Resilient Liquid Democracy: Mitigating Voting Power Imbalances via Secure Delegation Networks

arXiv:2607.01730 · cs.CR, cs.CY · Submitted 2026-07-02 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Security Radio. Generated commentary on the latest security and cryptography papers.

Nadia: Hi, I'm Nadia.

Elias: And I'm Elias, here with Priya.

Nadia: We're looking at the paper: "Resilient Liquid Democracy".

Elias: Here's the gist of the paper. The gist The proposed mechanism introduces a liquid democracy protocol that combines timed-release cryptography with ranked multi-delegation and fallback ballots to mitigate vulnerabilities like herding, coercion, and delegate failure while maintaining full auditability<ref:2607.01730#pg6> How it works The system separates delegation formation during the voting phase from delegation resolution during the reveal phase to enforce secrecy during formation and transparency during the latter<ref:2607.01730#pg5> Delegation choices are sealed with decentralized timed-release encryption while they are being formed and revealed in full for tallying, and each voter names ranked backup delegates together with a personal fallback ballot so that delegate failures do not silence them<ref:2607.01730#pg4> The protocol involves a submission phase where each voter encrypts their policy tuple using the Timed-Release Encryption (TRE) scheme parameterized by the reveal time Tr<ref:2607.01730#pg5> This results in a ciphertext Ci = TREncTr(Πi; ri), which is posted to a public ledger before the submission deadline<ref:2607.01730#pg5> The system hides delegation decisions only during the phase in which voters are still making them, preventing influence from visible delegation counts and contemporaneous proof of how a voter delegated<ref:2607.01730#pg5> Security Guarantees The design provides two formal security guarantees: pre-reveal secrecy and resubmission receipt-freeness<ref:2607.01730#pg6> Pre-reveal secrecy ensures that no efficient adversary learns anything about how a voter delegated or voted before the reveal time Tr, which suppresses popularity-driven herding and removes the evidentiary basis for bribery during formation<ref:2607.01730#pg5> Receipt-freeness via resubmission means that even a voter forced to hand over a “receipt” can still cast a different, true policy before Tr, defeating coercion of the form “prove to me how you voted”<ref:2607.01730#pg6> Robustness Against Failure and Herding The system extends standard liquid democracy with ranked multi-delegation and fallback ballots to improve robustness against delegate failure<ref:2607.01730#pg5> The ranked delegation mechanism transforms delegation from a single-point-of-failure structure into a more resilient routing system<ref:2607.01730#pg5> Ranked delegation with a personal fallback ballot caps the loss at an arithmetic floor of about 3%, with the residual loss governed by the fraction of delegators who provide fallback ballots<ref:2607.01730#pg4> This robust design keeps vote loss near 3% even when 30% of delegates are targeted, versus 26% for a single-delegate design under targeted attacks<ref:2607.01730#pg5> Empirical Findings Experiments grounded in real voting data show that the mechanism keeps voting power dispersed where transparent formation concentrates it into few hands, keeps outcomes stable under herding where transparent outcomes become a lottery on arrival order, and cuts vote loss under targeted delegate failures from 26% to about 3%<ref:2607.01730#pg4> Delegation helps only when abstention is large and systematically unrepresentative and enough abstainers use the channel; otherwise it is at best neutral and can harm the outcome<ref:2607.01730#pg4> The sealed formation produces more decentralized and robust delegation network, while transparent formation concentrates voting power in proportion to the herding strength, which translates into vote loss under targeted delegate failures rising from 26% under sealed formation to 49% under strong herding for single delegation<ref:2607.01730#pg4> Delegation can improve outcomes by routing votes toward better-informed agents, but it concentrates risk<ref:2607.01730#pg5> Ranked multi-delegation combined with a personal fallback ballot turns delegation from a fragile single-point-of-failure structure into a resilient routing mechanism that preserves both participation and outcome accuracy, even under targeted attacks<ref:2607.01730#pg5> The value of sealed formation is not a direct accuracy gain; it reduces power concentration and the resulting vulnerability to targeted delegate failures, and keeps outcomes stable under herding rather than dependent on the arbitrary order in which voters act<ref:2607.01730#pg5> More broadly, the paper argues for a specific timing of transparency rather than for secrecy as a permanent principle<ref:2607.01730#pg5> Future work will extend this work in several directions, including field deployments in live civic platforms or DAOs<ref:2607.01730#pg5> and integrating the mechanism with privacy-preserving decentralized identity systems<ref:2607.01730#pg5> Page 36 --- The gist The proposed mechanism introduces a liquid democracy protocol that combines timed-release cryptography with ranked multi-delegation and fallback ballots to mitigate vulnerabilities like herding, coercion, and delegate failure while maintaining full auditability<ref:2607.01730#pg6> Robustness Against Failure and Herding The system extends standard liquid democracy with ranked multi-delegation and fallback ballots to improve robustness against delegate failure<ref:2607.01730#pg5> Ranked delegation with a personal fallback ballot caps the loss at an arithmetic floor of about 3%, with the residual loss governed by the fraction of delegators who provide fallback ballots<ref:2607.01730#pg4> This robust design keeps vote loss near 3% even when 30% of delegates are targeted, versus 26% for a single-delegate design under targeted attacks<ref:2607.01730#pg5> Delegation can improve outcomes by routing votes toward better-informed agents, but it concentrates risk<ref:2607.01730#pg5> Ranked multi-delegation combined with a personal fallback ballot turns delegation from a fragile single-point-of-failure structure into a resilient routing mechanism that preserves both participation and outcome accuracy, even under targeted attacks<ref:2607.01730#pg5> The value of sealed formation is not a direct accuracy gain; it reduces power concentration and the resulting vulnerability to targeted delegate failures, and keeps outcomes stable under herding rather than dependent on the arbitrary order in which voters act<ref:2607.01730#pg5> More broadly, the paper argues for a specific timing of transparency rather than for secrecy as a permanent principle<ref:2607.01730#pg5> Future work will extend this work in several directions, including field deployments in live civic platforms or DAOs and integrating the mechanism with privacy-preserving decentralized identity systems<ref:2607.

Paper summary: Nadia: So, this paper, "Resilient Liquid Democracy: Mitigating Voting Power Imbalances via Secure Delegation Networks," it’s looking at liquid democracy—where you can vote directly or delegate your power to someone else.

Elias: Exactly. The main problem they point out is that when those delegations are public as they form, it invites herding based on popularity and makes coercion easier to verify, which leaves the whole election fragile if popular delegates just decide to abstain at the last minute.

Priya: From a data perspective, I’m wondering what this means for real-world voting systems. They claim this mechanism keeps voting power dispersed where transparent formation would concentrate it into just a few hands.

Nadia: Right, and they're addressing that by proposing a new liquid democracy mechanism that uses decentralized timed-release encryption while those delegation choices are being formed, but then they reveal them all at the tallying phase.

Elias: That timing of transparency is key; it means the choices are sealed during formation to stop influence from visible counts, but revealed for the final count to keep it auditable. They also add ranked backup delegates and a personal fallback ballot for each voter.

Priya: So, if a voter chooses a delegate who turns out to be unreliable later on, they don't just lose their vote; they have that ranked backup or the personal fallback option to use instead.

Nadia: That’s right, and they’re quantifying the effect of this robustness against failure in their experiments grounded in real voting data. They found that vote loss under targeted delegate failures drops from twenty-six percent down to about three percent if you have that ranked multi-delegation setup with a fallback ballot.

Elias: That drop from twenty-six to three is significant because it shows how this design handles those specific failure scenarios better than older single-delegate systems. The mechanism keeps the vote loss near three percent even when thirty percent of delegates are targeted.

Priya: What about the trade-off? They mentioned that delegation helps only when abstention is large and systematically unrepresentative and enough abstainers use the channel; otherwise, it’s at best neutral or can actually hurt the outcome.

Nadia: That’s a fair caveat, Priya; it means delegation isn't a magic fix for every situation. It really works when you have those specific conditions for abstention happening.

Elias: And they also looked at the opposite effect; transparent formation concentrates voting power in proportion to how popular the delegates are, which can lead to vote loss under targeted delegate failures rising from twenty-six percent up to forty-nine percent when there is strong herding for single delegation.

Priya: So, it seems like this paper isn't just about making things safer; it’s also about showing how different formation styles affect the final outcome stability.

Nadia: Exactly, and the core of their work is proving two formal security guarantees: pre-reveal secrecy and resubmission receipt-freeness for the formation phase of this liquid democracy protocol.

Elias: Pre-reveal secrecy means no efficient adversary learns anything about how a voter delegated or voted before that reveal time Tr, which suppresses popularity-driven herding and removes the evidence for bribery during formation.

Priya: And receipt-freeness means that even if someone forces a voter to hand over a proof of delegation, that person still can't change it and cast their true preference before Tr. That defeats the coercion of proving how you voted.

Nadia: The implication for us is that we can use liquid democracy in ways that are more resilient against manipulation during the setup phase, provided we manage the timing of when information becomes public.

Elias: This paper, "Resilient Liquid Democracy: Mitigating Voting Power Imbalances via Secure Delegation Networks," is essentially arguing for a specific timing of transparency rather than making secrecy a permanent rule.

Priya: It shifts the focus from hiding everything to finding that sweet spot where you get the benefit of delegation without letting popularity drive the whole system into chaos.

Nadia: So, to wrap up, this work shows how combining ranked multi-delegation with a personal fallback ballot turns delegation into a more resilient routing network.

Elias: It preserves both participation and outcome accuracy even under targeted attacks because it reduces power concentration and the resulting vulnerability to delegate failures.

Priya: The real contribution seems to be showing that the benefit of delegation isn't guaranteed; it depends entirely on how large and unrepresentative the abstention is when you use this specific protocol.

Conclusion: Nadia: So we’re wrapping up on this paper, "Resilient Liquid Democracy: Mitigating Voting Power Imbalances via Secure Delegation Networks."

Elias: It really boils down to taking liquid democracy and making it tougher against manipulation and failure by changing how delegation choices are revealed.

Priya: The authors are looking at how to keep the voting power dispersed instead of letting it get concentrated in a few hands when people delegate their votes.

Nadia: They introduce this new method using timed-release encryption during the formation phase, but then they reveal everything for the final tallying so it’s still auditable.

Elias: The authors claim this design provides two formal security guarantees: pre-reveal secrecy and resubmission receipt-freeness.

Priya: Pre-reveal secrecy means no one learns how a voter delegated or voted before the reveal time Tr, which is supposed to stop popularity from driving herding.

Nadia: And receipt-freeness means even if someone tries to force a voter to give them proof of delegation, that person can't change it before Tr, so coercion doesn't work.

Elias: The main point they drive home is that ranked multi-delegation combined with a personal fallback ballot turns delegation from this fragile single-point-of-failure structure into something more resilient.

Priya: They show in experiments that this setup caps vote loss at about three percent, even when thirty percent of delegates are targeted by bad actors.

Nadia: It’s a significant drop from the twenty-six percent loss we saw in single-delegate designs under targeted attacks.

Elias: The authors also noted that transparent formation can concentrate power based on herding strength, which leads to vote loss jumping up to forty-nine percent under strong herding for single delegation.

Priya: So the value here isn't just about making things secret; it’s about reducing that power concentration and the resulting vulnerability to targeted failures.

Nadia: They argue for a specific timing of transparency rather than trying to make secrecy permanent across the board.

Elias: The implication is that delegation can still improve outcomes by routing votes, but we have to manage the risk introduced by how we handle those delegation choices during the process.

Priya: We need to figure out if this level of robustness is practical for real civic platforms or decentralized autonomous organizations where these systems are actually deployed.

Zhuolun Li, Evangelos Pournaras

University of Leeds

cs.CR, cs.CY

Submitted: 2026-07-02

Updated: 2026-10-19

Project page: https://gpapasot.github.io/ld-comsoc/paper.pdf

License: http://creativecommons.org/licenses/by/4.0/

The gist: The gist The proposed mechanism introduces a liquid democracy protocol that combines timed-release cryptography with ranked multi-delegation and fallback ballots to mitigate vulnerabilities like

Key concepts

Timed-Release Encryption (TRE)
This cryptographic scheme is used to seal delegation choices during the voting phase. It encrypts the policy tuple using a time parameter (Tr), meaning the delegation decisions remain secret until a specific reveal time, preventing external parties from observing or influencing delegation choices in real-time.
Ranked Multi-Delegation
Instead of choosing one delegate, voters select multiple backup delegates ranked by preference. This transforms delegation into a resilient routing system rather than a single point of failure. It allows the system to route voting power through several paths, making it harder for targeted failures to silence a voter.
Fallback Ballot
Each voter is required to name ranked backup delegates along with a personal fallback ballot. This ensures that even if the primary delegation fails or is compromised, the voter still has an option to cast their vote directly or through a pre-selected alternative, capping potential loss at an arithmetic floor of about 3%.

Terminology

Summary

The gist The proposed mechanism introduces a liquid democracy protocol that combines timed-release cryptography with ranked multi-delegation and fallback ballots to mitigate vulnerabilities like herding, coercion, and delegate failure while maintaining full auditability<ref:2607.01730#pg6>

How it works

The system separates delegation formation during the voting phase from delegation resolution during the reveal phase to enforce secrecy during formation and transparency during the latter<ref:2607.01730#pg5> Delegation choices are sealed with decentralized timed-release encryption while they are being formed and revealed in full for tallying, and each voter names ranked backup delegates together with a personal fallback ballot so that delegate failures do not silence them<ref:2607.01730#pg4>

The protocol involves a submission phase where each voter encrypts their policy tuple using the Timed-Release Encryption (TRE) scheme parameterized by the reveal time Tr<ref:2607.01730#pg5> This results in a ciphertext Ci = TREncTr(Πi; ri), which is posted to a public ledger before the submission deadline<ref:2607.01730#pg5> The system hides delegation decisions only during the phase in which voters are still making them, preventing influence from visible delegation counts and contemporaneous proof of how a voter delegated<ref:2607.01730#pg5>

Security Guarantees

The design provides two formal security guarantees: pre-reveal secrecy and resubmission receipt-freeness<ref:2607.01730#pg6> Pre-reveal secrecy ensures that no efficient adversary learns anything about how a voter delegated or voted before the reveal time Tr, which suppresses popularity-driven herding and removes the evidentiary basis for bribery during formation<ref:2607.01730#pg5> Receipt-freeness via resubmission means that even a voter forced to hand over a “receipt” can still cast a different, true policy before Tr, defeating coercion of the form “prove to me how you voted”<ref:2607.01730#pg6>

Robustness Against Failure and Herding

The system extends standard liquid democracy with ranked multi-delegation and fallback ballots to improve robustness against delegate failure<ref:2607.01730#pg5> The ranked delegation mechanism transforms delegation from a single-point-of-failure structure into a more resilient routing system<ref:2607.01730#pg5> Ranked delegation with a personal fallback ballot caps the loss at an arithmetic floor of about 3%, with the residual loss governed by the fraction of delegators who provide fallback ballots<ref:2607.01730#pg4> This robust design keeps vote loss near 3% even when 30% of delegates are targeted, versus 26% for a single-delegate design under targeted attacks<ref:2607.01730#pg5>

Empirical Findings

Experiments grounded in real voting data show that the mechanism keeps voting power dispersed where transparent formation concentrates it into few hands, keeps outcomes stable under herding where transparent outcomes become a lottery on arrival order, and cuts vote loss under targeted delegate failures from 26% to about 3%<ref:2607.01730#pg4> Delegation helps only when abstention is large and systematically unrepresentative and enough abstainers use the channel; otherwise it is at best neutral and can harm the outcome<ref:2607.01730#pg4> The sealed formation produces more decentralized and robust delegation network, while transparent formation concentrates voting power in proportion to the herding strength, which translates into vote loss under targeted delegate failures rising from 26% under sealed formation to 49% under strong herding for single delegation<ref:2607.01730#pg4>

Conclusion

Delegation can improve outcomes by routing votes toward better-informed agents, but it concentrates risk<ref:2607.01730#pg5> Ranked multi-delegation combined with a personal fallback ballot turns delegation from a fragile single-point-of-failure structure into a resilient routing mechanism that preserves both participation and outcome accuracy, even under targeted attacks<ref:2607.01730#pg5> The value of sealed formation is not a direct accuracy gain; it reduces power concentration and the resulting vulnerability to targeted delegate failures, and keeps outcomes stable under herding rather than dependent on the arbitrary order in which voters act<ref:2607.01730#pg5> More broadly, the paper argues for a specific timing of transparency rather than for secrecy as a permanent principle<ref:2607.01730#pg5> Future work will extend this work in several directions, including field deployments in live civic platforms or DAOs<ref:2607.01730#pg5> and integrating the mechanism with privacy-preserving decentralized identity systems<ref:2607.01730#pg5>

--- Page 36 ---

The gist The proposed mechanism introduces a liquid democracy protocol that combines timed-release cryptography with ranked multi-delegation and fallback ballots to mitigate vulnerabilities like herding, coercion, and delegate failure while maintaining full auditability<ref:2607.01730#pg6>

Robustness Against Failure and Herding

The system extends standard liquid democracy with ranked multi-delegation and fallback ballots to improve robustness against delegate failure<ref:2607.01730#pg5> Ranked delegation with a personal fallback ballot caps the loss at an arithmetic floor of about 3%, with the residual loss governed by the fraction of delegators who provide fallback ballots<ref:2607.01730#pg4> This robust design keeps vote loss near 3% even when 30% of delegates are targeted, versus 26% for a single-delegate design under targeted attacks<ref:2607.01730#pg5>

Conclusion

Delegation can improve outcomes by routing votes toward better-informed agents, but it concentrates risk<ref:2607.01730#pg5> Ranked multi-delegation combined with a personal fallback ballot turns delegation from a fragile single-point-of-failure structure into a resilient routing mechanism that preserves both participation and outcome accuracy, even under targeted attacks<ref:2607.01730#pg5> The value of sealed formation is not a direct accuracy gain; it reduces power concentration and the resulting vulnerability to targeted delegate failures, and keeps outcomes stable under herding rather than dependent on the arbitrary order in which voters act<ref:2607.01730#pg5> More broadly, the paper argues for a specific timing of transparency rather than for secrecy as a permanent principle<ref:2607.01730#pg5> Future work will extend this work in several directions, including field deployments in live civic platforms or DAOs and integrating the mechanism with privacy-preserving decentralized identity systems<ref:2607.

Improvements for AI systems

  1. A mechanism that prevents popularity-driven herding during delegation formation by using decentralized timed-release encryption, ensuring that delegation choices are sealed with decentralized timed-release encryption while they are being formed. This suppresses influence, as it limits how early or well-known delegates can attract support, which removes the evidentiary basis that bribery and coercion strategies rely on during the formation phase.

  2. A voting system that maintains robustness against delegate failure by implementing ranked multi-delegation and fallback ballots, ensuring votes are preserved even when delegates abstain or delegation cycles occur. This prevents the vulnerability of single-delegate systems, which cause a vote loss that can rise from 26% to about 3% under targeted failures.

  3. A system that optimizes representational accuracy by dynamically adjusting voting rules based on the information regime, as experiments show delegation improves representational accuracy only when abstention is large and systematically unrepresentative. Specifically, using a median representative ballot can increase accuracy to 0.827 under certain conditions, demonstrating how the choice of delegation style moderates outcome quality.

  4. A system that manages power concentration by isolating the effect of transparency on network structure, as it shows Transparent formation concentrates voting power in proportion to the herding strength, which leads to outcomes becoming a lottery on arrival order rather than a stable tally. The sealed regime keeps voting power dispersed, with Sealing removes the signal that drives both effects.

Abstract

Liquid democracy lets voters either vote directly or delegate their voting power to another trusted participant. It is often deployed in political parties, municipal participation platforms, and the governance of decentralized autonomous organizations. Existing approaches make delegations publicly visible as they form, which results in herding driven by popularity, makes coercion verifiable, and leaves the election fragile when popular delegates abstain. We propose a novel liquid democracy mechanism that removes these vulnerabilities, while keeping the tally fully auditable. Delegation choices are sealed with decentralized timed-release encryption, while they are formed and before revealed in full for tallying. Each voter determines ranked backup delegates together with a personal fallback ballot so that delegation failures do not result in abstaining. Experiments grounded in real-world voting data show that the mechanism keeps voting power dispersed, whereas transparent formation concentrates it into few hands. Strikingly, vote loss under targeted delegate failures falls from 26% to 3%.

Sources

Related papers