AI 연구 논문 데일리

FrugalNeRF: Fast Convergence for Few-shot Novel View Synthesis without Learned Priors

SAM2Long: Enhancing SAM 2 for Long Video Segmentation with a Training-Free Memory Tree

CompassJudger-1: All-in-one Judge Model Helps Model Evaluation and Evolution

AutoTrain: No-code training for state-of-the-art models

PUMA: Empowering Unified MLLM with Multi-granular Visual Generation

Baichuan Alignment Technical Report

SemiEvol: Semi-supervised Fine-tuning for LLM Adaptation

Pangea: A Fully Open Multilingual Multimodal LLM for 39 Languages

RM-Bench: Benchmarking Reward Models of Language Models with Subtlety and Style

Meta-Chunking: Learning Efficient Text Segmentation via Logical Perception

Pre-training Distillation for Large Language Models: A Design Space Exploration

Alchemy: Amplifying Theorem-Proving Capability through Symbolic Mutation

Ichigo: Mixed-Modal Early-Fusion Realtime Voice Assistant

Zero-shot Model-based Reinforcement Learning using Large Language Models

Selecting Influential Samples for Long Context Alignment via Homologous Models' Guidance and Contextual Awareness Measurement

How Many Van Goghs Does It Take to Van Gogh? Finding the Imitation Threshold

Agent-to-Sim: Learning Interactive Behavior Models from Casual Longitudinal Videos

CBT-Bench: Evaluating Large Language Models on Assisting Cognitive Behavior Therapy

In-context learning and Occam's razor

Router-Tuning: A Simple and Effective Approach for Enabling Dynamic-Depth in Transformers

Hallucination Detox: Sensitive Neuron Dropout (SeND) for Large Language Model Training

Cross-Lingual Auto Evaluation for Assessing Multilingual LLMs

DM-Codec: Distilling Multimodal Representations for Speech Tokenization