Evaluating and Steering Modality Preferences in Multimodal Large Language Model Paper β’ 2505.20977 β’ Published 6 days ago β’ 1 β’ 1
view post Post 255 May highlights from Chinaβs open source ecosystem π₯ zh-ai-community/may-2025-open-works-from-the-chinese-community-681a3494145f2914dc679b7cβ¨ DeepSeek dropped R1 updates- Both R1 & 8B distralled smol model β¨ Bytedance goes big on open source: - BAGEL, Dolphin, Seedcoder, Dream0...β¨ Multimodal is on fire!- HuyuanCustom / HunyuanVideo-Avatar / HunyuanPortrait - MiniMax: SynLogic / Orsta-7B- Xiaomi: MiMo VL - Alibaba Wan: Wan2.1-VACE- OpenGVlab: ZeroGUI - StepFun: ACE-Step-v1/Step1X-3Dβ¨ Specialized models/datasets excels- Alibaba Qwen: World PM 72B - BAAI:RobotBrain (MLLM for robotic)- HiThink Research: BizFinBench (dataset)- OpenBMB: Ultra FineWeb (dataset)- Bilibili: Index-anisora (Anime/ACG)- Skywork:Matrix-Game (game)More awesome releases: Alibaba QwenLong-L1-32B, SkyWork OR1, OpenS2V-5M etc... See translation π₯ 1 1 π 1 1 + Reply
π May 2025 - Open works from the Chinese community Collection 43 items β’ Updated about 11 hours ago β’ 7
One-RL-to-See-Them-All Collection One RL to See Them All: Visual Triple Unified Reinforcement Learning. GitHub: https://github.com/MiniMax-AI/One-RL-to-See-Them-All β’ 5 items β’ Updated 6 days ago β’ 27
SynLogic Collection Synthesizing Verifiable Reasoning Data at Scale for Learning Logical Reasoning and Beyond β’ 2 items β’ Updated 6 days ago β’ 1
π May 2025 - Open works from the Chinese community Collection 43 items β’ Updated about 11 hours ago β’ 7
π June 2025 - Open works from the Chinese community Collection 1 item β’ Updated about 11 hours ago
view post Post 443 MiMo-VL π₯ smol & mighty vision language model by Xiaomi XiaomiMiMo/mimo-vl-68382ccacc7c2875500cd212β¨ 7B with RL & SFT β¨ Native resolution ViT for fine grained perceptionβ¨ MORL = smarter alignment across perception, grounding & reasoning See translation π₯ 1 1 + Reply
π May 2025 - Open works from the Chinese community Collection 43 items β’ Updated about 11 hours ago β’ 7
π May 2025 - Open works from the Chinese community Collection 43 items β’ Updated about 11 hours ago β’ 7
ZeroGUI Collection ZeroGUI: Automating Online GUI Learning at Zero Human Cost β’ 3 items β’ Updated 4 days ago β’ 1