Unverified50% confidenceBenchmarkExact time
After incorporating EvoMem, all baseline agents showed broad improvement, with Chain Level gains notably larger than Step Level gains: in the Terminal scenario, Chain Level improved by approximately 6 percentage points (Step Level only 2.4%), in the SWE scenario by 2 percentage points, and in the Persona-Mem scenario by 3.2 percentage points.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/16/2026
First Seen
Valid until: 12/15/2026
Sources
Related Entities
Related Claims
UnverifiedExperiments by the Long-Term team showed that using the same model with only adjustments to rule constraints and the memory system raised the score from 52.8% to 66.5%, jumping the ranking from Top 30 to Top 5.65% similarUnverifiedScale AI的研究团队发现，部分模型在被污染的基准上的表现比在新编写的同等难度题目上高出10-15个百分点61% similarUnverifiedIn enterprise knowledge base scenarios, using semantic chunking strategies typically yields recall accuracy more than 15% higher than fixed-length chunking, and introducing a Reranker re-ranking model can further improve accuracy by 10–30%59% similarUnverifiedCompared to extra high, the Max level consumes up to twice as many tokens, but delivers only a 4% to 10% performance improvement across various benchmarks.59% similarUnverifiedTop AI models' scores on standard benchmarks see significant leaps every 6–9 months.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/924247API
curl https://kongchang.com/api/v1/knowledge/claims/924247MCP
get_claim(id=924247)