based visual-language (I-VL) pre-training has shown great success for learning joint visual-textual representations from large-scale web data, revealing remarkable ability for zero-shot generalisation. This paper presents a simple but strong baseline to efficiently adapt the pre-trained I-VL model, and exploit its powerful ability for resource-hungry video understanding tasks, with minimal training. Specifically, we propose to optimise a few random vectors, termed as continuous prompt vectors, that convert video-related tasks into the same format as the pre-training objectives. In addition, to bridge the gap between static images and videos, temporal information is encoded with lightweight Transformers stacking on top of frame-wise visual features. Experimentally, we conduct extensive ablation studies to analyse the critical components. On 10 public benchmarks of action recognition, action localisation, and text-video retrieval, across closed-set, few-shot, and zero-shot scenarios, we achieve competitive or state-of-the-art performance to existing methods, despite optimising significantly fewer parameters.

Recent Text-to-Video (T2V) models have demonstrated powerful capability in visual simulation of real-world geometry and physical laws, indicating its potential as implicit world models. Inspired by this, we explore the feasibility of leveraging the video generation prior for viewpoint planning from given 4D scenes, since videos internally accompany dynamic scenes with natural viewpoints. To this end, we propose a two-stage paradigm to adapt pre-trained T2V models for viewpoint prediction, in a compatible manner. First, we inject the 4D scene representation into the pre-trained T2V model via an adaptive learning branch, where the 4D scene is viewpoint-agnostic and the conditional generated video embeds the viewpoints visually. Then, we formulate viewpoint extraction as a hybrid-condition guided camera extrinsic denoising process. Specifically, a camera extrinsic diffusion branch is further introduced onto the pre-trained T2V model, by taking the generated video and 4D scene as input. Experimental results show the superiority of our proposed method over existing competitors, and ablation studies validate the effectiveness of our key technical designs. To some extent, this work proves the potential of video generation models toward 4D interaction in real world.

Vision-language (VL) pre-training has recently received considerable attention. However, most existing end-to-end pre-training approaches either only aim to tackle VL tasks such as image-text retrieval, visual question answering (VQA) and image captioning that test high-level understanding of images, or only target region-level understanding for tasks such as phrase grounding and object detection. We present FIBER (Fusion-In-the-Backbone-based transformER), a new VL model architecture that can seamlessly handle both these types of tasks. Instead of having dedicated transformer layers for fusion after the uni-modal backbones, FIBER pushes multimodal fusion deep into the model by inserting cross-attention into the image and text backbones, bringing gains in terms of memory and performance. In addition, unlike previous work that is either only pre-trained on image-text data or on fine-grained data with box-level annotations, we present a two-stage pre-training strategy that uses both these kinds of data efficiently: (i) coarse-grained pre-training based on image-text data; followed by (ii) fine-grained pre-training based on image-text-box data. We conduct comprehensive experiments on a wide range of VL tasks, ranging from VQA, image captioning, and retrieval, to phrase grounding, referring expression comprehension, and object detection. Using deep multimodal fusion coupled with the two-stage pre-training, FIBER provides consistent performance improvements over strong baselines across all tasks, often outperforming methods using magnitudes more data. Code is available at https://github.com/microsoft/FIBER.

Contrastive Language-Image Pretraining (CLIP) stands out as a prominent method for image representation learning. Various neural architectures, spanning Transformer-based models like Vision Transformers (ViTs) to Convolutional Networks (ConvNets) like ResNets, are trained with CLIP and serve as universal backbones across diverse vision tasks. Despite utilizing the same data and training objectives, the effectiveness of representations learned by these architectures raises a critical question. Our investigation explores the differences in CLIP performance among these backbone architectures, revealing significant disparities in their classifications. Notably, normalizing these representations results in substantial performance variations. Our findings showcase a remarkable possible synergy between backbone predictions that could reach an improvement of over 20% through informed selection of the appropriate backbone. Moreover, we propose a simple, yet effective approach to combine predictions from multiple backbones, leading to a notable performance boost of up to 6.34\%. We will release the code for reproducing the results.

Vision-Language Models (VLMs) are increasingly required to process unbounded video streams in applications such as video-call assistants, live commentary, and embodied robots. An ideal streaming system should support proactive interaction, long-horizon memory, and real-time processing, while resting on a VLM backbone capable of handling diverse in-the-wild streaming tasks. However, existing VLMs excel at offline video understanding but fall short in streaming capabilities and lack dedicated infrastructure for streaming deployment. We address this gap on three fronts. (i) For backbone capability, we construct Streaming-Train-248K, a streaming dataset paired with a novel training objective for adapting VLMs to streaming interaction and understanding. (ii) For real-world deployment, we introduce Streaming Harness, a plug-and-play system that endows any VLM with three core abilities: proactive interaction (per-second response decisions), long-term memory (12-hour context retention), and real-time processing (sub-second latency). (iii) To drive continued community progress on streaming capabilities, we design Streaming-Eval, a benchmark that reflects models' capabilities across diverse in-the-wild scenarios. Extensive experiments demonstrate consistent gains from our approach across all core capabilities required for streaming video understanding. We will open-source our data, code, and benchmark to advance the community's shift from offline video understanding to deployable streaming intelligence.

Video foundation models achieve strong performance across many video understanding tasks, but typically require large-scale pre-training on massive video datasets, resulting in substantial data and compute costs. In contrast, modern image foundation models already provide powerful spatial representations. This raises an important question: can competitive video models be built by reusing these spatial representations and pre-training only for temporal reasoning? We take initial steps toward exploring a lightweight training paradigm that freezes a pre-trained image foundation model and trains only a recurrent temporal module to process streaming video. By reusing an image foundation model as a spatial encoder, this approach could significantly reduce the amount of video data and compute required compared to end-to-end video pre-training. In this work, we explore the feasibility of this approach before investing in computing for video pre-training. Our empirical findings across multiple video understanding tasks suggest that strong temporal performance can emerge without large-scale video pre-training, motivating future work on recurrent video foundation models obtained by pre-training a temporal module on top of a frozen image foundation model. Code: https://github.com/tue-mps/towards-video-image-frozen .

We introduce SolarWM, a fully open foundation for building interactive video world models from data preparation through long-horizon inference. Training across heterogeneous data sources and video backbones is challenging: datasets differ in temporal scale, camera geometry, visual quality, motion, and captioning styles, while video generators use distinct representations and architectures. Naive data mixing and model-specific implementations therefore produce inconsistent supervision and make results difficult to reproduce and compare. SolarWM addresses this coupling with a reconfigurable multi-source data engine and a backbone-native adaptation framework. The engine converts 1.43 million canonical clips from 10 datasets into a unified, frame-aligned contract covering visual observations, metric camera geometry, captions, quality metadata, selection decisions, and provenance, while decoupling source processing from mixture construction. Under shared camera-conditioning, training, and inference interfaces, we instantiate four 5B--33B models based on Wan2.2, LTX-2.5, and MiniMax-H3 while preserving their native representations and objectives. A unified three-stage recipe combines bidirectional adaptation, teacher-forced autoregressive initialization, and distribution matching distillation. The resulting causal models enable real-time interaction over rollouts ranging from minutes to hours after being trained on only 5s sequences. By releasing the resulting data, pipeline, recipes, weights, and framework, SolarWM provides a reproducible and extensible foundation for interactive world-model research.

Self-supervised learning has emerged as a powerful paradigm for label-free model pretraining, particularly in the video domain, where manual annotation is costly and time-intensive. However, existing self-supervised approaches employ diverse experimental setups, making direct comparisons challenging due to the absence of a standardized benchmark. In this work, we establish a unified benchmark that enables fair comparisons across different methods. Additionally, we systematically investigate five critical aspects of self-supervised learning in videos: (1) dataset size, (2) model complexity, (3) data distribution, (4) data noise, and (5) feature representations. To facilitate this study, we evaluate six self-supervised learning methods across six network architectures, conducting extensive experiments on five benchmark datasets and assessing performance on two distinct downstream tasks. Our analysis reveals key insights into the interplay between pretraining strategies, dataset characteristics, pretext tasks, and model architectures. Furthermore, we extend these findings to Video Foundation Models (ViFMs), demonstrating their relevance in large-scale video representation learning. Finally, leveraging these insights, we propose a novel approach that significantly reduces training data requirements while surpassing state-of-the-art methods that rely on 10% more pretraining data. We believe this work will guide future research toward a deeper understanding of self-supervised video representation learning and its broader implications.

A short video succeeds not simply because of what it shows, but because of how it schedules attention -- yet current multimodal models lack the structural grammar to parse or produce this organization. Existing models can describe scenes, answer event-centric questions, and read on-screen text, but they are far less reliable at identifying timeline-grounded units such as hooks, cut rationales, shot-induced tension, and platform-facing packaging cues. We propose SV6D (Structured Video in Six Dimensions), inspired by professional storyboard practice in film and television production, a representation framework that decomposes internet-native video into six complementary structural dimensions -- subject, aesthetics, camera language, editing, narrative, and dissemination -- with each label tied to physically observable evidence on the timeline. We formalize a unified optimization objective over SV6D that combines Hungarian-matched temporal alignment, dimension-wise semantic label distance, and quality regularization. Building on this framework, we present Leum-VL-8B, an 8B video-language model that realizes the SV6D objective through an expert-driven post-training pipeline, further refined through verifiable reinforcement learning on perception-oriented tasks. Leum-VL-8B achieves 70.8 on VideoMME (w/o subtitles), 70.0 on MVBench, and 61.6 on MotionBench, while remaining competitive on general multimodal evaluations such as MMBench-EN. We also construct FeedBench, a benchmark for structure-sensitive short-video understanding. Our results indicate that the missing layer in video AI is not pixel generation but structural representation: grounded on the timeline, linked to visible evidence, and directly consumable by downstream workflows such as editing, retrieval, recommendation, and generation control, including text-heavy internet video formats with overlays and image-text layouts.

Understanding motion in video is a fundamental challenge for visual learning, as frame-to-frame change entangles two sources of dynamics: camera motion and object motion. This decomposition has remained underexplored in representation learning, partly because these factors are tightly coupled in natural videos and difficult to supervise separately. Yet recovering it is important for learning robust motion representations that separate meaningful object dynamics from camera-induced variation. We study whether such structured motion representations can be recovered from frozen features of a pretrained image vision transformer. We propose the Structured Dynamics Model (SDM), which explicitly separates the dominant source of temporal change from residual dynamics through future-feature prediction, rather than representing video change with a single entangled latent or with unstructured, spatially dense transition tokens. Training combines self-supervised learning on real video with weak supervision of scene dynamics on synthetic Kubric data. We evaluate SDM on ProbeMotion, a new evaluation suite spanning synthetic and real videos with camera motion, object motion, and combined dynamics. SDM outperforms backbone baselines using global CLS or average-pooled features, and compares favorably to strongly supervised representations such as VGGT on several probes, despite using substantially weaker supervision. These results suggest that pretrained image models can be readily repurposed into structured video-dynamics representations, providing a useful inductive bias for learning and analyzing latent video dynamics.

In recent years, there has been a proliferation of spatiotemporal foundation models in different scientific disciplines. While promising, these models are often domain-specific and are only assessed within the particular applications for which they are designed. Given that many tasks can be represented as video modeling problems, video foundation models (ViFMs) hold considerable promise as general-purpose domain-agnostic approaches. However, it is not known whether the knowledge acquired on large-scale but potentially out-of-domain data can be effectively transferred across diverse scientific disciplines, and if a single, pretrained ViFM can be competitive with domain-specific baselines. To address this, we introduce SciVid, a comprehensive benchmark comprising five *Sci*entific *Vid*eo tasks, across medical computer vision, animal behavior, and weather forecasting. We adapt six leading ViFMs to SciVid using simple trainable readout modules, establishing strong baselines and demonstrating the potential for effective transfer learning. Specifically, we show that state-of-the-art results can be obtained in several applications by leveraging the general-purpose representations from ViFM backbones. Furthermore, our results reveal the limitations of existing ViFMs, and highlight opportunities for the development of generalizable models for high-impact scientific applications. We release our code at https://github.com/google-deepmind/scivid to facilitate further research in the development of ViFMs.

Learning from multiple modalities, such as audio and video, offers opportunities for leveraging complementary information, enhancing robustness, and improving contextual understanding and performance. However, combining such modalities presents challenges, especially when modalities differ in data structure, predictive contribution, and the complexity of their learning processes. It has been observed that one modality can potentially dominate the learning process, hindering the effective utilization of information from other modalities and leading to sub-optimal model performance. To address this issue the vast majority of previous works suggest to assess the unimodal contributions and dynamically adjust the training to equalize them. We improve upon previous work by introducing a multi-loss objective and further refining the balancing process, allowing it to dynamically adjust the learning pace of each modality in both directions, acceleration and deceleration, with the ability to phase out balancing effects upon convergence. We achieve superior results across three audio-video datasets: on CREMA-D, models with ResNet backbone encoders surpass the previous best by 1.9% to 12.4%, and Conformer backbone models deliver improvements ranging from 2.8% to 14.1% across different fusion methods. On AVE, improvements range from 2.7% to 7.7%, while on UCF101, gains reach up to 6.1%.

Large-scale video-text pretraining achieves strong performance but depends on noisy, synthetic captions with limited semantic coverage, often overlooking implicit world knowledge such as object motion, 3D geometry, and physical cues. In contrast, masked video modeling (MVM) directly exploits spatiotemporal structures but trails text-supervised methods on general tasks. We find this gap arises from overlooked architectural issues: pixel-level reconstruction struggles with convergence and its low-level requirement often conflicts with semantics, while latent prediction often encourages shortcut learning. To address these, we disentangle the traditional encoder-decoder design into an Encoder-Predictor-Decoder (EPD) framework, where the predictor acts as a latent world model, and propose InternVideo-Next, a two-stage pretraining scheme that builds a semantically consistent yet detail-preserving latent space for this world model. First, conventional linear decoder in pixel MVM enforces the predictor output latent to be linearly projected to, thus separable in pixel space, causing the conflict with semantic abstraction. Our Stage 1 proposes a conditional diffusion decoder and injects reliable image-level semantic priors to enhance semantics and convergence, thus bridging pixel-level fidelity with high-level semantic abstraction. Stage 2 further learns world knowledge by predicting frozen Stage 1 targets within this space, mitigating shortcut learning. Trained on public, unlabeled videos, InternVideo-Next achieves state-of-the-art results across benchmarks and provides a scalable path toward general video representation learning.

In this paper, we explore a new domain for video-to-video translation. Motivated by the availability of animation movies that are adopted from illustrated books for children, we aim to stylize these videos with the style of the original illustrations. Current state-of-the-art video-to-video translation models rely on having a video sequence or a single style image to stylize an input video. We introduce a new problem for video stylizing where an unordered set of images are used. This is a challenging task for two reasons: i) we do not have the advantage of temporal consistency as in video sequences; ii) it is more difficult to obtain consistent styles for video frames from a set of unordered images compared to using a single image. Most of the video-to-video translation methods are built on an image-to-image translation model, and integrate additional networks such as optical flow, or temporal predictors to capture temporal relations. These additional networks make the model training and inference complicated and slow down the process. To ensure temporal coherency in video-to-video style transfer, we propose a new generator network with feature warping layers which overcomes the limitations of the previous methods. We show the effectiveness of our method on three datasets both qualitatively and quantitatively. Code and pretrained models are available at https://github.com/giddyyupp/wait.

Recent video and language pretraining frameworks lack the ability to generate sentences. We present Multimodal Video Generative Pretraining (MV-GPT), a new pretraining framework for learning from unlabelled videos which can be effectively used for generative tasks such as multimodal video captioning. Unlike recent video-language pretraining frameworks, our framework trains both a multimodal video encoder and a sentence decoder jointly. To overcome the lack of captions in unlabelled videos, we leverage the future utterance as an additional text source and propose a bidirectional generation objective -- we generate future utterances given the present mulitmodal context, and also the present utterance given future observations. With this objective, we train an encoder-decoder model end-to-end to generate a caption from raw pixels and transcribed speech directly. Our model achieves state-of-the-art performance for multimodal video captioning on four standard benchmarks, as well as for other video understanding tasks such as VideoQA, video retrieval and action classification.

Customizing text-to-image (T2I) models has seen tremendous progress recently, particularly in areas such as personalization, stylization, and conditional generation. However, expanding this progress to video generation is still in its infancy, primarily due to the lack of customized video data. In this work, we introduce Still-Moving, a novel generic framework for customizing a text-to-video (T2V) model, without requiring any customized video data. The framework applies to the prominent T2V design where the video model is built over a text-to-image (T2I) model (e.g., via inflation). We assume access to a customized version of the T2I model, trained only on still image data (e.g., using DreamBooth or StyleDrop). Naively plugging in the weights of the customized T2I model into the T2V model often leads to significant artifacts or insufficient adherence to the customization data. To overcome this issue, we train lightweight Spatial Adapters that adjust the features produced by the injected T2I layers. Importantly, our adapters are trained on "frozen videos" (i.e., repeated images), constructed from image samples generated by the customized T2I model. This training is facilitated by a novel Motion Adapter module, which allows us to train on such static videos while preserving the motion prior of the video model. At test time, we remove the Motion Adapter modules and leave in only the trained Spatial Adapters. This restores the motion prior of the T2V model while adhering to the spatial prior of the customized T2I model. We demonstrate the effectiveness of our approach on diverse tasks including personalized, stylized, and conditional generation. In all evaluated scenarios, our method seamlessly integrates the spatial prior of the customized T2I model with a motion prior supplied by the T2V model.

Traditional audio-visual methods rely on independent audio and visual backbones, which is costly and not scalable. In this work, we investigate using an audio-visual siamese network (AVSiam) for efficient and scalable audio-visual pretraining. Our framework uses a single shared vision transformer backbone to process audio and visual inputs, improving its parameter efficiency, reducing the GPU memory footprint, and allowing us to scale our method to larger datasets and model sizes. We pretrain our model using a contrastive audio-visual matching objective with a multi-ratio random masking scheme, which enables our model to process larger audio-visual instance batches, helpful for contrastive learning. Unlike prior audio-visual methods, our method can robustly handle audio, visual, and audio-visual inputs with a single shared ViT backbone. Furthermore, despite using the shared backbone for both modalities, AVSiam achieves competitive or even better results than prior methods on AudioSet and VGGSound for audio-visual classification and retrieval. Our code is available at https://github.com/GenjiB/AVSiam