Post

This Thread may be Removed Anytime!

Twitter may remove this content at anytime! Save it as PDF for later use!

More from @omarsar0

elvis

@omarsar0

May 3

A Survey of AI Agent Protocols

5 things that stood out to me about this report:

Agent Internet Ecosystem

Here is what the layered architecture of the agent internet ecosystem looks like as it stands. It shows different layers, like the Agent Internet, the Protocol Layer, and the Application Layer.

Timeline

The report provides an overview of LLMs, the agent frameworks, agent protocols, and popular applications from 2019 till now. It's not complete, but it provides a rough overview of the progress. It's still early days for agents, and stronger LLMs and protocols are key.

Read 6 tweets

elvis

@omarsar0

May 1

Small reasoning models are here!

Microsoft just released Phi-4-Mini-Reasoning to explore small reasoning language models for math.

Let's find out how this all works:

Phi-4-Mini-Reasoning

The paper introduces Phi-4-Mini-Reasoning, a 3.8B parameter small language model (SLM) that achieves state-of-the-art mathematical reasoning performance, rivaling or outperforming models nearly TWICE its size.

Unlocking Reasoning

They use a systematic, multi-stage training pipeline to unlock strong reasoning capabilities in compact models, addressing the challenges posed by their limited capacity.

Uses large-scale distillation, preference learning, and RL with verifiable rewards.

Read 8 tweets

elvis

@omarsar0

Apr 30

Universal RAG

RAG is dead, they said.

Then you see papers like this and it gives you a better understanding of the opportunities and challenges ahead.

Lots of great ideas in this paper. I've summarized a few below:

What is it?

UniversalRAG is a framework that overcomes the limitations of existing RAG systems confined to single modalities or corpora. It supports retrieval across modalities (text, image, video) and at multiple granularities (e.g., paragraph vs. document, clip vs. video).

Modality-aware routing

To counter modality bias in unified embedding spaces (where queries often retrieve same-modality results regardless of relevance), UniversalRAG introduces a router that dynamically selects the appropriate modality (e.g., image vs. text) for each query.

Read 7 tweets

elvis

@omarsar0

Apr 29

Building Production-Ready AI Agents with Scalable Long-Term Memory

Memory is one of the most challenging bits of building production-ready agentic systems.

Lots of goodies in this paper.

Here is my breakdown:

What does it solve?

It proposes a memory-centric architecture for LLM agents to maintain coherence across long conversations and sessions, solving the fixed-context window limitation.

The solution:

Introduces two systems: Mem0, a dense, language-based memory system, and Mem0g, an enhanced version with graph-based memory to model complex relationships.

Both aim to extract, consolidate, and retrieve salient facts over time efficiently.

Read 9 tweets

elvis

@omarsar0

Apr 29

A Survey of Efficient LLM Inference Serving

This one provides a comprehensive taxonomy of recent system-level innovations for efficient LLM inference serving.

Great overview for devs working on inference.

Here is what's included:

Instance-Level Methods

Techniques like model parallelism (pipeline, tensor, context, and expert parallelism), offloading (e.g., ZeRO-Offload, FlexGen, TwinPilots), and request scheduling (inter- and intra-request) are reviewed...

Novel schedulers like FastServe, Prophet, and INFERMAX optimize decoding with predicted request lengths. KV cache optimization covers paging, reuse (lossless and semantic-aware), and compression (e.g., 4-bit quantization, compact encodings).

Read 5 tweets

elvis

@omarsar0

Apr 27

265 pages of everything you need to know about building AI agents.

5 things that stood out to me about this report:

1. Human Brain and LLM Agents

Great to better understand what differentiates LLM agents from human/brain cognition, and what inspirations we can get from the way humans learn and operate.

2. Definitions

There is a nice, detailed, and formal definition for what makes up an AI agent. Most of the definitions out there are too abstract.

Read 8 tweets

Share this page!

Enter URL or ID to Unroll

elvis

Try unrolling a thread yourself!

More from @omarsar0

elvis

elvis

elvis

elvis

elvis

elvis

Did Thread Reader help you today?

Don't want to be a Premium member but still want to support us?

Send Email!