Selected work

Representative Papers

RovoDev Code Reviewer: A Large-Scale Online Evaluation of LLM-based Code Review Automation at Atlassian

Jan 03, 2026arXiv.org

This work proposes a fine-tuning-free, large language model (LLM)-driven approach to address the need for high-quality, context-aware, and goal-directed automated code review comments in enterprise settings. By leveraging prompt engineering, contextual retrieval, and a comment quality filtering mechanism, the authors developed and deployed RovoDev Code Reviewer—an integrated system within Atlassian Bitbucket. Evaluation over a one-year period in a real-world industrial environment demonstrates that 38.7% of the system’s automatically generated comments led developers to modify their code, resulting in a 30.8% reduction in average pull request (PR) cycle time and a 35.6% decrease in manual reviewer comments. The system also effectively identified actionable code defects, confirming its practicality and effectiveness without requiring model fine-tuning.

2 citationsRead paper

VIGIL: Verifier-Informed Gated Improvement Loop for Spreadsheet Question Answering

Oct 03, 2026

This study addresses the challenge of enabling enterprise agents to continuously improve from delayed feedback in spreadsheet question answering without compromising existing system behavior. Building upon the FiCo framework, this work proposes the VIGIL mechanism and a dual-gated improvement loop. By freezing the base model and retriever, the approach validates SQL candidates and employs gated updates on lightweight calibrators and selectors, thereby achieving bounded and auditable continual learning. Experimental results demonstrate that the proposed method attains a forward accuracy of 82.4% and improves the average accuracy on a rigorous test set from 80.3% to 86.7%. These findings effectively validate the advantages of bounded adaptation strategies in preserving system stability and controllability during continuous deployment.

0 citationsRead paper

MRVQ: One Resident Index for Dimension- and Rate-Elastic Vector Search

Oct 02, 2026

This study addresses the excessive memory overhead in dense retrieval caused by maintaining multiple indices for varying dimensionalities and bitrates. To overcome this limitation, we propose Matryoshka Residual Vector Quantization (MRVQ), a method that optimizes frozen embeddings through post-training processing by truncating residual stages or coordinates. This approach enables a single index to flexibly accommodate diverse dimensionality-rate combinations, achieving versatile multi-purpose indexing. Experimental results demonstrate that MRVQ reduces memory consumption by 17.8× to 22× compared to maintaining three independent indices. Although its retrieval quality is marginally lower than that of optimal single-rate configurations, it significantly outperforms conventional Product Quantization (PQ) methods. Ultimately, MRVQ facilitates efficient and elastic vector search with minimal memory footprint.

0 citationsRead paper

FICO: Find-Then-Compute for Corpus-Level Spreadsheet Question Answering

Oct 02, 2026

This study addresses the challenges of inaccurate workbook localization and incomplete computation in question answering over spreadsheet collections by proposing FiCo, a method that innovatively integrates semantic source selection with schema-based precise calculation. Specifically, FiCo accurately identifies data sources through document summary retrieval and similar workbook disambiguation, subsequently executing constrained SQL queries over the entire selected tables to perform computations. This design effectively overcomes performance bottlenecks caused by erroneous source selection. Experimental evaluations demonstrate that FiCo achieves an accuracy of 76.2% on the DataBench dataset, representing a 9.9% improvement over baseline methods, and attains 79.7% on the MiMoTable dataset. These results indicate that the proposed approach significantly outperforms existing prefix-based Retrieval-Augmented Generation (RAG) methods.

0 citationsRead paper

Beyond Private Training: The New Landscape of AI Privacy

Sep 16, 2026

本文提出TSD-AUDIT框架，旨在解决向量检索系统中删除项仍被计算距离的问题，通过确保仅对有效向量进行评分并提供可验证的查询证书来实现安全删除。

0 citationsRead paper

Recent publications

Latest Papers

VIGIL: Verifier-Informed Gated Improvement Loop for Spreadsheet Question Answering

Oct 03, 2026

This study addresses the challenge of enabling enterprise agents to continuously improve from delayed feedback in spreadsheet question answering without compromising existing system behavior. Building upon the FiCo framework, this work proposes the VIGIL mechanism and a dual-gated improvement loop. By freezing the base model and retriever, the approach validates SQL candidates and employs gated updates on lightweight calibrators and selectors, thereby achieving bounded and auditable continual learning. Experimental results demonstrate that the proposed method attains a forward accuracy of 82.4% and improves the average accuracy on a rigorous test set from 80.3% to 86.7%. These findings effectively validate the advantages of bounded adaptation strategies in preserving system stability and controllability during continuous deployment.

0 citationsRead paper

MRVQ: One Resident Index for Dimension- and Rate-Elastic Vector Search

Oct 02, 2026

This study addresses the excessive memory overhead in dense retrieval caused by maintaining multiple indices for varying dimensionalities and bitrates. To overcome this limitation, we propose Matryoshka Residual Vector Quantization (MRVQ), a method that optimizes frozen embeddings through post-training processing by truncating residual stages or coordinates. This approach enables a single index to flexibly accommodate diverse dimensionality-rate combinations, achieving versatile multi-purpose indexing. Experimental results demonstrate that MRVQ reduces memory consumption by 17.8× to 22× compared to maintaining three independent indices. Although its retrieval quality is marginally lower than that of optimal single-rate configurations, it significantly outperforms conventional Product Quantization (PQ) methods. Ultimately, MRVQ facilitates efficient and elastic vector search with minimal memory footprint.

0 citationsRead paper

FICO: Find-Then-Compute for Corpus-Level Spreadsheet Question Answering

Oct 02, 2026

This study addresses the challenges of inaccurate workbook localization and incomplete computation in question answering over spreadsheet collections by proposing FiCo, a method that innovatively integrates semantic source selection with schema-based precise calculation. Specifically, FiCo accurately identifies data sources through document summary retrieval and similar workbook disambiguation, subsequently executing constrained SQL queries over the entire selected tables to perform computations. This design effectively overcomes performance bottlenecks caused by erroneous source selection. Experimental evaluations demonstrate that FiCo achieves an accuracy of 76.2% on the DataBench dataset, representing a 9.9% improvement over baseline methods, and attains 79.7% on the MiMoTable dataset. These results indicate that the proposed approach significantly outperforms existing prefix-based Retrieval-Augmented Generation (RAG) methods.

0 citationsRead paper

Beyond Private Training: The New Landscape of AI Privacy

Sep 16, 2026

本文提出TSD-AUDIT框架，旨在解决向量检索系统中删除项仍被计算距离的问题，通过确保仅对有效向量进行评分并提供可验证的查询证书来实现安全删除。

0 citationsRead paper

Split the Labor: Separating Evidence Interpretation from Decision Aggregation

Aug 14, 2026

This study addresses the issues of counting scale drift and decision inconsistency arising from the coupling of interpretation and aggregation in multi-source evidence reasoning. We propose a decoupling framework that constructs an evidence quadruplet interface and introduces calibrated log-likelihood ratio pooling to achieve generic rectification at the arithmetic level, thereby effectively separating evidence interpretation from decision aggregation. By integrating sequential encoders with tree ensemble models, the proposed method achieves an AUPRC of 0.921 on longitudinal corpora, significantly outperforming handcrafted baselines (0.805). These results validate the effectiveness of the separation architecture in enhancing the reliability of multi-source reasoning, offering a novel paradigm for complex evidence aggregation.

0 citationsRead paper