Unverified50% confidenceBenchmarkExact time
On logic game benchmarks, PIVOT raises the average avg@8 from 19.5% for the best baseline to 35.6%.
1
Sources
50%
Confidence
Long-term
Relevance
10/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOn the avg@16 metric for math tasks, BranchOPD improves over Vanilla OPD by 18.55 percentage points.64% similarUnverifiedAt 16× compression, AdaMem's substring match improves by up to 3.2 percentage points (5.5%) over the uniform allocation baseline, with an average relative gain of 3.4%.63% similarUnverifiedQwen3.8-27B achieves a pass@20 of 89.35% and an all-20 of 7.50% in ThinkingBox.63% similarUnverifiedOne meta-agent variant achieved the highest Avg@3 reward on five of the six benchmarks.63% similarUnverifiedIn EvoArena testing, GPT-5.5, which achieved the highest Step Level score, had an accuracy of just over 50%, while Gemini-3.1 Pro and GLM-5.1, which performed best at the Chain Level, reached only approximately 28%–29%.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1115278API
curl https://kongchang.com/api/v1/knowledge/claims/1115278MCP
get_claim(id=1115278)