Metraipolis › archive › dc_ea4b0ee452ff

Unraveling the Real Working Mechanism and Inherent Flaws of GAE: A Method for Interpreting Transformer Processes from an Economic Perspective — arXiv:2609.07213

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.07213
License (as archived): CC0-1.0
Archived: 2026-10-16T15:12:43.183Z
SHA-256 of the archived text: b099aa255588ae07d689c5273bd7303d43acce580502f1a2e0bf8aaa9f6e7871
Stable URL of this page: https://metraipolis.com/documento/dc_ea4b0ee452ff

How to cite

Unraveling the Real Working Mechanism and Inherent Flaws of GAE: A Method for Interpreting Transformer Processes from an Economic Perspective — arXiv:2609.07213 — arXiv (en), https://arxiv.org/abs/2609.07213 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ea4b0ee452ff (Metraipolis archive dc_ea4b0ee452ff, sha256 b099aa255588ae07d689c5273bd7303d43acce580502f1a2e0bf8aaa9f6e7871)

BibTeX

@misc{dcea4b0ee452ff,
  title = {Unraveling the Real Working Mechanism and Inherent Flaws of GAE: A Method for Interpreting Transformer Processes from an Economic Perspective — arXiv:2609.07213},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.07213},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ea4b0ee452ff; sha256 b099aa255588ae07d689c5273bd7303d43acce580502f1a2e0bf8aaa9f6e7871},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.07213. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We observe a phenomenon that current algorithmic research in the field of explainable artificial intelligence primarily pursues better performance on several proxy metrics. On the one hand, these proxy metrics themselves are more or less flawed and cannot properly measure the quality of methods. On the other hand, metric-oriented research approaches often lead to the neglect of the rationality and interpretability of the methods themselves. Explainable artificial intelligence is abbreviated as XAI. The metric-driven research paradigm has resulted in a lack of interpretability of the relevant XAI methods themselves. Accordingly, there is a need for interpretability research on XAI methods, which can be playfully referred to as XXAI. This paper is one of our works on XXAI. This paper takes Generic Attention-model Explainability (GAE), a widely influential model interpretation method , or rather, XAI method that represents an important technical route, as the research object, and explores the real working mechanism and flaws of this method as well as the technical route it represents. Based on the conclusions of this study, it may be necessary to re-examine or verify GAE-related methods and their domain applications. We argue that GAE is an interpretation method that focuses on the attention process. After pointing out the working mechanism and flaws of GAE, we propose Cumulative Asset Holdings (CAH), a more reasonable Transformer interpretation method integrating both process-based and feature-based ideas from an economic zero-sum games perspective. In addition, it is worth noting that our method is applicable to models with special tokens, where existing methods may suffer from limitations. The model simplification research method and the analysis of additive operations adopted in this study may provide inspiration for other research works in XAI.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.07213 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T14:49:57.428Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).