Integrated Gradients
PulseAugur coverage of Integrated Gradients — every cluster mentioning Integrated Gradients across labs, papers, and developer communities, ranked by signal.
8 day(s) with sentiment data
-
New method enhances interpretability of microbiome transformer models
Researchers have developed a new method for interpreting the decisions made by transformer models, specifically those used in microbiome analysis like BiomeGPT. This approach, called Signed Integrated Gradients, address…
-
New Spectral Integrated Gradients method improves AI feature attribution
Researchers have introduced Spectral Integrated Gradients (SIG), a novel feature attribution method designed to improve upon existing techniques like Integrated Gradients (IG). SIG addresses the limitations of IG's stan…
-
Interpretable ML predicts traffic congestion impacted by COVID-19
Researchers have developed interpretable machine learning models to predict traffic congestion in Alameda County, California, considering the unique impacts of the COVID-19 pandemic. By incorporating variables related t…
-
New framework fools AI explainability auditors by embedding evasion logic
Researchers have developed a new framework called "Crushing the Evidence" that can fool white-box explainable AI (XAI) auditors. This dual-penalty evasion technique embeds evasion logic directly into model parameters, a…
-
New research tackles LLM sycophancy with benchmarks and attribution methods
Two new research papers address the issue of sycophancy in large language models, where models tend to agree with users even when it contradicts factual information. The first paper, MedPRESS, introduces a multi-turn be…
-
AI models for ECG classification may rely on visual artifacts, not patient data
Researchers have analyzed shortcut learning and the Clever Hans effect in CNN-based ECG image classification. The study created six feature sets, including raw images, waveform-only images, and images with artificial ar…
-
New framework enables exact Aumann-Shapley attribution for GNNs
Researchers have developed APEX, a novel framework designed to provide exact Aumann-Shapley attributions for graph neural networks (GNNs). This framework utilizes a specialized GNN architecture called PolyGIN, which mai…
-
New hybrid model combines CoLES and State Space Models for user transaction analysis
Researchers have developed a novel hybrid approach for user-centric modeling of transactional event sequences, combining contrastive representation learning (CoLES) with State Space Models (SSMs). This method addresses …
-
New framework assesses AI forecasting model robustness against weather prediction errors
A new framework for evaluating the robustness of AI forecasting models in photovoltaic (PV) power generation has been developed. This framework addresses the challenge of numerical weather prediction (NWP) errors, which…
-
Deep learning models show resilience to weather forecast errors in PV power prediction
A new study evaluates the robustness of various deep learning models, including PatchTST, GRU, N-HITS, and LightGBM, when subjected to errors in numerical weather prediction (NWP) data. The research introduces a physica…
-
New IG-Lens method precisely attributes token probability across transformer layers
Researchers have developed IG-Lens, a novel method for precisely attributing the probability of a predicted token to specific layers within decoder-only transformer models. Unlike existing tools that offer approximate o…
-
New DiffIG method offers controllable AI explanations
Researchers have introduced Diffusion Integrated Gradients (DiffIG), a new method for generating explanations in artificial intelligence. DiffIG reformulates path generation as a conditional generative modeling problem,…
-
New Diffusion-based Method Enhances AI Explainability
Researchers have introduced Diffusion Integrated Gradients (DiffIG), a new method for generating attribution paths in explainable AI. Unlike existing approaches that use fixed or hand-crafted paths, DiffIG treats path g…
-
Protein language models fail to recover allergen epitopes, study finds
A new study has found that residue-level attributions in protein language models do not accurately recover allergen epitopes, despite the models' robustness in protein-level allergenicity prediction. Researchers develop…
-
Protein language models' allergen explanations lack biological grounding
A new study published on arXiv questions the biological relevance of explanations provided by protein language models used in allergenicity classification. While models like ESM-2 and DeepPlantAllergy demonstrate strong…
-
New explainability method analyzes sociopsychological text markers
Researchers have applied the integrated gradient (IG) method to analyze sociopsychological semantic markers in text, moving beyond simple sentiment analysis. This technique reveals which specific words contribute to cla…
-
New Weighted Integrated Gradients method enhances AI feature attribution reliability
Researchers have introduced Weighted Integrated Gradients (WG), a novel method to improve the reliability of feature attribution in explainable AI, particularly for computer vision models. Unlike existing methods like E…
-
AI explainability audit probes drug-target interaction models
A new research paper explores the explainability of black-box drug-target interaction (DTI) prediction models, specifically auditing the BridgeDPI architecture. The study employs a combination of gradient-based attribut…
-
Visual-TCAV offers new explainability for image classification models
Researchers have developed Visual-TCAV, a new framework for explaining image classification models. This method combines local saliency maps with concept-based attribution, addressing limitations of existing techniques.…
-
New method corrects attribution patching errors in language models
Researchers have developed a new method to improve the accuracy of attribution patching, a technique used to understand how different parts of a language model contribute to its behavior. The current method, a first-ord…