Unverified50% confidenceBenchmarkExact time
ABACUS programs achieve 99.67% macro-average accuracy on the three BIG-Bench Hard tasks.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/20/2026
First Seen
Valid until: 11/3/2026
Sources
Related Entities
Related Claims
UnverifiedPrecision Mode achieved an extraction accuracy rate of 94.7% on a benchmark covering approximately 9,000 complex documents.68% similarUnverifiedThe Top3 configuration achieved an accuracy of 97.148%, a weighted F1 of 97.055%, and an MCC of 0.9675, comprehensively outperforming the Full21 configuration using all 21 metrics (accuracy 95.999%, weighted F1 95.521%, MCC 0.9549).66% similarUnverifiedBullet 在 SWE-bench Verified 基准测试中取得 95.8% 的成绩，位列前三，平均每个任务耗时 119 秒66% similarUnverifiedUnsloth's 1-bit build of QWEN 3 235B claims to retain 77% accuracy, but its actual score on long tasks is only 8%.66% similarUnverifiedThe complete system achieved an accuracy of 77.3% across six simulator-derived report fields.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073918API
curl https://kongchang.com/api/v1/knowledge/claims/1073918MCP
get_claim(id=1073918)