[KongchangAI]
ModelUUE框架

UUE

A lightweight unlearning framework for LLMs that decouples forgetting and utility on adapters through orthogonal null-space projection and convex untargeted optimization.

Timeline (last 90 days)

Nov 16

The UUE framework reformulates unlearning as a controllable model-editing problem, achieving decoupling of forgetting and utility on lightweight adapters through orthogonal null-space projection and convex no-target optimization.

Unverified50%
Nov 16

UUE introduces a convex no-target objective function that destroys undesirable knowledge by pushing the forget-set embeddings toward their antipodes, and this objective function has a closed-form analytical solution.

Unverified50%
Nov 16

UUE is implemented through lightweight, pluggable adapters and requires no modification to the frozen backbone model at deployment.

Unverified50%
Nov 16

Experiments on the TOFU and WMDP benchmarks show that UUE and UUE-L significantly outperform existing methods in unlearning effectiveness.

Unverified50%
Nov 16

The paper currently reports only benchmark results; robustness and adversarial unlearning effects in actual deployment remain to be verified.

Unverified50%

All Facts (5)

Source Articles