Unverified50% confidenceBenchmarkExact time
In public health benchmarks, an ordinary AI model scored higher than Tabibu in the developer's own re-run, 0.522 versus 0.426 (non-official leaderboard data), and the developer believes the reason is that Tabibu's safety layer lowered its score in the benchmark.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/16/2026
First Seen
Valid until: 10/30/2026
Sources
Related Entities
Related Claims
Unverified在某些测试条件下，主流AI编程工具推荐不存在包名的概率可达5%~20%73% similarUnverifiedLevel 2 (role + background prompt) scored lower on the QA test than Level 1 (single-line instruction), with an average score of 22.2 vs 24.7, contrary to the experimenter's hypothesis that each level would surpass the previous one.72% similarUnverified在小样本LoRA训练中，rank值设置存在偏差-方差权衡，实践中rank=4至32是常见的安全区间71% similarUnverifiedThe gap between selection and generation exceeds 50 percentage points, revealing that reliability is far lower than benchmarks suggest when users need the model to directly write out kinship terms.70% similarUnverifiedIn real-environment A/B testing, the post-trained checkpoint reduced the tool-call failure rate by 21.2% relatively compared to the earlier version.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1043739API
curl https://kongchang.com/api/v1/knowledge/claims/1043739MCP
get_claim(id=1043739)