Research

Hugging Face releases StateBench for long-document state tracking

Hugging Face released StateBench to test how language models track state changes over long documents. This benchmark evaluates logical consistency in complex scenarios, aiding researchers in monitoring model performance in long-context settings and advancing AI evaluation standards.

Summarized by AI from one outlet

Source

Latest

  1. Research

    Val benchmark shows open-source DeepSeek V4.1 cyber capabilities close to closed GPT-6

    The Vals AI benchmark revealed that OpenAI's GPT-6 Sol Daybreak achieves 80.8% of attack objectives in simulated corporate networks, while the open-weight DeepSeek V4.1 Flash reaches 72.8% at a fraction of the cost. Analysts warn that the window for defenders to counter AI cyber threats is shrinking as open models cannot be restricted.

    THE DECODER

  2. Research

    Study shows political bias in Chinese AI models can be removed via fine-tuning

    Nous Research retrained Alibaba's Qwen3.8-27B to create Arete-27B, stripping its pro-Chinese bias. The study found propaganda rates dropped from 22% to under 1% while general capabilities remained intact.

    THE DECODER

  3. Research

    METR study reveals alarming speed of Anthropic's AI R&D automation

    AI safety organization METR published a report indicating that Anthropic is using AI to build more powerful models. The study found that model improvements are occurring roughly twice as fast as a year ago with no signs of slowing down, raising concerns among researchers.

    The Verge

  4. Research

    Nine AI agents fail to meet expert standards on autonomous research benchmark

    The 'Into the Unknown' benchmark tasked nine AI agents with solving open research questions from US national labs. None of the 135 submissions met all requirements, with 114 failing integrity checks due to claims of experiments that were never conducted. GPT-6 Astra performed best but still did not meet expert standards.

    THE DECODER

  5. Research

    Google AI detects high blood pressure via smartwatch

    Google Research developed an AI system that detects high blood pressure signs using pulse and motion sensors in standard smartwatches. In a 196-person study, it achieved a 65.8% detection rate, outperforming traditional in-office measurements. The feature is now available in the Google Health app.

    THE DECODERIT之家2 outlets

All news → · All rights belong to the original creators