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 19

Submitted by

zz1358m

Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements

NationalUniversityofSingapore

National University of Singapore

Submitted by

andy-yang

FreeToken: Efficient Edge-Native MoE Serving with Bandwidth-Adaptive Execution

UCBerkeley

University of California, Berkeley

Submitted by

williamium

Demystifying Agent Skills: Why They Work-Until They Don't

UCSanDiego

University of California at San Diego

Submitted by

taesiri

ASI-Bench: At the Dawn of Artificial Superintelligence

·
42 authors

Submitted by

zwq2018

Embodied-Navigator: Point, Think, Memorize, and Align for Efficient Navigation

OmniAI-ZJU

Submitted by

hzy46

Agent Lightning v1.0: Towards Harnessed Agentic RL

microsoft

Submitted by

ChocoWu

V-RAE: Rethinking Video Latent Spaces for Generation

NationalUniversityofSingapore

National University of Singapore

Submitted by

woody-woody

AVA-Encoder: Towards Agent-Native Video Representation Learning

QwenBusinessUnit

Qwen Business Unit

Submitted by

YuxinJiang

LEGO-RL: Harness-Native Reinforcement Learning for Coding Agents

Lego-X

Submitted by

sjy92

EDITBRIDGE: Towards Faithful and Efficient Ultra-High-Resolution Image Editing

·
8 authors

Submitted by

Tong98Zhang

DiSCO: Defending text-to-image generation through distribution-guided contrastive prompt optimization

kaust-generative-ai

KAUST Center of Excellence in Generative AI

Submitted by

nielsr

MoE-ViE: Mixture of Experts Vision Encoder for Efficient Image and Video Understanding

Submitted by

Owos

Dynamic Multi-Byte Prediction With Hierarchical Language Models

ohiostate

The Ohio State University

Submitted by

FireCRT

CoinVE-200K: A Large-Scale High-Quality Dataset for Compositional Instruction-Guided Video Editing

tencent

Submitted by

Yakundur

Harness the Memory: A Holistic Evaluation of Memory Substrates in Memory Agents

·
15 authors

Submitted by

Silin-Chen

SkillForge: Self-Distilling Agents for Project-Specific Issue Resolution

SJTU

Shanghai Jiao Tong University

2

Submitted by

CyberPegasus

From Corpora to Co-Evolving Capabilities: Capability-Centric Data Design for Generalist Image Generation

alibabagroup

Submitted by

taesiri

Abra: Scaling Diffusion Image Training

lumaai

Submitted by

taesiri

StartupBench: Benchmarking General-Purpose Agents on Market-Validated End-to-End Workflows

ByteDance-Seed

Submitted by

Beckham808

HarnessRisk: A Lifecycle-Oriented Benchmark for Agent Harness Safety

·
7 authors

Submitted by

ysng

Energy-Guided Flow Matching

·
9 authors

Submitted by

jisx

Cross-Model Memory Transfer via Target-Side Reader Adaptation

OLAResearchX

Omni Language AI Research

Submitted by

daypu

MathForm: Scaling Mathematical Autoformalization with Knowledge Retrieval and Verification-Guided Refinement

openbmb

Submitted by

Franck-Dernoncourt

Personalized Auto-Research: Towards a True AI Co-Scientist

·
8 authors

Submitted by

Zonghao2025

Security Assessment of DeepSeek Harness with A.I.G: Evaluating Resistance to Indirect Prompt Injection

tencent

Submitted by

taesiri

aDSL: Agentic 3D Creation via Joint Agent-Program Design

·
6 authors

Submitted by

zeyu-zheng

The Problem Is the Problem: Towards Scalable Mathematical Discovery

CarnegieMellonU

Carnegie Mellon University

Submitted by

Franck-Dernoncourt

Unifying Graph Neural Networks Through a Common Layer Equation

·
12 authors

Submitted by

qian43

GS-Voxel: Fitting-Free Structured Latents for Large-Scale 3DGS Generation

acvlab

Alibaba AMAP CV Lab

3

Submitted by

Bowieee

From Sequence to Structure: Relational Uncertainty Propagation for LLM Agents

ICIP

Submitted by

Genghan

PTXBench: Benchmark and Adapt LLMs for GPU Kernel Optimization with Architecture-specific PTX

StanfordUniversity

Stanford University

Submitted by

nielsr

PixRestore: Unified Image Restoration via Pixel Diffusion Transformer

·
9 authors

Submitted by

hamzashafiq

CardioState-JEPA: Delay-Aware Cross-Modal Learning of a Shared Cardiac Representation

·
6 authors