Shap
PulseAugur coverage of Shap — every cluster mentioning Shap across labs, papers, and developer communities, ranked by signal.
14 day(s) with sentiment data
-
AI transparency tool uses SHAP and ELI5 for explainable decisions
Researchers have developed an interactive application to demystify complex AI models, particularly in sensitive fields like healthcare and finance where trust is paramount. The tool utilizes techniques such as XGBoost, …
-
New RuleSHAP method uncovers injected behaviors in LLMs
Researchers have developed a new method called RuleSHAP to better detect and understand injected behaviors in large language models (LLMs). This technique combines global SHAP aggregates with rule induction, significant…
-
AI framework enhances cyber risk analytics for US critical infrastructure
Researchers have developed a new framework for assessing cyber risks and model reliability in U.S. critical infrastructure. This framework utilizes machine learning classifiers like XGBoost, Random Forest, and Decision …
-
New XAI Rubric Highlights Causal AI Need for Autonomous Driving Safety
A new rubric for Explainable AI (XAI) in autonomous driving safety has been proposed, highlighting a significant gap between current XAI methods and the evidence required by safety standards. The proposed rubric, derive…
-
SleepExplain model achieves 94% accuracy in sleep stage classification
Researchers have developed a new model called SleepExplain for classifying sleep stages from EEG data. This model utilizes ensemble methods like XGBoost and Gradient Boosting, achieving high accuracy rates of up to 94.3…
-
New framework uses SHAP and LLMs to explain teaching quality scores
Researchers have developed a new framework to interpret how automated scoring models assign quality ratings to complex language performances, such as classroom transcripts. This framework combines model-agnostic Shapley…
-
New framework tests ML explanation faithfulness without ground truth
Researchers have developed a new framework using metamorphic testing to evaluate the trustworthiness of machine learning model explanations. This approach, dubbed the "Rashomon Set," assesses explanation faithfulness wi…
-
New AI framework boosts phishing detection with explainability
Researchers have developed a new framework using DistilBERT, a lightweight Transformer model, to enhance the detection of sophisticated phishing emails. This framework incorporates adversarial training techniques to imp…
-
AI models show promise for early Alzheimer's detection
Researchers are developing advanced AI models for early Alzheimer's disease detection using various data sources. One study proposes a multilingual approach using transformer models on speech data, achieving an 82% F1 s…
-
New XAI framework uses D'Hondt rule for tabular data attribution
Researchers have introduced DhondtXAI, a novel framework for explainable AI (XAI) specifically designed for tabular data. This method utilizes the D'Hondt rule, a proportional representation system, to attribute feature…
-
AI model finds energy-saving drag reduction strategies
Researchers have developed a novel method combining Multi-Agent Deep Reinforcement Learning (MARL) and eXplainable Deep Learning (XDL) to significantly reduce drag in turbulent flows. This approach utilizes SHAP (SHaple…
-
New AI framework boosts IoT intrusion detection accuracy
Researchers have developed XAI-SOH-FL, a new framework designed to improve intrusion detection in heterogeneous IoT environments. This enhanced system integrates adaptive aggregation and explainable AI to address limita…
-
New model offers interpretable anomaly detection for physiological sensors
Researchers have developed a new framework called the Distilled Explanation Model (DEM) for anomaly detection in physiological sensor data. This three-stage model aims to provide both high accuracy and interpretable exp…
-
New framework identifies partial differential equations with false discovery rate control
Researchers have developed a new data-driven framework called KO-PDE-IDENT for discovering partial differential equations (PDEs) from noisy data. This method uses knockoff filters to control the false discovery rate, ad…
-
ExplainReduce method synthesizes local AI explanations into global insights
Researchers have developed a method called ExplainReduce to generate global explanations for complex machine learning models by synthesizing numerous local explanations. This technique reduces a large set of local appro…
-
Speech analysis framework aids mental health clinical decisions
Researchers have developed a framework for analyzing speech features to aid in clinical decision-making for mental health care. This system uses perceptually grounded acoustic and linguistic characteristics, such as pro…
-
BOHM method offers zero-cost AI system attribution using routing weights
Researchers have introduced BOHM, a novel method for attributing contributions within compound AI systems that utilize hierarchical routing. Unlike traditional Shapley-based methods, BOHM leverages existing routing weig…
-
ML model struggles with visibility prediction due to data shifts
Researchers have developed a machine learning framework for predicting atmospheric visibility in six South Korean cities, addressing challenges like imbalanced data and distribution shifts. The study employed techniques…
-
Research proves feature ranking impossible under collinearity
A new research paper published on arXiv demonstrates that no feature ranking method can be simultaneously faithful, stable, and complete when features are collinear. The study proves this impossibility and quantifies it…
-
New metric quantifies AI explanation fragility in cybersecurity
This paper introduces a novel metric, the Explanability Fragility Score, to quantify instability in AI explanations within cybersecurity intrusion detection systems. The research demonstrates that multicollinearity, a s…