Today Twitter feels like the Cambrian Explosion in AI
MACHINE LEARNING
-
Generative AI for Self-Adaptive Systems: Research and Implementation
By
–
Generative AI for Self-Adaptive Systems: State of the Art and Research Roadmap https://
dl.acm.org/doi/10.1145/36
86803
… https://
github.com/545659928/GenA
I4SAS
…
provide researchers and practitioners a comprehensive snapshot that outlines the potential benefits and challenges of employing GenAI’s within SAS. -
Hugging Face API for Daily Papers Documentation
By
–
There is also an API: https://
huggingface.co/api/daily_pape
rs
… Docs: https://
huggingface.co/docs/hub/en/api -
Molmo and PixMo: Open-Weight Multimodal Models with Open Data
By
–
Molmo and PixMo Open Weights and Open Data for State-of-the-Art Multimodal Models discuss: https://
huggingface.co/papers/2409.17
146
… Today's most advanced multimodal models remain proprietary. The strongest open-weight models rely heavily on synthetic data from proprietary VLMs to achieve good -
DreamWaltz-G: Expressive 3D Gaussian Avatars from 2D Diffusion
By
–
DreamWaltz-G
— AK (@_akhaliq) 26 septembre 2024
Expressive 3D Gaussian Avatars from Skeleton-Guided 2D Diffusion
discuss: https://t.co/szvheoe5MM
Leveraging pretrained 2D diffusion models and score distillation sampling (SDS), recent methods have shown promising results for text-to-3D avatar generation.… pic.twitter.com/BP9ckpAZFPDreamWaltz-G Expressive 3D Gaussian Avatars from Skeleton-Guided 2D Diffusion discuss: https://
huggingface.co/papers/2409.17
145
… Leveraging pretrained 2D diffusion models and score distillation sampling (SDS), recent methods have shown promising results for text-to-3D avatar generation. -
TalkinNeRF: Animatable Neural Fields for Full-Body Talking Humans
By
–
TalkinNeRF
— AK (@_akhaliq) 26 septembre 2024
Animatable Neural Fields for Full-Body Talking Humans
discuss: https://t.co/ous6cqa59g
We introduce a novel framework that learns a dynamic neural radiance field (NeRF) for full-body talking humans from monocular videos. Prior work represents only the body pose or… pic.twitter.com/vnFyv4F3eJTalkinNeRF Animatable Neural Fields for Full-Body Talking Humans discuss: https://
huggingface.co/papers/2409.16
666
… We introduce a novel framework that learns a dynamic neural radiance field (NeRF) for full-body talking humans from monocular videos. Prior work represents only the body pose or -
Synchronizing Dual Hands for Physics-Based Dexterous Guitar Playing
By
–
Synchronize Dual Hands for Physics-Based Dexterous Guitar Playing
— AK (@_akhaliq) 26 septembre 2024
discuss: https://t.co/ErvqGKKmH3
We present a novel approach to synthesize dexterous motions for physically simulated hands in tasks that require coordination between the control of two hands with high temporal… pic.twitter.com/weIaYp3XA5Synchronize Dual Hands for Physics-Based Dexterous Guitar Playing discuss: https://
huggingface.co/papers/2409.16
629
… We present a novel approach to synthesize dexterous motions for physically simulated hands in tasks that require coordination between the control of two hands with high temporal -
Language Model Optimization Report from Google Internship 18 Years Ago
By
–
Language model optimization report I did 18 years ago at Google to fulfill Waterloo coop requirements. Dusted it off after all these years. Written as a 19 year old intern with the incredible SMT team. Have been working on Machine Learning since then.
-
Small Models, Big Moment in AI Development
By
–
Some of these models may be small but this moment is big.
-
Llama 3.2 Vision Models Compete with Leading Closed Models
By
–
Evaluating performance on extensive human evaluations and benchmarks, results suggest that Llama 3.2 vision models are competitive with leading closed models on image recognition + a range of visual understanding tasks.