5 42 43

Pengxiang Li

pengxiang

pixeli

AI & ML interests

Video generation, Image editing, AD

Recent Activity

upvoted a paper 4 days ago

SPAM: Spike-Aware Adam with Momentum Reset for Stable LLM Training

upvoted a paper 7 days ago

rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking

authored a paper 9 days ago

TrackDiffusion: Tracklet-Conditioned Video Generation via Diffusion Models

View all activity

Organizations

None yet

pengxiang's activity

upvoted a paper 4 days ago

SPAM: Spike-Aware Adam with Momentum Reset for Stable LLM Training

Paper • 2501.06842 • Published 6 days ago • 14

upvoted a paper 7 days ago

rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking

Paper • 2501.04519 • Published 10 days ago • 230

upvoted 2 papers 9 days ago

InfiGUIAgent: A Multimodal Generalist GUI Agent with Native Reasoning and Reflection

Paper • 2501.04575 • Published 10 days ago • 22

Towards System 2 Reasoning in LLMs: Learning How to Think With Meta Chain-of-Though

Paper • 2501.04682 • Published 10 days ago • 83

upvoted a paper 15 days ago

OS-Genesis: Automating GUI Agent Trajectory Construction via Reverse Task Synthesis

Paper • 2412.19723 • Published 22 days ago • 79

upvoted 2 papers about 1 month ago

Mix-LN: Unleashing the Power of Deeper Layers by Combining Pre-LN and Post-LN

Paper • 2412.13795 • Published about 1 month ago • 19

Training Large Language Models to Reason in a Continuous Latent Space

Paper • 2412.06769 • Published Dec 9, 2024 • 74

upvoted 3 papers about 2 months ago

ShowUI: One Vision-Language-Action Model for GUI Visual Agent

Paper • 2411.17465 • Published Nov 26, 2024 • 78

DreamMix: Decoupling Object Attributes for Enhanced Editability in Customized Image Inpainting

Paper • 2411.17223 • Published Nov 26, 2024 • 5

MagicDriveDiT: High-Resolution Long Video Generation for Autonomous Driving with Adaptive Control

Paper • 2411.13807 • Published Nov 21, 2024 • 11

upvoted a paper 2 months ago

EgoVid-5M: A Large-Scale Video-Action Dataset for Egocentric Video Generation

Paper • 2411.08380 • Published Nov 13, 2024 • 25

upvoted a collection 2 months ago

🍃 MINT-1T

Collection

Data for "MINT-1T: Scaling Open-Source Multimodal Data by 10x: A Multimodal Dataset with One Trillion Tokens" • 13 items • Updated Jul 24, 2024 • 58

upvoted 2 papers 4 months ago

Training Language Models to Self-Correct via Reinforcement Learning

Paper • 2409.12917 • Published Sep 19, 2024 • 136

Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researchers

Paper • 2409.04109 • Published Sep 6, 2024 • 44

upvoted 4 papers 5 months ago

upvoted a paper 6 months ago

MoMa: Efficient Early-Fusion Pre-training with Mixture of Modality-Aware Experts

Paper • 2407.21770 • Published Jul 31, 2024 • 22

upvoted an article 6 months ago

Article

Mixture of Depth is Vibe

•

Apr 22, 2024

• 44