AI RESEARCH
DeltaLogic: Minimal Premise Edits Reveal Belief-Revision Failures in Logical Reasoning Models
arXiv CS.AI
•
ArXi:2604.02733v1 Announce Type: new Reasoning benchmarks typically evaluate whether a model derives the correct answer from a fixed premise set, but they under-measure a closely related capability that matters in dynamic environments: belief revision under minimal evidence change. We