Majid Fekri PRO

mjfk

·

AI & ML interests

None yet

Recent Activity

published an article 24 days ago

What Do Memory Benchmarks Actually Measure? (Hint: Not Storage)

upvoted an article about 1 month ago

Evaluating Agent Memory Honestly: Benchmarks, Failure Modes, and an Open Interchange Format

published an article about 1 month ago

Evaluating Agent Memory Honestly: Benchmarks, Failure Modes, and an Open Interchange Format

View all activity

Organizations

upvoted an article about 1 month ago

Article

Evaluating Agent Memory Honestly: Benchmarks, Failure Modes, and an Open Interchange Format

mjfk

•

Jul 20

• 1

upvoted a paper 4 months ago

Memanto: Typed Semantic Memory with Information-Theoretic Retrieval for Long-Horizon Agents

Paper • 2604.22085 • Published Apr 23 • 16

upvoted a paper 7 months ago

From HNSW to Information-Theoretic Binarization: Rethinking the Architecture of Scalable Vector Search

Paper • 2601.11557 • Published Dec 16, 2025 • 4