Vexorium-Agent 用 GPT-6.1 与 Gemini 4 分步编译门控，Lean 证明重构赛获 86.31 分

View PDF HTML (experimental)

Abstract:Our closed-track entry (Arena username Vexorium-Agent) scored 86.31 points on the final board (L = 77.07, H = 81.86, Z = 100) over the 60 final problems, against 33.33 for the reference proofs, 8th at the deadline (top: 92.57), for US\$166.33 at list prices with every call counted (per problem: mean US\$2.772, at most US\$2.874; US\$3 cap). Three practices carry the entry. First, each paid stage runs the model that measured best per dollar for it among the 9 we tested in our own harness: GPT-6.1 Sol the early sessions, GPT-6.1 Terra the later ones and the tail at 70% of Sol's cost per turn, and Gemini 4 Argon the review and repair; Sol also runs the style rewrite. Second, what our development and research runs learned ships as compiler-checked programmatic rewrite rules rather than prompt text (+0.93 points from rules fixed before the release; +0.26 and +0.30, in sample, from passes built or switched on after the release). Third, every call is recorded and reviewed before each change (all 9 board scores came within -0.063 to +0.015 points of our meter's prediction). The harness, gate and many of these rules also serve our open-track entry. We report the full bill and the spread of three runs.

Comments:	20 pages. Technical report of team Vexorium-Agent for the closed track of the Lean Refactor Arena, a competition at the NeurIPS 2026 Workshop on AI for Verifiable Coding
Subjects:	Software Engineering (cs.SE); Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO)
Cite as:	arXiv:2611.13011 [cs.SE]
	(or arXiv:2611.13011v1 [cs.SE] for this version)
	https://doi.org/10.48550/arXiv.2611.13011 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Bobber Cheng [view email]
[v1] Mon, 9 Nov 2026 03:12:45 UTC (65 KB)