Visuomotor policies learned from demonstrations often overfit to nuisance visual factors in raw RGB observations, resulting in brittle behavior under appearance shifts such as background changes and object recoloring. We propose a task-aware observation interface that canonicalizes visual input into a shared representation, improving robustness to out-of-distribution (OOD) appearance changes without modifying or fine-tuning the policy. Given an RGB image and an open-vocabulary specification of task-relevant entities, we use SAM3 to segment the target object and robot/gripper. We construct an L0 observation by repainting segmented entities with predefined semantic colors on a constant background. For tasks requiring stronger geometric cues, we further inject monocular depth from Depth Anything 3 into the segmented regions via depth-guided overwrite, yielding a unified semantic–geometric observation (L1) that remains a standard 3-channel, image-like input. We evaluate on RoboMimic (Lift), ManiSkill YCB grasping under clutter, four RLBench tasks under controlled appearance shifts, and two real-world Franka tasks (ReachX and CloseCabinet). Across benchmarks and policy backbones (Flow Matching Policy and SmolVLA), our interface preserves in-distribution performance while substantially improving robustness under OOD visual shifts.

This report presents the results and findings of the first edition of the Short-Films 20K (SF20K) Competition, held in conjunction with the SLoMO Workshop at ICCV 2025. The competition is designed to advance story-level video understanding beyond short-clip action recognition, introducing an open-ended video question-answering task built on a corpus of amateur short films. This setup ensures that models must rely on multimodal understanding rather than memorization of popular movies. Evaluation is conducted using the SF20K-Test benchmark (95 movies, 979 question-answer pairs) and scored via LLM-QA-Eval, an automated judge based on GPT-4.1-nano. The competition attracted 22 teams and 286 submissions across two tracks: a Main Track with unrestricted model size and a Special Track limited to models under 8 billion parameters. The winning team achieved 65.7

For artistic applications, video generation requires fine-grained control over both performance and cinematography, i.e., the actor's motion and the camera trajectory. We present ActCam, a zero-shot method for video generation that jointly transfers character motion from a driving video into a new scene and enables per-frame control of intrinsic and extrinsic camera parameters. ActCam builds on any pretrained image-to-video diffusion model that accepts conditioning in terms of scene depth and character pose. Given a source video with a moving character and a target camera motion, ActCam generates pose and depth conditions that remain geometrically consistent across frames. We then run a single sampling process with a two-phase conditioning schedule: early denoising steps condition on both pose and sparse depth to enforce scene structure, after which depth is dropped and pose-only guidance refines high-frequency details without over-constraining the generation. We evaluate ActCam on multiple benchmarks spanning diverse character motions and challenging viewpoint changes. We find that, compared to pose-only control and other pose and camera methods, ActCam improves camera adherence and motion fidelity, and is preferred in human evaluations, especially under large viewpoint changes. Our results highlight that careful camera-consistent conditioning and staged guidance can enable strong joint camera and motion control without training. Project page: https://elkhomar.github.io/actcam/.

Vision-and-Language Navigation (VLN) has long been constrained by the limited diversity and scalability of simulator-curated datasets, which fail to capture the complexity of real-world environments. To overcome this limitation, we introduce a large-scale video-instruction framework derived from web-based room tour videos, enabling agents to learn from natural human walking demonstrations in diverse, realistic indoor settings. Unlike existing datasets, our framework integrates both open-ended description-enriched trajectories and action-enriched trajectories reconstructed in 3D, providing richer spatial and semantic supervision. A key extension in this work is the incorporation of implicit geometry representations, which extract spatial cues directly from RGB frames without requiring fragile 3D reconstruction. This approach substantially improves data utilization, alleviates reconstruction failures, and unlocks large portions of previously unusable video data. Comprehensive experiments across multiple VLN benchmarks (CVDN, SOON, R2R, and REVERIE) demonstrate that our method not only sets new state-of-the-art performance but also enables the development of robust zero-shot navigation agents. By bridging large-scale web videos with implicit spatial reasoning, this work advances embodied navigation towards more scalable, generalizable, and real-world applicable solutions.

Monocular 3D scene reconstruction has recently seen significant progress. Powered by the modern neural architectures and large-scale data, recent methods achieve high performance in depth estimation from a single image. Meanwhile, reconstructing and decomposing common scenes into individual 3D objects remains a hard challenge due to the large variety of objects, frequent occlusions and complex object relations. Notably, beyond shape and pose estimation of individual objects, applications in robotics and animation require physically-plausible scene reconstruction where objects obey physical principles of non-penetration and realistic contacts. In this work we advance object-level scene reconstruction along two directions. First, we introduceMessyKitchens, a new dataset with real-world scenes featuring cluttered environments and providing high-fidelity object-level ground truth in terms of 3D object shapes, poses and accurate object contacts. Second, we build on the recent SAM 3D approach for single-object reconstruction and extend it with Multi-Object Decoder (MOD) for joint object-level scene reconstruction. To validate our contributions, we demonstrate MessyKitchens to significantly improve previous datasets in registration accuracy and inter-object penetration. We also compare our multi-object reconstruction approach on three datasets and demonstrate consistent and significant improvements of MOD over the state of the art. Our new benchmark, code and pre-trained models will become publicly available on our project website: https://messykitchens.github.io/.

Intuitive physics understanding in video diffusion models plays an essential role in building general-purpose physically plausible world simulators, yet accurately evaluating such capacity remains a challenging task due to the difficulty in disentangling physics correctness from visual appearance in generation. To the end, we introduce LikePhys, a training-free method that evaluates intuitive physics in video diffusion models by distinguishing physically valid and impossible videos using the denoising objective as an ELBO-based likelihood surrogate on a curated dataset of valid-invalid pairs. By testing on our constructed benchmark of twelve scenarios spanning over four physics domains, we show that our evaluation metric, Plausibility Preference Error (PPE), demonstrates strong alignment with human preference, outperforming state-of-the-art evaluator baselines. We then systematically benchmark intuitive physics understanding in current video diffusion models. Our study further analyses how model design and inference settings affect intuitive physics understanding and highlights domain-specific capacity variations across physical laws. Empirical results show that, despite current models struggling with complex and chaotic dynamics, there is a clear trend of improvement in physics understanding as model capacity and inference settings scale up.

Vision-Language-Action models (VLAs) are emerging as powerful tools for learning generalizable visuomotor control policies. However, current VLAs are mostly trained on large-scale image–text–action data and remain limited in two key ways: (i) they struggle with pixel-level scene understanding, and (ii) they rely heavily on textual prompts, which reduces their flexibility in real-world settings. To address these challenges, we introduce PixelVLA, the first VLA model designed to support both pixel-level reasoning and multimodal prompting with text and visual inputs. Our approach is built on a new visuomotor instruction tuning framework that integrates a multiscale pixel-aware encoder with a visual prompting encoder. To train PixelVLA effectively, we further propose a two-stage automated annotation pipeline that generates Pixel-160K, a large-scale dataset with pixel-level annotations derived from existing robot data. Experiments on three standard VLA benchmarks and two VLA model variants show that PixelVLA improves manipulation success rates by $10.1\%\sim28.7\%$ over OpenVLA, while requiring only $1.5\%$ of its pretraining cost. These results demonstrate that PixelVLA can be integrated into existing VLAs to enable more accurate, efficient, and versatile robot control in complex environments. The dataset and code will be released as open source.

As language models are increasingly deployed for complex autonomous tasks, their ability to reason accurately over longer horizons becomes critical. An essential component of this ability is planning and managing a long, complex chain-of-thought (CoT). We introduce LongCoT, a scalable benchmark of 2,500 expert-designed problems spanning chemistry, mathematics, computer science, chess, and logic to isolate and directly measure the long-horizon CoT reasoning capabilities of frontier models. Problems consist of a short input with a verifiable answer; solving them requires navigating a graph of interdependent steps that span tens to hundreds of thousands of reasoning tokens. Each local step is individually tractable for frontier models, so failures reflect long-horizon reasoning limitations. At release, the best models achieve <10% accuracy (GPT 5.2: 9.8%; Gemini 3 Pro: 6.1%) on LongCoT, revealing a substantial gap in current capabilities. Overall, LongCoT provides a rigorous measure of long-horizon reasoning, tracking the ability of frontier models to reason reliably over extended periods.

Articulated 3D objects are essential for interactive environments in embodied AI, robotics, and virtual reality, but reconstructing their structure and motion from sparse observations remains challenging. Existing approaches remain largely constrained by lack of supervised data or lack the priors needed to reliably recover articulation, hidden geometry, and internal object structure. We present the first debate-driven agentic approach to articulated 3D object reconstruction from text or image inputs that both grounds articulation reasoning in concrete motion and exposes the occluded geometry revealed under articulation. High-level agents reason about object semantics and motion using knowledge from vision-language and video models, while low-level agents estimate articulation parameters and interaction points; together, they engage in a two-round structured debate that first exploits global–local disagreement and then grounds the agents in freely generated video. The same video prior, conditioned on the agreed articulation, then drives each part through its motion to expose occluded interiors and geometry that cannot be inferred from a single static view. By combining agentic reasoning with a video generative prior, our approach jointly infers articulation and reconstructs complete 3D articulated objects, producing high-fidelity geometry, internal structure, and motion-consistent states beyond directly observed surfaces.

World Models (WMs) offer a promising mechanism for post-training Vision-Language-Action (VLA) policies by providing dynamics priors that improve generalization under task and scene variation. However, most WM-based post-training methods rely on pixel-space supervision, making policies sensitive to visual artifacts introduced by imperfect WM rollouts. We present World2Act, a latent-space post-training framework that transfers WM dynamics to the VLA policy without pixel-space supervision. World2Act operates in two stages: 1) it induces a shared video-action latent space by contrastively aligning WM-dynamics latents with action embeddings, and 2) it post-trains the VLA by guiding policy action representations toward WM-imagined dynamics rather than decoded pixels. Built on GR00T-N1.6, World2Act delivers absolute success-rate gains of up to +2.5

Recent video generation models have achieved remarkable progress and are now deployed in film, social media production, and advertising. Beyond their creative potential, such models also hold promise as world simulators for robotics and embodied decision making. Despite strong advances, current approaches still struggle to generate physically plausible object interactions and lack object-level control mechanisms. To address these limitations, we introduce KineMask, an approach for video generation that enables realistic rigid body control, interactions, and effects. Given a single image and a specified object velocity, our method generates videos with inferred motions and future object interactions. We propose a two-stage training strategy that gradually removes future motion supervision via object masks. Using this strategy we train video diffusion models (VDMs) on synthetic scenes of simple interactions and demonstrate significant improvements and generalization to rigid body and hand-object interactions in real scenes. Furthermore, KineMask integrates low-level motion control with high-level textual conditioning via predicted scene descriptions, leading to support for synthesis of complex dynamical phenomena. Our experiments show that KineMask generalizes to different VDMs and achieves strong improvements over recent models of comparable size. Ablation studies further highlight the complementary roles of low- and high-level conditioning in VDMs. Project Page: https://daromog.github.io/KineMask/

Recent progress in text-conditioned human motion generation has been largely driven by diffusion models trained on large-scale human motion data. Building on this progress, recent methods attempt to transfer such models for character animation and real robot control by applying a Whole-Body Controller (WBC) that converts diffusion-generated motions into executable trajectories. While WBC trajectories become compliant with physics, they may expose substantial deviations from original motion. To address this issue, we here propose PhysMoDPO, a Direct Preference Optimization framework. Unlike prior work that relies on hand-crafted physics-aware heuristics such as foot-sliding penalties, we integrate WBC into our training pipeline and optimize diffusion model such that the output of WBC becomes compliant both with physics and original text instructions. To train PhysMoDPO we deploy physics-based and task-specific rewards and use them to assign preference to synthesized trajectories. Our extensive experiments on text-to-motion and spatial control tasks demonstrate consistent improvements of PhysMoDPO in both physical realism and task-related metrics on simulated robots. Moreover, we demonstrate that PhysMoDPO results in significant improvements when applied to zero-shot motion transfer in simulation and for real-world deployment on a G1 humanoid robot.

Human motion generation has shown great advances thanks to the recent diffusion models trained on large-scale motion capture data. Most of existing works, however, currently target animation of isolated people in empty scenes. Meanwhile, synthesizing realistic human-object interactions in complex 3D scenes remains a critical challenge in computer graphics and robotics. One obstacle towards generating versatile high-fidelity human-object interactions is the lack of large-scale datasets with diverse object manipulations. Indeed, existing motion capture data is typically restricted to single people and manipulations of limited sets of objects. To address this issue, we propose an automatic motion extraction pipeline and use it to collect interaction-rich human motions. Our new dataset InterPose contains 73.8K sequences of 3D human motions and corresponding text captions automatically obtained from 45.8K videos with human-object interactions. We perform extensive experiments and demonstrate InterPose to bring significant improvements to state-of-the-art methods for human motion generation. Moreover, using InterPose we develop an LLM-based agent enabling zero-shot animation of people interacting with diverse objects and scenes.

Vision-Language-Action (VLA) models and world-action models have emerged as central paradigms for general-purpose robotic intelligence, yet their empirical progress remains constrained by the absence of evaluation protocols that are both physically realistic and diagnostically controlled. Simulator-centric benchmarks provide scale and reproducibility, but cannot fully capture the reality gap induced by perception noise, contact dynamics, latency, calibration error, and hardware constraints. Conversely, real-robot evaluations are often fragmented across platforms, scenes, objects, and scoring rules, making fair comparison and failure attribution difficult. We introduce ManipArena, a standardized real-robot evaluation framework for studying manipulation generalization under matched physical conditions. ManipArena comprises 20 tasks, 10,812 expert trajectories, 13.5M frames, and approximately 188 robot hours across tabletop and mobile manipulation. The framework combines schema-defined task variation, stratified in-domain, visualshift, and semantic-OOD trials, subtask-level partial-credit scoring, three-level language annotations, low-level motor signals, and paired real-to-sim environments reconstructed from physical scenes. Using ManipArena, we evaluate seven tabletop configurations spanning VLA and world-action-model policies. The results show that real-robot conclusions depend not only on architecture, but also on model provenance, fine-tuning regime, data sampling, and annotation granularity. ManipArena thus provides a reproducible and interpretable foundation for diagnosing capability boundaries and failure modes in embodied generalization.

Recent developments in vision-language models have significantly advanced video understanding. Existing datasets and tasks, however, have notable limitations. Most datasets are confined to short videos with limited events and narrow narratives. For example, datasets with instructional and egocentric videos often depict the activities of one person in a single scene. Although existing movie datasets offer richer content, they are often limited to short-term tasks, lack publicly available videos, and frequently encounter data leakage issues given the use of subtitles and other information about commercial movies during LLM pretraining. To address the above limitations, we propose Short-Films 20K (SF20K), the largest publicly available movie dataset. SF20K consists of 20,143 amateur films, amounting to 3,582 hours of video, with an average of 12 minutes per movie. We accompany this dataset with SF20K-Test, a manual, open-ended question answering benchmark. SF20K-Test consists of 95 movies and 979 question-answer pairs. Our extensive analysis of SF20K-Test reveals limited data leakage, emphasizes the need for long-term reasoning, and demonstrates the strong performance of recent VLMs. Finally, we show that instruction tuning on the large-scale dataset substantially improves model performance, paving the way for future progress in long-term video understanding.

In this paper, we propose an efficient contact planner for quadrupedal robots to navigate in extremely constrained environments such as stepping stones. The main difficulty in this setting stems from the mixed nature of the problem, namely discrete search over the steppable patches and continuous trajectory optimization. To speed up the discrete search, we study the properties of the transitions from one contact mode to another. In particular, we propose to learn a dynamic feasibility classifier and a target adjustment network. The former predicts if a contact transition between two contact modes is dynamically feasible. The latter is trained to compensate for misalignment in reaching a desired set of contact locations, due to imperfections of the low-level control. We integrate these learned networks in a Monte Carlo Tree Search (MCTS) contact planner. Our simulation results demonstrate that training these networks with offline data significantly speeds up the online search process and improves its accuracy.

Most existing Vision-Language-Action (VLA) models rely primarily on RGB information, while ignoring geometric cues crucial for spatial reasoning and manipulation. In this work, we introduce GLaD, a geometry-aware VLA framework that incorporates 3D geometric priors during pretraining through knowledge distillation. Rather than distilling geometric features solely into the vision encoder, we align the LLM's hidden states corresponding to visual tokens with features from a frozen geometry-aware vision transformer (VGGT), ensuring that geometric understanding is deeply integrated into the multimodal representations that drive action prediction. Pretrained on the Bridge dataset with this geometry distillation mechanism, GLaD achieves 94.1

Reasoning is a fundamental capability for solving complex multi-step problems, particularly in visual contexts where sequential step-wise understanding is essential. Existing approaches lack a comprehensive framework for evaluating visual reasoning and do not emphasize step-wise problem-solving. To this end, we propose a comprehensive framework for advancing step-by-step visual reasoning in large language models (LMMs) through three key contributions. First, we introduce a visual reasoning benchmark specifically designed to evaluate multi-step reasoning tasks. The benchmark presents a diverse set of challenges with eight different categories ranging from complex visual perception to scientific reasoning with over 4k reasoning steps in total, enabling robust evaluation of LLMs' abilities to perform accurate and interpretable visual reasoning across multiple steps. Second, we propose a novel metric that assesses visual reasoning quality at the granularity of individual steps, emphasizing both correctness and logical coherence. The proposed metric offers deeper insights into reasoning performance compared to traditional end-task accuracy metrics. Third, we present a new multimodal visual reasoning model, named LlamaV-o1, trained using a multi-step curriculum learning approach, where tasks are progressively organized to facilitate incremental skill acquisition and problem-solving. The proposed LlamaV-o1 is designed for multi-step reasoning and learns step-by-step through a structured training paradigm. Extensive experiments show that our LlamaV-o1 outperforms existing open-source models and performs favorably against close-source proprietary models. Compared to the recent Llava-CoT, our LlamaV-o1 achieves an average score of 67.3 with an absolute gain of 3.8% across six benchmarks while being 5 times faster during inference scaling. Our benchmark, model, and code are publicly available.

Existing Large Multimodal Models (LMMs) generally focus on only a few regions and languages. As LMMs continue to improve, it is increasingly important to ensure they understand cultural contexts, respect local sensitivities, and support low-resource languages, all while effectively integrating corresponding visual cues. In pursuit of culturally diverse global multimodal models, our proposed All Languages Matter Benchmark (ALM-bench) represents the largest and most comprehensive effort to date for evaluating LMMs across 100 languages. ALM-bench challenges existing models by testing their ability to understand and reason about culturally diverse images paired with text in various languages, including many low-resource languages traditionally underrepresented in LMM research. The benchmark offers a robust and nuanced evaluation framework featuring various question formats, including true/false, multiple choice, and open-ended questions, which are further divided into short and long-answer categories. ALM-bench design ensures a comprehensive assessment of a model's ability to handle varied levels of difficulty in visual and linguistic reasoning. To capture the rich tapestry of global cultures, ALM-bench carefully curates content from 13 distinct cultural aspects, ranging from traditions and rituals to famous personalities and celebrations. Through this, ALM-bench not only provides a rigorous testing ground for state-of-the-art open and closed-source LMMs but also highlights the importance of cultural and linguistic inclusivity, encouraging the development of models that can serve diverse global populations effectively. Our benchmark is publicly available.