Automated report

current/2025-01-02 HUNYUANPROVER: A Scalable Data Synthesis Framework and Guided Tree Search for Automated Theorem Proving.yaml

@@ -0,0 +1,9 @@
+date: "2025-01-02"
+author: Yang Li
+title: 'HUNYUANPROVER: A Scalable Data Synthesis Framework and Guided Tree Search for Automated Theorem Proving'
+thumbnail: ""
+link: https://huggingface.co/papers/2412.20735
+summary: HunyuanProver is a language model that helps computers solve math problems by generating more data and using smart search algorithms. It's really good at this and can solve problems that other computers can't. The creators will share their data with others to help more people learn and solve math problems....
+opinion: placeholder
+tags:
+    - ML

current/2025-01-02 OS-Genesis: Automating GUI Agent Trajectory Construction via Reverse Task Synthesis.yaml

@@ -0,0 +1,9 @@
+date: "2025-01-02"
+author: Qiushi Sun
+title: 'OS-Genesis: Automating GUI Agent Trajectory Construction via Reverse Task Synthesis'
+thumbnail: ""
+link: https://huggingface.co/papers/2412.19723
+summary: OS-Genesis is a new way to make data for training GUI agents. It lets the agents explore and interact with the environment first, then figures out what tasks they were doing. This helps make better and more diverse data, which makes the agents perform better on challenging tasks....
+opinion: placeholder
+tags:
+    - ML

current/2025-01-02 Xmodel-2 Technical Report.yaml

@@ -0,0 +1,9 @@
+date: "2025-01-02"
+author: Wang Qun
+title: Xmodel-2 Technical Report
+thumbnail: ""
+link: https://huggingface.co/papers/2412.19638
+summary: Xmodel-2 is a large language model designed for reasoning tasks, with a unified set of hyperparameters for different scales and WSD learning rate scheduler for efficient training. It achieves state-of-the-art performance on complex tasks and is available on GitHub....
+opinion: placeholder
+tags:
+    - ML