Ethereum Developers Propose EIP-8288 for Memory Pool Signature and Proof Aggregation

Ethereum (ETH) developers have proposed a solution where signatures and cryptographic proofs are verified and aggregated within the memory pool, with blocks containing recursive STARKs that encompass these elements. The draft EIP-8288 aims to reduce the burden of repeatedly including large cryptographic data like quantum-resistant signatures in blocks.
Written by Vitalik Buterin and Thomas Kater on June 3, EIP-8288 (In-mempool signature and proof aggregation) is currently a draft. The proposal explains that hash-based signatures under review are approximately 2–3KB in size, requiring up to 150,000–200,000 gas for direct verification—a rough estimate from examples, not final figures.
The memory pool is the space where transactions wait before being included in a block. The design has transactions declare verification dependencies instead of original signature and proof data. Nodes verify these dependencies before forwarding transactions, while memory pool nodes and block builders aggregate multiple proofs. A single recursive STARK within blocks confirms all transaction dependencies are valid.
STARKs are cryptographic proofs verifying correct computation execution, with recursive methods bundling the validity of other proofs into one. This approach does not eliminate data but shifts where verification and aggregation occur. It also aims to reduce the burden of adding separate verification features for new signature schemes.
The draft builds on EIP-8141 frame transactions, which allow accounts to choose their own validation logic and signature methods. It also proposes including recursive STARKs in the FOCIL list for compatibility. The background involves cost concerns: quantum-resistant signatures and privacy proofs can increase data size and verification costs compared to existing methods.
Instead of adding separate functionality for each new cryptographic method, EIP-8288 suggests wrapping signatures and proofs with STARKs for aggregation at the memory pool stage. However, this may increase verification and aggregation work in the memory pool, potentially raising exposure to denial-of-service attacks, as noted in the draft.
The proposal also suggests limiting the number of signature/STARK dependencies per transaction. STARK proofs themselves could exceed 128KB or reach up to 512KB for faster generation—both design estimates. The choice of instruction set for proof programs is under discussion, with comparisons between EVM, RISC-V, and Ethereum-specific sets. The draft notes that while RISC-V may offer efficiency in tooling, long-term maintenance of instruction rules and versions would be required.
Buterin’s earlier proposal on September 27 included recursive STARKs and signature aggregation as part of Ethereum’s long-term vision but did not specify an upgrade schedule or feature list. EIP-8288 remains a draft; adoption and fork timing are unconfirmed.
Korean Source
This article is an English localization of a Korean-language crypto news report. Original headline: 2~3KB 서명을 메모리풀서 검증·집계, 이더리움 EIP-8288 초안