Stars
A curated collection of AI agent research papers released in 2026, covering agent engineering, memory, evaluation, workflows, and autonomous systems.
[CVPR 2026 Oral] Official implementation for ChordEdit: One-Step Low-Energy Transport for Image Editing
UniRL is a Framework for Unified Multimodal Model Reinforcement Learning
Implementation of "MusicDET: Zero-Shot AI-Generated Music Detection", ICML 2026
MemCoT is a memory-driven Chain-of-Thought framework that enables scalable long-context reasoning through iterative memory evolution and active retrieval.
[TKDE] This repository is the official implementation of the TKDE 2025 "Fuzzy Granule Density-Based Outlier Detection with Multi-Scale Granular Balls".
[NeurIPS2023] Exploring Diverse In-Context Configurations for Image Captioning
Open source implementation of the paper "MM-Vid: Advancing Video Understanding with GPT-4V(ision)".
[AAAI2025] Video Repurposing from User Generated Content: A Large-scale Dataset and Benchmark
[ICLR 2026] On the Generalization of SFT: A Reinforcement Learning Perspective with Reward Rectification.
[CVPR2025] Number it: Temporal Grounding Videos like Flipping Manga
MMSpec: Benchmarking Speculative Decoding for Vision-Language Models
Auto-Rubric as Reward: From Implicit Preference to Explicit Generative Criteria
MMaDA - Open-Sourced Multimodal Large Diffusion Language Models (dLLMs with block diffusion, mixed-CoT, unified RL)
MUG-V 10B: High-efficiency Training Pipeline for Large Video Generation Models
[CVPR2025] "AniMo: Species-Aware Model for Text-Driven Animal Motion Generation"
[ICLR 2026] This repository is the official implementation of "EasyTune: Efficient Reinforcement Fine-Tuning for Diffusion-Based Motion Generation".
Official implementation of "MoMask: Generative Masked Modeling of 3D Human Motions (CVPR2024)"
Elevate your AI research writing, no more tedious polishing ✨
A PyTorch native library for training speculative decoding models
Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)
PyTorch code and models for VJEPA2 self-supervised learning from video.