EviScope: Paired Counterfactual Evidence Diagnostics for Faithful and Efficient Grounded Language Models
Read the original on arXiv Computation and Language →EviScope is a new paired counterfactual benchmark that evaluates grounded language models by fixing the question while manipulating evidence—adding, removing, distracting, or contradicting it. The v1.1 dataset includes 40 four‑condition quartets with repaired counterfactual claims and span‑level support labels for automated assessment. Experiments on Qwen2.5‑7B, Llama 3.1 8B, and Gemini 3.5 Flash show that paired metrics reveal grounding behaviors hidden by simple answer accuracy, such as unsupported answers, conflict blindness, and incorrect non‑answer actions.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.