9th January 2024 - Link Blog
WikiChat: Stopping the Hallucination of Large Language Model Chatbots by Few-Shot Grounding on Wikipedia. This paper describes a really interesting LLM system that runs Retrieval Augmented Generation against Wikipedia to help answer questions, but includes a second step where facts in the answer are fact-checked against Wikipedia again before returning an answer to the user. They claim “97.3% factual accuracy of its claims in simulated conversation” on a GPT-4 backed version, and also see good results when backed by LLaMA 7B.
The implementation is mainly through prompt engineering, and detailed examples of the prompts they used are included at the end of the paper.
Recent articles
- My fireside chat about agentic engineering at the Pragmatic Summit - 14th March 2026
- Perhaps not Boring Technology after all - 9th March 2026
- Can coding agents relicense open source through a “clean room” implementation of code? - 5th March 2026