Sundararajan et al. (2017) proposed Integrated Gradients, an axiomatic attribution method that assigns a contribution score to each input feature for token-level importance analysis in neural NLP and LLM outputs.

Authors

• Person: Not available Organization: arXiv
  A Survey on the Theory and Mechanism of Large Language Models

Sources

• A Survey on the Theory and Mechanism of Large Language Models arxiv.org arXiv via serper

Referenced by nodes (1)

• LLM outputs concept