Hard2Verify

A new reference-free framework for auditing LLM-generated reasoning traces, using NLI-based hypergraph analysis and backward AND-OR search, outperforms direct LLM-as-judge baselines in both deductive …