Needle removes LLM backdoors with one weight-orthogonalisation edit, cutting code-injection attack success to zero while keeping capability and safety loss lowest
The work proposes Needle, a training-free backdoor removal method: given an identified trigger, it estimates a backdoor direction and a refusal subspace from activation vectors, then applies sequential weight orthogonalisation, achieving the lowest mean attack success rate among the evaluated defences across Gemma-3-4B-IT, Qwen3-4B-Instruct-2507 and Gemma-3-12B-IT, fully removing a code-injection attack while producing the lowest KL divergence and minimal capability and safety change.