L-001 L-003

Candidate Resignation Monotonicity in Approval-Based Committee Elections

Source: cs.GT updates on arXiv.org — https://arxiv.org/abs/2608.06156 Date read: 2026-09-02 Connected to: L-001, L-003 Kind: content Escalation: store-only Escalation rationale:

What this is

A computational social choice paper that formalizes "resignation monotonicity"—a structural property of voting rules governing committee selection. It demonstrates that many standard approval-based committee (ABC) election rules fail this property and connects the failure to prior work on dynamic candidate sets. This is a technical result in mechanism design, not a primary theoretical or empirical challenge to protocol laws.

What I took from it

The paper is competent but narrow in scope: it identifies a formal inconsistency in existing voting rules (failure to preserve elected members when others resign) and proves which rules satisfy or violate the property. The contribution is primarily negative (showing failure) rather than mechanistic.

The connection to L-001 (ossification under adoption) is tangential—the paper does not examine why adopted rules resist modification when their properties prove problematic; it only catalogs which properties they fail. Similarly, L-003 (formalization ratchet) is not substantially engaged: the paper does not trace how informal committee practices become formalized under pressure, nor does it investigate whether formalization itself drives the failures observed. The work is a domain-specific constraint study, not a law investigation.

Research connections

  • L-001: Tangential. The paper shows that adopted ABC rules fail resignation monotonicity, but does not investigate whether adoption resistance prevents correction of known flaws—the mechanism behind ossification.
  • L-003: Tangential. The paper formalizes voting rules but does not examine how this formalization changes the coordination substrate or drives unintended protocol rigidity.
  • none (other seeds)

Seed

Seed title: none

Seed type: —

Seed text: —