M.E. student in Computational Science & Engineering at Harvard UniversityI am a master's student in Computational Science and Engineering at Harvard University, supported by the Amazon Robotics Day One Fellowship, and I am doing my master's thesis with Prof. Debora Marks at Harvard Medical School.
My research is in AI for science, especially generative models for molecular design in biology. I currently build foundation-model-based methods to design RNA that binds target proteins. Before that, I pretrained protein language models for antibody discovery at Pfizer. Earlier, I worked on world models for robotics, computer vision, and soft robotics at Harvard, CMU, and WPI.
I am applying to Ph.D. programs for Fall 2027.
Education
-
Harvard UniversityM.E. in Computational Science & Engineering
Amazon Day One FellowAug. 2025 - May 2027 -
Worcester Polytechnic InstituteB.S. in Computer ScienceAug. 2021 - May 2025
Experience
-
Harvard Medical School, Marks LabMaster's Thesis Student (Prof. Debora Marks)Sep. 2026 - present
-
AmazonRobotic Data Science InternMay 2026 - present -
Harvard UniversityResearch Student (Prof. Yilun Du)Jan. 2026 - Sep. 2026
-
Pfizer, AI for Drug DiscoveryIntern (Jun. - Dec. 2025)
Co-op (Sep. - Dec. 2024)2024 - 2025 -
Carnegie Mellon University, Robotics InstituteResearch Assistant (Prof. Howie Choset)Jul. 2024 - Sep. 2024 -
Beth Israel Deaconess Medical CenterResearch Student (Prof. Feng Tian)Sep. 2023 - May 2024 -
Worcester Polytechnic InstituteResearch Assistant (Profs. Ziming Zhang,
Markus Nemitz, Jacob Whitehill)May 2022 - Jun. 2024 -
Olympus Corporation of the AmericasR&D Co-opJan. 2023 - Jul. 2023
Honors & Awards
-
Amazon Fulfillment Technologies & Robotics Day One FellowshipSep. 2025
-
WPI Presidential Scholarship (full four years)2021 - 2025
-
MIT Math Prize for GirlsOct. 2021
-
American Mathematics Competition, National Top 5%Jan. 2021
News
Selected Publications (view all )
The Missing Primitive: Diagnosing and Repairing Mathematical Reasoning in Large Language Models
Shuo Xing*, Zilin Dai*, Chengyuan Qian, Fangzhou Lin, Wenjing Chen, Ping He, Pan Lu, Alvaro Velasquez, Mohit Bansal, Zhengzhong Tu (* equal contribution)
arXiv preprint 2026
We take a first step toward systematically studying mathematical understanding in LLMs. We introduce the notion of Mathematical Primitive to probe structural mathematical understanding and propose Prim, a benchmark that evaluates mathematical reasoning along four dimensions: Discovery, Generation, Digestion, and Execution. Our diagnosis shows that solution accuracy masks distinct capability profiles and that Discovery is the dominant bottleneck. Building on these findings, we introduce Absorb, a primitive-privileged self-distillation framework that consistently improves mathematical reasoning across model scales and challenging benchmarks.
The Missing Primitive: Diagnosing and Repairing Mathematical Reasoning in Large Language Models
Shuo Xing*, Zilin Dai*, Chengyuan Qian, Fangzhou Lin, Wenjing Chen, Ping He, Pan Lu, Alvaro Velasquez, Mohit Bansal, Zhengzhong Tu (* equal contribution)
arXiv preprint 2026
We take a first step toward systematically studying mathematical understanding in LLMs. We introduce the notion of Mathematical Primitive to probe structural mathematical understanding and propose Prim, a benchmark that evaluates mathematical reasoning along four dimensions: Discovery, Generation, Digestion, and Execution. Our diagnosis shows that solution accuracy masks distinct capability profiles and that Discovery is the dominant bottleneck. Building on these findings, we introduce Absorb, a primitive-privileged self-distillation framework that consistently improves mathematical reasoning across model scales and challenging benchmarks.
AdaptFuse: Training-Free Sequential Preference Inference via Externalized Bayesian Computation
Fangzhou Lin*, Zilin Dai*, Shuo Xing, Peiran Li, Siyuan Yang, Qianwen Ge, Kazunori Yamada, Yang Zhou, Ziming Zhang, Haichong Zhang, Zhengzhong Tu (* equal contribution)
arXiv preprint 2026
LLMs struggle to accumulate evidence across multiple rounds of user interaction in a manner consistent with Bayesian inference. We propose AdaptFuse, a training-free framework that externalizes probabilistic computation from the LLM: a symbolic module maintains a Bayesian posterior over a discrete hypothesis set, while a frozen LLM contributes semantic reasoning via multi-sample Dirichlet aggregation, and the two are combined through entropy-adaptive fusion. Across flight recommendation, hotel recommendation, and web shopping, AdaptFuse consistently outperforms both prompting baselines and fine-tuned Bayesian Teaching models, without storing or training on sensitive user data.
AdaptFuse: Training-Free Sequential Preference Inference via Externalized Bayesian Computation
Fangzhou Lin*, Zilin Dai*, Shuo Xing, Peiran Li, Siyuan Yang, Qianwen Ge, Kazunori Yamada, Yang Zhou, Ziming Zhang, Haichong Zhang, Zhengzhong Tu (* equal contribution)
arXiv preprint 2026
LLMs struggle to accumulate evidence across multiple rounds of user interaction in a manner consistent with Bayesian inference. We propose AdaptFuse, a training-free framework that externalizes probabilistic computation from the LLM: a symbolic module maintains a Bayesian posterior over a discrete hypothesis set, while a frozen LLM contributes semantic reasoning via multi-sample Dirichlet aggregation, and the two are combined through entropy-adaptive fusion. Across flight recommendation, hotel recommendation, and web shopping, AdaptFuse consistently outperforms both prompting baselines and fine-tuned Bayesian Teaching models, without storing or training on sensitive user data.
Frozen Language Anchors as Noise-Robust Priors for Domain Generalization under Label Noise: An Empirical Study
Zilin Dai, Lehong Wang, Fangzhou Lin, Ziming Zhang, Wang Lu
European Conference on Computer Vision (ECCV) E.T. 2026
Real-world machine learning applications often face both distribution shift and label noise. We study semantic representations from frozen language models as domain-invariant anchors, and use a weighted loss that adjusts each sample's contribution based on its similarity to the corresponding language anchor, making models more robust to noisy labels while generalizing across unseen domains.
Frozen Language Anchors as Noise-Robust Priors for Domain Generalization under Label Noise: An Empirical Study
Zilin Dai, Lehong Wang, Fangzhou Lin, Ziming Zhang, Wang Lu
European Conference on Computer Vision (ECCV) E.T. 2026
Real-world machine learning applications often face both distribution shift and label noise. We study semantic representations from frozen language models as domain-invariant anchors, and use a weighted loss that adjusts each sample's contribution based on its similarity to the corresponding language anchor, making models more robust to noisy labels while generalizing across unseen domains.
DeformX: A Versatile Co-Simulation Framework for Deformable Linear Objects
Yi Yang*, Xiang Fei*, Lehong Wang*, Chenhao Li, Zilin Dai, Henry Kou, Lu Li, Howie Choset (* equal contribution)
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2026
DeformX couples a Cosserat rod physics engine with NVIDIA Isaac Sim to simulate deformable linear objects such as cables and ropes with physically faithful dynamics, self-collision, and contact with free-form meshes, while rendering them realistically through mesh skinning. Fine-tuning SAM3 on DeformX data improves real-image wire segmentation by 10.2% mAP@75, and a rope-swinging policy trained purely in simulation transfers to a real UR5e arm.
DeformX: A Versatile Co-Simulation Framework for Deformable Linear Objects
Yi Yang*, Xiang Fei*, Lehong Wang*, Chenhao Li, Zilin Dai, Henry Kou, Lu Li, Howie Choset (* equal contribution)
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2026
DeformX couples a Cosserat rod physics engine with NVIDIA Isaac Sim to simulate deformable linear objects such as cables and ropes with physically faithful dynamics, self-collision, and contact with free-form meshes, while rendering them realistically through mesh skinning. Fine-tuning SAM3 on DeformX data improves real-image wire segmentation by 10.2% mAP@75, and a rope-swinging policy trained purely in simulation transfers to a real UR5e arm.
WireSeg-32K: A Physics-Grounded Synthetic Dataset for Wire Instance Segmentation
Zilin Dai, Lehong Wang, Yi Yang, Xiang Fei
Synthetic Data for Computer Vision Workshop @ CVPR 2026
Wires and cables are hard to segment because they are thin, highly deformable, and frequently self-occluded, and instance-level annotations are expensive. We present WireSeg-32K, a synthetic dataset with 32,000 RGB images, instance masks, depth maps, and a real-world test set, generated by coupling Cosserat-rod dynamics with photorealistic Isaac Sim rendering. LoRA fine-tuning SAM3 on WireSeg-32K alone improves real-world mAP@75 by 10.2%.
WireSeg-32K: A Physics-Grounded Synthetic Dataset for Wire Instance Segmentation
Zilin Dai, Lehong Wang, Yi Yang, Xiang Fei
Synthetic Data for Computer Vision Workshop @ CVPR 2026
Wires and cables are hard to segment because they are thin, highly deformable, and frequently self-occluded, and instance-level annotations are expensive. We present WireSeg-32K, a synthetic dataset with 32,000 RGB images, instance masks, depth maps, and a real-world test set, generated by coupling Cosserat-rod dynamics with photorealistic Isaac Sim rendering. LoRA fine-tuning SAM3 on WireSeg-32K alone improves real-world mAP@75 by 10.2%.
From Flat to Form-Fitting: A Computational Geometry Approach to 3D Conformal Electronics Design and Rapid Prototyping
Lehong Wang, Zilin Dai, Yuchen Wu, Ye Jin, Colm Barry, Tina Tian, Fujun Ruan, Shricharan Shastry, Rahul Panat, Gary K. Fedder, Howie Choset, Lu Li
IEEE International Conference on Automation Science and Engineering (CASE) 2025
We present a software pipeline that maps existing 2D PCB designs onto arbitrary, highly curved 3D surfaces and generates printing motions for aerosol jet additive manufacturing, removing the need to hand-design 3D circuit traces in CAD and enabling rapid prototyping of conformal electronics.
From Flat to Form-Fitting: A Computational Geometry Approach to 3D Conformal Electronics Design and Rapid Prototyping
Lehong Wang, Zilin Dai, Yuchen Wu, Ye Jin, Colm Barry, Tina Tian, Fujun Ruan, Shricharan Shastry, Rahul Panat, Gary K. Fedder, Howie Choset, Lu Li
IEEE International Conference on Automation Science and Engineering (CASE) 2025
We present a software pipeline that maps existing 2D PCB designs onto arbitrary, highly curved 3D surfaces and generates printing motions for aerosol jet additive manufacturing, removing the need to hand-design 3D circuit traces in CAD and enabling rapid prototyping of conformal electronics.
Fully FDM-Printed Soft Hand Exoskeleton
Zilin Dai*, Lehong Wang*, Markus P. Nemitz (* equal contribution)
ICRA Workshop on Emerging Technologies in Smart Exoskeleton Systems 2024
We design parameterized soft pneumatic hand exoskeletons that can be printed entirely on standard FDM printers, making personalized assistive devices affordable and easy to fabricate.
Fully FDM-Printed Soft Hand Exoskeleton
Zilin Dai*, Lehong Wang*, Markus P. Nemitz (* equal contribution)
ICRA Workshop on Emerging Technologies in Smart Exoskeleton Systems 2024
We design parameterized soft pneumatic hand exoskeletons that can be printed entirely on standard FDM printers, making personalized assistive devices affordable and easy to fabricate.
Vision-based FDM Printing for Fabricating Airtight Soft Actuators
Zilin Dai*, Yijia Wu*, Haotian Liu, Lehong Wang, Markus P. Nemitz (* equal contribution)
IEEE International Conference on Soft Robotics (RoboSoft) 2024 (Oral)
Pneumatic soft robots are typically fabricated by manual molding. We propose a low-cost approach to improve the print quality of desktop fused deposition modeling by adding a webcam that monitors printing and detects and corrects defects such as holes or gaps, improving the air-tightness of printed pneumatic actuators without fine-tuning printing parameters.
Vision-based FDM Printing for Fabricating Airtight Soft Actuators
Zilin Dai*, Yijia Wu*, Haotian Liu, Lehong Wang, Markus P. Nemitz (* equal contribution)
IEEE International Conference on Soft Robotics (RoboSoft) 2024 (Oral)
Pneumatic soft robots are typically fabricated by manual molding. We propose a low-cost approach to improve the print quality of desktop fused deposition modeling by adding a webcam that monitors printing and detects and corrects defects such as holes or gaps, improving the air-tightness of printed pneumatic actuators without fine-tuning printing parameters.
In Search of Negative Moments: Multi-Modal Analysis of Teacher Negativity in Classroom Observation Videos
Zilin Dai, Andrew McReynolds, Jacob Richard Whitehill
International Conference on Educational Data Mining (EDM) 2023
We build automatic and semi-automatic systems that combine speech emotion recognition, language analysis, and facial expression detection to find moments of teacher negativity in classroom observation videos.
In Search of Negative Moments: Multi-Modal Analysis of Teacher Negativity in Classroom Observation Videos
Zilin Dai, Andrew McReynolds, Jacob Richard Whitehill
International Conference on Educational Data Mining (EDM) 2023
We build automatic and semi-automatic systems that combine speech emotion recognition, language analysis, and facial expression detection to find moments of teacher negativity in classroom observation videos.
