Unverified50% confidenceBenchmarkExact time
On the Pokemon dataset, with 30 generations, SpectralCanary achieves a model-level accuracy (fixed threshold) of 94.8%.
1
Sources
50%
Confidence
Long-term
Relevance
10/23/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOn the Pokemon dataset, across three real data transformations and two architecture families, SpectralCanary achieves the highest average TPR@1% FPR of 91.1% per generation.70% similarUnverifiedAccording to benchmark data from Hugging Face, the IQ3_S tier scores an average of 93.26, close to the full-precision base model's score of 93.12.67% similarUnverifiedThe archived score for the Faria detector task is a balanced accuracy of 29/30—that is, 29/30 injections detected and 0/30 false positives—and back-calculating through the scoring formula also yields 29 correct clean instances.63% similarUnverified在GPT-4发布初期，主流模型在GPQA上的准确率约为35-40%，人类领域专家基准线约为65%，随机猜测为25%63% similarUnverifiedIn agentgateway's testing, qwen3.8 (27.3B, Q4_K_M quantization, 262K context) had a valid answer rate of 90.0% (54/60), an accuracy rate among answered questions of 98.1% (53/54), and an overall accuracy rate of 88.3% (53/60).63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1088302API
curl https://kongchang.com/api/v1/knowledge/claims/1088302MCP
get_claim(id=1088302)