News
- [Mar 2024]
MIT News coverage of our ICLR 2024 paper on grounding language plans in demonstrations.
- [Mar 2024]
Invited talk at Brown University Robotics Seminar.
- [Mar 2024]
Invited talk at University of Utah.
- [Jan 2024]
Check out our Generative AI newsletter from MIT's work of the future group.
- [Oct 2023]
Temporal logic imitation was awarded the Best Student Paper at IROS 2023 Workshop.
- [Sep 2023]
Selected as Work of the Future Fellow in Generative AI.
- [Jan 2023]
PBS News coverage of our human robot interaction exhibition at MIT museum.
Research
I'm interested in human robot (or any AI system such as LLM) interaction--for example, how we can leverage past interaction data to personalize and robustify future robot executions under adversarial perturbations.
My PhD work applies continuous motion imitation to long-horizon manipulation tasks with discrete structures, aka interactive task and motion imitation.
|
|
Grounding Language Plans in Demonstrations through Counter-factual Perturbations
Yanwei Wang,
Tsun-Hsuan Wang,
Jiayuan Mao,
Michael Hagenow,
Julie Shah
arxiv /
code /
project page
ICLR 2024 (Spotlight, acceptance rate: 5%)
This work learns grounding classifiers for LLM planning. By locally perturbing a few human demonstrations, we augment
the dataset with more successful executions and failing counterfactuals. Our end-to-end explanation-based network is
trained to differentiate successes from failures and as a by-product learns classifiers that ground continuous states
into discrete manipulation mode families without dense labeling.
|
|
Temporal Logic Imitation: Learning Plan-Satisficing Motion Policies from Demonstrations
Yanwei Wang,
Nadia Figueroa, Shen Li, Ankit Shah, Julie Shah
arxiv
/
code
/
project page
CoRL 2022 (Oral, acceptance rate: 6.5%)
IROS 2023 Workshop ( Best Student Paper, Learning Meets Model-based Methods for Manipulation and Grasping Workshop)
We present a continuous motion imitation method that can provably satisfy any discrete plan specified by a Linear Temporal Logic (LTL) formula. Consequently, the imitator is robust to both task- and motion-level disturbances and guaranteed to achieve task success.
|
|
Improving Small Language Models on PubMedQA via Generative Data Augmentation
Zhen Guo,
Yanwei Wang,
Peiqi Wang,
Shangdi Yu
arxiv
KDD 2023 (Foundations and Applications in Large-scale
AI Models Pre-training, Fine-tuning, and Prompt-based Learning Workshop)
We prompt large language models to augment a domain-specific dataset to train specialized small language models that outperform the general-purpose LLM.
|
|
Visual Pre-training for Navigation: What Can We Learn from Noise?
Yanwei Wang,
Ching-Yun Ko,
Pulkit Agrawal
arxiv
/
code
/
project page
IROS 2023
NeurIPS 2022 Workshop (Synthetic Data for Empowering ML Research / Self-Supervised Learning)
By learning how to pan, tilt and zoom its camera to focus on random crops of a noise image, an embodied agent can pick up navigation skills in realistically simulated environments.
|
|
MIT Museum Interactive Robot Exhibition: Teach a Robot Motions
Nadia Figueroa,
Yanwei Wang,
Julie Shah
We installed an interactive exhibition at MIT Museum that allows non-robot-experts to teach a robot an inspection task using demonstrations. The robustness and compliance of the learned motion policy enables visitors (including kids) to physically perturb the system safely 24/7 without losing a success gaurantee.
|
|
Active Learning for Object Search
Yanwei Wang,
Todd Murphey
This research project studies the application of two information gain methods in a search problem: 1. Infotaxis - Search guided by entropy minimization 2. Ergodic exploration - Search guided by proportional coverage.
|
|
Active Sensing with Tactile Sensors
Yanwei Wang,
Mitra Hartmann
This research project studies how active sensing, i.e., choosing what data to collect, can improve data efficiency for decision-making under uncertainty. Inspired by the active whisking behavior of rats, we use simulated rat whisker sensory signals as a model for spatial-temporal data to learn policies that first collect observations and then classify object shapes.
|
|
Multi-agent Distributed Sensing and Control
Yanwei Wang,
Michael Rubinstein
This research project studies multi-agent distributed algorithms concerning coordination, segregation, and locomotion, with a hardware implementation of robust localization with cheap sensors on a low-cost underactuated system.
|
|
Analyzing Energy Efficiency of Bio-Inspired Wind Generator
Yanwei Wang,
Richard Wolfson
pdf
For my undergraduate physics thesis at Middlebury College, I did a computational fluid dynamics (CFD) simulation of Festo's Dual-Wing generator using COMSOL CFD package.
|
Projects
Miscellaneous class projects.
|
|