Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Kai Zheng

Kai Zheng

3 indexed papers

Recent (6 mo)
3
With code
0
Influential cites
0
Benchmarked
0

Publications per year

3
26

Top categories

ML×3AI×2NLP×2Databases×1Distributed×1Vision×1

Frequent co-authors

Junming Chen1×
Junyang Jiang1×
Xu Chen1×
Zibo Liang1×
Jieying Wang1×
Shuyuan Fan1×

Research Timeline

2026
VeriEvol: Scaling Multimodal Mathematical Reasoning via Verifiable Evol-Instruct

This paper proposes VeriEvol, a framework for scaling reinforcement learning for visual mathematical reasoning by decoupling prompt difficulty and answer reliability, and verifying data construction.

GIFT: Geometry-Informed Low-precision Gradient Communication for LLM Pretraining

The paper presents GIFT, a method for reducing communication volume in large language model pretraining by transforming gradients into a near-isotropic space before quantization.

DBA-Bench: A Production-Fidelity Benchmark for LLM-Based Database Operations Agents

The paper introduces DBA-Bench, a benchmark for evaluating database agents with production fidelity, outcome-first evaluation, and controlled scenario reproducibility.

Highlighted terms show continued research focus across papers

Papers

cs.DBcs.AIcs.CLEmpiricalRecentJul 24, 2026

DBA-Bench: A Production-Fidelity Benchmark for LLM-Based Database Operations Agents

Junming Chen, Junyang Jiang, Xu Chen, Zibo Liang +1 more

The paper introduces DBA-Bench, a benchmark for evaluating database agents with production fidelity, outcome-first evaluation, and controlled scenario reproducibility.

View →
cs.DCcs.LGEmpirical
Recent
Jul 8, 2026

GIFT: Geometry-Informed Low-precision Gradient Communication for LLM Pretraining

Jieying Wang, Shuyuan Fan, Mingkai Zheng, Zhao Zhang

The paper presents GIFT, a method for reducing communication volume in large language model pretraining by transforming gradients into a near-isotropic space before quantization.

View →
cs.AIcs.CLcs.CVEmpiricalRecentJun 22, 2026

VeriEvol: Scaling Multimodal Mathematical Reasoning via Verifiable Evol-Instruct

Haoling Li, Kai Zheng, Jie Wu, Can Xu +3 more

This paper proposes VeriEvol, a framework for scaling reinforcement learning for visual mathematical reasoning by decoupling prompt difficulty and answer reliability, and verifying data construction.

View →