Automated report

current/2024-01-24 AutoRT: Embodied Foundation Models for Large Scale Orchestration of Robotic Agents.yaml

@@ -1,12 +1,17 @@
-date: "2024-01-24"
 author: Michael Ahn
-title: 'AutoRT: Embodied Foundation Models for Large Scale Orchestration of Robotic Agents'
-thumbnail: https://cdn-uploads.huggingface.co/production/uploads/60f1abe7544c2adfd699860c/z1u5HcBr9b2neYGO7b7W7.mp4
+date: '2024-01-24'
 link: https://huggingface.co/papers/2401.12963
-summary: This paper presents AutoRT, a system that uses existing foundation models to scale up the deployment of operational robots in unseen scenarios with minimal human supervision. AutoRT utilizes vision-language and large language models to propose instructions to a fleet of robots, effectively collecting diverse data for robot learning....
 opinion: placeholder
+summary: This paper presents AutoRT, a system that uses existing foundation models
+  to scale up the deployment of operational robots in unseen scenarios with minimal
+  human supervision. AutoRT utilizes vision-language and large language models to
+  propose instructions to a fleet of robots, effectively collecting diverse data for
+  robot learning....
 tags:
-    - Robotics and Control
-    - Deep Learning
-    - Natural Language Processing
-    - Computer Vision
+- Robotics and Control
+- Deep Learning
+- Natural Language Processing
+- Computer Vision
+thumbnail: https://github.com/codingpot/hf-daily-paper-newsletter-tester/blob/main/assets/2401.12963.gif?raw=true
+title: 'AutoRT: Embodied Foundation Models for Large Scale Orchestration of Robotic
+  Agents'

current/2024-01-24 GALA: Generating Animatable Layered Assets from a Single Scan.yaml

@@ -1,11 +1,14 @@
-date: "2024-01-24"
 author: Taeksoo Kim
-title: 'GALA: Generating Animatable Layered Assets from a Single Scan'
-thumbnail: https://cdn-uploads.huggingface.co/production/uploads/60f1abe7544c2adfd699860c/2g0hPK7dYkqTD52QfhiNV.qt
+date: '2024-01-24'
 link: https://huggingface.co/papers/2401.12979
-summary: This paper introduces GALA, a framework that decomposes single-layer clothed 3D human meshes into separate layers using a pre-trained 2D diffusion model as a geometry and appearance prior. The outputs can be combined with other assets to create novel clothed human avatars with any pose....
 opinion: placeholder
+summary: This paper introduces GALA, a framework that decomposes single-layer clothed
+  3D human meshes into separate layers using a pre-trained 2D diffusion model as a
+  geometry and appearance prior. The outputs can be combined with other assets to
+  create novel clothed human avatars with any pose....
 tags:
-    - Graphics and Vision
-    - Computer Vision
-    - Deep Learning
+- Graphics and Vision
+- Computer Vision
+- Deep Learning
+thumbnail: https://github.com/codingpot/hf-daily-paper-newsletter-tester/blob/main/assets/2401.12979.gif?raw=true
+title: 'GALA: Generating Animatable Layered Assets from a Single Scan'

current/2024-01-24 Large-scale Reinforcement Learning for Diffusion Models.yaml

@@ -1,11 +1,15 @@
-date: "2024-01-24"
 author: Yinan Zhang
-title: Large-scale Reinforcement Learning for Diffusion Models
-thumbnail: https://cdn-uploads.huggingface.co/production/uploads/60f1abe7544c2adfd699860c/gi8z5i6oJ7XQDiUOYF5UP.mp4
+date: '2024-01-24'
 link: https://huggingface.co/papers/2401.12244
-summary: The paper presents a scalable algorithm that uses Reinforcement Learning to improve diffusion models, making them more aligned with human preferences, fairness, and compositionality over millions of images. This approach substantially outperforms existing methods and improves pretrained Stable Diffusion models, generating samples preferred by humans 80.3% of the time....
 opinion: placeholder
+summary: The paper presents a scalable algorithm that uses Reinforcement Learning
+  to improve diffusion models, making them more aligned with human preferences, fairness,
+  and compositionality over millions of images. This approach substantially outperforms
+  existing methods and improves pretrained Stable Diffusion models, generating samples
+  preferred by humans 80.3% of the time....
 tags:
-    - Reinforcement Learning
-    - Deep Learning
-    - Computer Vision
+- Reinforcement Learning
+- Deep Learning
+- Computer Vision
+thumbnail: https://github.com/codingpot/hf-daily-paper-newsletter-tester/blob/main/assets/2401.12244.gif?raw=true
+title: Large-scale Reinforcement Learning for Diffusion Models