🏗️ Building on HF

Zixi "Oz" Li

OzTianlu

9 32 46

https://github.com/lizixi-0x2F

lizixi-0x2F

AI & ML interests

My research focuses on deep reasoning with small language models, Transformer architecture innovation, and knowledge distillation for efficient alignment and transfer.

Recent Activity

liked a dataset about 11 hours ago

nvidia/LIBERO_LeRobot_v3

liked a model about 16 hours ago

NoesisLab/Particles-10M

updated a model about 17 hours ago

NoesisLab/Particles-10M

View all activity

Organizations

upvoted an article about 2 months ago

Article

Why ResNet is Explicit Euler, and What That Tells Us About Deep Learning

NoesisLab

•

Jun 13

• 1

upvoted 2 papers about 2 months ago

On the Geometry of On-Policy Distillation

Paper • 2606.07082 • Published Jun 5 • 75

Entropy as a Structural Prior: How a Log-Barrier on DiT Belief Space Drives Musical Diversity and Development

Paper • 2606.07207 • Published Jun 5 • 4

upvoted a paper 2 months ago

A Closed-Form Upper Bound for Admissible Learning-Rate Steps in Belief-Space Dynamics

Paper • 2605.06741 • Published May 7 • 1

upvoted a collection 3 months ago

DeepSeek-V4

Collection

7 items • Updated 6 days ago • 790

upvoted 2 articles 5 months ago

Article

Arcade-3B: SLM Optimization via Orthogonal Decoupling of Latent State Spaces

NoesisLab

•

Mar 15

• 1

Article

Arcade-3B: 基于隐藏层状态空间正交解耦的 SLM 优化

NoesisLab

•

Mar 15

• 1

upvoted 3 papers 5 months ago

CodeScaler: Scaling Code LLM Training and Test-Time Inference via Execution-Free Reward Models

Paper • 2602.17684 • Published Feb 4 • 22

Efficient RLVR Training via Weighted Mutual Information Data Selection

Paper • 2603.01907 • Published Mar 2 • 14

Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learning

Paper • 2504.13914 • Published Apr 10, 2025 • 6

upvoted a collection 5 months ago

Seed Flagship Model Released

Collection

contributed • 8 items • Updated Apr 13 • 3

upvoted a paper 5 months ago

DualPath: Breaking the Storage Bandwidth Bottleneck in Agentic LLM Inference

Paper • 2602.21548 • Published Feb 25 • 56

upvoted 2 articles 5 months ago

Article

Exploring New Frontiers of LLMs: Adaptive Dual-Search Distillation (ADS) and the 30B Model Open Beta

NoesisLab

•

Mar 1

• 2

Article

GGML and llama.cpp join HF to ensure the long-term progress of Local AI

ggerganov, ngxson, allozaur, lysandre, victor, julien-c

•

Feb 20

• 509

upvoted a collection 5 months ago

Kai Models Series

Collection

Kai Models Distilled via Adaptive Dual Search Distillation • 3 items • Updated Mar 2 • 2

upvoted a paper 5 months ago

Nacrith: Neural Lossless Compression via Ensemble Context Modeling and High-Precision CDF Coding

Paper • 2602.19626 • Published Feb 23 • 3

upvoted an article 5 months ago

Article

Shattering the Memory Wall: O(1) Inference and Causal Monoid State Compression in Spartacus-1B

OzTianlu

•

Feb 25

• 2

upvoted a collection 6 months ago

Spartacus Monoid Reasoning Models

Collection

O(1) Reasoning Models • 1 item • Updated Feb 25 • 2

upvoted an article 6 months ago

Article

The Optimal Architecture for Small Language Models

codelion

•

Dec 26, 2025

• 121

upvoted a collection 6 months ago

Geilim Smol Language Models

Collection

Geilim Smol Language Models • 2 items • Updated Mar 3 • 1

Zixi "Oz" Li

AI & ML interests

Recent Activity

Organizations

OzTianlu's activity

Why ResNet is Explicit Euler, and What That Tells Us About Deep Learning

Arcade-3B: SLM Optimization via Orthogonal Decoupling of Latent State Spaces

Arcade-3B: 基于隐藏层状态空间正交解耦的 SLM 优化

Exploring New Frontiers of LLMs: Adaptive Dual-Search Distillation (ADS) and the 30B Model Open Beta

GGML and llama.cpp join HF to ensure the long-term progress of Local AI

Shattering the Memory Wall: O(1) Inference and Causal Monoid State Compression in Spartacus-1B

The Optimal Architecture for Small Language Models