Unverified50% confidenceBenchmarkExact time
With DeepSeek V4 Pro as the backbone, SwarmRSI's proportion matching SOTA on NatureBench is 21.1 percentage points higher than Claude Code with the same backbone.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/20/2026
First Seen
Valid until: 1/18/2027
Sources
SwarmRSI: How a Swarm of Research Agents Achieves Recursive Self-Improvement
bilibili熊二等兵10/18/2026
Related Entities
Related Claims
UnverifiedWith DeepSeek V4 Pro as the backbone, SwarmRSI achieves an average score of 74.2 on OpenProblems, exceeding the previous SOTA of 67.1.72% similarUnverifiedWith DeepSeek V4 Pro as the backbone, SwarmRSI achieves a medal rate of 81.8% on MLEBench Lite.70% similarUnverified在Terminal Bench跑分上，DeepSeek V4 Pro得87.9分，Claude Opus 4.8得85分66% similarUnverifiedMiniCPM5-2B's score on SWE-bench beats 4B-class competitors with twice its parameter count.63% similarUnverifiedIn a needle-in-a-haystack experiment with a 128K Token context and about 16,000 key-value pairs, changing the position of just one key-value pair produced a maximum accuracy gap of 40.2% for DeepSeek-V4-Flash-Base, 34.8% for DeepSeek-V4-Pro-Base, and still 19.1% and 14.8% respectively after post-training optimization.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073049API
curl https://kongchang.com/api/v1/knowledge/claims/1073049MCP
get_claim(id=1073049)