Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.04168 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910041003524096 |
|---|---|
| author | Pan, Zhuoran Li, Yue Guan, Zhi Hu, Jianbin Chen, Zhong |
| author_facet | Pan, Zhuoran Li, Yue Guan, Zhi Hu, Jianbin Chen, Zhong |
| contents | The increasingly complex Web3 ecosystem and decentralized finance (DeFi) landscape demand ever higher levels of technical expertise and financial literacy from participants. The Intent-Centric paradigm in DeFi has thus emerged in response, which allows users to focus on their trading intents rather than the underlying execution details. However, existing approaches, including Typed-intent design and LLM-driven solver, trade off expressiveness, trust, privacy, and composability.
We present OMNIINTENT, a language-runtime co-design that reconciles these requirements. OMNIINTENT introduces ICL, a domain-specific Intent-Centric Language for precise yet flexible specification of triggers, actions, and runtime constraints; a Trusted Execution Environment (TEE)-based compiler that compiles intents into signed, state-bound transactions inside an enclave; and an execution optimizer that constructs transaction dependency graphs for safe parallel batch submission and a mempool-aware feasibility checker that predicts execution outcomes. Our full-stack prototype processes diverse DeFi scenarios, achieving 89.6% intent coverage, up to 7.3x throughput speedup via parallel execution, and feasibility-prediction accuracy up to 99.2% with low latency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_04168 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | OMNIINTENT: A Trusted Intent-Centric Framework for User-Friendly Web3 Pan, Zhuoran Li, Yue Guan, Zhi Hu, Jianbin Chen, Zhong Cryptography and Security The increasingly complex Web3 ecosystem and decentralized finance (DeFi) landscape demand ever higher levels of technical expertise and financial literacy from participants. The Intent-Centric paradigm in DeFi has thus emerged in response, which allows users to focus on their trading intents rather than the underlying execution details. However, existing approaches, including Typed-intent design and LLM-driven solver, trade off expressiveness, trust, privacy, and composability. We present OMNIINTENT, a language-runtime co-design that reconciles these requirements. OMNIINTENT introduces ICL, a domain-specific Intent-Centric Language for precise yet flexible specification of triggers, actions, and runtime constraints; a Trusted Execution Environment (TEE)-based compiler that compiles intents into signed, state-bound transactions inside an enclave; and an execution optimizer that constructs transaction dependency graphs for safe parallel batch submission and a mempool-aware feasibility checker that predicts execution outcomes. Our full-stack prototype processes diverse DeFi scenarios, achieving 89.6% intent coverage, up to 7.3x throughput speedup via parallel execution, and feasibility-prediction accuracy up to 99.2% with low latency. |
| title | OMNIINTENT: A Trusted Intent-Centric Framework for User-Friendly Web3 |
| topic | Cryptography and Security |
| url | https://arxiv.org/abs/2603.04168 |