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

Aug 18

Submitted by

manchery

HarnessEval-W: Agentifying the Evaluation of Visual Worlds

MirroS-Lab

Submitted by

yasNing

VibeWorlding: Can Multimodal Agents Construct 3D Open Worlds End-to-End?

tencent

Submitted by

XXsongLALA

ClawGym II: Exploring Black-Box RL on Agent Harness

RUC

Renmin University of China

Submitted by

lkeab

UI-Mate: Advancing Open-Weight Foundation GUI Agents with In-Context Demonstrations

Tencent-Hunyuan

Tencent Hunyuan

Submitted by

Jihuai

MOSS-VL Technical Report

OpenMOSS-Team

Submitted by

scyyc9

Large Discovery Models: Empirically-grounded Model-Based Open-Ended Search

Yangtze-ailab

Submitted by

williamium

Learn What's Left, Not What's Mastered: Saturation Aware Advantage Reweighting for Multi-Reward Policy Optimization

UCSanDiego

University of California at San Diego

Submitted by

curtis-sun

Agentic Transaction: Towards ACID-Compliant Agent Systems

Tsinghua University

Submitted by

DyJiang

An Empirical Study of Training Pixel-Space Text-to-Image Diffusion Models

Tongyi-MAI

1

Submitted by

Harold328

GenRouter: Unified Workflow Routing for Agentic Image Generation

·
7 authors

Submitted by

chelseyu

How Do Agents Fail on AutoResearch: End-to-End Diagnostic Evaluation on 100 Real-World Frontier Research Tasks

PrentisAI

1

Submitted by

guan-wang

Understanding Cognition-Induced Risks in Agentic AI Systems

Shanghai-AI-Laboratory

Shanghai AI Laboratory

Submitted by

ZhaoyangLyu

MegaParts: Scaling Part-Aware 3D Object Generation to 300 Parts via Token-Efficient Autoregressive Modeling

Shanghai-AI-Laboratory

Shanghai AI Laboratory

Submitted by

xiangfanwu

Ventor-QTest: Threat-Model-Driven Verification of Vendor-Hosted LLM APIs

tencent

Submitted by

CedarWang

R^3-Bench: LLMs Struggle with Resource-Rational Reasoning under Shared Budgets

HKUST-DSAIL

Data Science & Artificial Intelligence Lab at HKUST

Submitted by

Karl28

VideoGAIA: A Benchmark for General AI Assistants on Agentic Video Understanding

CUHK-CSE

The Chinese University of Hong Kong

Submitted by

chaofanma

TRACE-Bench: Decomposing and Diagnosing Multi-Reference Image Generation

SJTU

Shanghai Jiao Tong University

Submitted by

HigherHu

GRNEdit: Efficient General Video Editing from a New Binary-Evidence Perspective in Generative Refinement Networks

·
8 authors

Submitted by

taesiri

Improving the matrix multiplication exponent with modern optimization and AlphaEvolve

deepmind

Submitted by

caipeng328

NaviDC-OCR: Navigating Document Parsing Across Digital and Camera-Captured Documents

StarDoc-AI

Submitted by

Akrin-scitix

ENTLORE: A Graph-Grounded Benchmark for Latent Organizational Reasoning in Enterprise Question Answering

ScitiX1

Scitix (SGP) Tech PTE. LTD.

Submitted by

kwanY

AnyTalk: Speech Animation for Arbitrary Characters Leveraging a Video Generation Model

Submitted by

hmhm1229

ConceptFormer: Learning Adaptive Latent Concepts for Query-Document Alignment in Visual Document Retrieval

·
12 authors

Submitted by

hbx

PACE-Bench: Benchmarking Physics Adaptation via Code Evolution in Dynamic Environments

TsinghuaNLP

Tsinghua NLP Group

Submitted by

parsa-mz

Prior Audit-Repair Context Shifts LLM Verifier Thresholds Toward Leniency

·
2 authors

Submitted by

parsa-mz

Gathered, Not Admitted: How Attention Brings a Latent Variable into Verbalizable Form

UCSC

University of California, Santa Cruz

Submitted by

alana89

Advancing Open and Reproducible Relational Learning: RelArena-α, TabPFN-Rel and RPI

Prior-Labs

Submitted by

Murjani

Drive, Pack, Fly: The Travelling Thief Problem with Drone

·
2 authors

Submitted by

KanameYOkoYAMA

A Plug-and-Play 2D Motion Interface for Real-World Motion Language Models

·
2 authors

Submitted by

geyongtao

WorldRover: A Scalable Synthetic Video Data Engine for World Exploration with Rich Annotations

AlayaLab

Submitted by

An998

When Context Bites: Detecting RAG Poisoning via Document-Level Attention Collapse

·
6 authors

1

Submitted by

LeoYML

Position: AI Agents in Scientific Teams Should Be Studied as Human-Agent Systems

National Laboratory of the Rockies

Submitted by

iliashum

DumpsterCluster: From Dumpster Diving to Serving LLaMA-70B on $60 GPUs

Oxford

University of Oxford

Submitted by

1nnoh

HiFi-BRep: High-Fidelity Latent Representation for Robust B-Rep Generation

zju

Zhejiang University

Submitted by

aryadomain

Prototype-Rectified Iterative Self-supervised Manifold Denoising under Severe Acoustic Shift

·
4 authors