Unverified50% confidenceFactExact time
Confidence-weighted voting over the orderings yields 69.7% accuracy, compared with 69.1% for a single base pass, at 4.1x the computation.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/9/2026
First Seen
Valid until: 2/7/2027
Sources
Related Entities
Related Claims
UnverifiedSelf-Consistency通过多次采样推理路径后投票取多数答案,可进一步提升准确率约5-10个百分点72% similarUnverifiedEvidence-driven pre-improvement validation improved precision from 58.6% to 72.3%, and reduced unsupported objectives per round from 1.93 to 0.50.71% similarUnverifiedUnder a fixed inference budget, accuracy after implementing flow matching again reached about 76%; by increasing integration steps, sampling multiple trajectories, and moving intermediate results back to earlier flow time points for refinement, the success rate was pushed past 99%.70% similarUnverifiedFor an invoice classification model with 97% accuracy, when a reviewer examines 300 invoices per day, only about 9 will be wrong, and errors are buried among correct items, causing the reviewer's rejection rate to approach zero.70% similarUnverifiedIn the candidate fact mechanism, there are 796 candidate facts, and only 7 have been confirmed more than twice and thus adopted.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1149601API
curl https://kongchang.com/api/v1/knowledge/claims/1149601MCP
get_claim(id=1149601)