Unverified50% confidenceBenchmarkExact time
The authors conducted a chronological holdout test on Coinbase's LOB data, and the proposed variant remained competitive on the uncorrupted BTC and altcoin test intervals.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/16/2026
First Seen
Valid until: 1/14/2027
Sources
Related Entities
Related Claims
UnverifiedMemory benchmarks such as LongMemEval and LoCoMo measure whether the answer correctly reaches the model and how many context tokens are consumed, not document retrieval itself.64% similarUnverified多种子重复实验中可使用配对t检验或Wilcoxon符号秩检验比较两组准确率分布，其中Wilcoxon检验因不依赖正态性假设更适合小样本62% similarUnverified该研究采用LOSO（留一被试交叉验证）结合Wilcoxon符号秩检验的评估方法61% similarUnverifiedThe tau-bench benchmark focuses on whether an agent reaches the expected database state and requires consistent success across multiple repeated attempts.61% similarUnverifiedThe evaluation framework for this experiment adopts an anti-cheating public evaluation approach featuring timestamp locking, non-deletable results, constant comparison against paired baselines, and explicit labeling of statistical significance and sample limitations.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1046120API
curl https://kongchang.com/api/v1/knowledge/claims/1046120MCP
get_claim(id=1046120)