Align When They Want, Complement When They Need! Human-Centered Ensembles for Adaptive Human-AI Collaboration

TL;DR

A novel human-centered adaptive AI ensemble that strategically toggles between two specialist AI models - the aligned model and the complementary model - based on contextual cues is introduced, using an elegantly simple yet provably near-optimal Rational Routing Shortcut mechanism.

Abstract

In human-AI decision making, designing AI that complements human expertise has been a natural strategy to enhance human-AI collaboration, yet it often comes at the cost of decreased AI performance in areas of human strengths. This can inadvertently erode human trust and cause them to ignore AI advice precisely when it is most needed. Conversely, an aligned AI fosters trust yet risks reinforcing suboptimal human behavior and lowering human-AI team performance. In this paper, we start by identifying this fundamental tension between performance-boosting (i.e., complementarity) and trust-building (i.e., alignment) as an inherent limitation of the traditional approach for training a single AI model to assist human decision making. To overcome this, we introduce a novel human-centered adaptive AI ensemble that strategically toggles between two specialist AI models - the aligned model and the complementary model - based on contextual cues, using an elegantly simple yet provably near-optimal Rational Routing Shortcut mechanism. Comprehensive theoretical analyses elucidate why the adaptive AI ensemble is effective and when it yields maximum benefits. Moreover, experiments on both simulated and real-world data show that when humans are assisted by the adaptive AI ensemble in decision making, they can achieve significantly higher performance than when they are assisted by single AI models that are trained to either optimize for their independent performance or even the human-AI team performance.

Figures (8)

• Figure 1: Summary comparison of human-AI team performance across AI design paradigms. Standard AI optimizes independent accuracy; aligned and complementary AIs specialize in trust-building and error-correction, respectively; behavior-aware AI optimizes team loss under our human interaction model. Our proposed adaptive AI ensemble, including the more practical one using Rational Routing Shortcut (RRS), strategically toggles between aligned and complementary AI, and achieves the highest team accuracy. Results shown here correspond to the behavior-grounded evaluation on WoofNette data.
• Figure 2: Complementarity-alignment tradeoff increases sharply with specialization, exacerbated by imperfect human DM in alignment region ($\alpha$), illustrating that a single model cannot simultaneously optimize for trust and performance.
• Figure 3: Accuracy gain ($\Gamma_\text{team}$) of adaptive AI ensemble over single AI on the College Admissions data. The plots (left to right) empirically validate that: (i) gain increases with specialist divergence ($\|\theta_{m_a}^* - \theta_{m_c}^* \|$); (ii) gain scales with human accuracy ($\alpha$) in alignment region (i.e., the aligned group's feature regime), reflecting the reliability factor $\kappa$; (iii) gain peaks when the task mixture is balanced (i.e., aligned group fraction $p \approx 0.5$), matching the concave $p(1-p)$ dependence; and (iv) gain increases linearly with group certainty ($1 - \mathcal{H} / 2\log 2$), demonstrating robustness to routing uncertainty.
• Figure A1: Samples from College Admission dataset with $p=0.5$ and $\delta = 0.25$.
• Figure A2: Visualization of decision boundary of standard and specialist logistic regression models on College Admission dataset with $p=0.5$ and $\delta = 0.25$.

Theorems & Definitions (13)

• Lemma 1: Alignment Loss Sensitivity
• Theorem 2: Complementarity-Alignment Tradeoff
• Theorem 3: Near-Oracle Guarantee for RRS
• Theorem 4: Adaptive AI Performance Gain
• Proposition 5
• Corollary 6: Adaptive AI Performance Gain Under Uncertainty
• Lemma 7: Scalar logistic curvature
• proof
• Lemma 8: Hessian eigenvalue bounds for logistic loss
• proof