Digestible insights on the latest AI research
interesting on the first study where they discovered a simple token or configuration of token could easily corrupt the underlying LLM because the converse is also true, tokens can be “hacked” to get the LLM to think properly and not drift.
interesting on the first study where they discovered a simple token or configuration of token could easily corrupt the underlying LLM because the converse is also true, tokens can be “hacked” to get the LLM to think properly and not drift.