Unverified50% confidenceBenchmarkExact time
Training SafeTulu on the three model families Qwen, Llama, and Olmo reduces the GCG attack success rate by up to 22%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/28/2026
First Seen
Valid until: 1/26/2027
Sources
Related Entities
Related Claims
UnverifiedOn OpenClaw recruitment tasks across five models, this harness-layer guard reduced the violation rate from 65.28% to 6.67%.57% similarUnverifiedUnder held-out interface perturbations, the accuracy of the post-trained Agent drops by nearly 25%.56% similarUnverifiedThe SGUID study points out that in on-policy distillation, fewer than 25% of the retrieved skills can provide effective learning signals.56% similarUnverifiedAcross the Qwen3 model scale series, APSD reaches performance comparable to SDPO with an average of 2.6 times fewer training steps.55% similarUnverifiedGGUF quantization can reduce a model's memory footprint to approximately 25% of its original precision.55% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1105871API
curl https://kongchang.com/api/v1/knowledge/claims/1105871MCP
get_claim(id=1105871)