-
PUMA: Empowering Unified MLLM with Multi-granular Visual Generation
Paper • 2410.13861 • Published • 57 -
JanusFlow: Harmonizing Autoregression and Rectified Flow for Unified Multimodal Understanding and Generation
Paper • 2411.07975 • Published • 31 -
Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimization
Paper • 2411.10442 • Published • 81 -
Multimodal Autoregressive Pre-training of Large Vision Encoders
Paper • 2411.14402 • Published • 47
hongbin
L-Hongbin
AI & ML interests
None yet
Recent Activity
updated
a collection
1 day ago
MutiModal_Dataset
updated
a collection
1 day ago
MutiModal_Paper
commented on
a paper
4 days ago
70% Size, 100% Accuracy: Lossless LLM Compression for Efficient GPU
Inference via Dynamic-Length Float
Organizations
None yet
Collections
6
-
Search, Verify and Feedback: Towards Next Generation Post-training Paradigm of Foundation Models via Verifier Engineering
Paper • 2411.11504 • Published • 22 -
Top-nσ: Not All Logits Are You Need
Paper • 2411.07641 • Published • 22 -
Adaptive Decoding via Latent Preference Optimization
Paper • 2411.09661 • Published • 10 -
When Precision Meets Position: BFloat16 Breaks Down RoPE in Long-Context Training
Paper • 2411.13476 • Published • 16
models
None public yet
datasets
None public yet