open-source models
Open-source models are machine learning models made available to the public with accessible code and resources. These facilitate collaboration, experimentation, and rapid advancement in AI research by breaking down entry barriers.
- BMMR: A Large-Scale Bilingual Multimodal Multi-Discipline Reasoning Dataset
- CHOICE: Benchmarking the Remote Sensing Capabilities of Large Vision-Language Models
- EOC-Bench: Can MLLMs Identify, Recall, and Forecast Objects in an Egocentric World?
- Holistic Order Prediction in Natural Scenes
- Modeling the Economic Impacts of AI Openness Regulation
- RBench-V: A Primary Assessment for Visual Reasoning Models with Multimodal Outputs
- RTV-Bench: Benchmarking MLLM Continuous Perception, Understanding and Reasoning through Real-Time Video
- ResearchCodeBench: Benchmarking LLMs on Implementing Novel Machine Learning Research Code
- SWE-SQL: Illuminating LLM Pathways to Solve User SQL Issues in Real-World Applications
- The Rise of Parameter Specialization for Knowledge Storage in Large Language Models
- Unified Reinforcement and Imitation Learning for Vision-Language Models
- VMDT: Decoding the Trustworthiness of Video Foundation Models
- VideoLucy: Deep Memory Backtracking for Long Video Understanding
- macOSWorld: A Multilingual Interactive Benchmark for GUI Agents