Unverified50% confidenceFactExact time
By updating only about 5% of the parameters, SAFT can approach or even surpass full-parameter adversarial training in robustness.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/30/2026
First Seen
Valid until: 1/28/2027
Sources
Related Entities
Related Claims
UnverifiedThe models built by the agent surpass current state-of-the-art methods in accuracy on four benchmarks, and still maintain most of their accuracy when using only about 1% of the training labels.66% similarUnverifiedThe model outputs only a risk level (such as RISK_LEVEL: 5) with no explanation; under a prompt of about 7000 tokens, Opus's P50 latency is about 1.5 seconds.65% similarUnverifiedPerturbation-augmented training can only narrow the probe performance gap to -3.7 percentage points, with limited effectiveness.65% similarUnverifiedAny study comparing fine-tuning schemes with different numbers of trainable parameters should be wary of the shared learning rate trap; otherwise, conclusions that fewer parameters yield better results may need to be revisited.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1114011API
curl https://kongchang.com/api/v1/knowledge/claims/1114011MCP
get_claim(id=1114011)