Prefix-Tuning: Optimizing Continuous Prompts for Generation [PDF]
Fine-tuning is the de facto way of leveraging large pretrained language models for downstream tasks. However, fine-tuning modifies all the language model parameters and therefore necessitates storing a full copy for each task.
Xiang Lisa Li, Percy Liang
semanticscholar +1 more source
Hierarchical Text-Conditional Image Generation with CLIP Latents [PDF]
Contrastive models like CLIP have been shown to learn robust representations of images that capture both semantics and style. To leverage these representations for image generation, we propose a two-stage model: a prior that generates a CLIP image ...
A. Ramesh +4 more
semanticscholar +1 more source
DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation [PDF]
Large text-to-image models achieved a remarkable leap in the evolution of AI, enabling high-quality and diverse synthesis of images from a given text prompt.
Nataniel Ruiz +5 more
semanticscholar +1 more source
Survey of Hallucination in Natural Language Generation [PDF]
Natural Language Generation (NLG) has improved exponentially in recent years thanks to the development of sequence-to-sequence deep learning technologies such as Transformer-based language models. This advancement has led to more fluent and coherent NLG,
Ziwei Ji +11 more
semanticscholar +1 more source
CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation [PDF]
Pre-trained models for Natural Languages (NL) like BERT and GPT have been recently shown to transfer well to Programming Languages (PL) and largely benefit a broad set of code-related tasks.
Yue Wang +3 more
semanticscholar +1 more source
Competition-level code generation with AlphaCode [PDF]
Programming is a powerful and ubiquitous problem-solving tool. Systems that can assist programmers or even generate programs themselves could make programming more productive and accessible.
Yujia Li +29 more
semanticscholar +1 more source
ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillation [PDF]
Score distillation sampling (SDS) has shown great promise in text-to-3D generation by distilling pretrained large-scale text-to-image diffusion models, but suffers from over-saturation, over-smoothing, and low-diversity problems. In this work, we propose
Zhengyi Wang +6 more
semanticscholar +1 more source
I am grateful to my colleague and friend, Editor-in-Chief Sherwood Brown, for this opportunity to extoll Dr. Alan Green, MD (1953–2020) as a humanitarian and as a clinician scientist.
openaire +2 more sources
BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension [PDF]
We present BART, a denoising autoencoder for pretraining sequence-to-sequence models. BART is trained by (1) corrupting text with an arbitrary noising function, and (2) learning a model to reconstruct the original text.
M. Lewis +7 more
semanticscholar +1 more source
Imagen Video: High Definition Video Generation with Diffusion Models [PDF]
We present Imagen Video, a text-conditional video generation system based on a cascade of video diffusion models. Given a text prompt, Imagen Video generates high definition videos using a base video generation model and a sequence of interleaved spatial
Jonathan Ho +10 more
semanticscholar +1 more source

