TAA-k finds query-adaptive retrieval cutoffs by first using knee detection to isolate a candidate window around the relevance-to-noise transition, then applying EVT goodness-of-fit tests inside that window.
hub
Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation
20 Pith papers cite this work, alongside 1,564 external citations. Polarity classification is still indexing.
20
Pith papers citing it
1,564
external citations · Pith
abstract
Commonly used evaluation measures including Recall, Precision, F-Measure and Rand Accuracy are biased and should not be used without clear understanding of the biases, and corresponding identification of chance or base case levels of the statistic. Using these measures a system that performs worse in the objective sense of Informedness, can appear to perform better under any of these commonly used measures. We discuss several concepts and measures that reflect the probability that prediction is informed versus chance. Informedness and introduce Markedness as a dual measure for the probability that prediction is marked versus chance. Finally we demonstrate elegant connections between the concepts of Informedness, Markedness, Correlation and Significance as well as their intuitive relationships with Recall and Precision, and outline the extension from the dichotomous case to the general multi-class case.
hub tools
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
A matched-pair protocol and Accurate Differentiation Rate metric reveal that conventional LLM accuracy on SAT problems is often inflated by over-predicting satisfiability, while cross-representation agreement exceeds 80 percent for most models.
Derives simultaneous high-probability upper bounds on realized FDP for conformal p-values that hold for arbitrary post-hoc thresholds via an envelope on the null empirical distribution function.
A Tabu-based algorithm learns time-ordered causal graphs from time series by optimizing per-edge lags with a decomposable BIC score and explicit lag penalty.
HyPE detects harmful prompts as outliers in hyperbolic space and HyPS sanitizes them using explainable attribution, outperforming prior defenses in accuracy and robustness across datasets and adversarial scenarios.
ML4AVD research remains locked into binary function-level classification of C/C++ vulnerabilities because twelve pain points in the pipeline reinforce each other through feedback loops.
Evaluator-specific demonstrations with retrospective reasoning improve LLM simulation of individual preference judges by up to 9.9 points over a non-personalized base judge, while interface telemetry often degrades accuracy.
AMAR uses a transformer with learnable query embeddings for set-based prediction of concurrent activities from composite Wi-Fi CSI, combined with edge feature extraction and vector quantization for bandwidth-efficient deployment.
HFS-TriNet applies heuristic frame selection and a three-branch network (ResNet50, SAM-based with temporal attention, WTCR) to classify prostate cancer from TRUS videos.
TagCC anchors statistical tabular representations to LLM-derived textual semantic concepts via contrastive learning jointly optimized with a clustering objective, outperforming prior methods on benchmarks.
Introduces the first structured pulmonary knowledge graph LungKG and uses it to train Lung-R1, which reaches SOTA on EMR-based pulmonary diagnosis tasks.
Multi-query vector retrieval via anomalous pattern detection on shared dimensions improves retrieval as query-set size grows, especially from 1 to 8 queries.
Face-swapped deepfakes exhibit measurable higher-order temporal irregularities in facial dynamics that are most pronounced during emotional expressions, enabling above-chance ML detection and partial alignment with human perceptual judgments.
LLM-scored offer predicates aggregated by a Logic Tensor Network classify procurement documents about as accurately as BERT or LLM baselines while exposing auditable predicate and rule truth values.
MAC framework selects Pareto-optimal LLM agents and masks low cross-consistency outputs for adaptive collaboration in medical decision-making.
End-Net, a multiscale inception-based CNN, reaches 0.9761 accuracy on a balanced multi-class MRI dataset of Alzheimer, tumors, MS and controls and is deployed as a public web service.
Anatomy-aware descriptors from RV, myocardium and LV outperform model complexity in low-label 5-class cardiac pathology prediction on the ACDC MRI dataset.
A two-stage MLP detects urological events in vesical pressure signals with 84% accuracy for voiding versus non-voiding and 90% for abdominal versus detrusor overactivity on external validation data.
CPEMH is a new agentic framework that orchestrates AI agents to design, evaluate, and select prompts for stable and traceable behavior in foundation models applied to mental health screening from transcripts.
A multi-view evidential framework combines semantic and reasoning information to improve accuracy and provide trustworthy uncertainty estimates for mental health prediction on text data.
citing papers explorer
-
Tail-Aware Adaptive-k: Query-Adaptive Context Selection for Retrieval-Augmented Generation
TAA-k finds query-adaptive retrieval cutoffs by first using knee detection to isolate a candidate window around the relevance-to-noise transition, then applying EVT goodness-of-fit tests inside that window.
-
Satisfiability Solving with LLMs: A Matched-Pair Evaluation of Reasoning Capability
A matched-pair protocol and Accurate Differentiation Rate metric reveal that conventional LLM accuracy on SAT problems is often inflated by over-predicting satisfiability, while cross-representation agreement exceeds 80 percent for most models.
-
Everywhere Valid Bounds on False Discovery Proportions in Conformal Inference
Derives simultaneous high-probability upper bounds on realized FDP for conformal p-values that hold for arbitrary post-hoc thresholds via an envelope on the null empirical distribution function.
-
Time series causal discovery with variable lags
A Tabu-based algorithm learns time-ordered causal graphs from time series by optimizing per-edge lags with a decomposable BIC score and explicit lag penalty.
-
Harnessing Hyperbolic Geometry for Harmful Prompt Detection and Sanitization
HyPE detects harmful prompts as outliers in hyperbolic space and HyPS sanitizes them using explainable attribution, outperforming prior defenses in accuracy and robustness across datasets and adversarial scenarios.
-
Direction for Detection: A Survey of Automated Vulnerability Detection and all of its Pain Points
ML4AVD research remains locked into binary function-level classification of C/C++ vulnerabilities because twelve pain points in the pipeline reinforce each other through feedback loops.
-
PERSONAJUDGE: Simulating Individual Human Preference Judgments with Evaluator-Specific Demonstration Data
Evaluator-specific demonstrations with retrospective reasoning improve LLM simulation of individual preference judges by up to 9.9 points over a non-personalized base judge, while interface telemetry often degrades accuracy.
-
AMAR: Lightweight Attention-Based Multi-User Activity Recognition from Wi-Fi CSI
AMAR uses a transformer with learnable query embeddings for set-based prediction of concurrent activities from composite Wi-Fi CSI, combined with edge feature extraction and vector quantization for bandwidth-efficient deployment.
-
HFS-TriNet: A Three-Branch Collaborative Feature Learning Network for Prostate Cancer Classification from TRUS Videos
HFS-TriNet applies heuristic frame selection and a three-branch network (ResNet50, SAM-based with temporal attention, WTCR) to classify prostate cancer from TRUS videos.
-
Beyond Statistical Co-occurrence: Unlocking Intrinsic Semantics for Tabular Data Clustering
TagCC anchors statistical tabular representations to LLM-derived textual semantic concepts via contrastive learning jointly optimized with a clustering objective, outperforming prior methods on benchmarks.
-
Lung-R1: A Knowledge Graph-Guided LLM for Pulmonary Diagnostic Reasoning
Introduces the first structured pulmonary knowledge graph LungKG and uses it to train Lung-R1, which reaches SOTA on EMR-based pulmonary diagnosis tasks.
-
Retrieval with Multiple Query Vectors through Anomalous Pattern Detection
Multi-query vector retrieval via anomalous pattern detection on shared dimensions improves retrieval as query-set size grows, especially from 1 to 8 queries.
-
Interpretable facial dynamics as behavioral and perceptual traces of deepfakes
Face-swapped deepfakes exhibit measurable higher-order temporal irregularities in facial dynamics that are most pronounced during emotional expressions, enabling above-chance ML detection and partial alignment with human perceptual judgments.
-
From Large Language Model Predicates to Logic Tensor Networks: Neurosymbolic Offer Validation in Regulated Procurement
LLM-scored offer predicates aggregated by a Logic Tensor Network classify procurement documents about as accurately as BERT or LLM baselines while exposing auditable predicate and rule truth values.
-
MAC: Masked Agent Collaboration Boosts Large Language Model Medical Decision-Making
MAC framework selects Pareto-optimal LLM agents and masks low cross-consistency outputs for adaptive collaboration in medical decision-making.
-
A Deep Multiscale Neural Network for Accurate Neurological Disorder Detection from MRI Scans and Real-Time Web Deployment
End-Net, a multiscale inception-based CNN, reaches 0.9761 accuracy on a balanced multi-class MRI dataset of Alzheimer, tumors, MS and controls and is deployed as a public web service.
-
Which Anatomy Matters Under Limited Labels? A Data-Efficient Anatomy-Aware Benchmark for Cardiac Pathology Prediction
Anatomy-aware descriptors from RV, myocardium and LV outperform model complexity in low-label 5-class cardiac pathology prediction on the ACDC MRI dataset.
-
Automated Detection of Urological Events in Bladder Pressure Signals with a Two-Stage Machine Learning Framework Validated on External Datasets
A two-stage MLP detects urological events in vesical pressure signals with 84% accuracy for voiding versus non-voiding and 90% for abdominal versus detrusor overactivity on external validation data.
-
CPEMH: An Agentic Framework for Prompt-Driven Behavior Evaluation and Assurance in Foundation-Model Systems for Mental Health Screening
CPEMH is a new agentic framework that orchestrates AI agents to design, evaluate, and select prompts for stable and traceable behavior in foundation models applied to mental health screening from transcripts.
-
Beyond Semantics: An Evidential Reasoning-Aware Multi-View Learning Framework for Trustworthy Mental Health Prediction
A multi-view evidential framework combines semantic and reasoning information to improve accuracy and provide trustworthy uncertainty estimates for mental health prediction on text data.