attribution-methods

Tag

Cards List
#attribution-methods

@noahdgoodman: how should we find representations in neural nets? how about the way we find everything else these days — gradient dece…

X AI KOLs Following ↗ · 5d ago Cached

A new paper introduces Matryoshka Attribution, a method using gradient descent to identify responsible parts of neural networks, achieving top performance on the Mechanistic Interpretability Benchmark.

0 favorites 0 likes
#attribution-methods

Measuring Explainer Stability via Attribution Separability

arXiv cs.LG ↗ · 2026-08-05 Cached

The paper introduces a distribution-based framework to measure the stability of attribution methods (explainers) by quantifying the separability of feature rankings and identifying the maximum top-k ranking that remains reliable across stochastic runs.

0 favorites 0 likes
#attribution-methods

The Failures of Marginal Influence-Based Attribution Methods for Global Time Series Explanations

arXiv cs.LG ↗ · 2026-07-21 Cached

This paper proves that existing marginal influence-based attribution methods fundamentally fail to capture the conditional dependency structure of time series models, and proposes DAG-faithfulness as a new criterion for faithful explanations.

0 favorites 0 likes
#attribution-methods

GRALIS: A Unified Canonical Framework for Linear Attribution Methods via Riesz Representation

arXiv cs.LG ↗ · 2026-05-08 Cached

This arXiv preprint introduces GRALIS, a unified mathematical framework using Riesz Representation Theory to formalize and compare linear attribution methods like SHAP, LIME, and Integrated Gradients.

0 favorites 0 likes
#attribution-methods

TPA: Next Token Probability Attribution for Detecting Hallucinations in RAG

arXiv cs.CL ↗ · 2026-04-20 Cached

TPA proposes a novel method for detecting hallucinations in RAG systems by attributing next-token probabilities to seven distinct sources (Query, RAG Context, Past Token, Self Token, FFN, Final LayerNorm, Initial Embedding) and aggregating by Part-of-Speech tags. The approach achieves state-of-the-art performance across five LLMs including Llama2, Llama3, Mistral, and Qwen.

0 favorites 0 likes
← Back to home

Submit Feedback