Cusumano
PulseAugur coverage of Cusumano — every cluster mentioning Cusumano across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
New method distinguishes telemetry data errors with 96% accuracy
Researchers have developed a new method to distinguish between process changes, instrument faults, and data corruption in streaming telemetry data. This approach uses a combination of online detectors, including a Kalma…
-
New statistical method offers quicker change-point detection in generalized linear models
Researchers have developed a new statistical method for detecting changes in generalized linear models, particularly when the parameter after a change is unknown. This method, called a window-limited CUSUM procedure, us…
-
New DI-SCUSUM method improves change detection with diffusion scores
Researchers have developed a new method called diffusion-integrated score CUSUM (DI-SCUSUM) for quickest change detection. This training-free detector uses Gaussian noise to smooth density estimates and calculate Hyväri…
-
New benchmark detects coordinated AI agent reconnaissance attacks
Researchers have developed SwarmReconGuard, a new benchmark system designed to detect Distributed Collective Reconnaissance (DCR) by agent populations. DCR involves numerous agents acting benignly individually but colle…
-
AI search reliability tests fail across OpenAI, Gemini, and Claude
A recent analysis of AI search capabilities from OpenAI, Gemini, and Claude revealed significant reliability issues. Pre-registered tests for measuring AI search accuracy, conducted from May to September, failed to conf…
-
New CURA system detects AI agent failures with certified runtime alarms
A new system called CURA has been developed to monitor computer-use agents (CUAs) and detect failures in real-time. CURA operates by analyzing telemetry data, without needing access to the agent's internal workings or r…
-
New KAISEN pipeline enhances fairness auditing for clinical AI models
Researchers have developed KAISEN, a novel five-phase audit pipeline designed to improve the reproducibility and reliability of fairness assessments in clinical risk models. The pipeline addresses subgroup stratificatio…
-
Developers can now detect silent LLM API changes with new method · 2 sources tracked
Developers can now systematically detect when hosted large language models like gpt-x or Claude Yelnick silently change their behavior. The method involves establishing a frozen 'canary suite' of prompts and recording o…
-
New AI detectors ensure model reliability without labels
Researchers have developed two novel concept drift detectors, CFPT-FM and TabAutoDrift, designed to maintain the reliability of AI models in dynamic environments without requiring labeled data post-deployment. These met…
-
AI agent monitors flawed by wall-clock calibration, study finds
A new research paper, "Bistable by Construction: Wall-Clock-Calibrated State Monitors Have No Moment-Detection Regime at Agent Cadence," published on arXiv, identifies a critical flaw in runtime monitors for autonomous …
-
New research tackles LLM and VLM hallucinations with novel detection and correction methods
Researchers are developing novel methods to combat hallucinations in large language models (LLMs) and vision-language models (VLMs). One approach, Recurrent Attention-based Uncertainty Quantification (RAUQ), uses attent…