{
 "generated": "2026-11-08",
 "licence": "Free to reuse with a link to https://aiworldwire.org/signals/",
 "items": [
  {
   "date": "2026-11-08",
   "type": "Model",
   "title": "Qwen3.8-27B (Alibaba Qwen)",
   "url": "https://huggingface.co/Qwen/Qwen3.8-27B",
   "region": "China",
   "topic": "Multimodal open model",
   "why": "27B-parameter vision-language model built on Qwen3.5, with a native 262,144-token context extendable to about 1 million; Qwen's own coding and agent benchmark scores are on the card. The card lists an Apache-2.0 licence.",
   "article": "https://aiworldwire.org/articles/states-take-ai-in-house-abu-dhabi-russia-south-korea/"
  },
  {
   "date": "2026-11-08",
   "type": "Paper",
   "title": "EDGAR: Efficient Diffusion-Driven Gaussian Avatar Refinement via Synthetic Failure",
   "url": "https://arxiv.org/abs/2611.07909",
   "region": "South Korea",
   "topic": "Vision and graphics",
   "why": "Submitted 5 November; a one-step diffusion fix for artifacts in 3D head avatars, with code linked at the authors' project page. The abstract page shows no affiliations; the region is our inference from the scan's KAIST tag.",
   "article": "https://aiworldwire.org/articles/states-take-ai-in-house-abu-dhabi-russia-south-korea/"
  },
  {
   "date": "2026-11-08",
   "type": "Paper",
   "title": "Muir: Linear Attention with Sensitivity-Allocated Gating and Matched-Filter Readout",
   "url": "https://arxiv.org/abs/2611.08376",
   "region": "US/Canada",
   "topic": "Efficient architectures",
   "why": "Submitted 5 November; per its authors, a linear-attention design that cuts perplexity 1.8 to 2.0% against a gated delta-rule baseline from 166M to 1.34B parameters. The abstract page shows no affiliations; the region is our inference from the scan's Google tag.",
   "article": "https://aiworldwire.org/articles/states-take-ai-in-house-abu-dhabi-russia-south-korea/"
  },
  {
   "date": "2026-11-07",
   "type": "Model",
   "title": "ERNIE-5.5-Flash-Base (Baidu)",
   "url": "https://huggingface.co/baidu/ERNIE-5.5-Flash-Base",
   "region": "China",
   "topic": "Open-weights base model",
   "why": "280B-parameter mixture-of-experts base checkpoint with 14B active per token, released 5 November before any instruction tuning. The custom licence covers use only outside the EU, the UK and South Korea, and asks for \"Powered by ERNIE\" on commercial products and a notice or authorisation by revenue (US$50 million). Benchmarks are Baidu's own.",
   "article": "https://aiworldwire.org/articles/ai-stocks-crash-midterms-put-data-centres-in-play/"
  },
  {
   "date": "2026-11-07",
   "type": "Paper",
   "title": "SLA: Scaling Spatial-Language-Action Model via Continual Pre-training",
   "url": "https://arxiv.org/abs/2611.08373",
   "region": "China",
   "topic": "Robotics",
   "why": "Submitted 5 November; adds 3D spatial alignment to robot-policy pre-training, starting from a pretrained pi05 policy, with the extra parts used only in training so the inference design is unchanged. The abstract page shows no affiliations; the region is our inference from the HKUSTGZ organisation tag on Hugging Face.",
   "article": "https://aiworldwire.org/articles/ai-stocks-crash-midterms-put-data-centres-in-play/"
  },
  {
   "date": "2026-11-07",
   "type": "Paper",
   "title": "From Pixels to Actions - Towards Native Vision-Language-Action Policies (NEO-VLA)",
   "url": "https://arxiv.org/abs/2611.06918",
   "region": "Singapore",
   "topic": "Robotics",
   "why": "Submitted 4 November; an encoder-free robot-control model that, per its authors, beats encoder-based designs under controlled comparisons. The abstract page shows no affiliations; the region is our inference from the mmlab-ntu organisation tag on Hugging Face.",
   "article": "https://aiworldwire.org/articles/ai-stocks-crash-midterms-put-data-centres-in-play/"
  },
  {
   "date": "2026-11-05",
   "type": "Model",
   "title": "Unlimited-OCR-2 (Baidu)",
   "url": "https://huggingface.co/baidu/Unlimited-OCR-2",
   "region": "China",
   "topic": "Document parsing",
   "why": "3B-parameter image-text-to-text model for very long documents; Baidu released it on 27 October and published the paper on 2 November. The model page lists the MIT licence.",
   "article": "https://aiworldwire.org/articles/schools-rein-in-ai-australia-limits-google-ties-search/"
  },
  {
   "date": "2026-11-05",
   "type": "Paper",
   "title": "Latent Reasoning in Continuous Space for Unified Multimodal Models (LARC)",
   "url": "https://arxiv.org/abs/2610.42149",
   "region": "South Korea",
   "topic": "Image generation",
   "why": "Reasons over continuous hidden states rather than text tokens before generating images; submitted 30 October. Authors' affiliations are not shown on the abstract page, so the country is our inference from author names.",
   "article": "https://aiworldwire.org/articles/schools-rein-in-ai-australia-limits-google-ties-search/"
  },
  {
   "date": "2026-11-05",
   "type": "Paper",
   "title": "Behind Video Generation: Insights on Multimodal Understanding from Kling",
   "url": "https://arxiv.org/abs/2611.04916",
   "region": "China",
   "topic": "Video generation",
   "why": "Kling Team report arguing video quality depends on multimodal understanding across the data, conditioning, alignment and evaluation chain; submitted 3 November. Affiliation is not stated on the abstract page.",
   "article": "https://aiworldwire.org/articles/schools-rein-in-ai-australia-limits-google-ties-search/"
  }
 ]
}