Towards Scalable Pre-training of Visual Tokenizers for Generation
Paper • 2512.13687 • Published • 108
papers
-
-
MMGR: Multi-Modal Generative Reasoning
Paper • 2512.14691 • Published • 121 -
Coupling Experts and Routers in Mixture-of-Experts via an Auxiliary Loss
Paper • 2512.23447 • Published • 99 -
LiveTalk: Real-Time Multimodal Interactive Video Diffusion via Improved On-Policy Distillation
Paper • 2512.23576 • Published • 66 -
mHC: Manifold-Constrained Hyper-Connections
Paper • 2512.24880 • Published • 328 -
Youtu-LLM: Unlocking the Native Agentic Potential for Lightweight Large Language Models
Paper • 2512.24618 • Published • 155 -
PhyGDPO: Physics-Aware Groupwise Direct Preference Optimization for Physically Consistent Text-to-Video Generation
Paper • 2512.24551 • Published • 21 -
Let It Flow: Agentic Crafting on Rock and Roll, Building the ROME Model within an Open Agentic Learning Ecosystem
Paper • 2512.24873 • Published • 109 -
Improving Multi-step RAG with Hypergraph-based Memory for Long-Context Complex Relational Modeling
Paper • 2512.23959 • Published • 111 -
Agent Learning via Early Experience
Paper • 2510.08558 • Published • 276 -
Entropy-Adaptive Fine-Tuning: Resolving Confident Conflicts to Mitigate Forgetting
Paper • 2601.02151 • Published • 115 -
Thinking with Map: Reinforced Parallel Map-Augmented Agent for Geolocalization
Paper • 2601.05432 • Published • 170 -
MMFormalizer: Multimodal Autoformalization in the Wild
Paper • 2601.03017 • Published • 106 -
Qwen3-VL Technical Report
Paper • 2511.21631 • Published • 163 -
Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance
Paper • 2512.08765 • Published • 134 -
Less is More: Recursive Reasoning with Tiny Networks
Paper • 2510.04871 • Published • 516 -
Advancing Open-source World Models
Paper • 2601.20540 • Published • 135 -
OPUS: Towards Efficient and Principled Data Selection in Large Language Model Pre-training in Every Iteration
Paper • 2602.05400 • Published • 355 -
ASA: Training-Free Representation Engineering for Tool-Calling Agents
Paper • 2602.04935 • Published • 42 -
Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters
Paper • 2602.10604 • Published • 200 -
Kimi K2.5: Visual Agentic Intelligence
Paper • 2602.02276 • Published • 273 -
GeoAgent: Learning to Geolocate Everywhere with Reinforced Geographic Characteristics
Paper • 2602.12617 • Published • 20 -
MedXIAOHE: A Comprehensive Recipe for Building Medical MLLMs
Paper • 2602.12705 • Published • 68 -
DeepImageSearch: Benchmarking Multimodal Agents for Context-Aware Image Retrieval in Visual Histories
Paper • 2602.10809 • Published • 59 -
SLA2: Sparse-Linear Attention with Learnable Routing and QAT
Paper • 2602.12675 • Published • 59 -
Unified Latents (UL): How to train your latents
Paper • 2602.17270 • Published • 60 -
Code2World: A GUI World Model via Renderable Code Generation
Paper • 2602.09856 • Published • 201 -
Mobile-Agent-v3.5: Multi-platform Fundamental GUI Agents
Paper • 2602.16855 • Published • 51 -
World Craft: Agentic Framework to Create Visualizable Worlds via Text
Paper • 2601.09150 • Published • 19 -
VESPO: Variational Sequence-Level Soft Policy Optimization for Stable Off-Policy LLM Training
Paper • 2602.10693 • Published • 221 -
Does Your Reasoning Model Implicitly Know When to Stop Thinking?
Paper • 2602.08354 • Published • 266 -
A Very Big Video Reasoning Suite
Paper • 2602.20159 • Published • 525 -
On Data Engineering for Scaling LLM Terminal Capabilities
Paper • 2602.21193 • Published • 103 -
Query-focused and Memory-aware Reranker for Long Context Processing
Paper • 2602.12192 • Published • 58 -
HyTRec: A Hybrid Temporal-Aware Attention Architecture for Long Behavior Sequential Recommendation
Paper • 2602.18283 • Published • 57 -
ARLArena: A Unified Framework for Stable Agentic Reinforcement Learning
Paper • 2602.21534 • Published • 26 -
CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation
Paper • 2602.24286 • Published • 99 -
dLLM: Simple Diffusion Language Modeling
Paper • 2602.22661 • Published • 153 -
Geometry-Guided Reinforcement Learning for Multi-view Consistent 3D Scene Editing
Paper • 2603.03143 • Published • 145 -
AgentDoG: A Diagnostic Guardrail Framework for AI Agent Safety and Security
Paper • 2601.18491 • Published • 126 -
Video-Thinker: Sparking "Thinking with Videos" via Reinforcement Learning
Paper • 2510.23473 • Published • 87 -
Experiential Reinforcement Learning
Paper • 2602.13949 • Published • 75 -
DataFlex: A Unified Framework for Data-Centric Dynamic Training of Large Language Models
Paper • 2603.26164 • Published • 366 -
CARLA-Air: Fly Drones Inside a CARLA World -- A Unified Infrastructure for Air-Ground Embodied Intelligence
Paper • 2603.28032 • Published • 343 -
SKILL0: In-Context Agentic Reinforcement Learning for Skill Internalization
Paper • 2604.02268 • Published • 101 -
GEMS: Agent-Native Multimodal Generation with Memory and Skills
Paper • 2603.28088 • Published • 86 -
FIPO: Eliciting Deep Reasoning with Future-KL Influenced Policy Optimization
Paper • 2603.19835 • Published • 352 -
Agentic Reasoning for Large Language Models
Paper • 2601.12538 • Published • 204 -
Rethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model Capability
Paper • 2604.06628 • Published • 327 -
On the Generalization of SFT: A Reinforcement Learning Perspective with Reward Rectification
Paper • 2508.05629 • Published • 190 -
HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worlds
Paper • 2604.14268 • Published • 124 -
From Context to Skills: Can Language Models Learn from Context Skillfully?
Paper • 2604.27660 • Published • 166 -
SWE-Pruner: Self-Adaptive Context Pruning for Coding Agents
Paper • 2601.16746 • Published • 91 -
OpenSearch-VL: An Open Recipe for Frontier Multimodal Search Agents
Paper • 2605.05185 • Published • 102 -
Cosmos 3: Omnimodal World Models for Physical AI
Paper • 2606.02800 • Published • 102