Shapley Values
PulseAugur coverage of Shapley Values — every cluster mentioning Shapley Values across labs, papers, and developer communities, ranked by signal.
9 day(s) with sentiment data
-
New ConMem framework improves LLM inspection of manufacturing logs
Researchers have developed ConMem, a novel memory framework designed to enhance Large Language Model (LLM) performance in long-horizon manufacturing inspection tasks. ConMem addresses the limitations of existing systems…
-
New framework audits multimodal LLMs by identifying decision-driving modalities
Researchers have developed a new framework called Counterfactual Modality Attribution (CMA) to assess which modality, such as images or text, is primarily responsible for a multimodal large language model's (MLLM) predi…
-
New mechanism tackles false-name manipulation in ML data attribution
Researchers have introduced quotient semivalues as a novel mechanism to address false-name manipulation in machine learning data attribution. This method aims to provide more accurate valuations by clustering data and u…
-
ConMem framework improves LLM inspection accuracy with contribution-aware memory
Researchers have developed ConMem, a novel memory framework designed to enhance LLM-assisted equipment inspection by prioritizing valuable historical data. ConMem segments inspection logs into functional units and estim…
-
New credit assignment methods enhance AI search agent training · 3 sources tracked
Researchers have developed new methods for training long-horizon search agents, which are AI systems designed to perform complex, multi-step tasks. One approach, ABSeeker, uses Answer-Backtracked Credit Assignment (ABC)…
-
New LLM Auditing Methods Uncover Data Flaws and Steerability Issues
Two new research papers explore methods for auditing and understanding the behavior of large language models (LLMs). The first paper introduces a data auditing pipeline that uses influence scores to identify errors and …
-
AI framework optimizes FDM warpage detection with feature selection
Researchers have developed an Automated Data Processing (ADP) framework to optimize machine learning model and feature selection for predicting warpage in fused deposition modeling (FDM). The framework uses a reinforcem…
-
New framework attributes LLM reasoning path contributions using Shapley values
Researchers have developed a new reinforcement learning framework called Parallel Shapley to address the challenge of attributing rewards in multi-step reasoning for large language models (LLMs). This method treats each…
-
MinShap framework offers new approach to feature selection in ML
Researchers have introduced MinShap, a new framework designed to identify important and non-redundant features in machine learning models. Unlike traditional Shapley value methods that average feature contributions, Min…
-
New taxonomy unifies explainable AI feature attribution methods
A new survey paper published on arXiv introduces a mathematical taxonomy for local additive feature attribution methods, which are crucial for explainable AI. The paper organizes various methods, including Shapley, path…
-
New research tackles modality gaps and robustness in multimodal learning
Two new research papers explore methods to improve multimodal learning by addressing the challenges of modality gaps and robustness. The first paper introduces xNCE, a modification to contrastive learning that uses inte…
-
New research reveals data valuation distortions in machine learning
A new research paper titled "Validation-Induced Shapley Shifts: How Validation Structure Distorts Data Valuation" published on arXiv highlights a significant vulnerability in how machine learning data is valued. The stu…
-
New methodology measures lag relevance in time series forecasting models
Researchers have introduced a novel methodology for assessing lag relevance in machine learning models used for univariate time series forecasting. This approach leverages frameworks such as Ghost variables and Shapley …
-
OperatorSHAP offers fast, accurate Shapley value estimation for neural operators
Researchers have developed OperatorSHAP, a novel method for estimating Shapley values in neural operators. This approach addresses the computational cost and input limitations of existing explainability techniques like …
-
New Shapley-inspired k-means algorithm enhances feature weighting
Researchers have developed SHARK (Shapley Reweighted k-means), a novel feature-weighting method for clustering algorithms that avoids the need for additional hyperparameter tuning. This approach leverages Shapley values…
-
New Priority-Aware Shapley Value method enhances AI data valuation
Researchers have introduced Priority-Aware Shapley Value (PASV), a novel method for data valuation and feature attribution that addresses the limitations of traditional Shapley values. PASV incorporates precedence const…
-
New RuleSHAP framework enhances ML inference for epidemiological data
Researchers have developed a new framework called RuleSHAP to improve statistical inference for machine learning models in epidemiology. This framework integrates Bayesian regression, tree ensembles, and Shapley values …
-
New framework decodes modality contributions in audio-visual speech recognition
Researchers have developed Dr. SHAP-AV, a framework utilizing Shapley values to analyze how audio-visual speech recognition models balance acoustic and visual information. Experiments across six models and varying noise…
-
New Shapley Value method explains multimodal AI models
Researchers have developed a novel extension of Shapley Values to explain the behavior of multimodal multilingual models (MLLMs). This framework addresses the challenges of integrating text and audio data by treating th…
-
New Aumann-SHAP framework explains ML decisions via counterfactual geometry
Researchers have developed Aumann-SHAP, a new framework for explaining machine learning model decisions by analyzing counterfactual interactions. This method decomposes changes by focusing on a local hypercube between b…