Han Shen

Senior Research Engineer, Ant Group

Joined

September 2020

Names

Han Shen (Preferred)

Emails

****@rpi.edu (Confirmed)

,

****@gmail.com (Confirmed)

Personal Links

Career & Education History

Senior Research Engineer

Ant Group (antgroup.com)

2025 – Present

PhD student

Rensselaer Polytechnic Institute (rpi.edu)

2020 – 2025

Research Intern

IBM Thomas J. Watson Research Center (us.ibm.com)

2024 – 2024

Advisors, Relations & Conflicts

PhD Advisor

2020 – 2024

Expertise

Reinforcement learning

2020 – Present

Bilevel learning and optimization

2020 – Present

Publications

On Entropy Control in LLM-RL Algorithms
Han Shen
- ICLR 2026 Poster
- Readers: Everyone
Mitigating Forgetting in LLM Supervised Fine-Tuning and Preference Learning
Heshan Devaka Fernando, Han Shen, Parikshit Ram, Yi Zhou, Horst Samulowitz, Nathalie Baracaldo, Tianyi Chen
- Submitted to ICLR 2025
- Readers: Everyone
SEAL: Safety-enhanced Aligned LLM Fine-tuning via Bilevel Data Selection
Han Shen, Pin-Yu Chen, Payel Das, Tianyi Chen
- ICLR 2025 Poster
- Readers: Everyone
Principled Penalty-based Methods for Bilevel Reinforcement Learning and RLHF
Han Shen, Zhuoran Yang, Tianyi Chen
- ICML 2024 Poster
- Readers: Everyone
Mitigating Gradient Bias in Multi-objective Learning: A Provably Convergent Approach
Heshan Devaka Fernando, Han Shen, Miao Liu, Subhajit Chaudhury, Keerthiram Murugesan, Tianyi Chen
- 12 May 2023
- OpenReview Archive Direct Upload
- Readers: Everyone
Alternating Implicit Projected SGD and Its Efficient Variants for Equality-constrained Bilevel Optimization
Quan Xiao, Han Shen, Wotao Yin, Tianyi Chen
- 02 Feb 2023
- OpenReview Archive Direct Upload
- Readers: Everyone
On Penalty-based Bilevel Gradient Descent Method
Han Shen, Tianyi Chen
- ICML 2023 Poster
- Readers: Everyone
Mitigating Gradient Bias in Multi-objective Learning: A Provably Convergent Approach
Heshan Devaka Fernando, Han Shen, Miao Liu, Subhajit Chaudhury, Keerthiram Murugesan, Tianyi Chen
- Published: 01 Feb 2023, Last Modified: 14 Jan 2026
- ICLR 2023 notable top 5%
- Readers: Everyone
A Single-timescale Analysis for Stochastic Approximation with Multiple Coupled Sequences
Han Shen, Tianyi Chen
- Published: 31 Oct 2022, Last Modified: 15 Oct 2022
- NeurIPS 2022 Accept
- Readers: Everyone
Mitigating Gradient Bias in Multi-objective Learning: A Provably Convergent Stochastic Approach
Heshan Fernando, Han Shen, Miao Liu, Subhajit Chaudhury, Keerthiram Murugesan, Tianyi Chen
- 2022 (modified: 20 Nov 2022)
- CoRR 2022
- Readers: Everyone

View all 11 publications

Co-Authors