Author

Mao Zheng

Observed name: Zheng, Mao

Institution: Tencent (China)

Observed institution: Tencent (China)

OpenAlex ID: A5135849426

Distill What You Trust: Reliability-Aware Multi-Teacher On-Policy Distillation

2026 · Sun, Jie, Zheng, Mao, et al. · arXiv (Cornell University)

PDF

Data-Free On-Policy Distillation: How Far Can We Go Without External Data?

2026 · Li, Gengsheng, Zheng, Mao, et al. · arXiv (Cornell University)

PDF

CodeDelegator: Mitigating Context Pollution via Role Separation in Code-as-Action Agents

2026 · Tianxiang Fei, Cheng Chen, et al. · International Joint Conference on Artificial Intelligence

PDF

Simple-OPD: Demystifying Warm-up for On-policy Distillation

2026 · Liu, Tao, Wu, Taiqiang, et al. · arXiv (Cornell University)

PDF

EasyOPD: An Easy-to-use On-Policy Distillation Framework for Large Language Models

2026 · Sun, Jie, Zheng, Mao, et al. · arXiv (Cornell University)

PDF

Hy-MT2: A Family of Fast, Efficient and Powerful Multilingual Translation Models in the Wild

2026 · Zheng, Mao, Li, Zheng, et al. · arXiv (Cornell University)

PDF

SOD: Step-wise On-policy Distillation for Small Language Model Agents

2026 · Zhong, Qiyong, Zheng, Mao, et al. · arXiv (Cornell University)

PDF

Rubric-based On-policy Distillation

2026 · Fang, Junfeng, Hong, Zhepei, et al. · arXiv (Cornell University)

PDF

SimCT: Recovering Lost Supervision for Cross-Tokenizer On-Policy Distillation

2026 · Sun, Jie, Zheng, Mao, et al. · arXiv (Cornell University)

PDF

Unifying Group-Relative and Self-Distillation Policy Optimization via Sample Routing

2026 · Li, Gengsheng, Yang, Tianyu, et al. · arXiv (Cornell University)

PDF

PRISM: Probability Reallocation with In-Span Masking for Knowledge-Sensitive Alignment

2026 · Xu, Chenning, Zheng, Mao, et al. · arXiv (Cornell University)

PDF

A Survey of On-Policy Distillation for Large Language Models

2026 · Song, Mingyang, Zheng, Mao · arXiv (Cornell University)

PDF

Model Merging in the Era of Large Language Models: Methods, Applications, and Future Directions

2026 · Song, Mingyang, Zheng, Mao · arXiv (Cornell University)

PDF

PodBench: A Comprehensive Benchmark for Instruction-Aware Audio-Oriented Podcast Script Generation

2026 · Xu, Chenning, Zheng, Mao, et al. · arXiv (Cornell University)

PDF

Beyond the Illusion of Consensus: From Surface Heuristics to Knowledge-Grounded Evaluation in LLM-as-a-Judge

2026 · Mingyang Song, Mao Zheng · Conference on Empirical Methods in Natural Language Processing (EMNLP)

PDF

IFMTBench: A Comprehensive Benchmark for Multilingual Translation Instruction Following

2026 · Mingrui Sun, Mao Zheng, et al. · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)

PDF

HardMTBench: Stress-Testing Chinese-English Translation on Knowledge-Intensive Domains

2026 · Zheng Li, Mao Zheng, et al. · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)

PDF

Memory Beyond Recall: A Dual-Process Cognitive Memory System for Self-Evolving LLM Agents

2026 · Tianxiang Fei, Mingyang Song, et al. · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)

PDF

On-Policy Distillation with Curriculum Turn-level Guidance for Multi-turn Agents

2026 · Gengsheng Li, Mao Zheng, et al. · Conference on Empirical Methods in Natural Language Processing (EMNLP)

PDF

HY-MT1.5 Technical Report

2025 · Zheng, Mao, Li, Zheng, et al. · arXiv (Cornell University)

PDF

Hunyuan-MT Technical Report

2025 · Zheng, Mao, Li, Zheng, et al. · arXiv (Cornell University)

PDF

Works count: 21

Citation count:

H-index:

I10-index: