Unverified50% confidenceFactExact time
In the 2025 pilot phase, 9% of summaries were flagged as unsuitable for use, and in subsequent larger-scale sampling tests that proportion rose to 15%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/14/2026
First Seen
Valid until: 1/12/2027
Sources
Related Entities
Related Claims
UnverifiedThe 7 models that passed the surrogate test had a change rate of 15.9%, while the 6 models that did not pass had a rate as high as 25.8%, indicating that better-performing models are less susceptible to interference but still more so than the effect of prompt modification.69% similarUnverified该测试样本规模相对有限（5个模型、13个代码库），结论有待更大规模验证66% similarUnverifiedIn long-context testing, normal mode ultimately completed up to 114,000 tokens, MTP2 completed up to 81,000 tokens, and failed to complete at 98,000 tokens.66% similarUnverified19.4% (6,251 articles) of duplicate document contamination exists in the naive test set split.66% similarUnverifiedPerplexity reported in its research benchmarks that the new model achieved a 9% improvement in accuracy compared to the old model at half the cost.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1031013API
curl https://kongchang.com/api/v1/knowledge/claims/1031013MCP
get_claim(id=1031013)