Projects

Articles

Readings

About

Projects

Articles

Readings

About

Readings 📖

Update

For ML repositories/papers I find interesting visit:

My Github Repo

🔍 General ML Papers

TitleAuthorLinkNotes
Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth ApproachGeiping et al.ReadInteresting Architecture for scalable model building.
Imagine while Reasoning in Space: Multimodal Visualization-of-ThoughtLi et al.ReadHas the potential to thoroughly strengthen LLMs for robot control. Very interesting paper and results.
Autonomous Deep AgentYu et al.ReadThe deeper planning aspect is interesting. I'm not certain about method applicability.
Demystifying Long Chain-of-Thought Reasoning in LLMsYeo et al.ReadProvides strong intuition for CoT.
Contextual Document EmbeddingsMorris and RushReadStrong foundation for dataset curation for topic specific models.
MultiChallenge: A Realistic Multi-Turn Conversation Evaluation Benchmark Challenging to Frontier LLMsSirdeshmukh et al.ReadInteresting benchmark focusing on LLM continuity. Important for plan generation.

🎯 Reinforcement Learning Papers

TitleAuthorLinkNotes
Training Software Engineering Agents and Verifiers with SWE-GymPan et al.ReadServes as good background for how to build RL gym environment
ReKep: Spatio-Temporal Reasoning of Relational Keypoint Constraints for Robotic ManipulationHuang et al.ReadIncredible framework!

👷 Model Architecture

TitleAuthorLinkNotes
Understanding Multi-Head Latent AttentionEryk BanattReadGreat MLA explanation
Understanding Modern LLMs via DeepSeekEryk BanattReadComprehensive Deepseek guide.

📈 Finance

TitleAuthorLinkNotes
A Constrained Hierarchical Risk Parity Algorithm with Cluster-based Capital AllocationPfitzinger and KatzkeReadComprehensive HRP guide.

© 2025 Ali ElSheikh. All rights reserved.

Login

|

Github

|

Linkedin

|

Get in Contact

© 2025 Ali ElSheikh. All rights reserved.

Login

|

Github

|

Linkedin

|

Get in Contact