AI研究論文デイリー

UCFE: A User-Centric Financial Expertise Benchmark for Large Language Models

Web Agents with World Models: Learning and Leveraging Environment Dynamics in Web Navigation

NaturalBench: Evaluating Vision-Language Models on Natural Adversarial Samples

MagicTailor: Component-Controllable Personalization in Text-to-Image Diffusion Models

SeerAttention: Learning Intrinsic Sparse Attention in Your LLMs

FiTv2: Scalable and Improved Flexible Vision Transformer for Diffusion Model

DPLM-2: A Multimodal Diffusion Protein Language Model

Mini-Omni2: Towards Open-source GPT-4o with Vision, Speech and Duplex Capabilities

HART: Efficient Visual Generation with Hybrid Autoregressive Transformer

Diffusion Curriculum: Synthetic-to-Real Generative Curriculum Learning via Image-Guided Diffusion

Are AI Detectors Good Enough? A Survey on Quality of Datasets With Machine-Generated Texts

DAWN: Dynamic Frame Avatar with Non-autoregressive Diffusion Framework for Talking Head Video Generation

BiGR: Harnessing Binary Latent Codes for Image Generation and Improved Visual Representation Capabilities

Looking Inward: Language Models Can Learn About Themselves by Introspection

SHAKTI: A 2.5 Billion Parameter Small Language Model Optimized for Edge AI and Low-Resource Environments

How Do Training Methods Influence the Utilization of Vision Models?

Context is Key(NMF): Modelling Topical Information Dynamics in Chinese Diaspora Media

A Common Pitfall of Margin-based Language Model Alignment: Gradient Entanglement

Teaching Models to Balance Resisting and Accepting Persuasion

Montessori-Instruct: Generate Influential Training Data Tailored for Student Learning