Selected work

Representative Papers

Data Intelligence Agents: Interpreting, Modeling, and Querying Enterprise Data via Autonomous Coding Agents

Jun 17, 2026

Enterprise data integration often suffers from bottlenecks in data discovery, modeling, and querying due to inefficient handoffs among data owners, engineers, and analysts. This work proposes the first three-agent system centered on the Autonomous Coding Agent (ACA) abstraction, comprising a Data Interpreter, Schema Creator, and Query Generator. Leveraging an execution-driven architecture and a shared memory mechanism, the system enables end-to-end automated and auditable data workflows. It supports natural language–driven instructions, multi-dialect SQL generation, execution validation, and automatic repair. Evaluated across seven SQL benchmarks, the approach matches or surpasses state-of-the-art methods across four task categories and four SQL dialects, and has already been deployed in enterprise production environments.

0 citationsRead paper

Giving Sensors a Voice: Multimodal JEPA for Semantic Time-Series Embeddings

May 29, 2026

This work addresses the limited capability of existing methods to learn general-purpose representations for heterogeneous multivariate time series. The authors propose CHARM, a novel model that uniquely integrates textual descriptions of sensor channels with a channel-order equivariant Transformer encoder within a Joint Embedding Predictive Architecture (JEPA) to learn semantic-rich and noise-robust temporal representations. The approach innovatively introduces a description-aware gating mechanism and a tailored loss function to enhance the informativeness and temporal stability of learned embeddings, while enabling cross-dataset generalization and interpretable modeling of inter-channel relationships. Remarkably, CHARM achieves strong performance across anomaly detection, classification, and short- to long-term forecasting tasks using only linear probes, demonstrating the effectiveness of the proposed architecture.

0 citationsRead paper

NEMO: Execution-Aware Optimization Modeling via Autonomous Coding Agents

Jan 29, 2026

This work addresses the challenge of automatically translating natural language descriptions of decision problems into executable optimization models by proposing an execution-aware modeling framework based on Autonomous Coding Agents (ACA). The approach ensures code executability through a sandboxed environment and introduces novel coordination mechanisms—including asymmetric verification loops, external memory reuse, minimum Bayes risk decoding, and self-consistency—to significantly enhance modeling robustness and accuracy. The system supports both interactive and fully autonomous operation, achieving state-of-the-art performance across nine standard optimization benchmarks and substantially outperforming existing methods on multiple datasets. These results validate the effectiveness of an architecture that treats ACA as a first-class abstraction for automated optimization modeling.

0 citationsRead paper

FaCTR: Factorized Channel-Temporal Representation Transformers for Efficient Time Series Forecasting

Jun 05, 2025

To address architectural mismatch and over-parameterization in Transformers for time-series forecasting—caused by low input information density and complex cross-channel coupling—this paper proposes the lightweight Factorized Channel-Temporal Transformer (FACT). FACT models dynamic, symmetric cross-channel interactions via low-rank decomposition, and integrates temporal patch embedding, covariate encoding, and learnable gated fusion to jointly model static and dynamic multivariate dependencies. It introduces the first explicitly structured design for interpretable channel-wise influence analysis and self-supervised pretraining. Evaluated on 11 public benchmarks, FACT achieves state-of-the-art performance with a maximum parameter count of only ~400K—on average 50× smaller than comparable spatiotemporal Transformers—while significantly improving prediction accuracy, inference efficiency, and decision interpretability.

0 citationsRead paper

Time to Embed: Unlocking Foundation Models for Time Series with Channel Descriptions

May 20, 2025

Time-series foundation models remain underexplored; existing approaches are largely task-specific, rely heavily on hand-crafted features, and lack cross-task transferable representations. To address this, we propose CHARM—the first foundation embedding model for multivariate time series—introducing channel-wise textual descriptions as domain priors to construct a channel-order-agnostic semantic-temporal joint embedding architecture. We design a self-supervised training framework based on Joint Embedding Predictive Architecture (JEPA), integrating invariance regularization and interpretability-aware loss, alongside a novel time-series-specific data augmentation strategy. With only 7 million parameters, CHARM achieves state-of-the-art performance across diverse downstream tasks—including classification, anomaly detection, and forecasting—establishing a new benchmark for multivariate time-series representation learning.

0 citationsRead paper

Recent publications

Latest Papers

Data Intelligence Agents: Interpreting, Modeling, and Querying Enterprise Data via Autonomous Coding Agents

Jun 17, 2026

Enterprise data integration often suffers from bottlenecks in data discovery, modeling, and querying due to inefficient handoffs among data owners, engineers, and analysts. This work proposes the first three-agent system centered on the Autonomous Coding Agent (ACA) abstraction, comprising a Data Interpreter, Schema Creator, and Query Generator. Leveraging an execution-driven architecture and a shared memory mechanism, the system enables end-to-end automated and auditable data workflows. It supports natural language–driven instructions, multi-dialect SQL generation, execution validation, and automatic repair. Evaluated across seven SQL benchmarks, the approach matches or surpasses state-of-the-art methods across four task categories and four SQL dialects, and has already been deployed in enterprise production environments.

0 citationsRead paper

Giving Sensors a Voice: Multimodal JEPA for Semantic Time-Series Embeddings

May 29, 2026

This work addresses the limited capability of existing methods to learn general-purpose representations for heterogeneous multivariate time series. The authors propose CHARM, a novel model that uniquely integrates textual descriptions of sensor channels with a channel-order equivariant Transformer encoder within a Joint Embedding Predictive Architecture (JEPA) to learn semantic-rich and noise-robust temporal representations. The approach innovatively introduces a description-aware gating mechanism and a tailored loss function to enhance the informativeness and temporal stability of learned embeddings, while enabling cross-dataset generalization and interpretable modeling of inter-channel relationships. Remarkably, CHARM achieves strong performance across anomaly detection, classification, and short- to long-term forecasting tasks using only linear probes, demonstrating the effectiveness of the proposed architecture.

0 citationsRead paper

NEMO: Execution-Aware Optimization Modeling via Autonomous Coding Agents

Jan 29, 2026

This work addresses the challenge of automatically translating natural language descriptions of decision problems into executable optimization models by proposing an execution-aware modeling framework based on Autonomous Coding Agents (ACA). The approach ensures code executability through a sandboxed environment and introduces novel coordination mechanisms—including asymmetric verification loops, external memory reuse, minimum Bayes risk decoding, and self-consistency—to significantly enhance modeling robustness and accuracy. The system supports both interactive and fully autonomous operation, achieving state-of-the-art performance across nine standard optimization benchmarks and substantially outperforming existing methods on multiple datasets. These results validate the effectiveness of an architecture that treats ACA as a first-class abstraction for automated optimization modeling.

0 citationsRead paper

FaCTR: Factorized Channel-Temporal Representation Transformers for Efficient Time Series Forecasting

Jun 05, 2025

To address architectural mismatch and over-parameterization in Transformers for time-series forecasting—caused by low input information density and complex cross-channel coupling—this paper proposes the lightweight Factorized Channel-Temporal Transformer (FACT). FACT models dynamic, symmetric cross-channel interactions via low-rank decomposition, and integrates temporal patch embedding, covariate encoding, and learnable gated fusion to jointly model static and dynamic multivariate dependencies. It introduces the first explicitly structured design for interpretable channel-wise influence analysis and self-supervised pretraining. Evaluated on 11 public benchmarks, FACT achieves state-of-the-art performance with a maximum parameter count of only ~400K—on average 50× smaller than comparable spatiotemporal Transformers—while significantly improving prediction accuracy, inference efficiency, and decision interpretability.

0 citationsRead paper

Time to Embed: Unlocking Foundation Models for Time Series with Channel Descriptions

May 20, 2025

Time-series foundation models remain underexplored; existing approaches are largely task-specific, rely heavily on hand-crafted features, and lack cross-task transferable representations. To address this, we propose CHARM—the first foundation embedding model for multivariate time series—introducing channel-wise textual descriptions as domain priors to construct a channel-order-agnostic semantic-temporal joint embedding architecture. We design a self-supervised training framework based on Joint Embedding Predictive Architecture (JEPA), integrating invariance regularization and interpretability-aware loss, alongside a novel time-series-specific data augmentation strategy. With only 7 million parameters, CHARM achieves state-of-the-art performance across diverse downstream tasks—including classification, anomaly detection, and forecasting—establishing a new benchmark for multivariate time-series representation learning.

0 citationsRead paper