Topic
diagnostic
JustDiag! Diagnostic Justification Engine Enhances Accountability in Root Cause Analysis
JustDiag! is a diagnostic justification engine for root cause analysis that maintains an explicit process state over evidence, findings, competing hypotheses, conflicts, and next checks. Evaluated on 66 real-world incidents, it achieved stronger outcome and process scores compared to a control without diagnostic justification, though with slightly lower terminal completion due to calibrated non-closure.
New EEG Benchmark Promises Standardized Evaluation of Foundation Models
A new benchmark called EEG-FM-Bench aims to standardize evaluation of electroencephalography foundation models (EEG-FMs). It integrates 14 datasets across 10 paradigms and provides tools for gradient and representation analysis. Early experiments reveal critical insights about multi-task learning, pre-training efficiency, and model scaling.
New Diagnostic for Language-Driven Bandits Determines When Lightweight Models Beat LLMs
A new paper proposes LLMP-UCB, a bandit algorithm that uses repeated LLM inference for uncertainty estimates, but finds that lightweight numerical bandits on text embeddings often match or exceed LLM accuracy at lower cost. The authors also introduce a geometric diagnostic to guide when to use LLMs versus simpler models, offering a cost-performance tradeoff framework for AI decision systems.