AI研究论文每日精选

OmniSVG: A Unified Scalable Vector Graphics Generation Model

Hogwild! Inference: Parallel LLM Generation via Concurrent Attention

Skywork R1V: Pioneering Multimodal Reasoning with Chain-of-Thought

An Empirical Study of GPT-4o Image Generation Capabilities

COIG-P: A High-Quality and Large-Scale Chinese Preference Dataset for Alignment with Human Values

Less-to-More Generalization: Unlocking More Controllability by In-Context Generation

HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference

Generative Evaluation of Complex Reasoning in Large Language Models

HiFlow: Training-free High-Resolution Image Generation with Flow-Aligned Guidance

V-MAGE: A Game Evaluation Framework for Assessing Visual-Centric Capabilities in Multimodal Large Language Models

Tuning-Free Image Editing with Fidelity and Editability via Unified Latent Diffusion Model

Accelerate Parallelizable Reasoning via Parallel Decoding within One Sequence

Efficient Reinforcement Finetuning via Adaptive Curriculum Learning

CrossWordBench: Evaluating the Reasoning Capabilities of LLMs and LVLMs with Controllable Puzzle Generation

Leanabell-Prover: Posttraining Scaling in Formal Reasoning

Unified World Models: Coupling Video and Action Diffusion for Pretraining on Large Robotic Datasets

ProtoGCD: Unified and Unbiased Prototype Learning for Generalized Category Discovery