Hugging Face's logo
MLLM's picture
17 4

MLLM PRO

Anran-MLLM
ttkhty0's profile picture godx7's profile picture bitersun's profile picture
·

AI & ML interests

None yet

Recent Activity

reacted to theirpost with 👍 2 days ago
🚀 Introducing PerceptionDLM — the first multimodal diffusion LLM for parallel region perception! Most MLLMs are autoregressive, so captioning N regions costs N sequential passes. PerceptionDLM instead describes ALL masked regions in a single denoising process. 🧩 ✨ Highlights • ⚡ Up to 3.4× faster on dense multi-region captioning, with stable per-image latency • 🏆 PerceptionDLM-Base beats LLaDA-V on 15/16 multimodal benchmarks (new SOTA among open diffusion VLMs) • 📊 New benchmark: ParaDLC-Bench — jointly evaluates caption quality AND inference efficiency • 🔓 Code, models & benchmark all open-sourced 🤖 Models https://huggingface.co/MSALab/PerceptionDLM-Base https://huggingface.co/MSALab/PerceptionDLM 📊 Benchmark https://huggingface.co/datasets/MSALab/ParaDLC-Bench 📄 Paper: https://huggingface.co/papers/2606.19534 💻 Code: https://github.com/MSALab-PKU/PerceptionDLM Diffusion LLMs aren't just for text — they unlock efficient, parallel visual perception. 👁️✨ #multimodal #diffusion #VLM #perception
reacted to theirpost with 🔥 2 days ago
🚀 Introducing PerceptionDLM — the first multimodal diffusion LLM for parallel region perception! Most MLLMs are autoregressive, so captioning N regions costs N sequential passes. PerceptionDLM instead describes ALL masked regions in a single denoising process. 🧩 ✨ Highlights • ⚡ Up to 3.4× faster on dense multi-region captioning, with stable per-image latency • 🏆 PerceptionDLM-Base beats LLaDA-V on 15/16 multimodal benchmarks (new SOTA among open diffusion VLMs) • 📊 New benchmark: ParaDLC-Bench — jointly evaluates caption quality AND inference efficiency • 🔓 Code, models & benchmark all open-sourced 🤖 Models https://huggingface.co/MSALab/PerceptionDLM-Base https://huggingface.co/MSALab/PerceptionDLM 📊 Benchmark https://huggingface.co/datasets/MSALab/ParaDLC-Bench 📄 Paper: https://huggingface.co/papers/2606.19534 💻 Code: https://github.com/MSALab-PKU/PerceptionDLM Diffusion LLMs aren't just for text — they unlock efficient, parallel visual perception. 👁️✨ #multimodal #diffusion #VLM #perception
upvoted a paper 4 days ago
PerceptionDLM: Parallel Region Perception with Multimodal Diffusion Language Models
View all activity

Organizations

None yet