Explore the complete news stream by source, topic and date.
A Reorganization of the Organization In 1990, a 17-year-old science student made an important decision. He gave up the opportunity to be directly admitted to Tsinghua University’s automotive engineering program and instead chose to major in speech synthesis technology at the University of Science and Technology of China. Nine years later, he and a dozen or so fellow USTC students founded “Garage....”
arXiv:2610.33008v1 Announce Type: new Abstract: Diffusion large language models (dLLMs) enable flexible generation through iterative masked denoising, but incur substantial inference cost because token representations are repeatedly recomputed across denoising steps. Existing cache-based methods reduce this cost by reusing intermediate results across denoising steps and recomputing only a subset of positions at each step, chosen either on a fixed schedule or by heuristic signals of change. Our trajectory experiments instead point to decoding trajectory as the more relevant quantity. Cached generation remains accurate when it stays close to the full-denoising trajectory, with errors concentrated around unmask-order deviations and reduced as longer reference prefixes are recovered. This sugg
arXiv:2610.33062v1 Announce Type: new Abstract: Unified large multimodal models (ULMMs) are able to perform visual reasoning through interleaved text-image generation, but generating explicit intermediate images incurs significant computational overhead. While visual latent reasoning addresses this, most existing methods rely on vision-language models (VLMs) incapable of visual generation. We propose Oddin, a novel framework enabling latent visual reasoning in ULMMs. Oddin replaces intermediate images with compact latent tokens processed entirely by the understanding branch, reserving the generation branch for the final output image. We train the model via offline self-distillation with three objectives, final image generation losses, text distillation losses, and cross-entropy losses for
arXiv:2610.31721v1 Announce Type: new Abstract: Optimizing high-dimensional physical systems under expensive evaluations and complex constraints is a central challenge in science and engineering. In offline model-based optimization (MBO), methods must improve designs using only a fixed dataset and no test-time access to the true objective. However, existing methods typically suffer from conservatism, unconstrained surrogate exploitation, or function as static inference-time samplers that cannot refine candidates against precise physical objectives. In this paper, we propose Differentiable Diffusion Conditioning (D2C) for offline MBO, which leverages the fact that deterministic sampling makes the reverse denoising chain differentiable with respect to the conditioning signal. D2C turns a fro
arXiv:2610.31765v1 Announce Type: new Abstract: Generating multiple videos of a dynamic scene from a single reference video requires synthesizing both observed and unobserved content consistently across viewpoints. Existing camera-control methods for video diffusion typically rely on architectural mechanisms, including camera embeddings, cross-view attention, and geometric conditioning. Although effective for static scene content, these approaches often struggle with the fine-grained, dynamic geometry that must remain coherent across views and over time. We synchronize the diffusion process itself by sampling the Gaussian noise prior in a canonical 3D space derived from the source video and rasterizing it into target cameras, rather than sampling independently in each 2D image plane. Our m
arXiv:2610.31819v1 Announce Type: new Abstract: Speculative decoding is a technique that accelerates the inference speed of autoregressive large language models (LLMs) by sampling from a faster, lower quality draft model. Traditionally, a fundamental characteristic of speculative decoding is that it is lossless with respect to the verifier distribution, guaranteeing speed without sacrificing quality. This differs from recently developed diffusion LLMs (dLLMs), which are notably able to achieve substantial speedups by degrading the quality of their outputs with less diffusion effort. Because diffusion models are able to sample faster than autoregressive models by generating tokens in parallel, they are considered strong candidates for draft models in speculative diffusion. While it is widel
arXiv:2610.32042v1 Announce Type: new Abstract: Cross-lingual dubbing aims to convert source speech into target-language speech that remains synchronized with the original video. The key to this task is to produce a faithful translation that is naturally speakable within the source timing, which is non-trivial because languages differ in duration, rhythm, and articulation. In this paper, we propose X-Dubbing, which formulates the task as a joint problem of time-aware speech-to-text translation and video-aware text-to-speech. Specifically, our method consists of a Dubbing Translator that generates temporally structured scripts and a Voice Actor that renders them in synchrony with the source video, both grounded in a language-agnostic timing budget. We further refine the translator with dubb
arXiv:2610.31747v1 Announce Type: new Abstract: Existing 3D talking avatar methods mainly rely on parametric face models including 3DMM or FLAME, whose compact motion representations limit their ability to synthesize extreme expressions and to model accessories such as headwear. We introduce AudioLift, a two-stage framework that uses generated multi-view videos as the intermediate representation for audio-driven avatar generation without parametric face models or per-instance optimization. Our framework combines a synchronized multi-view video generation stage with an inconsistency-robust Gaussian reconstruction stage to recover coherent full-head geometry from a single portrait image. Extensive experiments demonstrate that AudioLift produces 3D talking avatars capable of capturing more ex
Thorsten Sløk, Chief Economist at Apollo Global Management, analyzed the S&P 493 (the S&P 500 excluding seven tech giants) and found that non-tech firms’ profit margins are on par with per-capita revenue, with no evidence that AI has boosted productivity. He noted that AI lacks an understanding of key business factors such as accountability, trust, and corporate culture, and therefore has yet to deliver financial returns.
Why it matters：This view is based on empirical observations of the mainstream listed‑company universe (S&P 493), citing specific data ranges and limitations (excluding seven technology firms such as Alphabet and Amazon), and aligns with the findings of several recent studies, offering a verifiable analytical framework with clearly defined attribution boundaries.
Canva has launched its first in-house AI image-generation model, Canva Craft, which employs a two-stage architecture—planning followed by generation—to enhance design accuracy, consistency, and controllability. It supports 2K resolution and over 30 aspect ratios, catering to use cases such as social media, marketing content, and digital art.
Canva has officially launched its first in-house image-generation model, Canva Craft, which employs a two-stage generative architecture, supports 2K resolution and output in 30 aspect ratios, and emphasizes text readability and layer editability. The model was trained on millions of human-designed templates within the platform and has already been integrated into Dream Lab and the API; going forward, it will be accessible via MCP to third-party chatbots such as ChatGPT and Gemini.
AWS has released a technical solution that leverages the Kokoro-82M text-to-speech model hosted on SageMaker, using bidirectional WebSocket to enable AI agents to generate text and synthesize speech simultaneously, significantly reducing end-to-end speech response latency. The solution offers two audio back‑transmission methods—Base64‑SSE and unified binary transmission—and is explicitly labeled as “proof-of-value,” meaning it has not yet reached production‑ready status.
The AWS blog introduces an empirical, automated scaling method for deploying text-to-speech (TTS) models on SageMaker real-time endpoints, eliminating guesswork and basing the approach on two measured metrics—Q_max (the maximum sustainable concurrent requests per instance) and T_total (the time required for planned scaling)—to create a reproducible, verifiable scaling strategy.
IBM has released three open-source speech models in the Granite Speech 5.0 series—granite-speech-5.0-2b (high-precision multilingual ASR/MT), granite-speech-5.0-470m-turboctc (ultra-fast edge-side multilingual transcription), and the updated granite-speech-4.1-2b-plus (enhanced multi-speaker conversational modeling)—all licensed under Apache 2.0 and compatible with mainstream inference frameworks.
Mistral will launch Voxtral 3 by October 26, 2026—a family of end-to-end SDKs and models for multilingual real-time voice interaction, covering three major modules: ASR (including speaker separation), the inference layer, and TTS. The solution emphasizes data sovereignty and modular deployment, though all performance claims are based on the vendor’s internal testing and have not been independently verified by a third party.
Starting in October 2026, YouTube Music will gradually roll out generative AI music content labeling worldwide, continuing the practice already in place on the main YouTube platform for AI-generated videos. This measure does not entail removing or restricting content; rather, it responds to users’ demand for greater transparency regarding the origins of creative works.
Welcome FLUX.3 Image Klein in Diffusers
Algorithms & Theory
Gemini 3.8 text-to-speech says hello
