Manuscript

GPU-Accelerated Conic Quadratic Programming with Local Linear Convergence under Strict Complementarity

PDHCG-CQP extends GPU-accelerated first-order optimization to conic quadratic programs, with local linear convergence guarantees and multi-GPU execution.

Aug 10, 2026

A Curvature-Aware Rank-Adaptive Distributed Augmented-Lagrangian Solver for Large-Scale SDPs

Jul 20, 2026

OSDN: Improving Delta Rule with Provable Online Preconditioning in Linear Attention

May 1, 2026

PDHCG-II: An Enhanced Version of PDHCG for Large-Scale Convex QP

Feb 1, 2026

D-PDLP: Scaling PDLP to Distributed Multi-GPU Systems

Jan 1, 2026

PDHCG: A Scalable First-Order Method for Large-Scale Competitive Market Equilibrium Computation

Jun 1, 2025

BenLOC: A Benchmark for Learning to Configure MIP Optimizers

Jun 1, 2025

OptPipe: Memory- and Scheduling-Optimized Pipeline Parallelism for LLM Training

Jan 1, 2025

FMIP: Joint Continuous-Integer Flow For Mixed-Integer Linear Programming

Jan 1, 2025

A Restarted Primal-Dual Hybrid Conjugate Gradient Method for Large-Scale Quadratic Programming

Jan 1, 2025