Jaward Sesay

Jaward

AI & ML interests

I like to train large deep neural nets too 🧠🤖💥 | First Paper (AutoAgents: A Framework for Automatic Agent Generation) Accepted @ IJCAI 2024 | Role Model Karpathy

Recent Activity

replied to their post 4 days ago

Amazing work👏 Introduces Dream 7B - a discrete diffusion reasoning model, fully opensourced with weights on 🤗 - it outperforms existing non-autoregressive models and matches or beats frontier autoregressive of similar size on reasoning tasks. Models: - base: https://huggingface.co/Dream-org/Dream-v0-Base-7B - SFT: https://huggingface.co/Dream-org/Dream-v0-Instruct-7B Code: https://github.com/HKUNLP/Dream Project: https://hkunlp.github.io/blog/2025/dream/

posted an update 4 days ago

posted an update 6 days ago

Implements from first-principle recently proposed dynamic tanh as alternative to layernorm. Specifically, we trained a nanoGPT (0.8 M params) on tiny shakespeare with conventional layernorm, RMSNorm and dynamic tanh, then compared performances. Observed performance seems to match or is stable for α = 0.5~ 1.5, might outperform if trained longer. Code: https://github.com/Jaykef/ai-algorithms/blob/main/Dynamic_Tanh.ipynb Background music by 周子珺

View all activity

Organizations

Jaward's activity

upvoted 2 papers 21 days ago

Being-0: A Humanoid Robotic Agent with Vision-Language Models and Modular Skills

Paper • 2503.12533 • Published 22 days ago • 63

ReCamMaster: Camera-Controlled Generative Rendering from A Single Video

Paper • 2503.11647 • Published 24 days ago • 131

upvoted a paper 23 days ago

Transformers without Normalization

Paper • 2503.10622 • Published 25 days ago • 153

upvoted a paper 30 days ago

HybridNorm: Towards Stable and Efficient Transformer Training via Hybrid Normalization

Paper • 2503.04598 • Published Mar 6 • 18

upvoted a paper about 1 month ago

Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs

Paper • 2503.01743 • Published Mar 3 • 83

upvoted 2 papers 2 months ago

Process Reinforcement through Implicit Rewards

Paper • 2502.01456 • Published Feb 3 • 60

OmniHuman-1: Rethinking the Scaling-Up of One-Stage Conditioned Human Animation Models

Paper • 2502.01061 • Published Feb 3 • 208

upvoted an article 2 months ago

Article

Open-R1: a fully open reproduction of DeepSeek-R1

Jan 28

• 835

upvoted a paper 3 months ago

Evolving Deeper LLM Thinking

Paper • 2501.09891 • Published Jan 17 • 114

upvoted a collection 3 months ago

Cosmos

Collection

The collection of Cosmos models • 31 items • Updated 4 days ago • 279

upvoted 4 papers 5 months ago

Multimodal Autoregressive Pre-training of Large Vision Encoders

Paper • 2411.14402 • Published Nov 21, 2024 • 46

LLaVA-o1: Let Vision Language Models Reason Step-by-Step

Paper • 2411.10440 • Published Nov 15, 2024 • 123

AgentStore: Scalable Integration of Heterogeneous Agents As Specialized Generalist Computer Assistant

Paper • 2410.18603 • Published Oct 24, 2024 • 32

GPT-4o System Card

Paper • 2410.21276 • Published Oct 25, 2024 • 85

upvoted 4 papers 6 months ago

upvoted a collection 6 months ago

Emu3

Collection

Emu3: Next-Token Prediction is All You Need • 7 items • Updated Feb 13 • 70

upvoted a paper 6 months ago

LLaVA-3D: A Simple yet Effective Pathway to Empowering LMMs with 3D-awareness

Paper • 2409.18125 • Published Sep 26, 2024 • 34