Unverified50% confidenceTradeoffExact time
训练时应用差分隐私噪声（DP-SGD）可有效降低记忆化程度，但会带来个性化效果的损耗
1
Sources
50%
Confidence
Long-term
Relevance
7/19/2026
First Seen
Sources
Related Claims
Unverified差分隐私训练、去重预处理和机器遗忘等缓解记忆化的技术方案均会在不同程度上牺牲模型性能78% similarUnverifiedDifferential Privacy (DP) clips training gradients and injects calibrated Gaussian noise, allowing the influence of any single training data point's presence or absence on model output to be mathematically bounded.75% similarUnverifiedMulti-head Latent Attention significantly reduces memory usage and computational overhead during inference by compressing the KV cache into a low-dimensional latent space.65% similarUnverifiedDifferential privacy was proposed by Cynthia Dwork in 2006, with the core idea of adding random noise to gradients or outputs during data statistics or model training.64% similarVerifiedRAG technology can break through the temporal limitations of model training data and significantly reduce the probability of hallucination occurrences.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/558514API
curl https://kongchang.com/api/v1/knowledge/claims/558514MCP
get_claim(id=558514)