@debuinu.com
Reposted
I asked ChatGPT to tell me about the history of the Super Mario Bros any% world record.
I think my job is safe for a while...
I think my job is safe for a while...
January 18, 2025 at 12:54 AM
I asked ChatGPT to tell me about the history of the Super Mario Bros any% world record.
I think my job is safe for a while...
I think my job is safe for a while...
Cool paper where they use a conv/natten-like scanning method of attention to dramatically speed up FLUX, especially with large sizes.
arxiv.org/abs/2412.16112
arxiv.org/abs/2412.16112
CLEAR: Conv-Like Linearization Revs Pre-Trained Diffusion Transformers Up
Diffusion Transformers (DiT) have become a leading architecture in image generation. However, the quadratic complexity of attention mechanisms, which are responsible for modeling token-wise relationsh...
arxiv.org
December 23, 2024 at 9:44 PM
Cool paper where they use a conv/natten-like scanning method of attention to dramatically speed up FLUX, especially with large sizes.
arxiv.org/abs/2412.16112
arxiv.org/abs/2412.16112
*slaps 8b llama3 llm*
this baby can fit loads of generative neural images for diffusion model conditioning in just a little finetuning
arxiv.org/abs/2412.14164
this baby can fit loads of generative neural images for diffusion model conditioning in just a little finetuning
arxiv.org/abs/2412.14164
MetaMorph: Multimodal Understanding and Generation via Instruction Tuning
In this work, we propose Visual-Predictive Instruction Tuning (VPiT) - a simple and effective extension to visual instruction tuning that enables a pretrained LLM to quickly morph into an unified auto...
arxiv.org
December 20, 2024 at 4:47 AM
*slaps 8b llama3 llm*
this baby can fit loads of generative neural images for diffusion model conditioning in just a little finetuning
arxiv.org/abs/2412.14164
this baby can fit loads of generative neural images for diffusion model conditioning in just a little finetuning
arxiv.org/abs/2412.14164
Someone finally used the Chameleon weights for something arxiv.org/abs/2412.01824
X-Prompt: Towards Universal In-Context Image Generation in Auto-Regressive Vision Language Foundation Models
In-context generation is a key component of large language models' (LLMs) open-task generalization capability. By leveraging a few examples as context, LLMs can perform both in-domain and out-of-domai...
arxiv.org
December 3, 2024 at 2:29 PM
Someone finally used the Chameleon weights for something arxiv.org/abs/2412.01824