DLER: Doing Length pEnalty Right - Incentivizing More Intelligence per Token via Reinforcement Learning
Paper
• 2510.15110 • Published
• 17
multimodal
-
-
PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Model
Paper • 2510.14528 • Published • 118 -
Bee: A High-Quality Corpus and Full-Stack Suite to Unlock Advanced Fully Open MLLMs
Paper • 2510.13795 • Published • 59 -
UniME-V2: MLLM-as-a-Judge for Universal Multimodal Embedding Learning
Paper • 2510.13515 • Published • 12 -
SAIL-Embedding Technical Report: Omni-modal Embedding Foundation Model
Paper • 2510.12709 • Published • 13 -
HoneyBee: Data Recipes for Vision-Language Reasoners
Paper • 2510.12225 • Published • 11 -
Visual Spatial Tuning
Paper • 2511.05491 • Published • 52 -
DeepEyesV2: Toward Agentic Multimodal Model
Paper • 2511.05271 • Published • 45 -
NVIDIA Nemotron Nano V2 VL
Paper • 2511.03929 • Published • 30 -
SAIL-RL: Guiding MLLMs in When and How to Think via Dual-Reward RL Tuning
Paper • 2511.02280 • Published • 4 -
When Visualizing is the First Step to Reasoning: MIRA, a Benchmark for Visual Chain-of-Thought
Paper • 2511.02779 • Published • 59 -
Sample By Step, Optimize By Chunk: Chunk-Level GRPO For Text-to-Image Generation
Paper • 2510.21583 • Published • 31 -
Towards Universal Video Retrieval: Generalizing Video Embedding via Synthesized Multimodal Pyramid Curriculum
Paper • 2510.27571 • Published • 19