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

arXiv: https://arxiv.org/abs/2609.30571 · HTML: https://mlsanity.com/paper/2609.30571/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 25 references that are in the map.
- First posted on arXiv 24 Sep 2026.

## Builds on (most cited first, 17 of 17)

- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [Large Language Models Encode Clinical Knowledge](https://mlsanity.com/paper/2212.13138/) (2022)
- [PubMedQA: A Dataset for Biomedical Research Question Answering](https://mlsanity.com/paper/1909.06146/) (2019)
- [Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation](https://mlsanity.com/paper/2302.09664/) (2023)
- [Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLs](https://mlsanity.com/paper/2305.03111/) (2023)
- [Adversarial Examples for Evaluating Reading Comprehension Systems](https://mlsanity.com/paper/1707.07328/) (2017)
- [FinQA: A Dataset of Numerical Reasoning over Financial Data](https://mlsanity.com/paper/2109.00122/) (2021)
- [Generating Natural Language Adversarial Examples](https://mlsanity.com/paper/1804.07998/) (2018)
- [LegalBench: A Collaboratively Built Benchmark for Measuring Legal Reasoning in Large Language Models](https://mlsanity.com/paper/2308.11462/) (2023)
- [Can Generalist Foundation Models Outcompete Special-Purpose Tuning? Case Study in Medicine](https://mlsanity.com/paper/2311.16452/) (2023)
- [AdvPrompter: Fast Adaptive Adversarial Prompting for LLMs](https://mlsanity.com/paper/2404.16873/) (2024)
- [ContractNLI: A Dataset for Document-level Natural Language Inference for Contracts](https://mlsanity.com/paper/2110.01799/) (2021)
- [Stretching Sentence-pair NLI Models to Reason over Long Documents and Clusters](https://mlsanity.com/paper/2204.07447/) (2022)
- [DARG: Dynamic Evaluation of Large Language Models via Adaptive Reasoning Graph](https://mlsanity.com/paper/2406.17271/) (2024)
- [SECA: Semantically Equivalent and Coherent Attacks for Eliciting LLM Hallucinations](https://mlsanity.com/paper/2510.04398/) (2025)
- [Zero-Shot Question Answering over Financial Documents using Large Language Models](https://mlsanity.com/paper/2311.14722/) (2023)
- [SQLMorph: Query Mutation and Fine-Grained Metrics for Text-to-SQL Evaluation](https://mlsanity.com/paper/2609.08950/) (2026)

## Abstract (arXiv)

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 49.9% relative to single-pass baselines using the same feasibility checks. These results show that evolutionary search can produce substantially harder valid evaluation cases.