A No Free Lunch Theorem for Human-AI Collaboration

Authors: Kenny Peng, Nikhil Garg, Jon Kleinberg

AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical Our main result answers this question, mostly in the negative. In fact, Theorem 1 shows that it is difficult to ensure a much lower bar: producing binary classifications that are always at least as accurate as the worst individual agent. [...] To prove Theorem 1 (which we do in Section 4)
Researcher Affiliation Academia 1Cornell Tech 2Cornell University
Pseudocode No The paper contains mathematical definitions, propositions, and proofs. There are no structured pseudocode or algorithm blocks explicitly labeled or formatted as such.
Open Source Code No The paper does not contain any explicit statements about releasing source code or links to a code repository for the methodology described.
Open Datasets No The paper defines a theoretical 'collaboration setting' for its proofs but does not use or provide access information for any publicly available or open datasets for empirical evaluation.
Dataset Splits No The paper does not conduct empirical experiments or use specific datasets; therefore, no dataset splits are provided.
Hardware Specification No The paper is theoretical and does not describe any experiments; therefore, no specific hardware specifications are mentioned.
Software Dependencies No The paper is theoretical and does not describe any experiments; therefore, no software dependencies with version numbers are listed.
Experiment Setup No The paper is theoretical and does not describe any experiments; therefore, no experimental setup details or hyperparameters are provided.