Expired60% confidenceBenchmarkExact time
The fine-tuned YOLOv8s achieved a mAP@50 of 99.5% on the shahed category.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
7/12/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/10/2026(expired)
Sources
Multi-Sensor Kalman Fusion in Practice: A Complete Breakdown of a Drone Tracking System
redditr/learnmachinelearning7/11/2026
Related Claims
UnverifiedSWE-2 achieved a score of 50.0% on the FrontierCode 1.1 Main benchmark.60% similarUnverifiedHaiku 5.5 outperforms GPT-6 Luna by approximately 40% on Computer Use tasks.57% similarUnverifiedOn the Berkeley Function Calling Leaderboard v4, Jev scored 99.2% and CLM-8B scored 95.2%.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/490370API
curl https://kongchang.com/api/v1/knowledge/claims/490370MCP
get_claim(id=490370)