arXiv 24 Apr 2026 · cs.GT
arXiv:2604.22236 · PDF · DOI · OpenAlex · Extracted main text
Human decision-makers often face choices about complex cases with many potentially relevant features, but limited bandwidth to inspect and integrate all available information. In such settings, we study algorithms that highlight a small subset of case-specific features for human consideration, rather than producing a single prediction or recommendation. We model highlighting as a constrained information policy that selects a small number of features to reveal. A central issue is how humans interpret the algorithm's choice of features: a sophisticated agent correctly conditions on the selection rule, while a naive agent updates only on revealed feature values and treats the selection event as exogenous. We show that optimizing highlighting for sophisticated agents can be computationally intractable, even in simple discrete and binary settings, whereas optimizing for naive agents is tractable as long as the maximal bandwidth is fixed. We also show that a highlighting policy that is optimal for sophisticated agents can perform arbitrarily poorly when deployed to naive agents, motivating robust, implementable alternatives. We illustrate our framework in a calibrated empirical exercise based on the American Housing Survey. Overall, our results establish the value of highlighting a context-specific set of features rather than a fixed one as a practically appealing and computationally feasible tool for achieving human-algorithm complementarity.
appendix boundary found by appendix_command · 82% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Kamenica, Emir and Gentzkow, Matthew (2011) Bayesian persuasion | 0.843 | 3 | 3 | 100% |
| 2 | Aloise, Daniel and Deshpande, Amit and Hansen, Pierre and Popat, Pre… (2009) NP-hardness of Euclidean sum-of-squares clustering | 0.511 | 2 | 2 | 50% |
| 3 | Agarwal, Nikhil and Moehring, Alex and Wolitzky, Alexander (2025) Designing Human-AI Collaboration: A Sufficient-Statistic Approach | 0.405 | 1 | 1 | 100% |
| 4 | Alur, Rohan and Raghavan, Manish and Shah, Devavrat (2024) Human Expertise in Algorithmic Prediction | 0.405 | 1 | 1 | 100% |
| 5 | Angelova, Victoria and Dobbie, Will S. and Yang, Crystal (2023) Algorithmic Recommendations and Human Discretion | 0.405 | 1 | 1 | 100% |
| 6 | Athey, Susan C. and Bryan, Kevin A. and Gans, Joshua S (2020) The Allocation of Decision Authority to Human and Artificial Intelligence | 0.405 | 1 | 1 | 100% |
| 7 | Balakrishnan, Maya and Ferreira, Kris Johnson and Tong, Jordan (2025) Human-Algorithm Collaboration with Private Information: Naïve Advice-Weighting Behavior and Mitigation | 0.405 | 1 | 1 | 100% |
| 8 | Bastani, Hamsa and Bastani, Osbert and Sinchaisri, Wichinpong Park (2026) Improving human sequential decision making with reinforcement learning | 0.405 | 1 | 1 | 100% |
| 9 | Bordalo, Pedro and Coffman, Katherine and Gennaioli, Nicola and Shle… (2016) Stereotypes | 0.405 | 1 | 1 | 100% |
| 10 | Boyaci, Tamer and Canyakmaz, Caner and De Véricourt, Francis (2024) Human and machine: The impact of machine input on decision making under cognitive limitations | 0.405 | 1 | 1 | 100% |
Showing the top 10 of 29 scored citations.