Official code repository for "Pre-Deployment Advertisement Ranking under Data Scarcity via Context-Aware Criteria Generation with VLMs" (ACL 2026 Industry Track Oral)

Effective Dataset Distillation for Spatio-Temporal Forecasting with Bi-dimensional Compression (ICDE 26)

Published as a conference paper at AAAI 2026

Published at The Web Conference 2026

Official code repository of "ReFuGe: Feature Generation for Prediction Tasks on Relational Databases with LLM Agents" (WWW 2026)

Query language for blending SQL and local language models across structured + unstructured data, with type constraints.

[ACL'26 Findings] Official Repository of "Hybrid-Vector Retrieval for Visually Rich Documents: Combining Single-Vector Efficiency and Multi-Vector Accuracy"

A system for agentic LLM-powered data processing and ETL

A System for Optimized Semantic Computation

AIKU 2023-2R Project: Fake news generator

(EMNLP 2025) Weaver: A modular agentic pipeline that dynamically combines SQL and LLMs for advanced table-based question answering

TAG-Bench: A benchmark for table-augmented generation (TAG)

Optimized LLM-Powered Data Processing: up to 1000x speedups with fast, accurate query processing, that's as simple as writing Pandas code

This is the official implementation of the paper “Griffin: Towards a Graph-Centric Relational Database Foundation Model.”

Relational Transformer: Toward Zero-Shot Foundation Models for Relational Data

Distance-Decaying Core Decomposition

[ICDM'25] Identifying Group Anchors in Real-World Group Interactions Under Label Scarcity

[ICDM'25] Edge Probability Graph Models Beyond Edge Independency: Concepts, Analyses, and Algorithms

Implementation of Attributed Hypergraph Generation with Realistic Interplay Between Structure and Attributes

This repo contains the codebase for the paper "Unifying Generative and Dense Retrieval for Sequential Recommendation".

Multiplicity-Aware Hypergraph Reconstruction

Official code for "A Self-Supervised Mixture of Experts Framework for Multi-behavior Recommendation" (CIKM 25)

[SIGIR'25] Code of "Generative Recommender with End-to-End Learnable Item Tokenization".

GRID: Generative Recommendation with Semantic IDs

[NeurIPS 2025 Datasets and Benchmarks] Source code for the paper RDB2G-Bench: A Comprehensive Benchmark for Automatic Graph Modeling of Relational Databases

[ICML 2025] RelGNN: Composite Message Passing for Relational Deep Learning