ZK Hints
Daily Digest 每日摘要 View All 查看全部
@lagrangedev Foundation announced that the LA token airdrop will be registered from May 28 to June 2, 2025, LA is the utility token of the Lagrange Prover Network, supporting verifiable AI and other use cases. @lagrangedev Foundation宣布将于2025年5月28日至6月2日进行LA代币空投注册，LA是Lagrange Prover Network的实用代币，支持可验证AI等用例。
@cryptodavidw discussed Schwartz-Zippel circuits and their relationship to iterative constraint systems in their blog, including how to use challenge points to verify polynomial equalities, and its application in plonk permutations. @cryptodavidw 在博客中讨论了Schwartz-Zippel电路及其与迭代约束系统的关系，包括如何使用挑战点来验证多项式等式，以及这在plonk置换中的应用。
@zkv_xyz invites everyone to fill out the form to add ZK projects to the Map of ZK. @zkv_xyz 邀请大家填写表单，把ZK项目加入到Map of ZK中。
@zeroknowledgefm released the video playlist of zkSummit 13. @zeroknowledgefm 发布了zkSummit 13的视频合集。
@SuccinctLabs and @AntChainOpenLab collaborated to develop FPGA acceleration technology for SP1, achieving 20x performance improvement over CPU, marking SP1 as the first RISC-V zkVM with end-to-end FPGA hardware acceleration. @SuccinctLabs 与 @AntChainOpenLab 合作开发了SP1的FPGA加速技术，性能比CPU提升20倍，标志着SP1成为首个具备端到端FPGA硬件加速的RISC-V zkVM。
@0xteddav detailed Sum-Check protocol in their blog, a fundamental tool in zero-knowledge proofs and verifiable computation, allowing verifiers to efficiently check if the prover correctly calculates the sum of polynomial evaluations on the Boolean hypercube. @0xteddav 在博客中详细介绍了Sum-Check协议，这是零知识证明和可验证计算中的基础工具，允许验证者高效检查证明者是否正确计算了多项式在布尔超立方上的评估总和。
Zhao and Ding proposed a method in their paper to break Poseidon challenges with Graeffe transforms and complexity analysis by FFT lower bounds, significantly improving the attack efficiency, and open-sourced the related code. Note: The effective security of Poseidon has not been reduced. Zhao和Ding在论文中提出了一种利用Graeffe变换和FFT下界复杂性分析破解Poseidon挑战的方法，显著提高了攻击效率，并开源了相关代码。注：Poseidon的有效安全性并没有降低。
Eagen et al. proposed a new technique in their paper to speed up elliptic curve scalar multiplications in SN(T)ARK circuits, reducing proof time by 22% to 53% through integer lattice reduction or semi-extended Euclidean algorithm. Eagen等人在论文中提出了在SN(T)ARK电路中快速进行椭圆曲线标量乘法的新技术，通过整数格约简或半扩展欧几里得算法，将证明时间缩短了22%至53%。
Chalkias et al. proposed zkAt, a new cryptographic primitive, in their paper, for privacy-preserving authentication on public blockchains. zkAt uses zero-knowledge proofs to allow users to authenticate transactions while keeping the underlying authentication policy private. Chalkias等人在论文中提出了零知识认证器（zkAt），一种新的密码学原语，用于在公共区块链上实现隐私保护的认证。zkAt利用零知识证明使用户能够认证交易，同时保持底层认证策略的私密性。
Key Points: 要点：
- • This paper designs a binding scenario for authentication policies, making the circuit invisible but immutable, allowing users to verify their own private policy. 本文设计了一个认证策略绑定场景，使电路对外不可见但不可变，便于验证用户私下自设条件。
- • Using interpolation to generate an equivalent QAP, the verification logic remains unchanged but the structure is hidden. 利用插值生成等价 QAP，验证逻辑不变，但结构不可见。
- • zkAt modifies the Groth16 setup, making the vk independent of the circuit, hiding the policy. zkAt 修改 Groth16 setup，使 vk 与电路无关，隐藏策略。
- • vk and π do not leak policy information, achieving policy privacy. vk 和 π 不泄露策略信息，实现策略隐私。
- • zkAt policies are fixed, updates require zkAt+. zkAt 策略固定，更新需用 zkAt+。
- • zkAt+ uses recursive SNARKs to support privacy policy updates. zkAt+ 用递归 SNARK 支持隐私策略更新。
@argumentxyz team open-sourced Ix, a zero-knowledge proof-carrying code platform, which can compile Lean 4 programs into zk-SNARKs, implement program execution and type checking verification, significantly reducing the running cost of verification. @argumentxyz 团队开源了Ix平台，该平台能够将Lean 4程序编译成zk-SNARKs，实现程序的执行和类型检查验证，显著降低验证的运行成本。
@0xPolygon and @IrreducibleHW team open-sourced PetraVM, a general-purpose virtual machine based on the Binius proof system, supporting recursive proof verification and high-performance computation, through WebAssembly compilation and a custom high-level language PetraML. @0xPolygon 和 @IrreducibleHW 团队开源了PetraVM，一个基于Binius证明系统的通用虚拟机，支持递归证明验证和高效计算，通过WebAssembly编译和自定义高级语言PetraML实现高性能可验证计算。
Singh et al. proposed a method in their paper to reduce the communication complexity of the classical sum-check protocol using multilinear polynomial commitment schemes, achieving a proof cost of O((ℓ+d)n) and a communication complexity of O(ℓ + dlog log n). Singh等人在论文中提出了一种利用多线性多项式承诺方案减少经典sum-check协议通信复杂性的方法，实现了证明者成本为O((ℓ+d)n)和通信复杂度为O(ℓ + dlog log n)的突破。
Barbacovi et al. proposed a new scheme in their paper to enforce arbitrary constraints on Bitcoin transactions, while keeping some information private. This method also bypasses the non-Turing-complete nature of Bitcoin scripts, allowing the execution of unbounded constraints. Note: This scheme is executed on the BSV chain, not on the BTC chain, and BSV is the earliest Bitcoin fork chain to support Groth16. Barbacovi等人在论文中提出了一种利用zkSNARKs的新方案，能够在比特币交易上强制执行任意约束，同时保持部分信息私密。该方法还绕过了比特币脚本的非图灵完备性，允许执行无界约束。编者注：该方案是在BSV链上执行，不能在BTC链上执行，而BSV是最早支持Groth16的比特币分叉链。
@cknabs open-sourced lattirust, a library focused on lattice-based zero-knowledge succinct arguments, similar to lattigo but for arguments. @cknabs 开源了lattirust，这是一个类似于arkworks但专注于格基零知识简洁论证的库，类似于lattigo但用于论证。
@powdr_labs announced powdr-OpenVM, a compiler middleware that enhances @openvm_org with autoprecompiles for custom guest acceleration, showing >5x proof time improvement in end-to-end experiments. @powdr_labs 宣布了powdr-OpenVM，该编译器中间件通过自动预编译增强 @openvm_org，实现了端到端实验中证明时间超过5倍的提升。
@SuccinctLabs team introduced their next-generation zkVM SP1 Hypercube in their blog, using a new proof system architecture based on multilinear polynomials, with a cluster price as low as $300~400k, and can achieve a proof time of about 10 seconds for most Ethereum L1 blocks. @SuccinctLabs 团队在博客中介绍了其下一代zkVM SP1 Hypercube，采用全新的基于多线性多项式的证明系统架构，集群价格低至30~40万美元，对大多数以太坊L1区块可达到10s左右证明时间。
@VitalikButerin discussed the current progress of ZKP real-time proving in his tweet, pointing out key challenges such as worst-case proof time, missing formal verification, proof energy consumption, and L1 gaslimit extension. @VitalikButerin 在推文中讨论了ZKP实时证明技术的当前进展，指出了包括实时证明最坏情况、正式验证缺失、证明能耗及L1 gaslimit扩展需求等关键挑战。
@RiscZero announced that it will reduce the Ethereum block proof time to less than 12 seconds, and the cost of building a cluster can be as low as 120,000 USD, with the best performance currently being 20 seconds. @RiscZero 宣布将实现以太坊区块证明时间缩短至12秒以内，且构建集群的成本可低至12万美元，目前最好的性能是20秒。
@StarkWareLtd Exploration team introduced Broly in their blog, a POC that allows users to create Bitcoin inscriptions on Starknet without owning BTC or directly interacting with the Bitcoin network. @StarkWareLtd Exploration团队在博客中介绍了Broly，一个允许用户通过Starknet无需拥有BTC或直接与比特币网络交互即可创建比特币铭文的POC。
Veridise shares the initial results of their collaboration with Succinct using the Picus tool to verify the determinism of SP1 circuits, including successful verification of multiple SP1 circuits and identification of improvement points, emphasizing the importance of determinism in ZK circuit security. Veridise分享了与Succinct合作使用Picus工具验证SP1电路确定性的初步成果，包括成功验证多个SP1电路及识别改进点，强调了确定性在ZK电路安全中的重要性。
Key Points: 要点：
- • Veridise and Succinct collaborated to verify the determinism of SP1 circuits using the Picus tool Veridise与Succinct合作，用Picus工具验证SP1电路的确定性
- • Determinism verification can eliminate vulnerabilities in ZK circuits due to insufficient logical constraints 确定性验证可排除ZK电路中因逻辑约束不足导致的漏洞
- • Developed a tool to convert Plonky3 to LLZK, successfully verified 11 basic operations 开发Plonky3到LLZK的转换工具，成功验证11项基础操作
- • Current process has limitations such as modular constraints and input/output labeling 当前流程存在模块化约束、输入输出标注等局限性
- • Succinct plans to optimize the Plonky3 framework to support formal verification Succinct计划优化Plonky3框架以支持形式化验证
- • Targeting to extend formal verification to all SP1 circuits, becoming a regular development step 目标是将形式化验证扩展至SP1所有电路，成为开发常规环节
vlayer announced the launch of its mainnet, aiming to make smart contracts useful and relevant to billions of users through innovative features like Time Travel, Teleport, Email Proofs, and Web Proofs. vlayer宣布主网上线，旨在通过Time Travel、Teleport、Email Proofs和Web Proofs等创新功能，使智能合约对全球数十亿用户变得实用和相关。
Fermah and Reclaim have launched the private zkTLS and Confidential Proof Delegation (CPD), significantly enhancing the efficiency of zero-knowledge proofs and optimizing the performance of Web3 privacy applications. Fermah与Reclaim合作推出私有zkTLS及保密证明委托（CPD），显著提升零知识证明效率，优化Web3隐私应用性能。
Kobi Gurkan discusses specific issues in the software supply chain for ZK provers, emphasizing the increasing importance of supply chain security with the trend towards data localization and proof sharing. Kobi Gurkan在博客中讨论了ZK证明者在软件供应链中的特定问题，强调了随着数据本地化和仅共享证明的趋势，供应链安全问题日益重要。
Key Points: 要点：
- • ZK proof privacy relies on reproducible builds, but supply chains are complex and vulnerable ZK证明的隐私安全依赖可复现构建，但供应链复杂且易被篡改
- • Open-source audits are insufficient; need simplified trust models focusing on critical steps (e.g. builds) 开源审计不足，需简化信任模型并聚焦关键环节（如构建过程）
- • Desktop apps are more reproducible; iOS is hardest due to closed ecosystem 桌面程序可复现性较好，iOS因封闭生态最难验证
- • Solutions: Open build systems, standardized environments, verifiable distribution 改进方案：公开构建系统、标准化构建环境、支持验证分发
- • TEE reduces trust requirements but depends on reproducible builds TEE等技术可降低信任需求，但依赖可复现构建
- • Industry trend: Enterprises push end-to-end verification, requiring stronger supply chain security 行业趋势：企业正推动端到端验证，需加强全链路安全性
Thomas et al. explained the Valida instruction set architecture in their paper, designed for zkVM implementation, aiming to optimize fast and efficient proof execution. Thomas等人在论文中讲解了Valida指令集架构，专为zkVM实现设计，旨在优化快速高效的执行证明。
Key Points: 要点：
- • Instruction Set Architecture (ISA) optimized for zkVMs to enhance ZK proof efficiency 专为zkVM设计的指令集架构（ISA），优化零知识证明执行效率
- • No general-purpose registers; directly accesses RAM via stack offsets to reduce instruction count 无通用寄存器，直接通过栈偏移访问RAM，减少指令数量
- • Modular design; base version consists of specified chipset components 模块化设计，基础版本由指定芯片组构成
- • Harvard architecture, 32-bit little-endian, with only FP and PC as special-purpose registers 哈佛架构，32位小端序，仅含FP和PC两个专用寄存器
- • Interacts externally via read-only input and write-only output tapes, ideal for blockchain verifiable computation 通过输入/输出带与外部交互，适合区块链可验证计算场景
- • Compared to RISC-V, significantly reduces SNARK proving overhead and improves performance 相比RISC-V，显著降低SNARK证明开销，提升性能
Quantum Punks team shares their vision of quantum cryptography, exploring how quantum physics can build more secure systems and unique cryptographic methods, and the potential of quantum cryptography in protecting digital rights. Quantum Punks团队分享了他们对量子密码学的愿景，探讨了如何利用量子物理构建更安全的系统和独特的加密方法，以及量子密码学在保护数字权利方面的潜力。
Project Eleven (P11) is hosting the The QDay Prize competition, which will end on April 5, **2026**, aiming to challenge the use of Shor's algorithm on quantum computers to break the largest ECC keys, with a prize of 1 Bitcoin. Project Eleven (P11)主办的The QDay Prize竞赛将于**2026年**4月5日截止，旨在挑战使用Shor算法在量子计算机上破解最大ECC密钥，奖金为1比特币。
Pappas et al. proposed HydraProofs in their paper, the first vector commitment scheme that can generate all opening proofs for a vector of size N in O(N) time, and is directly compatible with zkSNARKs families, significantly improving prover time. Pappas等人在论文中提出了HydraProofs，首个能在O(N)时间内为大小为N的向量生成所有 opening 证明的向量承诺方案，并直接兼容于zkSNARKs家族，显著优化了证明者时间。
Key Points: 要点：
- • HydraProofs generates all opening proofs in O(N) time HydraProofs 实现 O(N) 时间生成全部 opening 证明
- • Directly compatible with zkSNARKs using multilinear encodings 直接兼容多项式编码的 zkSNARKs，无需电路中重建
- • Uses partitioning and folding to reduce HyperEval to linear time 通过分段与折叠技术将 HyperEval 降至线性时间
- • Supports efficient multi-user verification when combined with GKR 结合 GKR 实现多用户输入的有效验证
- • In verifiable secret sharing, proving is faster than the dealer's computation 在可验证秘密共享中，生成证明比秘密分发者计算更快
- • Enables robust aggregation verification in federated learning with low overhead 支持联邦学习中的鲁棒聚合验证，性能开销低
@RiscZero will generate ZK proofs for @MagicNewton, a platform that improves crypto user experience through verifiable AI agents. Each agent runs in a TEE, and RISC Zero generates proofs and posts them onchain for audit. @RiscZero 宣布将为 @MagicNewton 平台生成ZK证明，该平台通过可验证的AI代理改善加密用户体验。每个代理在TEE中运行，RISC Zero生成证明并上链，以供审计。
@0xteddav released a unique private dating app that uses privacy-based matching mechanisms, only revealing matches when successful, built using @NoirLang and @TACEO_IO's coSNARKS technology. @0xteddav 发布了一款独特的私人约会应用，通过隐私的匹配机制，仅在匹配成功时互相知晓，采用 @NoirLang 和 @TACEO_IO 的coSNARKS技术构建。
David Nevado et al. explore lattice-based signature aggregation techniques in the paper , proposing a method to replace BLS signatures in the post-quantum era, with a focus on the efficient aggregation and verification performance of Falcon signatures. David Nevado等人在论文中探讨了格基签名聚合技术，提出了一种在后量子时代替代BLS签名的方法，重点分析了Falcon签名的高效聚合及其验证性能。
Key Points: 要点：
- • BLS signatures are insecure in the quantum computing era, requiring post-quantum alternatives. BLS签名在量子计算时代不安全，需后量子替代方案
- • LaBRADOR is a lattice-based scheme that can aggregate Falcon signatures (quantum-resistant). LaBRADOR是格基方案，可聚合Falcon签名（抗量子）
- • 10k signature aggregation results: 74KB proof, 5.95 seconds generation, 2.65 seconds verification. 10k签名聚合结果：74KB证明，5.95秒生成，2.65秒验证
- • Verification bottleneck: 90% time spent on serial step simple_reduce. 验证瓶颈：90%时间在串行步骤simple_reduce
- • Compared to hash-based schemes (e.g., STARKs): verification is fast (106ms) but proofs are large (1.7MB). 对比哈希基方案（如STARKs）：验证快(106ms)但证明大(1.7MB)
- • Advantages of LaBRADOR: small proof size; disadvantages: slower verification than hash schemes. LaBRADOR优势：小证明体积；劣势：验证慢于哈希方案
- • Future directions: optimize parallel verification or combine with delegated proof techniques. 未来方向：优化并行验证或结合委托证明技术
nmohnblatt compares Ligerito and WHIR in the blog, discussing their similarities and differences in recursive structures, Sumcheck constraints, and code usage. nmohnblatt在博客中比较了Ligerito和WHIR两种多项式承诺方案，讨论了它们在递归结构、Sumcheck约束和代码使用上的相似性与差异。
Key Points: 要点：
- • WHIR supports only RS codes, with concise verification; Ligerito supports arbitrary linear codes, with broader applicability. WHIR仅支持RS码，验证简洁；Ligerito支持任意线性码，泛用性强
- • WHIR utilizes foldable properties, while Ligerito employs code switching technology. WHIR利用可折叠性优化，Ligerito采用码切换技术
- • WHIR has higher verification efficiency, while Ligerito generates proofs faster. WHIR验证效率更高，Ligerito证明生成更快
- • WHIR may produce smaller proofs, while Ligerito has better compatibility. WHIR可能生成更小证明，Ligerito兼容性更好
- • choose WHIR for verification efficiency, Ligerito for flexibility. 选择依据：验证效率选WHIR，灵活性选Ligerito
- • Blaze is a hybrid solution of both. Blaze是两者的混合方案
@zksecurityXYZ conducted a security assessment of the Rust p256 crate, finding no major issues, with the codebase being well-tested and architecturally sound. The assessment covered elliptic curve operations, ECDSA signature verification, and field arithmetic. This audit report can serve as a learning reference for p256 curves. @zksecurityXYZ 对Rust p256 crate进行了安全评估，未发现重大问题，代码库经过充分测试且架构良好。评估范围包括椭圆曲线操作、ECDSA签名验证和字段算术。本审计报告可以作为p256曲线的学习参考资料。
Key Points: 要点：
- • P-256 optimization: Utilize the a=-3 property to accelerate calculations P-256优化：利用a=-3特性加速计算
- • ECDSA security: Must use a deterministic nonce (RFC6979) ECDSA安全：必须用确定性nonce（RFC6979）
- • Performance key: Precomputing 16 points improves signature speed 性能关键：预计算16个点提升签名速度
- • Important vulnerability: Random number generation may produce invalid values 重要漏洞：随机数生成可能产生非法值
- • Optimization tip: Removing redundant operations can speed up by 16% 优化技巧：删除多余运算可提速16%
@VitalikButerin discussed the vision of simplifying the Ethereum L1 in the blog, including simplification strategies for the consensus layer and execution layer, and how to reduce complexity through shared protocol components. @VitalikButerin 在博客中讨论了简化以太坊L1的愿景，包括共识层和执行层的简化策略，以及如何通过共享协议组件减少复杂性。
@solana reported a potential vulnerability in the ZK ElGamal Proof program, allowing the construction of arbitrary proofs. A patch has been released and reviewed by a security company, with no known exploits currently. @solana 基金会报告了ZK ElGamal Proof程序中的一个潜在漏洞，允许构建任意证明。已发布补丁并由安全公司审查，目前无已知利用。
@zksecurityXYZ discussed that the error bound of Barrett reduction algorithm can be tighter than the traditional assumption, especially on moduli used in cryptography (e.g., NIST curves), with the quotient approximation error being at most 1. This improvement eliminates the need for the second subtraction in practice, resulting in a 14% performance boost in scalar multiplication for RustCrypto p256. @zksecurityXYZ 在博客中讨论了Barrett reduction算法的误差界限可以比传统假设的更紧，特别是在密码学中使用的模数（如NIST曲线）上，商近似误差最多为1。这一改进消除了实践中第二次减法的需要，使RustCrypto p256在标量乘法中实现了14%的性能提升。
Baweja et al. proposed FICS and FACS in their paper, aiming to improve the efficiency of succinct arguments based on IOPs. FICS reduces the verifier query complexity, and FACS achieves linear prover time, both supporting linear time encodable code classes including systematic LDPC codes and tensor codes. Baweja等人在论文中提出了FICS和FACS两种方案，旨在提高基于IOP的简洁论证的效率。FICS减少了验证者查询复杂度，FACS实现了线性证明者时间，两者均支持包括系统LDPC码和张量码在内的线性时间可编码码类。
Castillo et al. proposed the Trusted Compute Unit (TCU) framework in their paper, integrating zkVMs and TEEs to enable cross-technology verifiable computation composition, allowing decentralized applications (dApps) to offload computation tasks to TCU and obtain correctness proofs while ensuring data privacy and output integrity. Castillo等人在论文中提出了Trusted Compute Unit (TCU)框架，通过整合zkVM和TEE实现跨技术可验证计算组合，允许去中心化应用（dApp）将计算任务卸载至TCU并获取正确性证明，同时保障数据隐私与输出完整性。
Novakovic et al. proposed Ligerito, a small and concretely fast polynomial commitment scheme supporting univariate/multilinear polynomials, with proof size optimized to $log(N)^2/loglog(N)$ and fast proof generation (2^24-degree polynomial in 1.3 seconds) on consumer-grade hardware (e.g., M1 MacBook Pro). Novakovic等人在论文中提出了Ligerito，一种高效的多项式承诺与内积方案，支持单变量/多线性多项式，证明大小优化至$log(N)^2/loglog(N)$，并在消费级硬件（如M1 MacBook Pro）上实现快速证明生成（2^24项多项式仅需1.3秒）。
Google Wallet updates, introducing ZKP technology to support secure and convenient age and identity verification for residents of more UK and US states, expanding to 50 countries. Google Wallet 更新，引入ZKP技术，支持英国和美国更多州的居民安全便捷地验证年龄和身份，同时扩展至50个国家。
@aztecnetwork introduced its new public testnet in blog, highlighting the breakthroughs of PLONK proof systems and Noir programming language in achieving programmable privacy, and how to support developers to build fully privacy-protected applications. @aztecnetwork 在博客中介绍了其新推出的公共测试网，强调了PLONK证明系统和Noir编程语言在实现可编程隐私方面的突破，以及如何支持开发者构建完全隐私保护的应用程序。
@srinathtv discussed the importance of folding schemes and their performance issues in practical applications in blog, particularly highlighting the superiority of NeutronNova and Neo folding schemes. @srinathtv 在博客中讨论了折叠方案的重要性及其在实际应用中的性能问题，特别强调了NeutronNova和Neo折叠方案的优越性。
Key Points: 要点：
- • Folding reduces prover's cost by eliminating significant parts of the prover's work. 折叠方案通过消除证明者工作中的明显部分来降低证明成本。
- • Recursive proof composition becomes _practical and efficient_. 递归证明组合变得 _实用且高效_。
- • Implementation details—especially memory checking and MSM—make or break performance. 实现细节（尤其是内存检查和MSM）决定了性能。
- • Popular elliptic curve libraries often miss optimization opportunities, leaving performance on the table. 流行的椭圆曲线库经常错过优化机会，导致性能不佳。
@zksecurityXYZ explored variants of the KZG polynomial commitment scheme in blog, including its basic construction and batch processing techniques, and how to apply these techniques in zkSNARKs to improve efficiency. @zksecurityXYZ 在博客中探讨了KZG多项式承诺方案的变体，包括其基本构造和批处理技术，以及如何在zkSNARKs中应用这些技术以提高效率。
ZK Podcast episode 359 invited IBM Research researcher Vadim Lyubashevsky to discuss the role and development of lattice-based cryptography in zero-knowledge systems. ZK Podcast第359集邀请了IBM Research的研究科学家Vadim Lyubashevsky，讨论了基于格的密码学在零知识系统中的作用及其发展。