Archive

L1

Controlling How Long A Reasoning Model Thinks With Reinforcement Learning

Matryoshka Quantization

Janus-Pro

Unified Multimodal Understanding and Generation with Data and Model Scaling

DeepSeek-R1

Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps

DeepSeek-VL2

Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding

Gaze-LLE

Gaze Target Estimation via Large-Scale Learned Encoders

NVILA

Efficient Frontier Visual Language Models

Rotary Position Encoding

A figure among cyphers: Part-1

PaliGemma 2

A Family of Versatile VLMs for Transfer

Star Attention

Efficient LLM Inference over Long Sequences

AIMv2

Multimodal Autoregressive Pre-training of Large Vision Encoders

JanusFlow

Harmonizing Autoregression and Rectified Flow for Unified Multimodal Understanding and Generation

Cut Your Losses in Large-Vocabulary Language Models

The Super Weight in Large Language Models

Depth Pro

A Hitchhiker’s Guide to Scaling Law Estimation

OmniParser for Pure Vision Based GUI Agent

Normalized Transformer

What Matters for Model Merging at Scale?

Agent WorkFlow Memory