new

Get trending papers in your email inbox once a day!

Get trending papers in your email inbox!

Daily Papers

byAK and the research community

Apr 10

Submitted by

taesiri

SkillClaw: Let Skills Evolve Collectively with Agentic Evolver

·
8 authors

Submitted by

Chenyu-Zhou

Externalization in LLM Agents: A Unified Review of Memory, Skills, Protocols and Harness Engineering

·
21 authors

Submitted by

Zuyan

HY-Embodied-0.5: Embodied Foundation Models for Real-World Agents

Tencent-Hunyuan

Tencent Hunyuan

Submitted by

LZXzju

KnowU-Bench: Towards Interactive, Proactive, and Personalized Mobile Agent Evaluation

·
16 authors

Submitted by

alkxncda

LPM 1.0: Video-based Character Performance Model

·
25 authors

Submitted by

fenglinglwb

OpenSpatial: A Principled Data Engine for Empowering Spatial Intelligence

JoyFutureAcademy

Joy Future Academy

Submitted by

Gaojunyao

MegaStyle: Constructing Diverse and Scalable Style Dataset via Consistent Text-to-Image Style Mapping

tencent

Submitted by

zli12321

Graph of Skills: Dependency-Aware Structural Retrieval for Massive Agent Skills

Latitude-Attitude

University of Pennsylvania

Submitted by

dkliang

When Numbers Speak: Aligning Textual Numerals and Visual Instances in Text-to-Video Diffusion Models

H-EmbodVis

Submitted by

shilinyan

Act Wisely: Cultivating Meta-Cognitive Tool Use in Agentic Multimodal Models

Accio-Lab

Submitted by

Zigeng

DMax: Aggressive Parallel Decoding for dLLMs

NationalUniversityofSingapore

National University of Singapore

Submitted by

Bowieee

Towards Real-world Human Behavior Simulation: Benchmarking Large Language Models on Long-horizon, Cross-scenario, Heterogeneous Behavior Traces

·
14 authors

Submitted by

weizhoudb

Automating Database-Native Function Code Synthesis with LLMs

SJTU

Shanghai Jiao Tong University

Submitted by

jskarras

FIT: A Large-Scale Dataset for Fit-Aware Virtual Try-On

google

Submitted by

taesiri

MolmoWeb: Open Visual Web Agent and Open Data for the Open Web

·
16 authors

Submitted by

Z-MU-Z

OmniJigsaw: Enhancing Omni-Modal Reasoning via Modality-Orchestrated Reordering

zju

Zhejiang University

Submitted by

Nothing2Say

Beyond Stochastic Exploration: What Makes Training Data Valuable for Agentic Search

·
6 authors

Submitted by

QQTang1223

Flux Attention: Context-Aware Hybrid Attention for Efficient LLMs Inference

LCM-Lab

Long-Context Model Laboratory

Submitted by

xhluca

Structured Distillation of Web Agent Capabilities Enables Generalization

McGill-NLP

McGill NLP Group

Submitted by

taesiri

GameWorld: Towards Standardized and Verifiable Evaluation of Multimodal Game Agents

·
5 authors

Submitted by

isminoula

RewardFlow: Generate Images by Optimizing What You Reward

PLAN-Lab

Perception and LANguage Lab @UIUC

Submitted by

isminoula

Phantom: Physics-Infused Video Generation via Joint Modeling of Visual and Latent Physical Dynamics

PLAN-Lab

Perception and LANguage Lab @UIUC

Submitted by

taesiri

PokeGym: A Visually-Driven Long-Horizon Benchmark for Vision-Language Models

·
8 authors

Submitted by

taesiri

OpenVLThinkerV2: A Generalist Multimodal Reasoning Model for Multi-domain Visual Tasks

·
6 authors

Submitted by

h-h-y

Structural Graph Probing of Vision-Language Models

Northeastern

Northeastern University