Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding

An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.

The first high school physics Olympiad benchmark for evaluating (M)LLMs with step-level grading and human-level comparison.

Random network distillation on Montezuma's Revenge and Super Mario Bros.

✨✨ [ICLR 2026] R1-Reward: Training Multimodal Reward Model Through Stable Reinforcement Learning

[NeurIPS 2025] T2I-R1: Reinforcing Image Generation with Collaborative Semantic-level and Token-level CoT

[ICLR 2025] MLLM can see? Dynamic Correction Decoding for Hallucination Mitigation

[CVPR 2024 Highlight] Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding

✨✨[AAAI 2026] This is the official implementation of our paper "QuoTA: Query-oriented Token Assignment via CoT Query Decouple for Long Video Comprehension"

✨✨[NeurIPS 2025] This is the official implementation of our paper "Video-RAG: Visually-aligned Retrieval-Augmented Long Video Comprehension"

Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

The Next Step Forward in Multimodal LLM Alignment

MME-CoT: Benchmarking Chain-of-Thought in LMMs for Reasoning Quality, Robustness, and Efficiency

✨✨Long-VITA: Scaling Large Multi-modal Models to 1 Million Tokens with Leading Short-Context Accuracy

✨✨Woodpecker: Hallucination Correction for Multimodal Large Language Models

✨✨Latest Advances on Multimodal Large Language Models

A Robust and Versatile Monocular Visual-Inertial State Estimator

Game Boy emulator written in Python

Ready-to-use code and tutorial notebooks to boost your way into few-shot learning for image classification.

[TPAMI 2023] LibFewShot: A Comprehensive Library for Few-shot Learning.