
AI不再「炫技」,淘宝要让技术解决用户每一个具体问题 (opens in a new tab)
近日,在 CNCC2025 大会上,郑波首次公开了淘宝全模态大模型的最新进展,并系统介绍了多模态智能在淘宝 AIGX 技术体系的研究应用。另外,结合 AI 模型技术在淘宝应用中的实践,他认为,「狭义 AGI 很可能在 5-10 年内到来。」
Total 438 articles

近日,在 CNCC2025 大会上,郑波首次公开了淘宝全模态大模型的最新进展,并系统介绍了多模态智能在淘宝 AIGX 技术体系的研究应用。另外,结合 AI 模型技术在淘宝应用中的实践,他认为,「狭义 AGI 很可能在 5-10 年内到来。」

Shanghai AI Lab presents a wearable capture system using trackers, gloves and a camera, plus C³ motion reconstruction, and releases 100 hours of data and code. Reconstruction and multimodal generation are assessed separately; humanoid transfer is shown only in simulation.

Movens-5M pairs five million simulation-validated robot motions with text, video, speech and music. The team trains Movens-Zero and demonstrates it on a Unitree G1, using scaling experiments to examine gains from more data and larger motion experts.

The report covers Qwen4-Max's architecture, million-token context and long-horizon delivery examples. Work360 and AII show different cost reductions; complex open-ended tasks remain a weakness, while smaller models are planned for later open release.

Shishi reports average 2.48-times video-inference acceleration over Dense in launch tests, alongside GPU-memory scheduling, asset understanding and structured context for generation. Results apply to the demonstrated settings.

Researchers review more than 300 studies and nearly 70 benchmarks across video, embodied intelligence, multimodal retrieval and personal assistants, discussing compression, evidence traceability and the risks of outdated information.

The Accio-led team combines visually dependent tasks, process credit assignment and alternating training. Neos-35B averages 64.7, still 2.4 percentage points behind the comparison Gemini workflow.

The API handles yes/no, multiple-choice, and scoring tasks with multimodal input and shared context. Google reports test latency and pricing; real-world calibration and costs depend on application conditions.

DGP 为不同语言层动态分配视觉层权重。在论文的 LLaVA-1.5-7B 配置下,11 项指标均超过基线,团队也报告了额外参数和计算开销。

闫俊杰在 Scottish Mortgage 播客解释低推理成本如何支持后训练,并讨论 H3 开放权重和 M3 Pro 的开发进度。访谈亦区分年化收入运行率与财报收入。