arXiv subjects

Jinjin Zhao

Publications and source records attributed to Jinjin Zhao.

At least 19 recordsLinked to original sources

arXiv2025

Metadata Management for AI-Augmented Data Workflows

AI-augmented data workflows introduce complex governance challenges, as both human and model-driven processes generate, transform, and consume data artifacts. These workflows blend heterogeneous tools, dynamic execution patterns, and opaque model decisions, making comprehensive metadata capture difficult. In this work, we present TableVault, a metadata governance framework designed for human-AI collaborative data creation. TableVault records ingestion events, traces operation status, links execution parameters to their data origins, and exposes a standardized metadata layer. By combining database-inspired guarantees with AI-oriented design, such as declarative operation builders and lineage-aware references, TableVault supports transparency and reproducibility across mixed human-model pipelines. Through a document classification case study, we demonstrate how TableVault preserves detailed lineage and operational context, enabling robust metadata management, even in partially observable execution environments.

cs.DB↗

arXiv2025

Learning Lineage Constraints for Data Science Operations

Data science workflows often integrate functionalities from a diverse set of libraries and frameworks. Tasks such as debugging require data lineage that crosses library boundaries. The problem is that the way that "lineage" is represented is often intimately tied to particular data models and data manipulation paradigms. Inspired by the use of intermediate representations (IRs) in cross-library performance optimizations, this vision paper proposes a similar architecture for lineage - how do we specify logical lineage across libraries in a common parameterized way? In practice, cross-library workflows will contain both known operations and unknown operations, so a key design of XProv to link both materialized lineage graphs of data transformations and the aforementioned abstracted logical patterns. We further discuss early ideas on how to infer logical patterns when only the materialized graphs are available.

cs.DB↗

arXiv2025

Fast Capture of Cell-Level Provenance in Numpy

Effective provenance tracking enhances reproducibility, governance, and data quality in array workflows. However, significant challenges arise in capturing this provenance, including: (1) rapidly evolving APIs, (2) diverse operation types, and (3) large-scale datasets. To address these challenges, this paper presents a prototype annotation system designed for arrays, which captures cell-level provenance specifically within the numpy library. With this prototype, we explore straightforward memory optimizations that substantially reduce annotation latency. We envision this provenance capture approach for arrays as part of a broader governance system for tracking for structured data workflows and diverse data science applications.

cs.DB↗

arXiv2025

TableVault: Managing Dynamic Data Collections for LLM-Augmented Workflows

Large Language Models (LLMs) have emerged as powerful tools for automating and executing complex data tasks. However, their integration into more complex data workflows introduces significant management challenges. In response, we present TableVault - a data management system designed to handle dynamic data collections in LLM-augmented environments. TableVault meets the demands of these workflows by supporting concurrent execution, ensuring reproducibility, maintaining robust data versioning, and enabling composable workflow design. By merging established database methodologies with emerging LLM-driven requirements, TableVault offers a transparent platform that efficiently manages both structured data and associated data artifacts.

cs.DB↗

arXiv2024

A System for Quantifying Data Science Workflows with Fine-Grained Procedural Logging and a Pilot Study

It is important for researchers to understand precisely how data scientists turn raw data into insights, including typical programming patterns, workflow, and methodology. This paper contributes a novel system, called DataInquirer, that tracks incremental code executions in Jupyter notebooks (a type of computational notebook). The system allows us to quantitatively measure timing, workflow, and operation frequency in data science tasks without resorting to human annotation or interview. In a series of pilot studies, we collect 97 traces, logging data scientist activities across four studies. While this paper presents a general system and data analysis approach, we focus on a foundational sub-question in our pilot studies: How consistent are different data scientists in analyzing the same data? We taxonomize variation between data scientists on the same dataset according to three categories: semantic, syntactic, and methodological. Our results suggest that there are statistically significant differences in the conclusions reached by different data scientists on the same task and present quantitative evidence for this phenomenon. Furthermore, our results suggest that AI-powered code tools subtly influence these results, allowing student participants to generate workflows that more resemble expert data practitioners.

cs.HC↗

arXiv2024

Towards Causal Physical Error Discovery in Video Analytics Systems

Video analytics systems based on deep learning models are often opaque and brittle and require explanation systems to help users debug. Current model explanation system are very good at giving literal explanations of behavior in terms of pixel contributions but cannot integrate information about the physical or systems processes that might influence a prediction. This paper introduces the idea that a simple form of causal reasoning, called a regression discontinuity design, can be used to associate changes in multiple key performance indicators to physical real world phenomena to give users a more actionable set of video analytics explanations. We overview the system architecture and describe a vision of the impact that such a system might have.

cs.CV↗

arXiv2024

Data Makes Better Data Scientists

With the goal of identifying common practices in data science projects, this paper proposes a framework for logging and understanding incremental code executions in Jupyter notebooks. This framework aims to allow reasoning about how insights are generated in data science and extract key observations into best data science practices in the wild. In this paper, we show an early prototype of this framework and ran an experiment to log a machine learning project for 25 undergraduate students.

cs.HC↗

arXiv2024

Compression and In-Situ Query Processing for Fine-Grained Array Lineage

Tracking data lineage is important for data integrity, reproducibility, and debugging data science workflows. However, fine-grained lineage (i.e., at a cell level) is challenging to store, even for the smallest datasets. This paper introduces DSLog, a storage system that efficiently stores, indexes, and queries array data lineage, agnostic to capture methodology. A main contribution is our new compression algorithm, named ProvRC, that compresses captured lineage relationships. Using ProvRC for lineage compression result in a significant storage reduction over functions with simple spatial regularity, beating alternative columnar-store baselines by up to 2000x}. We also show that ProvRC facilitates in-situ query processing that allows forward and backward lineage queries without decompression - in the optimal case, surpassing baselines by 20x in query latency on random numpy pipelines.

cs.DB↗

arXiv2023

vONTSS: vMF based semi-supervised neural topic modeling with optimal transport

Recently, Neural Topic Models (NTM), inspired by variational autoencoders, have attracted a lot of research interest; however, these methods have limited applications in the real world due to the challenge of incorporating human knowledge. This work presents a semi-supervised neural topic modeling method, vONTSS, which uses von Mises-Fisher (vMF) based variational autoencoders and optimal transport. When a few keywords per topic are provided, vONTSS in the semi-supervised setting generates potential topics and optimizes topic-keyword quality and topic classification. Experiments show that vONTSS outperforms existing semi-supervised topic modeling methods in classification accuracy and diversity. vONTSS also supports unsupervised topic modeling. Quantitative and qualitative experiments show that vONTSS in the unsupervised setting outperforms recent NTMs on multiple aspects: vONTSS discovers highly clustered and coherent topics on benchmark datasets. It is also much faster than the state-of-the-art weakly supervised text classification method while achieving similar classification performance. We further prove the equivalence of optimal transport loss and cross-entropy loss at the global minimum.

cs.LG↗

arXiv2023

FFPDG: Fast, Fair and Private Data Generation

Generative modeling has been used frequently in synthetic data generation. Fairness and privacy are two big concerns for synthetic data. Although Recent GAN [\cite{goodfellow2014generative}] based methods show good results in preserving privacy, the generated data may be more biased. At the same time, these methods require high computation resources. In this work, we design a fast, fair, flexible and private data generation method. We show the effectiveness of our method theoretically and empirically. We show that models trained on data generated by the proposed method can perform well (in inference stage) on real application scenarios.

cs.LG↗

arXiv2023

Data Station: Delegated, Trustworthy, and Auditable Computation to Enable Data-Sharing Consortia with a Data Escrow

Pooling and sharing data increases and distributes its value. But since data cannot be revoked once shared, scenarios that require controlled release of data for regulatory, privacy, and legal reasons default to not sharing. Because selectively controlling what data to release is difficult, the few data-sharing consortia that exist are often built around data-sharing agreements resulting from long and tedious one-off negotiations. We introduce Data Station, a data escrow designed to enable the formation of data-sharing consortia. Data owners share data with the escrow knowing it will not be released without their consent. Data users delegate their computation to the escrow. The data escrow relies on delegated computation to execute queries without releasing the data first. Data Station leverages hardware enclaves to generate trust among participants, and exploits the centralization of data and computation to generate an audit log. We evaluate Data Station on machine learning and data-sharing applications while running on an untrusted intermediary. In addition to important qualitative advantages, we show that Data Station: i) outperforms federated learning baselines in accuracy and runtime for the machine learning application; ii) is orders of magnitude faster than alternative secure data-sharing frameworks; and iii) introduces small overhead on the critical path.

cs.DB↗

arXiv2022

Deep Transfer Learning with Graph Neural Network for Sensor-Based Human Activity Recognition

The sensor-based human activity recognition (HAR) in mobile application scenarios is often confronted with sensor modalities variation and annotated data deficiency. Given this observation, we devised a graph-inspired deep learning approach toward the sensor-based HAR tasks, which was further used to build a deep transfer learning model toward giving a tentative solution for these two challenging problems. Specifically, we present a multi-layer residual structure involved graph convolutional neural network (ResGCNN) toward the sensor-based HAR tasks, namely the HAR-ResGCNN approach. Experimental results on the PAMAP2 and mHealth data sets demonstrate that our ResGCNN is effective at capturing the characteristics of actions with comparable results compared to other sensor-based HAR models (with an average accuracy of 98.18% and 99.07%, respectively). More importantly, the deep transfer learning experiments using the ResGCNN model show excellent transferability and few-shot learning performance. The graph-based framework shows good meta-learning ability and is supposed to be a promising solution in sensor-based HAR tasks.

cs.CV↗

arXiv2022

Antiphase boundary in CH$_3$NH$_3$PbI$_3$ repels charge carriers while promotes fast ion migrations

Defects in organic-inorganic hybrid perovskites (OIHPs) greatly influence their optoelectronic properties. Identification and better understanding of defects existing in OIHPs is an essential step towards fabricating high-performance perovskite solar cells. However, direct visualizing the defects is still a challenge for OIHPs due to their sensitivity during electron microscopy characterizations. Here, by using low dose scanning transmission electron microscopy techniques, we observe the common existence of antiphase boundary (APB) in CH$_3$NH$_3$PbI$_3$ (MAPbI$_3$), resolve its atomic structure, and correlate it to the electrical/ionic activities and structural instabilities. Such an APB is caused by the half-unit-cell shift of [PbI$_6$]$_4$-octahedron along the [100]/[010] direction, leading to the transformation from corner-sharing [PbI$_6$]$_4$-octahedron in bulk MAPbI$_3$ into edge-sharing ones at the APB. Based on the identified atomic-scale configuration, we further carry out density functional theory calculations and reveal that the APB in MAPbI$_3$ repels both electrons and holes while serves as a fast ion-migration channel, causing a rapid decomposition into PbI$_2$ that is detrimental to optoelectronic performance. These findings provide valuable insights into the relationships between structures and optoelectronic properties of OIHPs and suggest that controlling the APB is essential for their stability.

physics.app-ph↗

arXiv2021

Atomic-scale imaging of CH3NH3PbI3 structure and its decomposition pathway

Understanding the atomic structure and structural instability of organic-inorganic hybrid perovskites is the key to appreciate their remarkable photoelectric properties and failure mechanism. Here, using low-dose imaging technique by direct-detection electron-counting camera in transmission electron microscope, we investigate the atomic structure and decomposition pathway of CH3NH3PbI3 (MAPbI3) at the atomic scale. We successfully image the atomic structure of perovskite in real space under ultra-low electron dose condition, and observe a two-step decomposition process, i.e. initial loss of MA followed by the collapse of perovskite structure into 6H-PbI2 with their critical threshold dose also determined. Interestingly, an intermediate phase (MA0.5PbI3) with locally ordered vacancies can robustly exist before perovskite collapses, enlightening strategies for prevention and recovery of perovskite structure during degradation. Associated with structure evolution, the bandgap gradually increases from ~1.6 eV to ~2.1 eV, and it is found that both C-N and N-H bonds can be destroyed under irradiation, releasing NH3 and leaving hydrocarbons. These findings enhance our understanding of the photoelectric properties and failure mechanism of MAPbI3, providing potential strategy into material optimization.

physics.app-ph↗

arXiv2021

Polar or nonpolar? That is not the question for perovskite solar cells

Perovskite solar cell (PSC) is one of the most promising next generation photovoltaic technologies, and there are considerable interests in the role of possible polarization of organic-inorganic halide perovskites (OIHPs) in photovoltaic conversion. The polarity of OIHPs, however, is still hotly debated. In this review, we examine recent literature on the polarity of OIHPs from both theoretical and experimental points of view, and argue that they can be both polar and nonpolar, depending on compositions, processing, and environments. Implications of OIHP polarity to photovoltaic conversion is also discussed, and effort in answering these questions continues to render us new insights. In the future, integrating local scanning probe with global macroscopic measurements in-situ will provide invaluable microscopic insight into the intriguing macroscopic phenomena, while synchrotron diffractions and scanning transmission electron microscopy on more stable samples may ultimately settle the debate.

physics.app-ph↗

arXiv2020

Knowledge Graph semantic enhancement of input data for improving AI

Intelligent systems designed using machine learning algorithms require a large number of labeled data. Background knowledge provides complementary, real world factual information that can augment the limited labeled data to train a machine learning algorithm. The term Knowledge Graph (KG) is in vogue as for many practical applications, it is convenient and useful to organize this background knowledge in the form of a graph. Recent academic research and implemented industrial intelligent systems have shown promising performance for machine learning algorithms that combine training data with a knowledge graph. In this article, we discuss the use of relevant KGs to enhance input data for two applications that use machine learning -- recommendation and community detection. The KG improves both accuracy and explainability.

cs.AI↗

arXiv2020

Transmission electron microscopy of organic-inorganic hybrid perovskites: myths and truths

Organic-inorganic hybrid perovskites (OIHPs) have attracted extensive research interest as a promising candidate for efficient and inexpensive solar cells. Transmission electron microscopy characterizations that can benefit the fundamental understanding and the degradation mechanism are widely used for these materials. However, their sensitivity to the electron beam illumination and hence structural instabilities usually prevent us from obtaining the intrinsic information or even lead to significant artifacts. Here, we systematacially investigate the structural degradation behaviors under different experimental factors to reveal the optimized conditions for TEM characterizations of OIHPs by using low-dose electron diffraction and imaging techniques. We find that a low temperature does not slow down the beam damage but instead induces a rapid amorphization for OIHPs. Moreover, a less severe damage is observed at a higher accelerating voltage. The beam-sensitivity is found to be facet-dependent that a (100) exposed MAPbI3 surface is more stable than (001) surface. With these guidance, we successfully acquire the atomic structure of pristine MAPbI3 and identify the characterization window that is very narrow. These findings are helpful to guide future electron microscopy characterization of these beam-sensitive materials, which are also useful for finding strategies to improve the stability and the performance of the perovskite solar cells.

physics.app-ph↗

arXiv2018

Ferroic Domains of Alternating Polar and Nonpolar Orders Regulate Photocurrent in Single Crystalline CH3NH3PbI3 Films Self-grown on FTO/TiO2 Substrate

Photovoltaic conversion efficiency (PCE) of halide perovskite solar cells has risen spectacularly, yet the very crystalline structure of CH3NH3PbI3 remains ambiguous after extensive researches, and its polar nature remains hotly debated. Here we present compelling evidences that CH3NH3PbI3 crystals self-grown on FTO/TiO2 substrate consist of ferroic domains with alternating polar and nonpolar orders, in contrast to previous experimental and theoretical expectations, and polar domains possess reduced photocurrent. It is found that polar and nonpolar orders of CH3NH3PbI3 can be distinguished from their distinct lateral piezoresponse, energy dissipation, first and second harmonic electromechanical couplings, and temperature variation, even though their difference in crystalline lattice is very subtle, and they possess two-way memory effect through cubic-tetragonal phase transition. These findings resolve key questions regarding polar nature of CH3NH3PbI3 and its implication on photovoltaics, reconcile contradictory data widely reported, and point a direction toward engineering ferroic domains for enhanced PCE.

cond-mat.mtrl-sci↗