Unverified50% confidenceBenchmarkExact time
In the experiments, the average cosine similarity between old facts and their replacements is 0.816, versus a corpus baseline of 0.372, with one pair reaching 1.000; after removing metadata filtering, the stale version becomes the nearest neighbor in 6 of 9 cases, with a median rank of 1.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAt the actual repetition levels of ordinary text, five models ranging from 1B to 13B parameters carry only a faint trace of what they have seen, with a rank correlation coefficient of approximately -0.08.72% similarUnverifiedDistillation training used 600 prompts (500 training / 100 validation), achieving a representation cosine similarity of 0.9959, a semantic correction cosine similarity of 0.9836, and a mixed-condition cosine similarity of 0.999958 relative to the teacher bridge at alpha=0.10.67% similarUnverifiedAblation analysis shows that fine-tuning on the segment paradigm yields a +5.29-point improvement on the test set, while replacing the first 4,500 transcript words with 2,000 retrieved words yields only a +1.55-point improvement.67% similarUnverifiedUsing only 64 paired segments, raw Recall@1 can be raised to 88.4%–90.3%, and token cosine similarity to about 0.82.66% similarUnverifiedWhen random forests were used on the main data-generating mechanism, the coverage of estimators using in-sample fitted propensity scores dropped to 0.12–0.48, severely under-covering; whereas after cross-fitting both nuisances, coverage recovered to 0.91–1.00.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1044515API
curl https://kongchang.com/api/v1/knowledge/claims/1044515MCP
get_claim(id=1044515)