Unverified50% confidenceFactExact time
In HDFS anomaly detection, Decision-1's raw output probabilities correlate 0.43 with event count and only 0.14 with labels; the model overreacts to lexical cues and log volume, and misjudges missing events as healthy.
1
Sources
50%
Confidence
Long-term
Relevance
10/13/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWhen labeled data and a stable schema are available, a logistic regression classifier reached 95.1% accuracy with 50 labeled blocks and 99.7% with 8000 blocks on the HDFS anomaly detection task, surpassing the 86% of the zero-shot LLM.76% similarUnverifiedIn HDFS anomaly detection, after using structured questioning (computing lifecycle_summary and changing the question to 'is the lifecycle complete'), Decision-1 reached 86% calibrated accuracy on 1200 test blocks (threshold 0.15), with AUC reaching 0.964.75% similarUnverifiedThe NPV of definitive benign cases (0.73) still lags behind the malignant precision (0.94), indicating that the model has insufficient confidence in ruling out malignancy.71% similarUnverifiedIn experiments on the Loghub HDFS_v1 dataset, Decision-1 initially used a simple prompt and reached only 49% accuracy (AUC 0.68), lower than the 71% rule-based baseline.69% similarUnverifiedAt a significance level of ε=0.09, 63% of nodules received a definitive single-label prediction, while the remaining 37% of cases were referred to clinicians for judgment.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1018536API
curl https://kongchang.com/api/v1/knowledge/claims/1018536MCP
get_claim(id=1018536)