AI 연구 논문 데일리

OmniHuman-1: Rethinking the Scaling-Up of One-Stage Conditioned Human Animation Models

The Differences Between Direct Alignment Algorithms are a Blur

Process Reinforcement through Implicit Rewards

Preference Leakage: A Contamination Problem in LLM-as-a-judge

AlignVLM: Bridging Vision and Language Latent Spaces for Multimodal Understanding

SafeRAG: Benchmarking Security in Retrieval-Augmented Generation of Large Language Model

SliderSpace: Decomposing the Visual Capabilities of Diffusion Models

MM-IQ: Benchmarking Human-Like Abstraction and Reasoning in Multimodal Models

DeepRAG: Thinking to Retrieval Step by Step for Large Language Models

Scaling Embedding Layers in Language Models

MakeAnything: Harnessing Diffusion Transformers for Multi-Domain Procedural Sequence Generation

AIN: The Arabic INclusive Large Multimodal Model

FastKV: KV Cache Compression for Fast Long-Context Processing with Token-Selective Propagation

ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning

The Jumping Reasoning Curve? Tracking the Evolution of Reasoning Performance in GPT-[n] and o-[n] Models on Multimodal Puzzles

Almost Surely Safe Alignment of Large Language Models at Inference-Time

RandLoRA: Full-rank parameter-efficient fine-tuning of large models

PhD Knowledge Not Required: A Reasoning Challenge for Large Language Models

Improving Transformer World Models for Data-Efficient RL

Improved Training Technique for Latent Consistency Models

Lifelong Sequential Knowledge Editing without Model Degradation

LongDPO: Unlock Better Long-form Generation Abilities for LLMs via Critique-augmented Stepwise Information

Learning to Generate Unit Tests for Automated Debugging

Language Models Prefer What They Know: Relative Confidence Estimation via Confidence Preferences

A Study on the Performance of U-Net Modifications in Retroperitoneal Tumor Segmentation

Current Pathology Foundation Models are unrobust to Medical Center Differences