• Zolt is a Zig zkVM prover that generates proofs verifiable by the unmodified upstream a16z/jolt verifier. Zero dependencies, zero FFI — all cryptography (field arithmetic, pairings, MSM, polynomial commitments) is implemented from scratch using only the Zig standard library.

• https://github.com/MatteoMer/zolt

• https://x.com/RobinSalen/status/2031388330412355997?s=20

• https://github.com/Plonky3/Plonky3

• Speaker: Dengji Ma

• Video link: https://youtu.be/HNhl_S4DSjk

• Paper link: https://eprint.iacr.org/2025/2258.pdf

• Speaker: Yingfei

• Video link: https://youtu.be/8KlNRypo_T8?si=uqM4hvaXvs1qm7LQ

• Paper link: https://eprint.iacr.org/2025/2124

• Git Repo: https://github.com/coset-io/baby-lattice-folding

• In this episode of the lean Ethereum miniseries, Nico Mohnblatt speaks with Thomas Coratger and Emile from the Ethereum Foundation about the design and implementation of LeanVM, a minimal zkVM created to support post-quantum signature aggregation on Ethereum’s consensus layer. They explain why the team chose a VM architecture over fixed circuits and how LeanVM takes inspiration from Cairo with just 4 opcodes and 2 precompiles to keep the instruction set extremely small and make formal verification easier. The conversation also covers LeanVM implementation choices like using Plonky3 and WHIR for efficient proving on CPUs, benchmarks for aggregation speed, and the role of Python specs in testing client interop. They share ongoing efforts to optimize low-level primitives and invite community input on the project.

• https://youtu.be/YWkyvTrwtQU?si=_SDs292p1cYwnkvC

• https://eprint.iacr.org/2026/499

• Youheng Wang, Rujia Li, Zhaoyang Xie, Kaikai Feng, Qingjie Chen, Yang Gao, Sisi Duan

• https://eprint.iacr.org/2026/498

• Yuanju Wei, Zhelei Zhou, Xinxuan Zhang, Songyu Wu, Binwu Xiang, Cheng Hong, Yi Deng

• https://eprint.iacr.org/2026/494

• Kexi Huang, Yanpei Guo, Wenjie Qu, Jiaheng Zhang

• https://eprint.iacr.org/2026/487

• Juraj Belohorec, Pavel Hubáček, Aleksi Kalsta, Kristýna Mašková

• https://eprint.iacr.org/2026/480

• Lior Goldberg, Maya Dotan, Ittay Dror, Gideon Kaempfer, Nir Levi, Noa Oved, Arad Reder, Anat Veredgorn, Noa Wolfgor

• https://eprint.iacr.org/2026/474

• Jonathan Bootle, Julia Guskind, Sikhar Patranabis, Katerina Sotiraki

• https://eprint.iacr.org/2026/471

• https://eprint.iacr.org/2025/2113

• 发表于 IEEE S&P 2026

• 摘要：zk-SNARK 使证明者能够在不泄露关于秘密见证任何额外信息的前提下，通过一个简洁且可高效验证的证明，使验证者确信某个陈述为真。制约 zk-SNARK 实际部署的一个重要障碍在于其较高的证明开销。基于这一动机，我们研究了服务器辅助的 zk-SNARK，其中客户端（即证明者）将其大部分计算工作外包给单个不受信任的服务器，而服务器既无法获知见证信息，甚至也无法获知证明本身。我们对这一概念进行了形式化定义，并展示了如何为广泛部署的 zk-SNARK 构造服务器辅助证明方案，包括 Nova、Groth16 和 Plonk。支撑我们设计的关键构件是一种新的原语——加密多标量乘法（encrypted multi-scalar multiplication, EMSM），其能够实现多标量乘法（MSM）的私有委托计算。我们基于学习奇偶校验带噪声（learning parity with noise, LPN）假设的若干变体构造了 EMSM，在该构造中，客户端执行群运算，而服务器的计算开销与明文 MSM 基本一致。我们实现并评估了所提出的构造。与本地证明相比，我们的方法可将客户端的计算量最多降低至原来的，并将证明延迟最多降低。

If you're interested in our ZK Insights or have ideas for similar content, we highly encourage everyone to head over to our GitHub repo and submit a Pull Request. Join forces with like-minded ZKPunks to co-create!

✨ Github repo link: https://github.com/ZKPunk-Org/zk-insights

Special thanks to: Yingfei, Hins