Yutong (Kelly) He
@electronickale
Followers
886
Following
1K
Media
16
Statuses
103
PhD student @mldcmu, I’m so delusional that doing generative modeling is my job
Pittsburgh, PA
Joined March 2021
✨ Love 4o-style image generation but prefer to use Midjourney? Tired of manual prompt crafting from inspo images?. PRISM to the rescue! 🖼️→📝→🖼️. We automate black-box prompt engineering—no training, no embeddings, just accurate, readable prompts from your inspo images! 1/🧵
3
31
84
🔥🔥🔥.
Tokenization has been the final barrier to truly end-to-end language models. We developed the H-Net: a hierarchical network that replaces tokenization with a dynamic chunking process directly inside the model, automatically discovering and operating over meaningful units of data
0
0
5
RT @RickyTQChen: Padding in our non-AR sequence models? Yuck. 🙅. 👉 Instead of unmasking, our new work *Edit Flows* perform iterative refine….
0
79
0
RT @FahimTajwar10: RL with verifiable reward has shown impressive results in improving LLM reasoning, but what can we do when we do not hav….
0
143
0
When the ddl is approaching and you are violently editing something you wrote a while ago
0
0
17
Check out our paper for all the details!.🔗 Project page: 📄 arXiv: 💻 Code: w/ @AlexRobey23 @smiurtitkii @yidingjiang @jnwilliams_cmu @pappasg69 @HamedSHassani @mittu1204 @rsalakhu @zicokolter 8/8.
github.com
The official implementation of PRISM: Automated Black-box Prompt Engineering for Personalized Text-to-Image Generation - KellyYutongHe/prism_demo
1
1
6
🧩 Multi-concept generation becomes intuitive too! PRISM lets you easily identify and combine different components from reference images into coherent scenes. 7/🧵
1
0
2
✏️ Want to tweak your generated image?. You can simply edit PRISM-generated prompts directly! Change outfits, poses or backgrounds, just modify the part of the prompt you want, no more mysterious embeddings or gibberish prompts! 6/🧵
1
0
4
📊 In our experiments, PRISM outperforms/matches baselines across popular text-to-image models (Stable Diffusion, DALL-E, Midjourney) on T2I personalization, creating human-readable prompts with superior visual accuracy—no more manual prompt engineering headaches! 5/🧵
1
0
2
🧠 Our solution: Inspired by LLM jailbreaking (yup, really), PRISM iteratively refines human-readable prompts via in-context learning from VLMs. A magic trio, Prompt Engineer Assistant (VLM), any T2I Generator, and Judge (another VLM), powers this feedback loop! 4/🧵
1
0
3
🤔 The problem?. Current personalized image generation needs model training, while automated prompt engineering often requires white-box access or produces unintelligible prompts that only work on specific models (looking at you 👀, 4o and Textual Inversion "<S*>" tokens). 3/🧵.
1
0
2
💡 PRISM automatically produces accurate, human-interpretable and transferable prompts that can capture concepts from your inspo images. And it only requires BLACK-BOX access to the text-to-image generative models! 2/🧵
1
0
4
Dear program chairs of all conferences, please don’t put a 5000 character limit on our rebuttal response, especially when the reviewers have more than ten 7500-character text boxes for them to write reviews, thank you so much.
2
0
30
RT @FahimTajwar10: Interacting with the external world and reacting based on outcomes are crucial capabilities of agentic systems, but exis….
0
94
0
RT @dylanjsam: Excited to share new work from my internship @GoogleAI !. Curious as to how we should measure the similarity between example….
0
41
0
RT @ssokota: Model-free deep RL algorithms like NFSP, PSRO, ESCHER, & R-NaD are tailor-made for games with hidden information (e.g. poker).….
0
60
0
RT @dylanjsam: To trust LLMs in deployment (e.g., agentic frameworks or for generating synthetic data), we should predict how well they wil….
0
40
0
RT @jmuiuc: A troubling incident unfolded at #NeurIPS2024, where a keynote speaker used a slide that perpetuated harmful stereotypes and ra….
0
30
0