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

May 25

Submitted by

taesiri

SkillOpt: Executive Strategy for Self-Evolving Agent Skills

MicrosoftResearch

Microsoft Research

Submitted by

Met4physics

Rethinking Cross-Layer Information Routing in Diffusion Transformers

RTP-LLM

3

Submitted by

Jinjing713

Lens: Rethinking Training Efficiency for Foundational Text-to-Image Models

microsoft

Submitted by

Ningyu

SciAtlas: A Large-Scale Knowledge Graph for Automated Scientific Research

University College London CS

Submitted by

giantPanda0906

StepAudio 2.5 Technical Report

·
101 authors

Submitted by

BBBBCHAN

See What I Mean: Aligning Vision and Language Representations for Video Fine-grained Object Understanding

AlibabaTongyiLab

Submitted by

taesiri

From Raw Experience to Skill Consumption: A Systematic Study of Model-Generated Agent Skills

MicrosoftResearch

Microsoft Research

Submitted by

taesiri

PiD: Fast and High-Resolution Latent Decoding with Pixel Diffusion

nvidia

Submitted by

Zuica96

PhotoFlow: Agentic 3D Virtual Photography Missions

Visionary-Laboratoary

Visionary-Laboratoary

Submitted by

zino1

VGenST-Bench: A Benchmark for Spatio-Temporal Reasoning via Active Video Synthesis

·
4 authors

Submitted by

syjian

RankE: End-to-End Post-Training for Discrete Text-to-Image Generation with Decoder Co-Evolution

Westlake-University

Westlake University

Submitted by

yuhangzang

ETCHR: Editing To Clarify and Harness Reasoning

internlm

Intern Large Models

Submitted by

taesiri

LLMs as Noisy Channels: A Shannon Perspective on Model Capacity and Scaling Laws

·
8 authors

Submitted by

INV-WZQ

SCOPE: Simulating Cross-game Operations in Playable Environments for FPS World Models

·
14 authors

Submitted by

lizizun

Geo-Align: Video Generation Alignment via Metric Geometry Reward

·
5 authors

Submitted by

Chtholly17

From Seeing to Thinking: Decoupling Perception and Reasoning Improves Post-Training of Vision-Language Models

·
9 authors

Submitted by

taesiri

GenRecon: Bridging Generative Priors for Multi-View 3D Scene Reconstruction

·
5 authors

Submitted by

a-F1

Rethinking Muon Beyond Pretraining: Spectral Failures and High-Pass Remedies for VLA and RLVR

OPTML-Group

OPTML Group @ MSU

Submitted by

jindongwang

LatentUMM: Dual Latent Alignment for Unified Multimodal Models

CarnegieMellonU

Carnegie Mellon University

Submitted by

goyalkaraniit

The Expense of Seeing: Attaining Trustworthy Multimodal Reasoning Within the Monolithic Paradigm

·
1 authors

1

Submitted by

wgcyeo

HINT-SD: Targeted Hindsight Self-Distillation for Long-Horizon Agents

kaist-ai

1

Submitted by

ShuhongZheng

Good Token Hunting: A Hitchhiker's Guide to Token Selection for Visual Geometry Transformers

uoft

University of Toronto

Submitted by

yflantmy

The Illusion of Reasoning: Exposing Evasive Data Contamination in LLMs via Zero-CoT Truncation

·
5 authors

Submitted by

HuskyDoge

Equilibrium Reasoners: Learning Attractors Enables Scalable Reasoning

locuslab

Submitted by

liboaccn

VaaWIT: Visual-Aware Adaptation of Large Language Models for Multilingual Web Image Translation

·
6 authors

Submitted by

liboaccn

Mix-MoE: Improving Multilingual Machine Translation of Large Language Models through Mixed MoEs

·
4 authors

Submitted by

Parkprogrammer

Self-Improving CAD Generation Agents with Finite Element Analysis as Feedback

SeoulNatlUniv

Seoul National University

1

Submitted by

yilmazkorkmaz

Next-Acceleration-Scale Prediction for Autoregressive MRI Reconstruction

JohnsHopkins

Johns Hopkins University

Submitted by

KevinQHLin

Soap2Soap: Long Cinematic Video Remaking via Multi-Agent Collaboration

showlab