Sparser, Faster, Lighter Transformer Language Models (Paper Walkthrough)

Generative Recommendation for Large-Scale Advertising (Paper Walkthrough)

LeWorldModel: Stable End-to-End JEPA from Pixels (Paper Walkthrough)

Doc-to-LoRA: Learning to Instantly Internalize Contexts (2min TLDR)

SAM 3D: 3Dfy Anything in Images (Paper Walkthrough)

ARC Is a Vision Problem! (Paper Walkthrough)

Diffusion Transformers with Representation Autoencoders (Paper Walkthrough)

Less is More: Recursive Reasoning with Tiny Networks (Paper Walkthrough)

Why Language Models Hallucinate (Paper Walkthrough)

Locality in Image Diffusion Models Emerges from Data Statistics (Paper Walkthrough)

Can LLMs Lie? Investigation beyond Hallucination (Paper Walkthrough) #hallucination

On the Theoretical Limitations of Embedding-Based Retrieval (Paper Walkthrough) #aipaper #arxiv

Make Your Own Daily Podcast of Research Paper Updates #aipaper #arxiv #podcast #techtips #vibevoice

How do we decide which AI paper to deep dive? #aipaper #arxiv

On the Edge of Memorization in Diffusion Models (Paper Walkthrough)

Exploiting Policy Idling for Dexterous Manipulation (Paper Walkthrough)

DINOv3 (Paper Walkthrough)

Genie: Generative Interactive Environments (Paper Walkthrough) #genie3 #deepmind

Seed-Prover: Deep and Broad Reasoning for Automated Theorem Proving (Paper Walkthrough)

Group Sequence Policy Optimization (Paper Walkthrough)

PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers

GUI-G^2: Gaussian Reward Modeling for GUI Grounding (Paper Walkthrough)

GraspGen: A Diffusion-based Framework for 6-DOF Grasping with On-Generator Training (Paper Walkthru)

For Perception Tasks: The Cost of LLM Pretraining by Next-Token Prediction Outweigh its Benefits

Dynamic Chunking for End-to-End Hierarchical Sequence Modeling (Paper Walkthrough)

Questioning Representational Optimism in Deep Learning Fractured Entangled Representation Hypothesis

Open-World Object Counting in Videos (Paper Walkthrough)

Temporal Chain of Thought: Long-Video Understanding by Thinking in Frames (Paper Walkthrough)

Chain-of-Thought Is Not Explainability (Paper Walkthrough)

Cache Me If You Can: How Many KVs Do You Need for Effective Long-Context LMs? (Paper Walkthrough)

Diffuse and Disperse: Image Generation with Representation Regularization (Paper Walkthrough)

FastVLM: Efficient Vision Encoding for Vision Language Models (Paper Walkthrough)

Mean Flows for One-step Generative Modeling (Paper Walkthrough)

Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures

AlphaEdit: Null-Space Constrained Knowledge Editing for Language Models (ICLR 2025 OutstandingPaper)

Learning Dynamics of LLM Finetuning (ICLR 2025 Outstanding Papers)

Welcome to the Era of Experience (David Silver, Richard S. Sutton)

Reasoning Models Can Be Effective Without Thinking (Paper Walkthrough)

A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to Reproducibility

One-Minute Video Generation with Test-Time Training (CVPR 2025)

Inference-Time Scaling for Generalist Reward Modeling (Paper Walkthrough)

Scaling Language-Free Visual Representation Learning (Paper Walkthrough)

ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learning (Paper Walkthrough)

Visual Jenga: Discovering Object Dependencies via Counterfactual Inpainting

Qwen2.5-Omni Technical Report (Paper Walkthrough)

Denoising Hamiltonian Network for Physical Reasoning (Paper Walkthrough)

What Makes a Reward Model a Good Teacher? An Optimization Perspective (Paper Walkthrough)

AudioX: Diffusion Transformer for Anything-to-Audio Generation (Paper Walkthrough)

DAPO: An Open-Source LLM Reinforcement Learning System at Scale (Paper Walkthrough)

Large-scale Pre-training for Grounded Video Caption Generation (Paper Walkthrough)

Transformers without Normalization (Paper Walkthrough)

Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models (ICLR 2025)

Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Models (Paper Walkthru)

L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning (Paper Walkthrough)

Transformers need glasses! Information over-squashing in language tasks (MLST latest, NeurIPS 2024)

DiffSensei: Bridging MultiModal LLMs & Diffusion Models for Customized Manga Generation (CVPR 2025)

GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control (Paper Walkthrough)

Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs

Visual-RFT: Visual Reinforcement Fine-Tuning—R1 Technique Expands to Multimodal, Fully Open Source!

Fractal Generative Models—Growing Images Like Nature Grows Trees (Paper Walkthrough)