AI 연구 논문 데일리

Beyond 'Aha!': Toward Systematic Meta-Abilities Alignment in Large Reasoning Models

System Prompt Optimization with Meta-Learning

Parallel Scaling Law for Language Models

OpenThinkIMG: Learning to Think with Images via Visual Tool Reinforcement Learning

EnerVerse-AC: Envisioning Embodied Environments with Action Condition

The CoT Encyclopedia: Analyzing, Predicting, and Controlling how a Reasoning Model will Think

End-to-End Vision Tokenizer Tuning

EWMBench: Evaluating Scene, Motion, and Semantic Quality in Embodied World Models

WorldPM: Scaling Human Preference Modeling

J1: Incentivizing Thinking in LLM-as-a-Judge via Reinforcement Learning

MLE-Dojo: Interactive Environments for Empowering LLM Agents in Machine Learning Engineering

Unilogit: Robust Machine Unlearning for LLMs Using Uniform-Target Self-Distillation

Style Customization of Text-to-Vector Generation with Image Diffusion Priors

PointArena: Probing Multimodal Grounding Through Language-Guided Pointing

Depth Anything with Any Prior

Achieving Tokenizer Flexibility in Language Models through Heuristic Adaptation and Supertoken Learning

Exploring the Deep Fusion of Large Language Models and Diffusion Transformers for Text-to-Image Synthesis

ReSurgSAM2: Referring Segment Anything in Surgical Video via Credible Long-term Tracking

3D-Fixup: Advancing Photo Editing with 3D Priors

QuXAI: Explainers for Hybrid Quantum Machine Learning Models

AdaptCLIP: Adapting CLIP for Universal Visual Anomaly Detection

AI Agents vs. Agentic AI: A Conceptual Taxonomy, Applications and Challenge

Real2Render2Real: Scaling Robot Data Without Dynamics Simulation or Robot Hardware

MetaUAS: Universal Anomaly Segmentation with One-Prompt Meta-Learning

Learning to Detect Multi-class Anomalies with Just One Normal Image Prompt

Few-Shot Anomaly-Driven Generation for Anomaly Classification and Segmentation

X-Sim: Cross-Embodiment Learning via Real-to-Sim-to-Real