Proof of Track 2023-10-07

Oct 07, 2023

Smart Contract Security

Takeaway: Automated security tools can prevent important vulnerability classes; Security tools should focus in detecting more vulnerabilities; Research should focus on improving semi-automated tools.

Compositional Formal Verification of Zero-Knowledge Circuits

See the recap.

STARK Future Directions

Current status of STARK ecosysmtem, and some future directions. Mostly info on StarkNet.

The MEV Workshop at the Science of Blockchain Conference 2023

In case you wanna watch. I didn't watch it.

🔍 Modular Summit (2023)

World Engine: Horizontally Scaling Rollups With Shards

This talks points out the loop-driven runtime of game, and games are perfectly suited with shards.

Anoma ❤️ Celestia: intent-centric rollups

Intent loves modular.

Avail: Architecture and Use Cases

Good to know the architecture and progress of Avail.

Modular State of the Union

The current state of modular. It's amazing to see how the narrative of modular has evolved so much.

Light nodes are not just a meme

It increases social scalability.

Shielded Transactions Are Rollups

This provides another good definition of rollup: base offloads execution to rollup, while rollup submits state root and some form of trust to base.

More sovereignty more security

We will always have software bugs, so we need to have more sovereignty.

TLDR: Wallets aren't the problem. Our mental model of how data is written to the chain is the problem; Many chains = users won't know (& should not know) what chains they are transacting on; Need better ways of interacting with more chains without creating silos.

Relayers in the Modular World

Modular stack has more than just 4 layers, and for each layer-to-layer communication, we need relayers. So a lot of relayers that lead to pricing problem and decentralization problem.

Towards a Theory of MEV Part II: Uncertainty

Tarun explains the uncertainty of MEV, and we can review it with fourier transformation that simple models can be used to explain complex phenomena.

Current State on Orderflow Block Building & the Latency Game

Takes: order flow through monopoly: it's all market maker; different types of builders: non-profit builder; strategic builder; searcher builder.

Builders and More Advanced Forms of Aggregation

Vitalik talks about types of aggregation (ERC-4337, BLS aggregation, Block Builder, ZK Proof Aggregation).

Ethereum PBS R&D Roadmap

Takes: mev-boost (out protocol, trusted, massive adopted (95%)); two-slot PBS (in protocol, weaken security); Payload-Timeliness Committee (in protocol, design: consensus-layer block produced without tx -> consensus-layer attestation remain same; builder reveal payload; ptc committee votes); optimistic relaying (out protocol, remove relay responsibilities); Proposer-enforced protocol commitments (in protocol, key idea: block validity used to enforce commitment, PBS can be enshrined). You MUST check out the beautiful graph at the end.

PBS on L2s

L2 offers a pefect place for experimentation of MEV with all the unique characteristics (sequencer == p + b, fast blocktime, less proposer, no tx pool). Also shared sequencers need to be considered as well.

🔍 ZK10 (2023)

ZKML Endgame: Specialized ZK Proving with GKR

Modulus announced Remainder (Robust GKR implementation of structured circuits + wiring predicate-based circuits, Built-in dataparallel circuit implementation, Open-source version coming out by year's end, Supports dataparallel, large-scale decision forest inference showcase in November).

Lurk: From zkDSLs to zkGPLs

Solve compiler bug with LEM. The Lurk Evaluation Model, or LEM, is an embedded zkDSL which allows for safe and efficient construction of universal circuits which can evaluate general-purpose programming languages under zero-knowledge (zkGPLs).

Optimizing recursive STARK verification in Polygon Miden VM

VM-based approach to recursive proof verification used in Polygon Miden VM. Polygon Miden always has the best presentation.

Lasso: An Engineering Heavy Overview

Structured tables with less waste. Jolt is the application of Lasso to VM instructions. Jolt-based VMs are simple, fast, and easily auditable.

Analysis of zkVM Designs

JUST WATCH IT.

ZK Vulnerabilities and Attacks

Do we have to be scared? Yes -> Different mental model for onboarding developers -> Vulnerabilities are complex and hard to catch!-> As opposed to smart contracts - much of the code is proprietary -> Automated security tooling has limitations - manual audits require extensive expertise along the whole SNARK stack; No -> Currently, most provers are permissioned - trust, that soundness bugs are not exploited -> Advanced DSL and tooling -> Great security researchers / auditing teams / education resources in the space.

ZKML with EZKL: Where we are and the future

ezkl is 10,000 faster in the past year.

powdr: a modular stack for zkVMs

Kinda like an LLVM for zkVM (with front end of high level languages and back end of proving frameworks).

Thanks for reading!

I'm not a financial advisor. Proof Of Track is for education and information purposes only. DYOR.

Subscribe Proof Of Track, follow me at @msfew_eth, or connect with me at sny.is.

Proof of Track 2023-10-07

🔍 Science of Blockchain Conference (2023)

🔍 Modular Summit (2023)

🔍 ZK10 (2023)

🔍 StarkNet Summit (2023)

Thanks for reading!

Discussion about this post