# Hang Yu 余航

Source: https://hangyu8123.github.io/

Ph.D. Candidate · Tufts University · HRI & Robot Learning

I am a Ph.D. candidate in Computer Science at Tufts University, advised by Dr. Elaine Short in the Assistive Agent and Behavior Learning lab. My work focus on Human-Robot Interaction, Robot Learning, VLAs, and Agentic Robotics. Most of my datasets are open-sourced and were collected in public spaces with real-world, non-expert users. Right now I am a research intern at ABB Robotics, working on building new VLA models for bimanual manipulation and exploring agentic robotics for industrial robotics.

[Email](mailto:hyu08@tufts.edu)   [Google Scholar](https://scholar.google.com/citations?user=j0qxKQIAAAAJ&hl=en&oi=ao)   [GitHub](https://github.com/HangYu8123)   [LinkedIn](https://www.linkedin.com/in/hang-yu-0343b2273/)   [CV (PDF)](https://hangyu8123.github.io/assets/cv.pdf)

## About

I am a Ph.D. candidate in Computer Science at Tufts University, advised by Dr. Elaine Short in the  [Assistive Agent and Behavior Learning lab (AABL)](https://aabl.cs.tufts.edu/) . I work on improving robot learning from multi-modular human teaching, including human feedback, human demonstrations, and outputs from VLM/LLM. Most of my data is open-sourced and collected from public spaces with real-world non-expert users.

Actively open to research collaborations in HRI, RLHF for robots, and human data collection methodology.

### Research Interests

Topics

Interactive RL • Learning from Human Feedback • Learning from Demonstrations • Human modeling for robot learning • Evaluation & datasets for HRI

Keywords

HRI, RLHF/RLAIF, IntRL, Preference learning, Human factors

![Human teaching a robot using kinesthetic teaching demonstrations for imitation learning](https://hangyu8123.github.io/assets/img/interests-1.png)

![Interface for robot learning from rich human feedback showing scalar and directional inputs](https://hangyu8123.github.io/assets/img/interests-2.jpg)

## News

- Present Internship I am a Research Intern at ABB Robotics , working on VLA for bimanual manipulation.

- Present Talk I gave a talk at Yale University .

- Aug 2025 Mentoring Co-authored three papers with my mentees at IEEE RO-MAN 2025 .

- May 2025 Award Selected for the Doctoral Consortium : "Enabling Robust Learning from Non-Experts" at IEEE ICRA 2025 .

- Aug 2024 Award Recipient of the Student Travel Award at IEEE RO-MAN 2024 .

- Oct 2023 Paper Published "From 'Thumbs Up' to '10 out of 10'" at IEEE/RSJ IROS 2023 .

## Education

Tufts University

#### Ph.D. in Computer Science (Human–Robot Interaction)

GPA 4.0/4.0 — Expected Mar. 2026
Advisor: Dr. Elaine Schaertl Short

Tufts University

#### M.S.E. in Computer Science

GPA 3.95/4.0 — Jan. 2021
Advisor: Dr. Elaine Schaertl Short

## Publications

### Conference Papers

![CHARM framework diagram showing human attribute modeling for reinforcement learning](https://hangyu8123.github.io/assets/img/papers/charm.png)

#### CHARM: Considering Human Attributes for Reinforcement Modeling

Qidi Fang*, Hang Yu* , Shijie Fang, Jindan Huang, Qiuyu Chen, Reuben M. Aronson, Elaine S. Short.

IEEE RO-MAN 2025 , Eindhoven, NL.

[arXiv](https://arxiv.org/abs/2506.13079)  [PDF](https://arxiv.org/pdf/2506.13079)  [Code](https://github.com/AABL-Lab/CHARM)

![Visual examples of systematic non-optimality in human demonstrations from the Sidetracks dataset](https://hangyu8123.github.io/assets/img/papers/sidetrack.jpg)

#### Demonstration Sidetracks: Categorizing Systematic Non-Optimality in Human Demonstrations

Shijie Fang*, Hang Yu* , Qidi Fang, Reuben M. Aronson, Elaine S. Short.

IEEE RO-MAN 2025 , Eindhoven, NL.

[arXiv](https://arxiv.org/abs/2506.11262)  [PDF](https://arxiv.org/pdf/2506.11262)  [Dataset / Code](https://github.com/AABL-Lab/Human-Demonstration-Sidetracks)

![Cozmo robot face display designs testing expressiveness and clarity for HRI](https://hangyu8123.github.io/assets/img/papers/eyes.png)

#### See What I Mean? Expressiveness and Clarity in Robot Display Design

Matthew Ebisu, Hang Yu , Reuben M. Aronson, Elaine S. Short.

IEEE RO-MAN 2025 , Eindhoven, NL.

[arXiv](https://arxiv.org/abs/2506.16643)  [PDF](https://arxiv.org/pdf/2506.16643)  [Code](https://github.com/mattufts/huamn_Cozmo_interaction)

![Human feedback signal interface showing progress bars for teaching robots](https://hangyu8123.github.io/assets/img/papers/progress.jpg)

#### How Much Progress Did I Make? An Unexplored Human Feedback Signal for Teaching Robots

Hang Yu , Qidi Fang, Shijie Fang, Reuben M. Aronson, Elaine S. Short.

IEEE RO-MAN 2024 , Pasadena, USA.

[arXiv](https://arxiv.org/abs/2407.06459)  [PDF](https://aabl.cs.tufts.edu/papers/roman2024_hang.pdf)  [Dataset](https://github.com/TeachingwithProgress/Non-Expert_Demonstrations)

![Comparison of scalar feedback versus '10 out of 10' feedback scales in interactive reinforcement learning](https://hangyu8123.github.io/assets/img/papers/steady.png)

#### From “Thumbs Up” to “10 out of 10”: Reconsidering Scalar Feedback in Interactive Reinforcement Learning

Hang Yu , Reuben M. Aronson, Katherine H. Allen, Elaine S. Short.

IEEE/RSJ IROS 2023 , Detroit, USA.

[arXiv](https://arxiv.org/abs/2311.10284)  [PDF](https://arxiv.org/pdf/2311.10284)

### Journal Publications

![Diagram illustrating Top-K Interesting Preference Rules Mining Based on MaxClique](https://hangyu8123.github.io/assets/img/papers/top-k.png)

#### Top-K Interesting Preference Rules Mining Based on MaxClique

Zheng Tan, Jing-lei Liu, Hang Yu , Wei Wei.

Expert Systems with Applications  143 (2020) 113043.

[DOI](https://doi.org/10.1016/j.eswa.2019.113043)  [ScienceDirect](https://www.sciencedirect.com/science/article/abs/pii/S0957417419307602)

![Framework for Contextual Preference Collaborative Measure Based on Belief System](https://hangyu8123.github.io/assets/img/papers/belief.png)

#### Contextual Preference Collaborative Measure Framework Based on Belief System

Hang Yu , Wei Wei, Zheng Tan, Jing-lei Liu.

Computer Science (计算机科学)  47(4):74–84, 2020.

[Journal](https://computerjournals.net/cn/simple_view_abstract.aspx?aid=1510D19A1395533AD2A8D5AF0923A594&jid=64A12D73428C8B8DBFB978D04DFEB3C1)

![MaxClique based Conditional Preference Mining diagram](https://hangyu8123.github.io/assets/img/papers/maxclique.png)

#### Conditional Preference Mining Based on MaxClique

Zheng Tan, Jing-lei Liu, Hang Yu .

Journal of Computer Applications (计算机应用)  37(11):3107–3114, 2017.

[Journal](https://computerjournals.net/cn/simple_view_abstract.aspx?aid=F1F705A1C719906AB225E536701F0222&jid=831E194C147C78FAAFCC50BC7ADD1732)

### Workshops & Doctoral Consortium

![Diagram for Human-Robot Interaction via Expressive Human Feedback](https://hangyu8123.github.io/assets/img/papers/icra_dc.jpg)

#### Doctoral Consortium: Human-Robot Interaction via Expressive Human Feedback

ICRA 2025 Doctoral Consortium

![PHIRL framework for Progress-Heuristicized Inverse Reinforcement Learning](https://hangyu8123.github.io/assets/img/papers/phirl.png)

#### PHIRL: Progress-Heuristicized Inverse Reinforcement Learning

RSS Workshop 2025

![Active Feedback Learning interface](https://hangyu8123.github.io/assets/img/papers/active.jpg)

#### Active Feedback Learning with Rich Feedback

HRI ’21 Companion  ·  [ACM DL](https://dl.acm.org/doi/10.1145/3434074.3447207)

## Experience & Service

Research

- Research Intern , ABB Robotics — VLA for bimanual manipulation (present)

- Graduate Researcher , AABL Lab, Tufts University — advised by Dr. Elaine Schaertl Short

Teaching Assistant

- Ethics for AI, Robotics, and HRI — Spring 2024, Spring 2025

- Human–Robot Interaction — Fall 2022

- Human–Computer Interaction — Spring 2021

- C++ Programming — Fall 2016

Reviewer / Program Committee

- HRI (2021–2024)

- ICRA (2023–2025)

- CoRL (2025)

- AAAI (2026 PC)

- AAMAS (2026 PC)

Honors & Awards

- 2025: ICRA Doctoral Consortium

- 2024: RO-MAN Travel Grant

- 2017: Yantai University Outstanding Student Scholarship

- 2016: Lanqiao Programming Competition National First Prize

## A Little More about Me

Photographs and a short film from outside the lab.

[Short film](https://hangyu8123.github.io/assets/img/life/video.mp4)

![Outside research photo 1](https://hangyu8123.github.io/assets/img/life/img1.jpg)

![Outside research photo 2](https://hangyu8123.github.io/assets/img/life/img2.jpg)

![Outside research photo 3](https://hangyu8123.github.io/assets/img/life/img3.jpg)

![Outside research photo 4](https://hangyu8123.github.io/assets/img/life/img4.jpg)

## Contact

- Email [hyu08@tufts.edu](mailto:hyu08@tufts.edu)

- Scholar [Google Scholar profile](https://scholar.google.com/citations?user=j0qxKQIAAAAJ&hl=en&oi=ao)

- GitHub [github.com/HangYu8123](https://github.com/HangYu8123)

- LinkedIn [hang-yu-0343b2273](https://www.linkedin.com/in/hang-yu-0343b2273/)

- CV [Download (PDF)](https://hangyu8123.github.io/assets/cv.pdf)

- Lab [AABL, Tufts University](https://aabl.cs.tufts.edu/)

Open to research collaborations in HRI, RLHF for robots, and human data collection methodology.