Value 还是 Attention：图注意力网络 LRP 可解释性方法研究

View PDF HTML (experimental)

Abstract:Graph Attention Networks (GATs) are increasingly used in domains where scientists need to read a model's reasoning, from medicine to materials science, yet the routing they learn remains opaque to most explainers. Layer-wise Relevance Propagation (LRP) follows the model's computation, but in an attention layer the relevance reaching a node is the sum of two mechanically different quantities: its content relevance through the value path and its routing relevance through the attention path. Existing rules either drop one or add them into a single number. We write a single product rule with free weights on the two factors that recovers current approaches as special cases. The value path measures what content a node supplied against a zero baseline; the attention path measures how the node shaped what was passed relative to its neighborhood, and its sign records whether the node's features moved the gating the way the prediction demanded. On a ground-truth classification task, we show that the same node receives negative, positive and near-zero relevance under three settings of the rule. The separated relevance maps yield a hypothesis about the model: it gains by not reading that node. We give reading rules for both paths and argue that they should be interpreted separately.

Comments:	Accepted at the XAI4Science Workshop @ NeurIPS 2026 (non-archival). 17 pages, 9 figures
Subjects:	Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Cite as:	arXiv:2611.08088 [cs.LG]
	(or arXiv:2611.08088v1 [cs.LG] for this version)
	https://doi.org/10.48550/arXiv.2611.08088 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Ken Schreiber [view email]
[v1] Thu, 5 Nov 2026 14:35:00 UTC (995 KB)