Metraipolis › archive › dc_9d508b95f86c

CHORUS: Complementary Experts for High-Coverage Testbench Stimulus Generation — arXiv:2608.10090

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.10090
License (as archived)
CC0-1.0
Archived
2026-10-19T06:08:10.545Z
SHA-256 of the archived text
9965751ceb8d6f6be1235256e570d2c7d1cff8389a8194f9efa0dda1f12b7ef2
Stable URL of this page
https://metraipolis.com/documento/dc_9d508b95f86c

How to cite

CHORUS: Complementary Experts for High-Coverage Testbench Stimulus Generation — arXiv:2608.10090 — arXiv (en), https://arxiv.org/abs/2608.10090 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9d508b95f86c (Metraipolis archive dc_9d508b95f86c, sha256 9965751ceb8d6f6be1235256e570d2c7d1cff8389a8194f9efa0dda1f12b7ef2)
BibTeX
@misc{dc9d508b95f86c,
  title = {CHORUS: Complementary Experts for High-Coverage Testbench Stimulus Generation — arXiv:2608.10090},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.10090},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9d508b95f86c; sha256 9965751ceb8d6f6be1235256e570d2c7d1cff8389a8194f9efa0dda1f12b7ef2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10090. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models (LLMs) have advanced code generation, where executable feedback provides a more reliable learning signal than textual imitation alone. Hardware verification is an important application of code generation and accounts for a substantial fraction of modern chip design effort, with high-coverage testbench stimulus generation as a key task. We present CHORUS, a post-training framework that pushes performance beyond what a conventional supervised fine-tuning (SFT)-to-reinforcement learning (RL) pipeline achieves. CHORUS builds on two observations. First, staged SFT produces behaviorally diverse checkpoints, and dense-reward RL turns them into strong experts with comparable aggregate performance but distinct task-level strengths. Second, these complementary strengths can be exploited through either training-free model merging or further post-training to outperform the best individual expert. By consolidating the resulting specialists into a single 4B model, CHORUS achieves 88.0% Pass@1 on CVDP-ECov, outperforming DeepSeek-R1 (671B) by 13.5 percentage points.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-12 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.10090 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-19T05:57:28.233Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).