Metraipolis › archive › dc_fa05f9ab9eb0
VeriHarness: Scaling Agentic Verification for Long-Horizon Tasks — arXiv:2610.00972
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.00972
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T11:54:50.023Z
- SHA-256 of the archived text
301839ca1706f8f2ec40dff10d4ed14413bb3669b405d9c032b5dd371081643c- Stable URL of this page
https://metraipolis.com/documento/dc_fa05f9ab9eb0
How to cite
VeriHarness: Scaling Agentic Verification for Long-Horizon Tasks — arXiv:2610.00972 — arXiv (en), https://arxiv.org/abs/2610.00972 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fa05f9ab9eb0 (Metraipolis archive dc_fa05f9ab9eb0, sha256 301839ca1706f8f2ec40dff10d4ed14413bb3669b405d9c032b5dd371081643c)
BibTeX
@misc{dcfa05f9ab9eb0,
title = {VeriHarness: Scaling Agentic Verification for Long-Horizon Tasks — arXiv:2610.00972},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.00972},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fa05f9ab9eb0; sha256 301839ca1706f8f2ec40dff10d4ed14413bb3669b405d9c032b5dd371081643c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.00972. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
As LLM agents undertake increasingly complex, long-horizon tasks, verifying their outputs becomes increasingly challenging. We study how verification capability can be strengthened with a fixed base model, without access to reference answers or grading rubrics at test time. Repeated sampling yields multiple rollouts that can contain complementary correct claims, but we need a reliable verification mechanism to determine which claims to trust. We first find that disagreement often exposes correct alternatives, while consensus can conceal errors. These observations motivate VeriHarness, which turns the underlying LLM a generator uses into an agentic verifier by giving it a workspace, evidence tools, and reusable verification skills. A disagreement resolver checks competing claims against environmental evidence, while a consensus challenger tests shared claims and searches for omitted requirements. Their findings guide the selection and revision of the final artifact. Across five long-horizon workspace benchmarks and two frontier models, VeriHarness achieves the highest selection scores among the evaluated baselines. Evidence-backed revision further improves average performance, bringing gains over a single rollout to 6.2 points with Gemini 3.5 Flash and 6.4 points with Claude Opus 4.8. We further show that verification skills can self-improve from failure feedback, demonstrating VeriHarness as a novel and critical approach for scaling long-horizon agentic verification. We release the full pool of approximately 26,000 rollouts across all five benchmarks and both models, produced at a cost of over $100,000, to support future research on agentic verification. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.MA). Categorie arXiv: Artificial Intelligence (cs.AI) ; Multiagent Systems (cs.MA). Pagina del lavoro: https://arxiv.org/abs/2610.00972 — 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 2026-10-06T10:17:30.569Z dalla pagina https://arxiv.org/catchup/cs.MA/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).