I can see how quantization or down sampling itself could be a fundamental way to address this. 1. Train normal full precision model. 2. Quantize down until pe

I can see how quantization or down sampling itself could be a fundamental way to address this. 1. Train normal full precision model. 2. Quantize down until performance is borderline and then perform the unlearning process. 3. Train/convert/upsample back to FP for subsequent tuning iterations. Seems like you can create an information bottleneck this way. The echos of the forgotten may have trouble fitting through something that narrow.