Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.18824 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910234434338816 |
|---|---|
| author | Islam, Mohammed Saidul Baghbanzadeh, Negin Kohankhaki, Farnaz Cheraghi, Afshin Kore, Ali Mehdi, Shayaan Dolatabadi, Elham Afkanpour, Arash |
| author_facet | Islam, Mohammed Saidul Baghbanzadeh, Negin Kohankhaki, Farnaz Cheraghi, Afshin Kore, Ali Mehdi, Shayaan Dolatabadi, Elham Afkanpour, Arash |
| contents | Evaluation of foundation models often rely on aggregate scores from benchmarks that lack comprehensive coverage and metadata for a fine-grained evaluation. We introduce a framework for automated benchmark generation. Our framework generates evaluation problems grounded in reference material, such as textbooks, producing benchmarks with broad coverage, rich metadata, and robustness to contamination. The pipeline employs a multi-agent architecture for problem generation and a solution-graph-driven strategy that significantly improves the reliability of ground truth solutions. Using the framework, we generate three benchmarks in Machine Learning, Corporate Finance, and Personal Finance. Expert review finds a significantly lower ground-truth error rate than previous benchmarks such as MMLU and GSM8K. Evaluation of 12 commercial and open-source models shows that our benchmarks achieve near-uniform competency coverage and surface performance differences across models that existing benchmarks fail to capture. We will open-source the framework and our curated benchmarks soon. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_18824 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Fine-Grained Benchmark Generation for Comprehensive Evaluation of Foundation Models Islam, Mohammed Saidul Baghbanzadeh, Negin Kohankhaki, Farnaz Cheraghi, Afshin Kore, Ali Mehdi, Shayaan Dolatabadi, Elham Afkanpour, Arash Machine Learning Artificial Intelligence Computation and Language Evaluation of foundation models often rely on aggregate scores from benchmarks that lack comprehensive coverage and metadata for a fine-grained evaluation. We introduce a framework for automated benchmark generation. Our framework generates evaluation problems grounded in reference material, such as textbooks, producing benchmarks with broad coverage, rich metadata, and robustness to contamination. The pipeline employs a multi-agent architecture for problem generation and a solution-graph-driven strategy that significantly improves the reliability of ground truth solutions. Using the framework, we generate three benchmarks in Machine Learning, Corporate Finance, and Personal Finance. Expert review finds a significantly lower ground-truth error rate than previous benchmarks such as MMLU and GSM8K. Evaluation of 12 commercial and open-source models shows that our benchmarks achieve near-uniform competency coverage and surface performance differences across models that existing benchmarks fail to capture. We will open-source the framework and our curated benchmarks soon. |
| title | Fine-Grained Benchmark Generation for Comprehensive Evaluation of Foundation Models |
| topic | Machine Learning Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2605.18824 |