Unverified50% confidenceFactExact time
The SAO team selected three high-difficulty benchmarks—SWE-Bench Verified, BeyondAIME, and IMOAnswerBench—for validation
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/12/2026
First Seen
Valid until: 12/11/2026
Sources
Related Entities
Related Claims
UnverifiedSWE-Bench推出了SWE-Bench Verified和SWE-Bench Multimodal变体，前者人工筛选确保可解性，后者加入截图等多模态上下文75% similarVerifiedSWE-bench Verified has been manually reviewed to ensure the quality and solvability of each instance70% similarUnverifiedSWE-bench Verified is one of the authoritative industry leaderboards for AI programming ability, with human annotators filtering out samples with ambiguous descriptions or unstable tests.68% similarUnverifiedThe official release includes benchmark evaluation data for Terminal Bench 2.1, SWE Bench Verified/Pro, and CLAW Eval, among others.66% similarUnverifiedAgent evaluation benchmark suites include GAIA, AgentBench, and SWE-bench.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/910348API
curl https://kongchang.com/api/v1/knowledge/claims/910348MCP
get_claim(id=910348)