2026-09-27 · America/Los_Angeles · 论文 · #8

Evidence-Grounded Auditing of Identification Assumptions in Climate-Policy Causal Evaluations

Difference-in-differences (DID) studies are widely used to evaluate climate policy, but assessing the evidence supporting their identification assumptions remains challenging. We introduce ARGUS, a structured language-model pipeline that audits reported evidence against an eleven-dimension assumption-implication-evidence rubric and abstains when relevant evidence cannot be retrieved. We evaluate ARGUS using injected flaws, economics papers, and a small pilot with reconciled labels. On the 11-flaw benchmark, ARGUS detects 73% of planted flaws, compared with 18% for a keyword-based pipeline.…

热度 14.0 / 100；排名与评分保留该期记录。