AI RESEARCH
Case-Grounded Evidence Verification: A Framework for Constructing Evidence-Sensitive Supervision
arXiv CS.AI
•
ArXi:2604.09537v1 Announce Type: cross Evidence-grounded reasoning requires than attaching retrieved text to a prediction: a model should make decisions that depend on whether the provided evidence s the target claim. In practice, this often fails because supervision is weak, evidence is only loosely tied to the claim, and evaluation does not test evidence dependence directly. We