AI研究論文每日精選

Learning to Reason under Off-Policy Guidance

Eagle 2.5: Boosting Long-Context Post-Training for Frontier Vision-Language Models

FlowReasoner: Reinforcing Query-Level Meta-Agents

ToolRL: Reward is All Tool Learning Needs

OTC: Optimal Tool Calls via Reinforcement Learning

X-Teaming: Multi-Turn Jailbreaks and Defenses with Adaptive Multi-Agents

UFO2: The Desktop AgentOS

SphereDiff: Tuning-free Omnidirectional Panoramic Image and Video Generation via Spherical Latent Representation

THOUGHTTERMINATOR: Benchmarking, Calibrating, and Mitigating Overthinking in Reasoning Models

StyleMe3D: Stylization with Disentangled Priors by Multiple Encoders on 3D Gaussians

Seeing from Another Perspective: Evaluating Multi-View Understanding in MLLMs

EasyEdit2: An Easy-to-use Steering Framework for Editing Large Language Models

LeetCodeDataset: A Temporal Dataset for Robust Evaluation and Efficient Training of Code LLMs

Uni3C: Unifying Precisely 3D-Enhanced Camera and Human Motion Controls for Video Generation

InfiGUI-R1: Advancing Multimodal GUI Agents from Reactive Actors to Deliberative Reasoners

LearnAct: Few-Shot Mobile GUI Agent with a Unified Demonstration Benchmark

An LMM for Efficient Video Understanding via Reinforced Compression of Video Cubes

DRAGON: Distributional Rewards Optimize Diffusion Generative Models

LookingGlass: Generative Anamorphoses via Laplacian Pyramid Warping

RainbowPlus: Enhancing Adversarial Prompt Generation via Evolutionary Quality-Diversity Search

TAPIP3D: Tracking Any Point in Persistent 3D Geometry

NEMOTRON-CROSSTHINK: Scaling Self-Learning beyond Math Reasoning

PROMPTEVALS: A Dataset of Assertions and Guardrails for Custom Production Large Language Model Pipelines

LoftUp: Learning a Coordinate-Based Feature Upsampler for Vision Foundation Models

RF-DETR Object Detection vs YOLOv12 : A Study of Transformer-based and CNN-based Architectures for Single-Class and Multi-Class Greenfruit Detection in Complex Orchard Environments Under Label Ambiguity

CoMotion: Concurrent Multi-person 3D Motion

Roll the dice & look before you leap: Going beyond the creative limits of next-token prediction

SilVar-Med: A Speech-Driven Visual Language Model for Explainable Abnormality Detection in Medical Imaging