Could someone suggest the best/must read research papers on evaluating each item in this mini course - Hallucinations, Data Leakage, Refusals and prompt injections and Passive/active monitoring?
Hallucinations -
Hallucinations-LLM.pdf (1.9 MB)
Prompt injection -
Prompt injection-LLM.pdf (937.2 KB)
About passive and active monitoring, are you looking for LLM which are monitored post production for their prompt behaviour?? or something else in particular because monitoring llm related research papers are many. I couldn’t find particularly categorisation based on passivity or actively looking at how llm behaviour related research paper. Probably if dig deeper one can find.
regards
DP
1 Like