Inductive Deductive Synthesis: Enabling AI to Generate Formally Verified Systems
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911707434057728 |
|---|---|
| author | Agarwal, Shubham Krentsel, Alexander Liu, Shu Cemri, Mert Cheng, Audrey Meng, Rui Pfister, Tomas Li, Chun-Liang Ratnasamy, Sylvia Parameswaran, Aditya Zaharia, Matei Stoica, Ion Lesani, Mohsen |
| author_facet | Agarwal, Shubham Krentsel, Alexander Liu, Shu Cemri, Mert Cheng, Audrey Meng, Rui Pfister, Tomas Li, Chun-Liang Ratnasamy, Sylvia Parameswaran, Aditya Zaharia, Matei Stoica, Ion Lesani, Mohsen |
| contents | AI agents increasingly excel at generating, testing, and refining code. However, they fall short on tasks requiring formal guarantees of full coverage that testing alone cannot provide. Distributed systems are a prime example: properties such as consistency between reads and writes must hold under every possible interleaving of events. Mechanized formal verification can guarantee such correctness, but typically demands months to years of expert effort. As evidence, even SOTA coding agents (Codex with GPT-5.4 and Claude Code with Opus 4.6) succeed on only 2/7 distributed key-value-store specifications. In this paper, we present the first effective approach to addressing this gap, Inductive Deductive Synthesis (IDS), which jointly and incrementally synthesizes implementation and proof, and learns from failed attempts to systematically try promising strategies. Built as an agentic LLM system, IDS achieves 7/7 in about 6.8 hours and $106 per spec on average, roughly 200x faster than expert effort and 17% cheaper than SOTA agents. IDS further incorporates performance feedback into the same loop, yielding implementations up to 3x faster than published verified systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_23109 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Inductive Deductive Synthesis: Enabling AI to Generate Formally Verified Systems Agarwal, Shubham Krentsel, Alexander Liu, Shu Cemri, Mert Cheng, Audrey Meng, Rui Pfister, Tomas Li, Chun-Liang Ratnasamy, Sylvia Parameswaran, Aditya Zaharia, Matei Stoica, Ion Lesani, Mohsen Artificial Intelligence Distributed, Parallel, and Cluster Computing Logic in Computer Science Programming Languages AI agents increasingly excel at generating, testing, and refining code. However, they fall short on tasks requiring formal guarantees of full coverage that testing alone cannot provide. Distributed systems are a prime example: properties such as consistency between reads and writes must hold under every possible interleaving of events. Mechanized formal verification can guarantee such correctness, but typically demands months to years of expert effort. As evidence, even SOTA coding agents (Codex with GPT-5.4 and Claude Code with Opus 4.6) succeed on only 2/7 distributed key-value-store specifications. In this paper, we present the first effective approach to addressing this gap, Inductive Deductive Synthesis (IDS), which jointly and incrementally synthesizes implementation and proof, and learns from failed attempts to systematically try promising strategies. Built as an agentic LLM system, IDS achieves 7/7 in about 6.8 hours and $106 per spec on average, roughly 200x faster than expert effort and 17% cheaper than SOTA agents. IDS further incorporates performance feedback into the same loop, yielding implementations up to 3x faster than published verified systems. |
| title | Inductive Deductive Synthesis: Enabling AI to Generate Formally Verified Systems |
| topic | Artificial Intelligence Distributed, Parallel, and Cluster Computing Logic in Computer Science Programming Languages |
| url | https://arxiv.org/abs/2605.23109 |