Paper: arXiv 2404.15489

Authors: Matthew Willetts, Christian Harrington

Abstract

Maximal Extractable Value (MEV) in Constant Function Market Making is fairly well understood. Does having dynamic weights, as found in liquidity boostrap pools (LBPs), Temporal-function market makers (TFMMs), and Replicating market makers (RMMs), introduce new attack vectors? In this paper we explore how inter-block weight changes can be analogous to trades, and can potentially lead to a multi-block MEV attack. New inter-block protections required to guard against this new attack vector are analysed. We also carry our a raft of numerical simulations, more than 450 million potential attack scenarios, showing both successful attacks and successful defense.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper employs advanced mathematical derivations for multi-block MEV analysis in dynamic AMMs, including price manipulation costs and arbitrage returns, while backing its claims with over 450 million numerical simulations testing attack viability under proposed guardrails.

Research Flowchart

  flowchart TD
  A["Research Goal: Are new MEV attack vectors introduced<br>by dynamic weights in AMMs?"] --> B["Methodology: Mathematical & Simulation Analysis"]
  B --> C["Input: 450M+ Simulated Scenarios"]
  C --> D{"Key Computational Process:<br>Analyze inter-block weight changes vs. trades"}
  D -- Attack Vector Detected --> E["Outcome 1: Discovery of Multi-Block MEV Attack"]
  D -- Defense Tested --> F["Outcome 2: Validation of Inter-Block Protections"]
  E & F --> G["Conclusion: Dynamic AMMs require<br>new specific MEV protections"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes to analyze multi‑block MEV attack vectors arising from inter‑block weight changes in dynamic AMMs and to develop corresponding inter‑block protection mechanisms, supported by extensive numerical simulations of attack scenarios.

Source sentence
In this paper we explore how inter-block weight changes can be analogous to trades, and can potentially lead to a multi-block MEV attack. New inter-block protections required to guard against this new attack vector are analysed.

Assumes 5

  • Pool weights change from block to block, causing quoted prices to change even if reserves are unchanged. model
    Source sentence
    This means that, even if the reserves of the pool are unchanged, the quoted prices of the pool change from block to block.
  • An increase in a token’s relative weight raises its quoted price relative to other tokens. model
    Source sentence
    If the weight of token i increases relative to token j, the price of token i increases relative to token j; conversely, a decrease in the relative weight of token i causes a decrease in its price, such that trading reduces holdings of token i and increases those of j.
  • External true market prices for tokens exist and are used as a reference. market
    Source sentence
    We assume the existence of true, external market prices for the tokens. mp,2 = mp is the market price of the second token in terms of the first. mp,1, the market price of the first token in terms of the second is simply the reciprocal of this: mp,1 = 1/mp,2 = 1/mp.
  • Initial pool state is constrained so that no arbitrage opportunity exists before the first trade. model
    Source sentence
    We constrain the initial state of the pool and market prices such that no arbitrage opportunity exists before the first trade (i.e. the quoted prices of the pool are within the no-arb region).
Show 1 more
  • Guardrail parameters (e.g., maximum trade size, minimum weight values, weight change limits) limit the feasibility of attacks. model
    Source sentence
    Note that naturally we would expect, all else being equal, small maximum trade sizes, large min-imum weight values, and small weight changes to be harder to attack.

Data

  • Synthetic attack simulations synthetic unknown
    Source sentence
    We also carry our a raft of numerical simulations, more than 450 million potential attack scenarios, showing both successful attacks and successful defense. Later in paper we describe a battery of in-silico experiments where we perform this attack on TFMM pools, optimising not only the attacking trades but the state (weights, weight changes, reserves) of the pools themselves so as to make the attack as good as

Costs

Unclear The classifier was not confident.

Finds 3

  • Inter‑block weight changes can be exploited to create a multi‑block MEV attack where an attacker trades before a weight shift to inflate price and then sells after the shift for profit. theoretical
    Source sentence
    Knowing that the upcoming change in pool weights, and thus quoted prices, will lead a pool to offering an above-market price when buying one of its constituents, an attacker could, prior to the weight change, trade with the pool to reduce its holdings of that asset. That trade would increase the quoted price of that asset, which would then stack with the weight-change-induced increase in quoted price.
  • For certain guardrail parameter settings, no profitable attack can be found even after optimizing pool state within those constraints. negative
    Source sentence
    For various settings of these guardrail parameters, attacks are not obtained for any settings of pool state when optimising the pool state (within the confines of the guardrails) for an attack to be found.
  • The paper presents the first public study of multi‑block MEV attacks on dynamic AMMs and their defenses, with relevance to LBPs and Replicating Market Makers. other
    Source sentence
    Colour is Z (attack return minus vanilla arbitrage return), scaled by initial pool value.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 8 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (4)

  1. Replicating market makers (2021) — link
  2. Balancer: A non-custodial portfolio manager, liquidity provider, and price sensor (2019)
  3. Fjord foundry
  4. Closed-form solutions for generic n-token

0 of 4 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.