TabMWP

anchor

Findings

• IC-1007LLMs cannot reliably self-verify or self-correct their own outputs without external tool feedback [eval]
• IC-1008The magnitude of CRITIC's improvement on mathematical program synthesis scales with Llama-2 model size [eval]
• IC-799WizardMath-70b scores lower than base Llama-2-70b on TabMWP (49.8% vs 57.5%), indicating degraded OOD generalization from rationale-based fine-tuning [eval]