LLMs struggle with financial reasoning but can outperform the market with human oversight.
problem Financial reasoning failures in LLM-generated stock market predictions.
method Evaluated four LLMs using three prompting strategies and compared to human oversight.
result LLMs require human oversight to fully realize their potential in financial markets.
Machine learning uses crowdworkers; determining their status as human subjects is tricky.
problem Determining the appropriate status of ML crowdworkers as human subjects.
method Investigation of natural language processing studies to expose challenges and propose solutions.
result Potential loophole in the U.S. Common Rule for ML research oversight.
AI agents manage portfolios, improving on human oversight.
problem Improving strategic asset allocation for institutional investors.
method 50 specialized agents produce capital market assumptions, construct portfolios, critique, and vote on each other's output.
result Meta-agent compares forecasts with realized returns and improves agent performance.
Paper introduces PHI to identify structurally distinct payment patterns in UK municipal procurement.
problem Vulnerability of public procurement to error, fraud, and corruption in high-volume transactions.
method Introduces Payment Heterogeneity Index (PHI) using Gaussian Mixture Model (GMM) and non-parametric statistics.
result Identifies a significant cohort with structurally distinct payment patterns, improving procurement oversight.
This paper proposes a method to use LLMs as auxiliary evaluators in place of human judges.
problem The need for cost-effective and scalable evaluation of AI systems.
method Formulates a two-stage sampling design with LLM evaluations and human ratings, using a doubly robust estimator.
result Proposes a method to determine optimal sample sizes for human and LLM ratings.
The paper studies how to allocate human validation in AI-assisted tasks to minimize errors.
problem Heterogeneous reliability of AI-generated signals across tasks, products, and customer segments.
method Tuned prediction-powered inference, upper confidence bounds policy, Neyman square-root rule.
result The proposed policy outperforms uniform and epsilon-greedy allocation, closing most of the gap to the oracle when reliability is heterogeneous.
For sophisticated reinforcement learning (RL) systems to interact usefully with real-world environments, we need to communicate complex goals to these systems. In this work, we explore goals defined in terms of (non-expert) human preferences between pairs of trajectory segments. We show that this approach can effective…
Survey of alignment techniques for large language models.
problem Ensuring large language models align with human values.
method Analysis of diverse alignment methods and training paradigms.
result Preference-based methods offer more flexibility for nuanced alignment.
LR-Robot accelerates SLRs by combining expert oversight and AI, revealing trends and patterns in financial research.
problem Manual SLRs are impractical due to the scale and complexity of modern financial research.
method Domain experts define taxonomies and constraints, LLMs execute classification, and human evaluation ensures reliability.
result AI can understand and synthesize literature, revealing trends and core research directions.
AI systems that explain their decisions can be monitored for harmful intentions.
problem Monitoring AI systems' decision-making processes for harmful intentions is imperfect and can miss some misbehavior.
method Monitoring the chain of thought (CoT) of AI systems that communicate in human language.
result CoT monitoring is a promising but fragile approach to AI safety.
LR-Robot automates SLRs with AI, expert oversight, and multidimensional analysis.
problem Efficient but contextually limited outputs from existing SLR frameworks.
method Human-in-the-loop process, structured knowledge sources, retrieval-augmented generation.
result Empirical demonstration of AI-driven literature synthesis in option pricing.
The 2007--2008 financial crisis has paved the way for the use of macroprudential policies in supervising the financial system as a whole. This paper views macroprudential oversight in Europe as a process, a sequence of activities with the ultimate aim of safeguarding financial stability. To conceptualize a process in t…
As machine learning systems move from computer-science laboratories into the open world, their accountability becomes a high priority problem. Accountability requires deep understanding of system behavior and its failures. Current evaluation methods such as single-score error metrics and confusion matrices provide aggr…
Interpretable classification models are built with the purpose of providing a comprehensible description of the decision logic to an external oversight agent. When considered in isolation, a decision tree, a set of classification rules, or a linear model, are widely recognized as human-interpretable. However, such mode…
AutoML systems are currently rising in popularity, as they can build powerful models without human oversight. They often combine techniques from many different sub-fields of machine learning in order to find a model or set of models that optimize a user-supplied criterion, such as predictive performance. The ultimate g…
Study shows GPT's earnings forecasts are human-like but not always accurate.
problem Information friction in AI-generated financial analysis.
method Examined GPT's earnings forecasts following corporate earnings releases and proposed a diagnostic framework.
result GPT's narrative attention is consistent and human-like but not always associated with higher forecast accuracy.
The rapid growth of text data has motivated the development of machine-learning based automatic text summarization strategies that concisely capture the essential ideas in a larger text. This study aimed to devise an extractive summarization method for A-133 Single Audits, which assess if recipients of federal grants a…
We present a formal model of human decision-making in explore-exploit tasks using the context of multi-armed bandit problems, where the decision-maker must choose among multiple options with uncertain rewards. We address the standard multi-armed bandit problem, the multi-armed bandit problem with transition costs, and …
This work tackles real-world robotic reinforcement learning challenges.
problem Limited success of reinforcement learning in real-world robotics.
method Proposes a system for autonomous real-world learning without instrumentation.
result Demonstrates a complete system that learns without human intervention.
STOOD-X detects out-of-distribution samples without distributional assumptions and provides explainable visualizations.
problem Challenges in OOD detection, including restrictive assumptions, scalability issues, and lack of interpretability.
method Two-stage methodology combining statistical nonparametric test and explainability enhancements.
result Achieves competitive performance in high-dimensional and complex settings, with explainability framework enabling human oversight.
We prove a structure theorem for closed topological manifolds of cohomogeneity one; this result corrects an oversight in the literature. We complete the equivariant classification of closed, simply connected cohomogeneity one topological manifolds in dimensions 5, 6, and 7 and obtain topological characterizations…
This thesis tackles bias in AI decision-making in banking.
problem Bias in AI-driven banking decisions.
method Understanding, mitigating, and accounting for bias in AI systems.
result Establishment of Responsible AI practices for fair decision-making.
Preterm birth is the most common cause of neonatal death. Current diagnostic methods that assess the risk of preterm birth involve the collection of maternal characteristics and transvaginal ultrasound imaging conducted in the first and second trimester of pregnancy. Analysis of the ultrasound data is based on visual i…
The paper assesses fairness in risk score models, focusing on epistemic value.
problem Fairness of risk score models in communicating uncertainty.
method Identified key fairness desiderata, developed metrics for quantitative assessment, and applied methodology in two case studies.
result Introduced a novel calibration error metric for meaningful comparisons between groups of different sizes.
This paper discusses the role of risk communication in macroprudential oversight and of visualization in risk communication. Beyond the soar in data availability and precision, the transition from firm-centric to system-wide supervision imposes vast data needs. Moreover, except for internal communication as in any orga…
AI models forget statistics' lesson: correlation doesn't imply causation.
problem AI models often produce flawed causal models due to ignoring correlation vs causation.
method Demonstrates examples of flawed AI models and proposes rethinking core models.
result Current efforts to make AI models ethical are insufficient.
Survey on finite group actions on CW-complexes homotopy to spheres.
problem Understanding finite group actions on CW-complexes homotopy equivalent to spheres.
method Survey of extensive literature on finite G-CW-complexes homotopy equivalent to spheres. result Finite G-CW-complexes homotopy equivalent to spheres have finite group actions. Unified AI detection framework for various artifacts.
problem Effective oversight and regulation of AI deployment.
method Unified detection framework based on Mahalanobis distance scores (MDS).
result Efficient and robust estimation of covariance matrix for positive samples.
No policy can simultaneously be fully autonomous, optimally calibrated, and helpful, proving a trilemma.
problem Proving impossibility of a policy achieving maximum helpfulness, optimal calibration, and full autonomy.
method Geometric proof showing that adding any non-affine autonomy incentive to a strictly proper scoring rule destroys strict properness.
result The Behavioral Credibility Trilemma: no policy can achieve all three goals simultaneously.
This paper uses deep learning to detect money laundering in cross-border transactions.
problem Detecting money laundering in cross-border transactions is challenging due to complexity and scale.
method Unsupervised learning models, including CNNs and hybrid CNNGRU architectures, were tested for anomaly detection.
result Hybrid Convolutional-Recurrent Neural Integration Model (CRNIM) showed superior performance.
Digital money could reduce germ spread during coronavirus.
problem Spreading of germs via paper money during coronavirus.
method Policy recommendations for mobile wallets, digital currencies, and data protection.
result Adopting digital money can help reduce germ spread.
Benchmark assesses LLMs' causal inference skills, revealing significant limitations.
problem Lack of rigorous evaluation of LLMs' causal inference capabilities.
method CausalPitfalls benchmark with structured challenges and grading rubrics.
result Significant limitations in current LLMs' statistical causal inference.
Model proposes how regulators should oversee complex algorithms in high-stakes applications.
problem Regulating complex algorithms used in high-stakes applications like lending, testing, and hiring.
method Proposes a model where regulators are limited in learning about complex algorithms with misaligned preferences, and explores different regulatory approaches.
result Complex algorithms can improve welfare, but regulation should focus on the source of incentive misalignment for optimal results.
Algorithmic insurance tackles financial risks from AI errors, proving CVaR-optimal thresholds reduce tail risk.
problem High-stakes AI errors lead to heterogeneous losses, challenging traditional insurance assumptions.
method Analyzed binary classification performance to tail risk exposure, using CVaR to quantify extreme losses.
result CVaR-optimal thresholds reduce tail risk up to 13-fold compared to accuracy maximization.
Classical stochastic gradient methods for optimization rely on noisy gradient approximations that become progressively less accurate as iterates approach a solution. The large noise and small signal in the resulting gradients makes it difficult to use them for adaptive stepsize selection and automatic stopping. We prop…
The proliferation of large data sets and Bayesian inference techniques motivates demand for better data sparsification. Coresets provide a principled way of summarizing a large dataset via a smaller one that is guaranteed to match the performance of the full data set on specific problems. Classical coresets, however, n…
No fair and strategy-proof automated market maker exists for more than two assets.
problem Designing a fair and strategy-proof automated market maker for multiple assets.
method Analyzing the weighted-product family of aggregation rules and their properties.
result No aggregation rule is both fair and strategy-proof for more than two assets.
Machine Learning (ML) is increasingly being used for computer aided diagnosis of brain related disorders based on structural magnetic resonance imaging (MRI) data. Most of such work employs biologically and medically meaningful hand-crafted features calculated from different regions of the brain. The construction of su…
LSBI approximates likelihood with linear functions for cosmological parameter estimation.
problem Estimating cosmological parameters from complex data.
method Sequential Linear Simulation-based Inference (LSBI) using Gaussian approximations.
result LSBI achieves convergence after 4-5 rounds of simulations, comparable to neural methods.
Reinforcement learning is a promising approach to developing hard-to-engineer adaptive solutions for complex and diverse robotic tasks. However, learning with real-world robots is often unreliable and difficult, which resulted in their low adoption in reinforcement learning research. This difficulty is worsened by the …
We embed arbitrary groups into regular graphs with prescribed automorphisms.
problem Embedding arbitrary groups into regular graphs with specific automorphisms.
method Constructing regular graphs with strong embeddings and automorphism groups isomorphic to any given finite group.
result For every d≥3 and every finite group G, there exists a d-regular graph Γ with a strong embedding β such that Aut(Γ)≅Aut(β(Γ))≅G. New framework detects crypto wash trading using liquidity measures.
problem Detecting and monitoring wash trading in crypto assets.
method Developed a new framework to detect wash trading through real-time liquidity fluctuation measures.
result Joint elevation in liquidity jump and diffusion indicates wash trading in crypto assets.
TraCeR uses transformers to analyze survival data with longitudinal covariates.
problem Handling longitudinal covariates and assessing model calibration in survival analysis.
method Transformer-based survival analysis framework with factorized self-attention architecture.
result TraCeR achieves significant performance improvements over state-of-the-art methods.
Study causal effects on humans in mixed human-AI systems with unobserved unit types.
problem Estimating causal effects on humans in systems with unobserved unit types and interaction networks.
method Assumed human-AI prior, causal message passing (CMP) framework, subpopulation analysis.
result Consistently recover human-specific causal effects using subpopulations with varying expected human composition and treatment exposure.
Hybrid Bayesian-conformal framework improves uncertainty quantification in healthcare predictions.
problem Jointly satisfying distribution-free coverage guarantees and risk-adaptive precision in clinical decision-making.
method Integrates Bayesian hierarchical random forests with group-aware conformal calibration, using posterior uncertainties to weight conformity scores.
result Achieves target coverage (94.3% vs 95% target) with adaptive precision, 21% narrower intervals for low-uncertainty cases.
When machine predictors can achieve higher performance than the human decision-makers they support, improving the performance of human decision-makers is often conflated with improving machine accuracy. Here we propose a framework to directly support human decision-making, in which the role of machines is to reframe pr…
Conformal prediction sets improve human decision making by quantifying model uncertainty.
problem Humans signal uncertainty and offer alternatives when unsure, but machine learning models often lack this feature.
method Conducted a randomized controlled trial with human subjects given conformal prediction sets.
result Human accuracy improves when given conformal prediction sets compared to fixed-size prediction sets.
This paper argues for more realistic human models in RL.
problem Current RL models oversimplify human feedback, ignoring personal, contextual, and dynamic aspects.
method Calls for interdisciplinary research on human feedback in RL.
result Realistic human models are needed for robust human-in-the-loop RL systems.