1. BenchmarksScore79

    RATIO: A Benchmark for Retrieval Across Typed Ideation Operations in Scientific Literature

    Researchers have introduced RATIO, a new large-scale benchmark designed to evaluate retrieval systems for scientific literature. RATIO focuses on "ideation moves" such as addressing problems, broadening concepts, and specifying details, aiming to improve AI's ability to draw inspiration from research papers.

    Source: RATIO: A Benchmark for Retrieval Across Typed Ideation Operations in Scientific Literature (arxiv.org/abs/2608.27394v1) Full analysis
  2. Research PapersScore79

    Visual Retrieval Heads: Uncovering How VLMs Ground Text to Images

    Researchers have identified Visual Retrieval Heads (VRHs), a small subset of attention heads in Vision-Language Models (VLMs) responsible for grounding text descriptions to specific image regions. These VRHs are crucial for accurate visual localization, with their masking significantly degrading performance.

    This report is based on research published on arXiv: 'Retrieval Heads Meet Vision: Uncovering How VLMs Locate and Extract Visual Information' by Chanho Park, Daehyeon Choi, Jihyun Lee, and Minhyuk Sung. Full analysis
  3. Funding/BusinessScore76

    OpenAI and Thailand's MHESI Launch AI Startup Accelerator

    OpenAI has partnered with Thailand's Ministry of Higher Education, Science, Research and Innovation (MHESI) to launch an eight-week accelerator program. This initiative aims to support 10 startups in the health, wellness, and education sectors, guiding them in transforming AI prototypes into reliable products.

    Source: OpenAI Full analysis