cs.AI, cs.CL, cs.LG

Inspection and Control of Self-Generated-Text Recognition Ability in Llama3-8b-Instruct

arXiv:2410.02064v3 Announce Type: cross
Abstract: It has been reported that LLMs can recognize their own writing. As this has potential implications for AI safety, yet is relatively understudied, we investigate the phenomenon, seeking to establish whe…