AI研究論文デイリー

Generative World Explorer

BlueLM-V-3B: Algorithm and System Co-Design for Multimodal Large Language Models on Mobile Devices

Search, Verify and Feedback: Towards Next Generation Post-training Paradigm of Foundation Models via Verifier Engineering

AnimateAnything: Consistent and Controllable Animation for Video Generation

Top-nσ: Not All Logits Are You Need

Drowning in Documents: Consequences of Scaling Reranker Inference

FitDiT: Advancing the Authentic Garment Details for High-fidelity Virtual Try-on

SlimLM: An Efficient Small Language Model for On-Device Document Assistance

StableV2V: Stablizing Shape Consistency in Video-to-Video Editing

Awaker2.5-VL: Stably Scaling MLLMs with Parameter-Efficient Mixture of Experts

Adaptive Decoding via Latent Preference Optimization

LLäMmlein: Compact and Competitive German-Only Language Models from Scratch

SmoothCache: A Universal Inference Acceleration Technique for Diffusion Transformers

Comprehensive and Practical Evaluation of Retrieval-Augmented Generation Systems for Medical Question Answering

VeGaS: Video Gaussian Splatting

Evaluating the role of `Constitutions' for learning from AI feedback