HARDEN: Constrained Evolutionary Search for Harder, Answer-Preserving Evaluation Cases

arXiv cs.AISeptember 28, 2026en

arXiv cs.AI

AI Global Wire

arXiv:2609.30571v1 Announce Type: new Abstract: Language models are often evaluated on curated benchmarks that underrepresent the complexity of enterprise deployments. We introduce HARDEN, a constrained evolutionary search method to adapt the input of existing evaluation cases into more challenging variants while keeping their expected outputs fixed. HARDEN searches along generated domain-specific complexity axes while enforcing feasibility constraints such as preserving task semantics, realism, and execution validity. Across FinQA, PubMedQA, and ContractNLI and three Qwen3.5 model scales (35B-A3B, 122B-A10B, and 397B-A17B), HARDEN reduces task-model accuracy by 22.7% on average and by up to

This is a short summary published by AI Global Wire. The full article is owned and hosted by arXiv cs.AI — open it there to read it in full.

• Forskning
• Företag