Unverified50% confidenceFactExact time
Increasing the number of online samples per prompt from 1 to 7 changes the LC win rate by at most 2.1 points; sweeping the margin from 1 to 5 changes the LC win rate by at most 2.4 points in the independent pipeline.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/4/2026
First Seen
Valid until: 2/2/2027
Sources
Related Entities
Related Claims
UnverifiedAblation 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.70% similarUnverifiedIn the specific experimental case, the overlapping-anchor scheme (n=912) showed the model at 55.2% versus the majority class at 51.1%, a net advantage of +4.1pp, about 2.4 standard errors; the non-overlapping-anchor scheme (n=163) showed 59.5% versus 54.0%, a net advantage of +5.5pp, about 1.4 standard errors.65% similarUnverifiedThe interaction effect between prompt breadth and response refresh reached 4.07 percentage points, with a 95% confidence interval of [2.00, 6.28].65% similarUnverifiedIn code review tests, Opus 5.5 improved recall on more difficult pull requests from a baseline of 54% to nearly 77%, but the number of comments increased by approximately 20%.65% similarUnverifiedA slot's score is computed as the sum of all attendees' obtainable scores as a percentage of the maximum score (number of people × 1), and slots are sorted from highest to lowest score.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1134215API
curl https://kongchang.com/api/v1/knowledge/claims/1134215MCP
get_claim(id=1134215)