Large language models stay accurate when 90% of a coding prompt is deleted but frequently ignore a single quantifier flip that changes the problem, so their robustness blurs harmless noise and meaning-changing edits.
On faith- fulness and factuality in abstractive sum- marization
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Small Edits, Big Consequences: Telling Good from Bad Robustness in Large Language Models
Large language models stay accurate when 90% of a coding prompt is deleted but frequently ignore a single quantifier flip that changes the problem, so their robustness blurs harmless noise and meaning-changing edits.