Recent advances in text-to-video (T2V) generation, exemplified by models such as Sora and Kling, have demonstrated strong potential for constructing world 3.Liquid motion 9.Vaposimulators.rization However, existing T2V models still struggle to understand abstract physical principles and to generate videos that faithfully obey physical laws.

Estimating scene lighting from a single image or video remains a longstand-ing challenge in computer vision and graphics. Learning-based approaches areconstrained by the scarcity of ground-truth HDR environment maps, which areexpensive to capture and limited in diversity. While recent generative modelsoffer strong priors for image synthesis, lighting estimation remains difficult dueto its reliance on indirect visual cues, the need to infer global (non-local) con-text, and the recovery of high-dynamic-range outputs. We propose LuxDiT, anovel data-driven approach that fine-tunes a video diffusion transformer to gen-erate HDR environment maps conditioned on visual input. Trained on a largesynthetic dataset with diverse lighting conditions, our model learns to infer il-lumination from indirect visual cues and generalizes effectively to real-worldscenes. To improve semantic alignment between the input and the predicted environment map, we introduce a low-rank adaptation finetuning strategy using a collected dataset of HDR panoramas.

Amazon's new AI search feature generates images in real time as shoppers type descriptions in the app, helping them find products they can picture but cannot name.

Human motion is inherently intentional, yet most motion modeling paradigms focus on low-level kinematics, overlooking the semantic and causal factors that drive behavior. Existing datasets further limit progress: they capture short, decontextualized actions in static scenes, providing little grounding for embodied reasoning. To address these limitations, we introduce Intend to Move (I2M), a large-scale, multimodal dataset for intention-grounded motion modeling. I2M contains 10.1 hours of two-person 3D motion sequences recorded in dynamic realistic home environments, accompanied by multi-view RGB-D video, 3D scene geometry, and language annotations of each participant's evolving intentions. Benchmark experiments reveal a fundamental gap in current motion models: they fail to translate high-level goals into physically and socially coherent motion. I2M thus serves not only as a dataset but as a benchmark for embodied intelligence, enabling research on models that can reason about, predict, and act upon the "why" behind human motion.

Diffusion models have demonstrated remarkable generative capabilities, and Conditional Flow Matching (CFM) has improved their inference efficiency by following optimal transport paths. However, CFM-based models still require multiple iterative sampling steps, which makes them unsuitable for real-time or streaming generation scenarios. In this paper, we introduce StreamFlow, a novel streaming generative model designed for real-time audio generation from discrete tokens. StreamFlow leverages a causal noising training framework along the time axis and predicts multi-time vector fields at once on each stream, enabling streaming inference with minimal latency. To further improve generalization, we propose Scale-DiT, a Diffusion Transformer architecture that enhances robustness by modeling, normalizing, and scaling feature differences prior to skip connections. This significantly improves the robustness and performance of DiT without increasing the parameter size.

The remarkable reasoning capability of large language models (LLMs) stems from cognitive behaviors that emerge through reinforcement with verifiable rewards. This work investigates how to transfer this principle to Multimodal LLMs (MLLMs) to unlock advanced visual reasoning. We introduce a two-stage paradigm built on Qwen2.5-VL-7B: a massive linguistic cold-start fine-tuning, followed by multimodal reinforcement learning (RL) spanning nearly 1,000 steps--surpassing all previous open-source efforts in scale. This pioneering work reveals three fundamental insights: 1) Behavior transfer emerges surprisingly early in cold start due to linguistic mental imagery.

Textual claims are often accompanied by images to enhance their credibility and spread on social media, but this also raises concerns about the spread of misinformation. Existing datasets for automated verification of image-text claims remain limited, as they often consist of synthetic claims and lack evidence annotations to capture the reasoning behind the verdict. In this work, we introduce AVERIMATEC, a dataset consisting of 1,297 real-world image-text claims. Each claim is annotated with question-answer (QA) pairs containing evidence from the web, reflecting a decomposed reasoning regarding the verdict. We mitigate common challenges in fact-checking datasets such as contextual dependence, temporal leakage, and evidence insufficiency, via claim normalization, temporally constrained evidence annotation, and a two-stage sufficiency check. We assess the consistency of the annotation in AVERIMATEC via inter-annotator studies, achieving a κ = 0.742 on verdicts and 74.7% consistency on QA pairs. We also propose a novel evaluation method for evidence retrieval and conduct extensive experiments to establish baselines for verifying image-text claims using open-web evidence.

'Have I been influenced, or is this actually me?' How personal taste fell out of fashion Our favourite music, clothes and books used to be markers of individuality - but the algorithm has made us all sheep. What music, films, clothes, art, books - anything, really - do you actually like? Do you find these questions more difficult to answer than you would have done 10 years ago? It has become impossible to ignore: personal taste has been seriously debased - if not completely destroyed - by technological advancement. We know the internet has radically altered the way we form our opinions and beliefs. Now we're waking up to another sobering truth: it has wrecked our capacity to form our own preferences. It used to go something like this. We experienced the outside world - including arts, culture and fashion - via a combination of community, geography, mass and specialist media, and serendipitous accidents.

Controllability, temporal coherence, and detail synthesis remain the most critical challenges in video generation. In this paper, we focus on a commonly used yet underexplored cinematic technique known as Frame In and Frame Out.

Transformers and their attention mechanism have been revolutionary in the field of Machine Learning. While originally proposed for the language data, they quickly found their way to the image, video, graph, etc. data modalities with various signal geometries. Despite this versatility, generalizing the attention mechanism to scenarios where data is presented at different scales from potentially different modalities is not straightforward. The attempts to incorporate hierarchy and multimodality within transformers are largely based on ad hoc heuristics, which are not seamlessly generalizable to similar problems with potentially different structures. To address this problem, in this paper, we take a fundamentally different approach: we first propose a mathematical construct to represent multi-modal, multi-scale data. We then mathematically derive the neural attention mechanics for the proposed construct from the first principle of entropy minimization. We show that the derived formulation is optimal in the sense of being the closest to the standard Softmax attention while incorporating the inductive biases originating from the hierarchical/geometric information of the problem. We further propose an efficient algorithm based on dynamic programming to compute our derived attention mechanism. By incorporating it within transformers, we show that the proposed hierarchical attention mechanism not only can be employed to train transformer models in hierarchical/multi-modal settings from scratch, but it can also be used to inject hierarchical information into classical, pre-trained transformer models post training, resulting in more efficient models in zero-shot manner.