AI 연구 논문 데일리

OMG-LLaVA: Bridging Image-level, Object-level, Pixel-level Reasoning and Understanding

Step-DPO: Step-wise Preference Optimization for Long-chain Reasoning of LLMs

MUMU: Bootstrapping Multimodal Image Generation from Text-to-Image Data

Simulating Classroom Education with LLM-Empowered Agents

SeaKR: Self-aware Knowledge Retrieval for Adaptive Retrieval Augmented Generation

Aligning Teacher with Student Preferences for Tailored Training Data Generation

LiveBench: A Challenging, Contamination-Free LLM Benchmark

Can LLMs Learn by Teaching? A Preliminary Study

Dataset Size Recovery from LoRA Weights

MoA: Mixture of Sparse Attention for Automatic Large Language Model Compression

Is Programming by Example solved by LLMs?

T-FREE: Tokenizer-Free Generative LLMs via Sparse Representations for Memory-Efficient Embeddings

AUTOHALLUSION: Automatic Generation of Hallucination Benchmarks for Vision-Language Models

Read Anywhere Pointed: Layout-aware GUI Screen Reading with Tree-of-Lens Grounding

Understand What LLM Needs: Dual Preference Alignment for Retrieval-Augmented Generation

ArzEn-LLM: Code-Switched Egyptian Arabic-English Translation and Speech Recognition Using LLMs

ResumeAtlas: Revisiting Resume Classification with Large-Scale Datasets and Large Language Models

Benchmarking Mental State Representations in Language Models