UNBIND: UNlearning By INference-time Directional Steering for Code LLMs

תקציר מקורי באנגליתarXiv:2609.35913v1 Announce Type: cross Abstract: Code large language models acquire programming capabilities from large code corpora, but can also memorize implementations that later require removal. Code unlearning is needed to control their continued reproduction when copyright or security concerns arise. However, targeted and retained code share computational patterns, creating a tension between forgetting specific implementations and preserving general programming ability. We propose \textbf{UNBIND}, a code unlearning framework that separately considers which hidden states correspond to the target code and how to suppress its reproduction. By constructing separate directions for these objectives, UNBIND achieves selective unlearning at inference time while keeping model weights fixed.