Gradient clipping is increasingly important in centralized learning (CL) and federated learning (FL). Many works focus on its optimization properties under strong assumptions involving Gaussian noise and standard smoothness. However, practical machine learning tasks often only satisfy weaker conditions, such as heavy-tailed noise and $(L_0, L_1)$-smoothness. To bridge this gap, we propose a high-probability analysis for clipped Stochastic Gradient Descent (SGD) under these weaker assumptions. Our findings show a better convergence rate than existing ones can be achieved, and our high-probability analysis does not rely on the bounded gradient assumption. Moreover, we extend our analysis to FL, where a gap remains between expected and high-probability convergence, which the naive clipped SGD cannot bridge. Thus, we design a new \underline{Fed}erated \underline{C}lipped \underline{B}atched \underline{G}radient (FedCBG) algorithm, and prove the convergence and generalization bounds with high probability for the first time. Our analysis reveals the trade-offs between the optimization and generalization performance. Extensive experiments demonstrate that \methodname{} can generalize better to unseen client distributions than state-of-the-art baselines.

Continual learning (CL) aims to incrementally train a model to a sequence of tasks while maintaining performance on previously seen ones. Despite effectiveness in mitigating forgetting, data storage and replay may be infeasible due to privacy or security constraints, and are impractical or unavailable for arbitrary pre-trained models. Data-free or examplar-free CL aims to continually update models with new tasks without storing previous data. In addition to regularizing updates, we employ model inversion to synthesize data from the trained model, anchoring learned knowledge through replay without retaining old data. However, model inversion in predictive models faces two key challenges.

Video diffusion models substantially boost the productivity of artistic workflows with high-quality portrait video generative capacity. However, prevailing pipelines are primarily constrained to single-shot creation, while real-world applications urge multiple shots with identity consistency and flexible content controllability. In this work, we propose EchoShot, a native and scalable multi-shot framework for portrait customization built upon a foundation video diffusion model. To start with, we propose shot-aware position embedding mechanisms within the video diffusion transformer architecture to model inter-shot variations and establish intricate correspondence between multi-shot visual content and their textual descriptions. This simple yet effective design enables direct training on multi-shot video data without introducing additional computational overhead. To facilitate model training within multi-shot scenarios, we construct PortraitGala, a large-scale and high-fidelity human-centric video dataset featuring cross-shot identity consistency and fine-grained captions such as facial attributes, outfits, and dynamic motions. To further enhance applicability, we extend EchoShot to perform reference image-based personalized multi-shot generation and long video synthesis with infinite shot counts. Extensive evaluations demonstrate that EchoShot achieves superior identity consistency as well as attribute-level controllability in multi-shot portrait video generation. Notably, the proposed framework demonstrates potential as a foundational paradigm for general multi-shot video modeling.

We introduce MIRAGE, a new benchmark for multimodal expert-level reasoning and decision-making in consultative interaction settings. Designed for the domain of agriculture, MIRAGE captures the full complexity of expert consultations by combining natural user queries, expert-authored responses, and image-based context, offering a high-fidelity benchmark for evaluating models on grounded reasoning, clarification strategies, and long-form generation in a real-world, knowledge-intensive domain. Grounded in over 35,000 real user-expert interactions, and curated through a carefully designed multi-step pipeline, MIRAGE spans diverse crop health, pest diagnosis, and crop management scenarios. The benchmark includes more than 7,000 unique biological entities, covering plant species, pests, and diseases, making it one of the most taxonomically diverse benchmarks available for vision-language models in real-world expert-guided domains. Unlike existing benchmarks that rely on well-specified user inputs, MIRAGE features underspecified, context-rich scenarios, requiring models to infer latent knowledge gaps and either proactively guide the interaction or respond. Our benchmark comprises two core components. The Single-turn Challenge to reason over a single user turn and image set, identify relevant entities, infer causal explanations, and generate actionable recommendations; and a Multi-Turn challenge for dialogue state tracking, goal-driven generation, and expert-level conversational decision-making. We evaluate more than 20 closed and open-source frontier vision-language models (VLMs), using three reasoning language models as evaluators, highlighting the significant challenges posed by MIRAGE in both single-turn and multi-turn interaction settings. Even the advanced GPT4.1 and GPT4o models achieve 44.6% and 40.9% accuracy, respectively, indicating significant room for improvement.

Dipu Tamang from Nepal is among more than 400,000 international students in Japan. When Dipu Tamang arrived in Japan from Nepal in 2024, he joined a growing stream of young people who see the country less as a traditional study destination and more as a structured route into work and long-term opportunity. The 22-year-old graduated from Shinjuku Heiwa Japanese Language School in March and now studies international business at a vocational college in Tokyo. He juggles part-time work as a convenience store clerk and hotel housekeeper to help cover his living expenses. "At first, I was interested in Japanese pop culture," he said. "Then I wanted to learn the language.

North Korea will'never' get nuclear recognition, EU and South Korea say European Commission President Ursula von der Leyen shakes hands with South Korean President Lee Jae Myung next to European Council President Antonio Costa on the day of an EU-South Korea summit in Brussels on Wednesday. South Korea and the European Union have said that North Korea will "never" be recognized as a nuclear-weapon state, reaffirming their commitment to denuclearization days after China and North Korea pledged closer ties at a summit that made no public mention of the issue. South Korean President Lee Jae Myung held talks with European Commission President Ursula von der Leyen and European Council President Antonio Costa on Wednesday in Brussels, where they agreed to step up defense ties, including efforts to facilitate the exchange of classified information. "The DPRK will never be accepted as a nuclear-weapon state," the EU and South Korea said in a joint statement, referring to North Korea's formal name, the Democratic People's Republic of Korea. In a time of both misinformation and too much information, quality journalism is more crucial than ever. By subscribing, you can help us get the story right.

Pre-training equips text-to-image (T2I) models with broad world knowledge, but this alone is often insufficient to achieve high aesthetic quality and alignment. Consequently, supervised fine-tuning (SFT) is crucial for further refinement. However, its effectiveness highly depends on the quality of the fine-tuning dataset.Existing public SFT datasets frequently target narrow domains (e.g., anime or specific art styles), and the creation of high-quality, general-purpose SFT datasets remains a significant challenge.Current curation methods are often costly and struggle to identify truly impactful samples.This challenge is further complicated by the scarcity of public general-purpose datasets, as leading models often rely on large, proprietary, and poorly documented internal data, hindering broader research progress.This paper introduces a novel methodology for creating general-purpose SFT datasets by leveraging a pre-trained generative model as an estimator of high-impact training samples. We apply this methodology to construct and release Alchemist, a compact (3,350 samples) yet highly effective SFT dataset. Experiments demonstrate that Alchemist substantially improves the generative quality of five public T2I models while preserving diversity and style. Additionally, we release the fine-tuned models' weights to the public.

Sparse-reward reinforcement learning (RL) can model a wide range of highly complex tasks. Solving sparse-reward tasks is RL's core premise -- requiring efficient exploration coupled with long-horizon credit assignment -- and overcoming these challenges is key for building self-improving agents with superhuman ability. We argue that solving complex and high-dimensional tasks requires solving simpler tasks that are to the target task. In contrast, most prior work designs strategies for selecting exploratory tasks with the objective of solving task, making exploration of challenging high-dimensional, long-horizon tasks intractable. We find that the sense of direction, necessary for effective exploration, can be extracted from existing reinforcement learning algorithms, without needing any prior information.

Although significant progress has been made in audio-driven talking head generation, text-driven methods remain underexplored. In this work, we present OmniTalker, a unified framework that jointly generates synchronized talking audio-video content from input text while emulating the target identity's speaking and facial movement styles, including speech characteristics, head motion, and facial dynamics. Our framework adopts a dual-branch diffusion transformer (DiT) architecture, with one branch dedicated to audio generation and the other to video synthesis. At the shallow layers, cross-modal fusion modules are introduced to integrate information between the two modalities. In deeper layers, each modality is processed independently, with the generated audio decoded by a vocoder and the video rendered using a GAN-based high-quality visual renderer. Leveraging DiT's in-context learning capability through a masked-infilling strategy, our model can simultaneously capture both audio and visual styles without requiring explicit style extraction modules. Thanks to the efficiency of the DiT backbone and the optimized visual renderer, OmniTalker achieves real-time inference at 25 FPS. To the best of our knowledge, OmniTalker is the first one-shot framework capable of jointly modeling speech and facial styles in real time. Extensive experiments demonstrate its superiority over existing methods in terms of generation quality, particularly in preserving style consistency and ensuring precise audio-video synchronization, all while maintaining efficient inference.

Pre-trained Vision-Language Models (VLMs) exhibit strong generalization capabilities, enabling them to recognize a wide range of objects across diverse domains without additional training. However, they often retain irrelevant information beyond the requirements of specific target downstream tasks, raising concerns about computational efficiency and potential information leakage. This has motivated growing interest in approximate unlearning, which aims to selectively remove unnecessary knowledge while preserving overall model performance. Existing approaches to approximate unlearning have primarily focused on {\em class unlearning}, where a VLM is retrained to fail to recognize specified object classes while maintaining accuracy for others. However, merely forgetting object classes is often insufficient in practical applications.