arXiv cs.CL
· Papers
Task Competence Is Not Instruction Following: Evaluating Instruction-Conflicting Behavior in Small Language Models
arXiv:2607.19608v1 Announce Type: new Abstract: Instruction tuning is meant to make language models follow user requests, yet it is unclear whether small models comply when an instruction conflicts with their usual task behavior. We study this across three tasks - multiple-choice question answering (MCQA), sentiment cl