Verified75% confidenceOpinionExact time
Zero retention does not equal zero learning—the fact that data is not stored does not mean that patterns and information extracted from it will not influence the model in some form
3
Sources
75%
Confidence
Long-term
Relevance
9/17/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAperture offers zero data retention, meaning request and response data are not stored.72% similarUnverifiedMultiplying the unlearning loss of the forget set by 0.0 causes the forget items to produce no gradients, meaning the model performs no real unlearning updates.71% similarUnverifiedFor users who can fit the entire model into VRAM, this caching optimization is of limited significance, as the bottleneck is not in data transfer.68% similarUnverifiedLossless compression conclusions reported on contaminated benchmarks are meaningless, because the model can directly retrieve answers from parametric memory rather than relying on attention to retrieve from context.67% similarUnverifiedThe charts and data tables shown in the article are all from the base model with no LoRA adapters attached, i.e., purely zero-shot results.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/928686API
curl https://kongchang.com/api/v1/knowledge/claims/928686MCP
get_claim(id=928686)