Current text-to-image (T2I) models often fail to account for diverse human experiences, leading to misaligned systems. We advocate for pluralism in AI alignment, where an AI understands and is steerable towards diverse, and often conflicting, human values. Our work provides three core contributions to achieve this in T2I models. First, we introduce a novel dataset for Diverse Intersectional Visual Evaluation (DIVE) -- the first multimodal dataset for pluralistic alignment. It enables deep alignment to diverse safety perspectives through a large pool of demographically intersectional human raters who provided extensive feedback across 1000 prompts, with high replication, capturing nuanced safety perceptions. Second, we empirically confirm demographics as a crucial proxy for diverse viewpoints in this domain, revealing significant, context-dependent differences in harm perception that diverge from conventional evaluations. Finally, we discuss implications for building aligned T2I models, including efficient data collection strategies, LLM judgment capabilities, and model steerability towards diverse perspectives. This research offers foundational tools for more equitable and aligned T2I systems.Content Warning: The paper includes sensitive content that may be harmful.

In recent years, we have witnessed a rapid growth in the deployment of Machine Learning (ML) models in complex real-world settings.ML models are being used to support decision-making across a wide range of domains, including healthcare [2,19,22,34], credit lending [5,14], criminal justice [1, 13], and employment [11,20].For example, in the criminal justice system, algorithmic recidivism risk scores inform pre-trial bail decisions for defendants [1].In credit lending, lenders routinely use credit-scoring models to assess the risk of default by applicants [14].The excitement around modern ML systems facilitating high-stakes decisions is fueled by the promise of these technologies to tap into large datasets, mine relevant statistical patterns within them, and utilize those patterns to make more accurate predictions at a lower cost and without suffering from the same cognitive biases and limitation as human decision-makers.Growing evidence, however, suggests that ML models are vulnerable to various biases [1], instability [8], and opaqueness [4].These observations have led to calls to preserve human involvement in high-stakes decision-making systems-with the hope of combining and amplifying the respective strengths of human cognition and ML models through carefully designed hybrid decision-making systems.Such systems consist of ML models and human experts jointly making decisions, and they are common in practice-including in the domains mentioned above.Researchers have proposed and tested various hybrid human-ML designs which vary, for instance, in the way decision-making power is distributed between humans and machines [6,10,16,30,31].However, empirical findings regarding the success and effectiveness of these proposals are mixed [15, and references therein].Simultaneously, a growing body of theoretical work has attempted to conceptualize and formalize these hybrid designs [3,9] and study optimal ways of aggregating human and ML judgments within

Large language models (LLMs) are increasingly becoming all-powerful and pervasive via deployment in sociotechnical systems. Yet these language models, be it for classification or generation, have been shown to be biased, behave irresponsibly, causing harm to people at scale. It is crucial to audit these language models rigorously before deployment. Existing auditing tools use either or both humans and AI to find failures. In this work, we draw upon literature in human-AI collaboration and sensemaking, and interview research experts in safe and fair AI, to build upon the auditing tool: AdaTest [36], which is powered by a generative LLM. Through the design process we highlight the importance of sensemaking and human-AI communication to leverage complementary strengths of humans and generative models in collaborative auditing. To evaluate the effectiveness of AdaTest++, the augmented tool, we conduct user studies with participants auditing two commercial language models: OpenAI’s GPT-3 and Azure’s sentiment analysis model. Qualitative analysis shows that AdaTest++ effectively leverages human strengths such as schematization, hypothesis testing. Further, with our tool, users identified a variety of failures modes, covering 26 different topics over 2 tasks, that have been shown in formal audits and also those previously under-reported.

Machine learning research has long focused on models rather than datasets, and prominent datasets are used for common ML tasks without regard to the breadth, difficulty, and faithfulness of the underlying problems. Neglecting the fundamental importance of data has given rise to inaccuracy, bias, and fragility in real-world applications, and research is hindered by saturation across existing dataset benchmarks. In response, we present DataPerf, a community-led benchmark suite for evaluating ML datasets and data-centric algorithms. We aim to foster innovation in data-centric AI through competition, comparability, and reproducibility. We enable the ML community to iterate on datasets, instead of just architectures, and we provide an open, online platform with multiple rounds of challenges to support this iterative development. The first iteration of DataPerf contains five benchmarks covering a wide spectrum of data-centric techniques, tasks, and modalities in vision, speech, acquisition, debugging, and diffusion prompting, and we support hosting new contributed benchmarks from the community. The benchmarks, online evaluation platform, and baseline implementations are open source, and the MLCommons Association will maintain DataPerf to ensure long-term benefits to academia and industry.

A number of applications including crowd-sourced labeling and peer review require aggregation of labels or evaluations sourced from multiple evaluators. There is often additional information available pertaining to the evaluators’ expertise. A natural approach for aggregation is to consider the widely studied Dawid-Skene model (or its extensions incorporating evaluators’ expertise), and employ the standard maximum likelihood estimator (MLE). While MLE is in general widely used in practice and enjoys a number of appealing theoretical guarantees, in this work we provide a negative result for the MLE. Specifically, we prove that the MLE is asymptotically inadmissible for a special case of evaluation aggregation with expertise level information. We show this by constructing an alternative estimator that we show is significantly better than the MLE in certain parameter regimes and at least as good elsewhere. Finally, simulations reveal that our findings may hold in more general conditions than what we theoretically analyze.

A number of applications require two-sample testing on ranked preference data. For instance, in crowdsourcing, there is a long-standing question of whether pairwise-comparison data provided by people is distributed identically to ratings-converted-to-comparisons. Other applications include sports data analysis and peer grading. In this paper, we design twosample tests for pairwise-comparison data and ranking data. For our two-sample test for pairwise-comparison data, we establish an upper bound on the sample complexity required to correctly test whether the distributions of the two sets of samples are identical. Our test requires essentially no assumptions on the distributions. We then prove complementary lower bounds showing that our results are tight (in the minimax sense) up to constant factors. We investigate the role of modeling assumptions by proving lower bounds for a range of pairwise-comparison models (WST, MST, SST, parameter-based such as BTL and Thurstone). We also provide tests and associated sample complexity bounds for partial (or total) ranking data. Furthermore, we empirically evaluate our results via extensive simulations as well as three real-world data sets consisting of pairwise-comparisons and rankings. By applying our two-sample test on real-world pairwise-comparison data, we conclude that ratings and rankings provided by people are indeed distributed differently.

Several strands of research have aimed to bridge the gap between artificial intelligence (AI) and human decision-makers in AI-assisted decision-making, where humans are the consumers of AI model predictions and the ultimate decision-makers in high-stakes applications. However, people's perception and understanding are often distorted by their cognitive biases, such as confirmation bias, anchoring bias, availability bias, to name a few. In this work, we use knowledge from the field of cognitive science to account for cognitive biases in the human-AI collaborative decision-making setting, and mitigate their negative effects on collaborative performance. To this end, we mathematically model cognitive biases and provide a general framework through which researchers and practitioners can understand the interplay between cognitive biases and human-AI accuracy. We then focus specifically on anchoring bias, a bias commonly encountered in human-AI collaboration. We implement a time-based de-anchoring strategy and conduct our first user experiment that validates its effectiveness in human-AI collaborative decision-making. With this result, we design a time allocation strategy for a resource-constrained setting that achieves optimal human-AI collaboration under some assumptions. We, then, conduct a second user experiment which shows that our time allocation strategy with explanation can effectively de-anchor the human and improve collaborative performance when the AI model has low confidence and is incorrect.

A number of applications require two-sample testing of pairwise comparison data. For instance, in crowdsourcing, there is a long-standing question of whether comparison data provided by people is distributed similar to ratings-converted-to-comparisons. Other examples include sports data analysis and peer grading. In this paper, we design a two-sample test for pairwise comparison data. We establish an upper bound on the sample complexity required to correctly distinguish between the distributions of the two sets of samples. Our test requires essentially no assumptions on the distributions. We then prove complementary information-theoretic lower bounds showing that our results are tight (in the minimax sense) up to constant factors. We also investigate the role of modeling assumptions by proving information-theoretic lower bounds for a range of pairwise comparison models (WST, MST, SST, parameter-based such as BTL and Thurstone).

This paper proposes a new approach to construct high quality space-filling sample designs. First, we propose a novel technique to quantify the space-filling property and optimally trade-off uniformity and randomness in sample designs in arbitrary dimensions. Second, we connect the proposed metric (defined in the spatial domain) to the quality metric of the design performance (defined in the spectral domain). This connection serves as an analytic framework for evaluating the qualitative properties of space-filling designs in general. Using the theoretical insights provided by this spatial-spectral analysis, we derive the notion of optimal space-filling designs, which we refer to as space-filling spectral designs. Third, we propose an efficient estimator to evaluate the space-filling properties of sample designs in arbitrary dimensions and use it to develop an optimization framework for generating high quality space-filling designs. Finally, we carry out a detailed performance comparison on two different applications in varying dimensions: a) image reconstruction and b) surrogate modeling for several benchmark optimization functions and a physics simulation code for inertial confinement fusion (ICF). Our results clearly evidence the superiority of the proposed space-filling designs over existing approaches, particularly in high dimensions.