Sequent-Prover：训练基于 SMT 的形式化可验证推理智能体

Authors:Debargha Ganguly, Weicong Chen, Mohsen Hariri, Vikash Singh, Sreehari Sankar, Shouren Wang, Wang Yang, Biyao Zhang, Chaoda Song, Alexander Nemecek, Kai Ye, Osama Zafar, Erman Ayday, Joe Stubbs, Srinivasan Iyengar, Vipin Chaudhary

View PDF HTML (experimental)

Abstract:We introduce Sequent-Prover, a family of verifier-using language-model agents, and the post-training framework that produces them. Agents reason in an executable SMT-LIB scratchpad: they write complete SMT-LIB programs, query Z3 across turns, inspect solver feedback, repair formalizations, and terminate with an explicit answer. Our recipe, verifier-conditioned trajectory distillation, bootstraps from a noisy teacher proposal policy by sampling multi-turn trajectories, retaining the successful tail that is correct, non-trivial, grounded, and executable, then applying supervised fine-tuning and trajectory-level preference optimization. Because solver success alone does not guarantee semantic faithfulness, we evaluate task accuracy, k-sample reliability, and SMT program quality. Across 37 benchmarks and 7,254 problems per model family, post-training raises the 4B model's per-attempt accuracy from 85.6% to 97.0% and four-sample all-correct reliability (pass^4}) from 64.1% to 91.7%. It also reduces average SMT calls from 4.10 to 2.52 and tokens per attempt from 18.97k to 12.72k. Sequent-Prover agents give answers backed by executable formal artifacts.

Subjects:	Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Logic in Computer Science (cs.LO)
Cite as:	arXiv:2610.15258 [cs.CL]
	(or arXiv:2610.15258v1 [cs.CL] for this version)
	https://doi.org/10.48550/arXiv.2610.15258 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Debargha Ganguly [view email]
[v1] Sun, 11 Oct 2026 01:10:19 UTC (2,866 KB)