PaperLens

@UC9mfC57IPmYp_ji-4uVU2vg - 399 subscribers

Welcome to PaperLens! This channel uses Google’s NotebookLM to turn complex research papers into clear, student‑friendly video overviews. Whether it’s AI, machine learning, or topics from my academic courses, you’ll find simplified explanations that save you time and help you actually understand the big ideas. Perfect for students, learners, and anyone interested in the latest developments in technology and research.

Home Videos Live Playlists

Optimizing LLM Agent Workflows: Static to Dynamic Graphs

Optimizing LLM Agent Workflows: Static to Dynamic Graphs PaperLens

65 views - 3 months ago

BIGMAS: Boosting LLM Reasoning with Brain-Inspired Agents

BIGMAS: Boosting LLM Reasoning with Brain-Inspired Agents PaperLens

149 views - 3 months ago

Building Terminal-Native AI Coding Agents

Building Terminal-Native AI Coding Agents PaperLens

81 views - 3 months ago

AutoHarness by Google DeepMind: Code for Better LLM Agents

AutoHarness by Google DeepMind: Code for Better LLM Agents PaperLens

157 views - 3 months ago

SkillNet: Create, Evaluate, and Connect AI Skills

SkillNet: Create, Evaluate, and Connect AI Skills PaperLens

21 views - 3 months ago

The Spike, the Sparse and the Sink in LLMs

The Spike, the Sparse and the Sink in LLMs PaperLens

139 views - 3 months ago

KARL: Knowledge Agents via Reinforcement Learning by Databricks

KARL: Knowledge Agents via Reinforcement Learning by Databricks PaperLens

47 views - 3 months ago

Memex(RL): Scaling Long-Horizon LLM Agents

Memex(RL): Scaling Long-Horizon LLM Agents PaperLens

28 views - 3 months ago

FlashAttention-4 Explained: Optimizing AI for Blackwell GPUs

FlashAttention-4 Explained: Optimizing AI for Blackwell GPUs PaperLens

45 views - 3 months ago

StructuredAgent: AND/OR Trees for Long-Horizon Web Tasks

StructuredAgent: AND/OR Trees for Long-Horizon Web Tasks PaperLens

15 views - 3 months ago

AgentIR: Reasoning-Aware Retrieval for Deep Research

AgentIR: Reasoning-Aware Retrieval for Deep Research PaperLens

27 views - 3 months ago

Think Harder or Know More? Adaptive Loops & Memory

Think Harder or Know More? Adaptive Loops & Memory PaperLens

7 views - 3 months ago

ALMA: Self-Improving Memory Designs for AI Agents

ALMA: Self-Improving Memory Designs for AI Agents PaperLens

68 views - 4 months ago

LLaDA2.1: Speeding Up Diffusion LLMs via Token Editing

LLaDA2.1: Speeding Up Diffusion LLMs via Token Editing PaperLens

20 views - 4 months ago

SKILLRL: Evolving LLM Agents via Recursive Skill-Augmented RL

SKILLRL: Evolving LLM Agents via Recursive Skill-Augmented RL PaperLens

132 views - 4 months ago

InftyThink+: Infinite-Horizon Reasoning via RL

InftyThink+: Infinite-Horizon Reasoning via RL PaperLens

28 views - 4 months ago

Agyn: A Multi-Agent System for Team-Based Autonomous Coding

Agyn: A Multi-Agent System for Team-Based Autonomous Coding PaperLens

61 views - 4 months ago

EchoJEPA: A Latent Predictive Foundation Model for Echocardiography

EchoJEPA: A Latent Predictive Foundation Model for Echocardiography PaperLens

74 views - 4 months ago

AdaptEvolve: Making Evolutionary AI Efficient

AdaptEvolve: Making Evolutionary AI Efficient PaperLens

8 views - 4 months ago

Gaia2: Benchmarking AI Agents in Dynamic Worlds

Gaia2: Benchmarking AI Agents in Dynamic Worlds PaperLens

56 views - 4 months ago

AgentArk: Distilling Multi-Agent Intelligence into One LLM

AgentArk: Distilling Multi-Agent Intelligence into One LLM PaperLens

73 views - 4 months ago

AgentSkiller: Scaling Generalist AI with Synthetic Data

AgentSkiller: Scaling Generalist AI with Synthetic Data PaperLens

26 views - 4 months ago

Solving Erdős Problems with Google DeepMind's Gemini

Solving Erdős Problems with Google DeepMind's Gemini PaperLens

235 views - 4 months ago

TinyLoRA: Learning to Reason in Just 13 Parameters

TinyLoRA: Learning to Reason in Just 13 Parameters PaperLens

119 views - 4 months ago

xMemory: Beyond RAG for Efficient AI Agent Memory

xMemory: Beyond RAG for Efficient AI Agent Memory PaperLens

103 views - 4 months ago

Scaling Small AI Agents with Strategy Auctions (SALE) | Meta Research

Scaling Small AI Agents with Strategy Auctions (SALE) | Meta Research PaperLens

12 views - 4 months ago

InfMem: System-2 Memory Control for Long-Context AI

InfMem: System-2 Memory Control for Long-Context AI PaperLens

29 views - 4 months ago

A-RAG: Scaling Agentic RAG via Hierarchical Interfaces

A-RAG: Scaling Agentic RAG via Hierarchical Interfaces PaperLens

100 views - 4 months ago

Agent Primitives: Reusable Latent Building Blocks for MAS

Agent Primitives: Reusable Latent Building Blocks for MAS PaperLens

51 views - 4 months ago

Accelerating Scientific Research with Google's Gemini

Accelerating Scientific Research with Google's Gemini PaperLens

99 views - 4 months ago

Heterogeneous Computing: The Future of AI Agent Inference

Heterogeneous Computing: The Future of AI Agent Inference PaperLens

60 views - 4 months ago

OpenScholar: Outperforming GPT-4o in Scientific Research

OpenScholar: Outperforming GPT-4o in Scientific Research PaperLens

207 views - 4 months ago

Kimi K2.5 Explained: Visual Agentic Intelligence & Agent Swarms

Kimi K2.5 Explained: Visual Agentic Intelligence & Agent Swarms PaperLens

90 views - 5 months ago

Token-Level Data Filtering: Shaping LLM Capabilities at Scale

Token-Level Data Filtering: Shaping LLM Capabilities at Scale PaperLens

38 views - 5 months ago

Anthropic Research: Does AI Assistance Hurt Skill Formation?

Anthropic Research: Does AI Assistance Hurt Skill Formation? PaperLens

183 views - 5 months ago

VibeTensor: NVIDIA’s Fully AI-Generated Deep Learning System

VibeTensor: NVIDIA’s Fully AI-Generated Deep Learning System PaperLens

191 views - 5 months ago

SDPO: Reinforcement Learning via Self-Distillation (Hübotter et al.)

SDPO: Reinforcement Learning via Self-Distillation (Hübotter et al.) PaperLens

133 views - 5 months ago

Meta’s Self-Improving Pretraining: Beyond Next-Token Prediction

Meta’s Self-Improving Pretraining: Beyond Next-Token Prediction PaperLens

40 views - 5 months ago

LingBot-World: Real-Time Open-Source World Simulator by Robbyant

LingBot-World: Real-Time Open-Source World Simulator by Robbyant PaperLens

464 views - 5 months ago

Insight Agents: Amazon’s LLM Multi-Agent System (SIGIR ’25)

Insight Agents: Amazon’s LLM Multi-Agent System (SIGIR ’25) PaperLens

64 views - 5 months ago

Communication Methods in Multi-Agent Reinforcement Learning (MARL)

Communication Methods in Multi-Agent Reinforcement Learning (MARL) PaperLens

71 views - 5 months ago

Team of Rivals: Isotopes AI’s Multi-Agent Architecture for Reliability

Team of Rivals: Isotopes AI’s Multi-Agent Architecture for Reliability PaperLens

50 views - 5 months ago

TTT-Discover: AI Sets New SOTA in Science via Test-Time Training

TTT-Discover: AI Sets New SOTA in Science via Test-Time Training PaperLens

98 views - 5 months ago

AI Reasoning: How Models Create a "Society of Thought"

AI Reasoning: How Models Create a "Society of Thought" PaperLens

20 views - 5 months ago

Agent Cognitive Compressor: Solving AI Memory Drift & Hallucinations

Agent Cognitive Compressor: Solving AI Memory Drift & Hallucinations PaperLens

115 views - 5 months ago

Terminal-Bench 2.0: Benchmarking AI Agents on Hard, Realistic CLI Tasks

Terminal-Bench 2.0: Benchmarking AI Agents on Hard, Realistic CLI Tasks PaperLens

590 views - 5 months ago

OneFlow: Single Agents Match Multi-Agent Performance (Amazon & UT Austin)

OneFlow: Single Agents Match Multi-Agent Performance (Amazon & UT Austin) PaperLens

19 views - 5 months ago

MCP-SIM: Self-Correcting Multi-Agent AI for Physics Simulation

MCP-SIM: Self-Correcting Multi-Agent AI for Physics Simulation PaperLens

231 views - 5 months ago

AI Agents vs. IDEs: Speed Gains Come with Tech Debt

AI Agents vs. IDEs: Speed Gains Come with Tech Debt PaperLens

112 views - 5 months ago

Efficient AI Agents: Optimizing Memory, Tools & Planning (2026 Survey)

Efficient AI Agents: Optimizing Memory, Tools & Planning (2026 Survey) PaperLens

30 views - 5 months ago

Beyond Entangled Planning: Task-Decoupled Planning (TDP) Overview

Beyond Entangled Planning: Task-Decoupled Planning (TDP) Overview PaperLens

24 views - 5 months ago

The State of Multi-Agent AI: Analyzing 42K Commits & Issues

The State of Multi-Agent AI: Analyzing 42K Commits & Issues PaperLens

72 views - 5 months ago

Learning Latent Action World Models from In-the-Wild Videos

Learning Latent Action World Models from In-the-Wild Videos PaperLens

82 views - 5 months ago

DroPE: Zero-Shot Long Context by Dropping Positional Embeddings

DroPE: Zero-Shot Long Context by Dropping Positional Embeddings PaperLens

36 views - 5 months ago

Dr. Zero: Meta’s New AI That Learns Without Human Data

Dr. Zero: Meta’s New AI That Learns Without Human Data PaperLens

168 views - 5 months ago

AgeMem: Unified Long & Short-Term Memory for LLM Agents

AgeMem: Unified Long & Short-Term Memory for LLM Agents PaperLens

59 views - 5 months ago

Focus Agent: Slime Mold-Inspired Context Compression for LLMs

Focus Agent: Slime Mold-Inspired Context Compression for LLMs PaperLens

34 views - 5 months ago

Agent-as-a-Judge: The Future of AI Evaluation (Paper Overview)

Agent-as-a-Judge: The Future of AI Evaluation (Paper Overview) PaperLens

95 views - 5 months ago

SimpleMem: Efficient Lifelong Memory for LLM Agents (30x Lower Cost)

SimpleMem: Efficient Lifelong Memory for LLM Agents (30x Lower Cost) PaperLens

197 views - 5 months ago

Ministral 3: Inside Mistral AI’s New Efficient Model Family

Ministral 3: Inside Mistral AI’s New Efficient Model Family PaperLens

320 views - 5 months ago