12 papers
Vision: Data-Centric Anchoring for Robust and Interpretable Agentic AI
Arun Vignesh Malarkkan, Xinyuan Wang, Yanjie Fu
Agentic AI systems built on large language models fail in two persistent ways that scaling does not fix: they break under distribution shift, and they cannot explain the decisions…
Invariant Reasoning Directions in Latent Trajectories of Language Models
Arun Vignesh Malarkkan, Manan Roy Choudhury, Utkarsh Byahut +3
Latent reasoning models perform multi-step inference directly in hidden-state space, yet the structure of these latent reasoning trajectories remains poorly understood. We show tha…
Causally-Guided Diffusion for Stable Feature Selection
Arun Vignesh Malarkkan, Xinyuan Wang, Kunpeng Liu +2
Feature selection is fundamental to robust data-centric AI, but most existing methods optimize predictive performance under a single data distribution. This often selects spurious…
FinRule-Bench: A Benchmark for Joint Reasoning over Financial Tables and Principles
Arun Vignesh Malarkkan, Manan Roy Choudhury, Guangwei Zhang +4
Large language models (LLMs) are increasingly applied to financial analysis, yet their ability to audit structured financial statements under explicit accounting principles remains…
CAFE: Causally-Guided Automated Feature Engineering with Multi-Agent Reinforcement Learning
Arun Vignesh Malarkkan, Wangyang Ying, Yanjie Fu
Automated feature engineering (AFE) enables AI systems to autonomously construct high-utility representations from raw tabular data. However, existing AFE methods rely on statistic…
Evolving Demonstration Optimization for Chain-of-Thought Feature Transformation
Xinyuan Wang, Kunpeng Liu, Arun Vignesh Malarkkan +1
Feature Transformation (FT) is a core data-centric AI task that improves feature space quality to advance downstream predictive performance. However, discovering effective transfor…
DELTA: Variational Disentangled Learning for Privacy-Preserving Data Reprogramming
Arun Vignesh Malarkkan, Haoyue Bai, Anjali Kaushik +1
In real-world applications, domain data often contains identifiable or sensitive attributes, is subject to strict regulations (e.g., HIPAA, GDPR), and requires explicit data featur…
Distribution Shift Aware Neural Tabular Learning
Wangyang Ying, Nanxu Gong, Dongjie Wang +5
Tabular learning transforms raw features into optimized spaces for downstream tasks, but its effectiveness deteriorates under distribution shifts between training and testing data.…
Causal Graph Profiling via Structural Divergence for Robust Anomaly Detection in Cyber-Physical Systems
Arun Vignesh Malarkkan, Haoyue Bai, Dongjie Wang +1
With the growing complexity of cyberattacks targeting critical infrastructures such as water treatment networks, there is a pressing need for robust anomaly detection strategies th…
Incremental Causal Graph Learning for Online Cyberattack Detection in Cyber-Physical Infrastructures
Arun Vignesh Malarkkan, Dongjie Wang, Haoyue Bai +1
The escalating threat of cyberattacks on real-time critical infrastructures poses serious risks to public safety, demanding detection methods that effectively capture complex syste…
Rethinking Spatio-Temporal Anomaly Detection: A Vision for Causality-Driven Cybersecurity
Arun Vignesh Malarkkan, Haoyue Bai, Xinyuan Wang +3
As cyber-physical systems grow increasingly interconnected and spatially distributed, ensuring their resilience against evolving cyberattacks has become a critical priority. Spatio…
A Survey on Data-Centric AI: Tabular Learning from Reinforcement Learning and Generative AI Perspective
Wangyang Ying, Cong Wei, Nanxu Gong +7
Tabular data is one of the most widely used data formats across various domains such as bioinformatics, healthcare, and marketing. As artificial intelligence moves towards a data-c…