Course Project

The course project will give the students a chance to explore deep generative modeling in greater detail. Course projects will be done in groups of up to 3 students and can fall into one or more of the following categories:


Project name Authors
Diffusion Jockey: Pre-trained Image Models for Controlled Audio Generation Varun Shenoy, Ben Auslin, Suppakit Waiwitlikhit
A healthcare agent (for low resource settings) Shreyas Lakhtakia
Caption Conditional Pokémon Generation Tony Sun, Channing Lee
EEG-to-Image: toward a more practical model Divija Hasteer, Prathik Rao, Adrien Specht
Generating High-Quality Brand Logos using Stable Diffusion Megan Dass, Pete Rushton, Myra Deng
Automating Code Review With Few-Shot In-Context Learning Yipeng Liu
Artist-Specific Finetuning for Generative Music Michael Xue, Edward Diller
Modeling an Inhospitable World: Forecasting climate change impacts on land use & cover using variational autoencoders Shayana Venukanthan, Divya Nagaraj
Traditional meets artificial: tile-based synthesis of scratchboard art using Stable Diffusion Miranda Li
Image Segmentation and Deep Generative In-painting Ehsan Shahrian, Jaisal Kothari, Kelvin Christian
Customized Doggie Portrait Liuxiao Zhang
Generating Galaxies for Strong Lens Simulations Sydney Erickson
Stable Diffusion Personalization Wen Xu
VirtuVogue: AI-Generated Garment & Virtual Try-On using Diffusion Models Keyuan Wu
Copyright Protection for Generative AI and Large Language Models Wei-Ning Chen
2D Game Asset Creation Using Stable Diffusion Zhouhua Xie
Few-Shot Video Generation with Recurrent Motion Prediction using Pre-trained Image Diffusion Models Ryan Li, Churan He, Yingying Chen
Asset Price Forecasting with Deep Generative Models: Predicting Daily Highs, Lows, and Closes for Targeted Trading Strategies Flor Lozano-Byrne, Masamine Someha, Dhruv Aggarwal
Learning Representations of NBA Player Performances Using Variational Autoencoder Jacob Deutsch
Towards Personalized Music Generation via Deep Reinforcement Learning Giancarlo Ricci
Fast Personal Stable Diffusion Nick Mecklenburg
NeRFBusters in the Stanford Wild Timothy Yao
Generative Diffusion Models for 3D Protein-Binding Molecule Structure Yilin Chen, Tianxiang Dai, Shuo Harper Hua
Direcet Preference Optimization for Diffusion Models Meihua Dang,
Increasing the Resolution of Low-Fidelity Physics Simulations Using Stable Diffusion Models Ali Lasemi, Marie-Jo Azzi, Joshua Barnett
Fusion video game characters Gautham Dixit
Generative Models for Synthetically Viable Materials Khalid El-Awady
Removing adverse weather impacts from autonomous vehicle camera systems with generative techniques Chunming Peng, Yun Kun Zhang, Jason Alexander Chan
StartGAN: Analyzing the Impact of Discriminator Pre-Training on Final Generator Performance Sri Jaladi, Andy Liang, Abhinav Sinha
SceneStitch: Garment Try-On in a World of Changing Horizons Zhengjia Huang,Li Tian,Jing Gong
Caption Conditional Pokémon Generation Tony Sun, Channing Lee
Project Milestone: Detecting the Shapes in a Slide from an Image using a Multimodal Generative Model Hashem Elezabi, Ahmed Sharaf
ACT + Diffusion Policy Yihe Tang, Max Sobol Mark, Avidesh F Marajh
Textual Inversion for 3D Generative Models Virginia Adams
Customizing and Enhancing Stable Diffusion for Style LoRA and Subject-Specific Generation Katherine Chen
ADISS: Ad-Hoc Indoor Scene Synthesis with 3D Reconstructed Objects Mofan Zhang, Martin JJ. Bucher, Shuojia Fu
Employing Deep Generative Models in Drug Discovery and Modeling Single-Cell Responses Denis Tolkunov, Billur Engin, Chi Wai Lau
Unconditional Discrete Diffusion Model for 3D Building Generation. Alberto Tono
Generative model of the joint distribution of protein sequence and structure for protein design Jinho Kim,Tianyu Lu,Talal Wadatalla
3D-aware Instruction-guided 3D Editing Mingjian Jiang, Xinyi Zhao, Zheng Wang
The Platonian "Form" in Variational Autoencoders: Dissecting Latent Space Properties Junyang Sun, Bi Tian Yuan
Diffusion-LM for Sentence Embeddings Maggie Wu
FACETIME: Age-Guided Child Face Generation from Parental Images Anirudh Sriram, Shoaib Mohammed
Generative Machine Learning for Modeling Viscoelasticity in Soft Tissues Jeremy McCulloch
Video Chapterization Using Generative Reconstruction Siqi Zuo
BachBot - Bridging Classical Elegance with AI Innovation Kyuil Lee, Dezhi Yu, Yongkang Huang
Zero-shot Image-to-3D Generation Using Diffusion Guidance and GAN Frank Zhao
Target-aware Molecule Generation via Direct Preference Optimization Siyi Carrie Gu
Diffusing Across the Spider-Verse Anuj Shetty, Huram-Abi Nzia Yotchoum, Daniel Sorvisto
3D Model Editing via Text Commands Fang Shu, Jian Xu, Wai Yam Gordon Tsai
Stable Diffusion and Beyond: text-to-image-to-video Jayadev Rajan, Arihant Jain
Generative Adversarial Networks in Anomaly Detection: Multivariate Time Series Analysis for Power System Rafael Carvalho Marques
Customization of Stable Diffusion Models with Different Methods Ang Li, Haishan Gao, Tommy Li
Correcting Human Posture in 2D Images with Stable Diffusion Will Baxley
Stable Diffusion and Beyond Xian Wu
Predicting Three-Dimensional RNA Structure with Graph-Based Diffusion Models Kaya Güvendi
Applying LoRA to Stable Diffusion with Impressionist Landscape Art Jensen Hwa
Exploring the Noise Space of Latent Diffusion Models Medhanie Irgau, Chinmay Lalgudi
Motion Interpolation via Autoregressive Joint Pose Sampling for Physics-Based Character Control Michael Piseno
Stable Diffusion with Extension to Text-to-Video Generation Yasuhiro Yoshida, Thomas Cheung, Qi Han
Generating VNG Characters with Pose Control Chenyang Dai, Jingwei Ma
A retraction-map based approach to image generation using Lipschitz contraction operators Joshua Kazdan, Rex Shen, Hao Sun
GPTTT: Teaching LLMs how to play Tic-Tac-Toe Mehul Arora
Inference-time Soft-Constrained Decoding For Autoregressive Language Models Atem Aguer
Multimodal Gen-AI for Fundamental Investment Research Hai Wang, Ting-Yu Chang
Search Methods for Image Generation Adam Klein
Synthetic data to train better spam filters Kai Wang
Diffusion-based Falsification of Cyber-Physical Systems Harrison Delecki, Marc Schlichting
Stable Diffusion and Beyond ChunWei Chan
Generating a building’s digital twin from a single street view image Kevin Mayer
RadGANvolve: CT-based synthetic MRI generation with GANs Reine Dayekh, Maayane Attias, Sushmita Belede
Generative Modeling for Immersive Game Scenes Yusef Qazi, Weidi Pan, Tushar Gupta
Subject-specific Customization of Video Diffusion Models Xinzhi Fan, Shaoheng Zhou, Masoud Charkhabi
Identity-Preserving Human Face Editing with Parameter-Efficient Training Ji Hun Wang, Youjin Song, Andrew Hyungmin Lee
A Generative Approach to Reconstruct Perceived Text with MEG Signals Shubh Khanna, Abel John
Plan My Trip Manolo Alvarez
Conditional Latent Diffusion Models for Accelerated MRI Yonatan Urman, Ashwin Kumar, Zachary Shah
Thumbs Up: Enhancing Accessibility of Content Creation Through Thumbnail Generation Sreya Halder, Lauren Kong, Jasmine Bilir
OccFacto: Controllable Part-Based Mesh Generation with Occupancy Networks Aniketh Iyengar, William Pan
Controlling Stable Diffusion for Outfit Swapping Ran Geng
Adoption of LLMs in Corporate Environments Yuwen Qiu, Cheng Shen, Saurabh Malik
Learned Optimizer with Diffusion Models Apoorv Srivastava, Chaitanya Pankajkumar Patel
Controllable Audio Generation with Auto-Regressive Self-Attention Modeling and Neural Compression William Shabecoff, Andrew Romero
On Cross-modal Information Retrieval using Latent-variable Spaces from Brain Activities Albert Cai, Omar Abul-Hassan
Accelerated Molecular Conformation Generation with Fast Diffusion Model Sampling Cesar Lema, Neal Vaidya, Romain Lacombe
Electripedia Tony Liu
Exploring Fine-Tuning Methods for Stable Diffusion Pura Peetathawatchai
Integrating Affective Dynamics into Deepfake Technology for Enhanced Emotional Realism in Talking Head Synthesis Aditya Bora, Nikhil Suresh
GANs for synthetic returns of correlated securities Henrique Monteiro
seg-zero: Diffusion Models for Controllable Human Face Editing Fan Pu Zeng, Jitong Zhou, Zhiyu Xie
Generative Pretraining of Medical AI Models Rui Deng, Bohan Li
Stable Diffusion and Beyond Nithin Akkati, Mubarak Ali Seyed Ibrahim
Abstract Graph Generation with DiGress Angelo Gagliardi
Fine-Tuning Stable Diffusion for Conditional Generative Stylist Application Miria Feng
Personalized Text-to-Image Generation with Identity Preservation Bofei Zhu
Generating Bubbles for Fluid Animations Kangrui Xue
3D-Aware Human Image Generation with Diffusion Paris Zhang
It's From Bits: Binary Generative Flows Tom Saberhagen
Diffusion Evolution: Evolving Pokemon with Stable Diffusion Pranay Agrawal, Josh Singh, Chethan Bhateja
Hare in Tortoise Shell (HITS): Improving Diffusion TTS using Non-Autoregressive Mel-Spectrograms latents Tomas Coghlan, Parth Sarthi, Salman Abdullah
Stable Diffusion and Beyond Ilias Kasoumis
Effectiveness of Generative Self-Supervised Learning on Neural Inertial Navigation Anson Wang
Comparative Study of Generative Models on mono-track music generation in the context of other tracks Krithika Ramakrishnan
Generative Models for Image Compression Spencer Graves
Generating 2D Materials with Deep Graph-based Models Giulia Socolof, Sathya Edamadaka
manipulativeLMs: Social Reasoning and Manipulative Behavior in Language Models Chinmaya Andukuri
Learned Chaos in Generative Models Through Reservoir Computing Quinn McIntyre, Dhruv Pai, Andres Carranza
You Text, We Sketch: Text-Guided Diffusion for Vectorized Sketch Generation Yash Shah, Samy Cherfaoui
MeGAN: Medical Imagery Generation via Learned Augmentations Xiluo He, Tom Nguyen
Prompt-based image generation of whole slide images (WSI) for tumor microenvironment characterization Adam Chun, Ekin Tiu
Teaching a language model a causal system through passive observations Ayush Chakravarthy
Physics Based Character Animation via diffusion models Takara Truong
Diffusion and Meta-learning for Molecule Conformation Generation Rafael Basto, Armando Borda, Evelyn Choi
Generative Models for Improving lung pathologies classification using X-rays Images Kuniaki Iwanami, Juan Pablo Triana Martinez
StarCraft 2 Fog Removal with Generative Models Cynthia Li, Yang Fan
Generating Accurate Quotations of Court Cases Sam Gensburg
Efficient Recovery of Text from Embeddings Ramgopal Venkateswaran
Geometric Latent Conditional Diffusion Models for 3D Molecule Generation Aksh Garg, Isaac Ju, Sanjay Nagaraj
Planning with Generative World Models in Robotics Hengyuan Hu, Jonathan Yang, Tianyu Hua
Cinematic Image Generation with Diffusion Models Arjun Karanam, Raghav Ganesh, Ronak Malde
Efficient Generative Modeling of Long-Horizon Videos Gabrael Levine, Jaden Clark
Extending FlowSite to Protein Sidechain Flexibility Milena Mathew
Tic-Tac-Train: Guiding ChatGPT to play tic-tac-toe optimally with logical reasoning Denise Lee, Benjamin Martinez, Samuel Do
End-to-End Semantic Anomaly Detection using a Visual Language Model Matt Foutter
Stable Diffusion and Beyond Ihor Leshchyshyn Jean Rodmond Jr Laguerre, Zouberou Sayibou
Deep Generative Models for Technical Analysis Michael Cho, Michael Marcotte
Facial Editing through ControlNet Jason Shu-Yang Wang, Yiming Ni
Adapting Text-to-Image Models for Video Generation Using Few Shot Learning Samir Agarwala, Hong Ju Jeon, Jared Watrous
Towards Layout-Aware Screenshot Generation Yuan Gao
Meta-Learning Online Adaptation of General Language Models Nathan Hu
Portfolio Diffusion: Generative modelling for investment portfolio construction Andrey Popov
Variational Autoencoder for Wildfire Risk Analysis In the Context of Financial Industry's Climate Scenario Analysis. Jonathan Hague
Improving Patch-Based Diffusion Models to Restore Images with Adverse Weather Conditions Kenny Dao, Viraj Mehta, Jeremy Tian
Diffusion-Based Generation of Previously Unseen Cell Biology Ayush Agrawal
Accelerating Parallel Diffusion Model Inference with Rectified Flows Sahil Jain
Sketch-to-CAD: Generating Computer Aided Design Models from Sketches Ghadi Nehme
Targeted Machine Unlearning for Diffusion Models Varun Desai, Nipun Argawala, Juan Muneton Gallego
Exploring Text-to-Image and Text-to-Text Generation Using Diffusion Models Roy Yuan, Constance Horng
Efficient Diffusion Model for Audio Inpainting Jadon Geathers, Kiran Bhat
Deep generative model based 3D image reconstruction for Electrical Impedance Tomography Haijing Zhang
Redessian Models - A Novel Regularisation Method for Disentangled Representation Learning Jubayer Hamid, Andy Tang, Sheryl Hsu
Tic-Tac-Toe for LLMs Jihee Suh
Stable Diffusion and Beyond Aniruddh Ramrakhyani
A Comparative Study of Generative AI Techniques for Music Durga Malladi, Ranajit Gangopadhyay, Jacob Smith
Towards Better Facial Image Editing Carson Murtuza Lanier, Charlotte Nicks
Revision Models: Exploring Sequence Regeneration in Language Models Jay Martin, Manan Patel, Shrey Verma
Consistency Accelerated Robot Inference Aaditya Prasad
Complex Pattern Recognition in LLMs: Using Language Models to Identify "Disruptive Innovation" Matthew Wise
Mitigating Out-of-Distribution Inference Errors in Score-Based Generative Models Derek Chong, Matthew Jin, Jenny Xu
Instruct-3D-to-3D: Editing 3D Scenes with Instructions Chi Yo Tsai, Eliza Huang, Fan Xie
A Bond-Prediction-Focused Approach to Novel Molecular Generation of Large Molecules Bay Foley-Cox, David Wendt
VisioNet: Vision-to-Audio Generation with Large Language Model Yingke Wang, Han Bai, Haoming Zou
Voxel Diffusion Brendon Matusch, James Zheng
A Comparative Study of Voice Synthesis Techniques Afshin Babveyh
Finetuning text to image diffusion models for transparent object image synthesis Kevin Lin
Drawing Wizard: Image-to-image Translation with Conditional GANs Yujie Gao
Sequence (re)-generation with different orderings Nikil Ravi, Emil Biju, Sagnik Bhattacharya
FermiGPT: Combining programmatic reasoning of LLMs with Retrieval-Augmented Generation Sidhart Krishnan
Audio Reactive AI Generated DJ Visuals Adam Zhao, Hannah Kim, Matt Reed
Robust Text-based Image Editing through Cross-Attention in Diffusion Models Yining Mao
Mistral 7B Chess Ozan Adiguzel
Multi-Singer Voice Interpolation: A Deep Generative Approach to Hybrid Voice Creation in Singing Guinness Chen
Subject Specific Fine-tuning to Create Synthetic Datasets for Computer Vision in Sports Arnav Gangal
When At First You Don’t Succeed: Knowing When to Try Again in Novel Test-Time Scenarios Maximilian Du
Style2Art: Artwork Generation with Style and Subject Jiacheng (JC) Hu
Improving Human Hands in Stable Diffusion Matthew Hayes
EEG-to-fMRI Generation for Data Synthesis in Brain Sciences and Diagnosis Binxu Li, Jiaqi Wu
Study of Semiconductor Wafer Patterns with Deep Generative models Chuanqi Chen