arXiv subjects

Sijie Dong

Publications and source records attributed to Sijie Dong.

4 recordsLinked to original sources

arXiv2026

BENCHCOMPASS: From Scores to Signals for Training and Harness Decisions in Payment-Domain LLMs

Payment operations are a critical financial infrastructure, but the value of large language models in this domain remains unclear because payment rules change quickly, evidence is fragmented, and decisions depend on transaction state, participant role, region, and payment rail. Existing benchmarks do not isolate whether failures come from missing payment-rule knowledge, poor use of supplied evidence, or brittleness under imperfect harness inputs. We introduce BENCHCOMPASS, a payment-domain benchmark whose construction pipeline builds scenario-grounded tasks from typed evidence packs, applies LLM-based quality checks, creates task-input attack variants, and reserves final item admission for domain experts. The release contains an expert-reviewed Pro benchmark covering payment knowledge, context-grounded scenario reasoning, and Attacked Open robustness, plus a lower-assurance Normal pool for inspection and future curation. Across 16 model variants, BENCHCOMPASS shows qualitatively different failure modes: missing parametric payment knowledge, incomplete reasoning over supplied rules, and failure to reject plausible but invalid workflows. The benchmark remains unsaturated: the best frontier model reaches 89.6% on Open Context-Grounded Reasoning and 81.7% under attacked inputs, while a representative 32B open-weight model reaches 69.8% and 42.6%. Benchmark data and code are available at https://github.com/ant-intl/BenchCompass.

cs.AI↗

arXiv2026

STAGE: Controlled Objective Admission for Multi-Preference LLM Alignment

Multi-preference alignment is often framed as scalarization: combine reward dimensions, then optimize. This leaves a temporal decision underspecified: when should each preference dimension enter policy optimization? We propose \methodname, a stability-guided active-set controller for controlled objective admission. \methodname starts from a small active set, retains admitted objectives, and expands when reward-deviation gates indicate low recent deviation or a patience budget is exhausted. A probing phase estimates a hard-to-easy order, and adaptive weighting emphasizes underperforming active dimensions. Automatic evaluations with 15 training preferences and 16 held-out benchmark columns show that \methodname obtains higher averages than simultaneous scalarization and shared-budget adapted baselines. Component ablations and expansion dynamics further support cumulative retention, gated admission, and probing-derived ordering as useful design choices in this setting. These results position objective-entry timing as a concrete control variable in reward-vector RLHF.

cs.CL↗

arXiv2025

Automated Data Quality Validation in an End-to-End GNN Framework

Ensuring data quality is crucial in modern data ecosystems, especially for training or testing datasets in machine learning. Existing validation approaches rely on computing data quality metrics and/or using expert-defined constraints. Although there are automated constraint generation methods, they are often incomplete and may be too strict or too soft, causing false positives or missed errors, thus requiring expert adjustment. These methods may also fail to detect subtle data inconsistencies hidden by complex interdependencies within the data. In this paper, we propose DQuag, an end-to-end data quality validation and repair framework based on an improved Graph Neural Network (GNN) and multi-task learning. The proposed method incorporates a dual-decoder design: one for data quality validation and the other for data repair. Our approach captures complex feature relationships within tabular datasets using a multi-layer GNN architecture to automatically detect explicit and hidden data errors. Unlike previous methods, our model does not require manual input for constraint generation and learns the underlying feature dependencies, enabling it to identify complex hidden errors that traditional systems often miss. Moreover, it can recommend repair values, improving overall data quality. Experimental results validate the effectiveness of our approach in identifying and resolving data quality issues. The paper appeared in EDBT 2025.

cs.DB↗

arXiv2024

Data Quality Awareness: A Journey from Traditional Data Management to Data Science Systems

Artificial intelligence (AI) has transformed various fields, significantly impacting our daily lives. A major factor in AI success is high-quality data. In this paper, we present a comprehensive review of the evolution of data quality (DQ) awareness from traditional data management systems to modern data-driven AI systems, which are integral to data science. We synthesize the existing literature, highlighting the quality challenges and techniques that have evolved from traditional data management to data science including big data and ML fields. As data science systems support a wide range of activities, our focus in this paper lies specifically in the analytics aspect driven by machine learning. We use the cause-effect connection between the quality challenges of ML and those of big data to allow a more thorough understanding of emerging DQ challenges and the related quality awareness techniques in data science systems. To the best of our knowledge, our paper is the first to provide a review of DQ awareness spanning traditional and emergent data science systems. We hope that readers will find this journey through the evolution of data quality awareness insightful and valuable.

cs.DB↗