Test AI On Your Terms, Not Theirs
The New York Times reports that AI is getting more powerful, but its hallucinations are getting worse. It turns out that non-reasoning models are improving, ...
Recent Posts
Vectara Hallucination Leaderboard
Hallucination leaderboard for top 25 LLMs providing factually consistent answers based on short text RAG extraction.
Organizations pursuing artificial intelligence face a stark reality. WSJ
While AI experimentation abounds, achieving enterprise-wide results demands methodical implementation and systematic scaling. Despite 78% of companies using ...
TWIX: Reconstructing Structured Data from Templatized Documents. UC Berkeley EPIC Lab
Extracting good data out of complex PDFs is a fundamental challenge and will take multiple approaches. This UC Berkeley team turned the problem upside down, ...