NLP

  1. LLM factual errors are often recall failures, Google says

    LLM factual errors are often recall failures, Google says

    Google study reframes factuality as a retrieval problem Google Research says many factual mistakes in frontier large language models may come from failed recall rather than missing knowledge. The team introduced a behavioral framework called knowledge profiling to separate whether a fact is...
  2. Science One Framework tests evidence chains for AI research

    Science One Framework tests evidence chains for AI research

    Google's AI research prototype makes verifiability the main test Google Research has introduced Science One Framework, an experimental prototype for autonomous scientific research that is designed around evidence chains rather than retrospective fact checking. The project addresses a narrow but...
Top