Tag
Cloudflare 发布 clef-omni,支持音频和视频输入,被用于 PII 检测、恶意域名识别等场景;同时 clef-flash 价格降至每百万输入 token 0.038 美元,具备较高性价比。
Kandinsky Lab presents Kandinsky 6.0 Video, a family of foundation diffusion models (Lite 3B and Pro 29B) for synchronized text-to-audio-video and image-to-audio-video generation with lip-synced 44kHz audio and Full-HD super-resolution. The models use a dual-stream CrossDiT architecture with bidirectional cross-attention, outperform Kandinsky 5.0, and are released with code, checkpoints, and diffusers integration under the MIT license.
Salt++ 是一个两阶段后训练框架,通过 CausalSelf-Flow 和上下文对齐的自回归 DMD,解决少步流式音视频生成中教师强制与因果分布蒸馏的上下文错配问题,在 480p 4-step 设定下相对 OmniForcing 提升视觉质量 57%、运动质量 45%。
提出 Adaptive Reward Routing,在联合音频-视频扩散模型的前向过程强化学习(DiffusionNFT)中动态调整更新位置与奖励权重:利用双向跨注意力响应定位跨模态影响,并以偏好先验加残差修正的方式协调多个竞争奖励,从而提升模态质量、语义一致性与音视频同步。
This article provides a technical description of SMPTE/EBU timecode synchronization systems, focusing on the Bi-Phase Mark LTC timecode and related standards, and discusses its application in music technology with a mention of the TS1 tape sync product.
OmniScope is a training-free token compression framework for omnimodal LLMs that estimates audio and video relevance separately using the query as a shared anchor, achieving up to 3.53x prefill speedup and over 15% GPU memory reduction with minimal accuracy loss.
OmniVAE is a jointly trained audio-video VAE that uses segment-level contrastive learning and feature distillation to align latent spaces, improving joint generation quality and synchronization in text-to-audio-video generation.
Thinking Machines Lab, founded by ex-OpenAI executives, releases its first open-weight AI model, Inkling, a 975-billion-parameter model capable of reasoning, coding, and processing audio, video, and text.
AVOC introduces a retrieval-inspired token compression method for omni-modal LLMs that effectively handles hour-long audio-video inputs by selecting informative tokens based on relevance, importance, and diversity. The framework achieves state-of-the-art results on long-form audio-video understanding benchmarks, surpassing prior methods by significant margins.
UnityShots is a memory-driven multi-shot audio-video generation system that maintains consistent subject appearance and audio across video cuts using fixed-size long-term and short-term memory slots with boundary-conditioned gates and discrete cut-type priors. It outperforms open-source baselines on cross-shot coherence metrics and matches closed-source systems.
StreamChar is a streaming framework for real-time audio-video generation of character animation, using an LLM orchestrator and joint audio-video DiT with two-stage distillation and memory mechanisms to maintain long-horizon consistency and visual quality.