Introduces an instance-level influence-based XAI method for dysarthria severity assessment that explains predictions by computing per-utterance influence scores from training samples and validates them via controlled deletion experiments.
Estimating training data influence by tracing gradient descent
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 1
dataset 1
citation-polarity summary
representative citing papers
SALT is a subspace-adaptive plug-in for GRPO that decomposes group-relative coefficients into shared and residual channels using mini-batch Gram geometry and amplifies residuals to mitigate signed cancellation in RLVR.
Target-aligned data selection via normalized endpoint loss drop on a validation-induced reference path achieves competitive performance with reduced computational overhead.
Influence scoring can use only forward passes: CountSketch-compressed outer products of the LM-head residual and final hidden state give accurate attribution and valuation from 14M to 32B parameters.
DUET is a global-to-local method that optimizes LLM training data mixtures via Bayesian optimization guided by influence-based selection and feedback from unseen evaluation tasks, with a regret bound showing convergence to the optimal mixture.
citing papers explorer
-
Towards Dys-XAI: Influence-Based Explanations for Dysarthria Severity Assessment
Introduces an instance-level influence-based XAI method for dysarthria severity assessment that explains predictions by computing per-utterance influence scores from training samples and validates them via controlled deletion experiments.
-
SALT: When More Rollouts Don't Help in Group-Based Policy Optimization and How to Make Them Matter
SALT is a subspace-adaptive plug-in for GRPO that decomposes group-relative coefficients into shared and residual channels using mini-batch Gram geometry and amplifies residuals to mitigate signed cancellation in RLVR.
-
Let the Target Select for Itself: Data Selection via Target-Aligned Paths
Target-aligned data selection via normalized endpoint loss drop on a validation-induced reference path achieves competitive performance with reduced computational overhead.
-
Sketching the Readout of Large Language Models for Scalable Data Attribution and Valuation
Influence scoring can use only forward passes: CountSketch-compressed outer products of the LM-head residual and final hidden state give accurate attribution and valuation from 14M to 32B parameters.
-
DUET: Optimizing Training Data Mixtures via Feedback from Unseen Evaluation Tasks
DUET is a global-to-local method that optimizes LLM training data mixtures via Bayesian optimization guided by influence-based selection and feedback from unseen evaluation tasks, with a regret bound showing convergence to the optimal mixture.