RAGCharacter localizes poisoned character spans in RAG evidence via prompt-conditioned counterfactual masking and achieves the best accuracy-over-attribution trade-off across tested attacks and models.
why should i trust you?
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Accuracy and understandability can be co-optimised for feature selection in tabular-data explanations while maintaining high classification performance.
citing papers explorer
-
Needle-in-RAG: Prompt-Conditioned Character-Level Traceback of Poisoned Spans in Retrieved Evidence
RAGCharacter localizes poisoned character spans in RAG evidence via prompt-conditioned counterfactual masking and achieves the best accuracy-over-attribution trade-off across tested attacks and models.
-
Improving Explanations: Applying the Feature Understandability Scale for Cost-Sensitive Feature Selection
Accuracy and understandability can be co-optimised for feature selection in tabular-data explanations while maintaining high classification performance.