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

November

Submitted by

lkdhy

Thinking with Video: Video Generation as a Promising Multimodal Reasoning Paradigm

OpenMOSS-Team

Submitted by

taesiri

Kandinsky 5.0: A Family of Foundation Models for Image and Video Generation

·
25 authors

Submitted by

taesiri

Lumine: An Open Recipe for Building Generalist Agents in 3D Open Worlds

ByteDance-Seed

Submitted by

taesiri

MiroThinker: Pushing the Performance Boundaries of Open-Source Research Agents via Model, Context, and Interactive Scaling

·
54 authors

Submitted by

lz1001

General Agentic Memory Via Deep Research

BAAI

Beijing Academy of Artificial Intelligence

Submitted by

ambud26

Souper-Model: How Simple Arithmetic Unlocks State-of-the-Art LLM Performance

Submitted by

taesiri

SAM 3: Segment Anything with Concepts

Submitted by

JC-Chen

P1: Mastering Physics Olympiads with Reinforcement Learning

·
28 authors

Submitted by

YiZhouDenseHub

Tiny Model, Big Logic: Diversity-Driven Optimization Elicits Large-Model Reasoning Ability in VibeThinker-1.5B

WeiboAI

Submitted by

jinjieni

Diffusion Language Models are Super Data Learners

NationalUniversityofSingapore

National University of Singapore

Submitted by

RazinAleks

One Small Step in Latent, One Giant Leap for Pixels: Fast Latent Upscale Adapter for Your Diffusion Models

·
3 authors

Submitted by

taesiri

TiDAR: Think in Diffusion, Talk in Autoregression

nvidia

Submitted by

jiaruz2

Latent Collaboration in Multi-Agent Systems

Gen-Verse

Submitted by

kuznetsoffandrey

GigaEvo: An Open Source Optimization Framework Powered By LLMs And Evolution Algorithms

AIRI-Institute

AIRI - Artificial Intelligence Research Institute

Submitted by

taesiri

SAM 3D: 3Dfy Anything in Images

Submitted by

YifanJ

VIDEOP2R: Video Understanding from Perception to Reasoning

amazon

5

Submitted by

richardxp888

Agent0: Unleashing Self-Evolving Agents from Zero Data via Tool-Integrated Reasoning

UNC-ChapelHill

University of North Carolina at Chapel Hill

Submitted by

fuvty

Think-at-Hard: Selective Latent Iterations to Improve Reasoning Language Models

nics-efc

Tsinghua-NICS-EFC

Submitted by

aarashfeizi

Grounding Computer Use Agents on Human Demonstrations

ServiceNow

Submitted by

YunxinLi

Uni-MoE-2.0-Omni: Scaling Language-Centric Omnimodal Large Model with Advanced MoE, Training and Data

HIT-TMG

Submitted by

tttonyalpha

Don't Blind Your VLA: Aligning Visual Representations for OOD Generalization

·
5 authors

Submitted by

KevinQHLin

VCode: a Multimodal Coding Benchmark with SVG as Symbolic Visual Representation

CSU-JPG

Submitted by

AdinaY

Depth Anything 3: Recovering the Visual Space from Any Views

ByteDance-Seed

Submitted by

LibraTree

GeoVista: Web-Augmented Agentic Visual Reasoning for Geolocalization

Tencent-Hunyuan

Tencent Hunyuan

Submitted by

menik1126

DoPE: Denoising Rotary Position Embedding

·
7 authors

Submitted by

taesiri

V-Thinker: Interactive Thinking with Images

·
13 authors

Submitted by

kcz358

OpenMMReasoner: Pushing the Frontiers for Multimodal Reasoning with an Open and General Recipe

lmms-lab

Submitted by

Hush-cd

HaluMem: Evaluating Hallucinations in Memory Systems of Agents

MemTensor

Submitted by

didiforhugface

AutoEnv: Automated Environments for Measuring Cross-Environment Agent Learning

·
15 authors

Submitted by

Kushnareva

Unveiling Intrinsic Dimension of Texts: from Academic Abstract to Creative Story

·
8 authors

Submitted by

taesiri

ThinkMorph: Emergent Properties in Multimodal Interleaved Chain-of-Thought Reasoning

·
8 authors

Submitted by

zzqsmall

Every Activation Boosted: Scaling General Reasoner to 1 Trillion Open Language Foundation

inclusionAI

2

Submitted by

taesiri

Scaling Agent Learning via Experience Synthesis

metaresearch

Submitted by

jiannanx

PAN: A World Model for General, Interactable, and Long-Horizon World Simulation

·
34 authors

Submitted by

GuoxinChen

IterResearch: Rethinking Long-Horizon Agents via Markovian State Reconstruction

·
16 authors

Submitted by

ChenMnZ

INT v.s. FP: A Comprehensive Study of Fine-Grained Low-bit Quantization Formats

ByteDance-Seed

Submitted by

univanxx

Multimodal Evaluation of Russian-language Architectures

·
18 authors

Submitted by

taesiri

Reasoning via Video: The First Evaluation of Video Models' Reasoning Abilities through Maze-Solving Tasks

·
11 authors

Submitted by

hammh0a

AraLingBench A Human-Annotated Benchmark for Evaluating Arabic Linguistic Capabilities of Large Language Models

IVUL-KAUST

Image and Video Understanding Lab

Submitted by

cccczshao

Continuous Autoregressive Language Models

tencent

Submitted by

yejunliang23

Part-X-MLLM: Part-aware 3D Multimodal Large Language Model

Tencent-Hunyuan

Tencent Hunyuan

Submitted by

QiushiSun

OS-Sentinel: Towards Safety-Enhanced Mobile GUI Agents via Hybrid Validation in Realistic Workflows

hkunlp

NLP Group of The University of Hong Kong

Submitted by

Lingaaaaaaa

MMaDA-Parallel: Multimodal Large Diffusion Language Models for Thinking-Aware Editing and Generation

ByteDance

Submitted by

nielsr

Back to Basics: Let Denoising Generative Models Denoise

MIT

Massachusetts Institute of Technology

Submitted by

zehongma

DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation

PekingUniversity

Peking University

Submitted by

zoeyuchao

π_RL: Online RL Fine-tuning for Flow-based Vision-Language-Action Models

RLinf

4

Submitted by

hamishivi

DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research

rl-research

Submitted by

KaiiWuu1993

A Style is Worth One Code: Unlocking Code-to-Style Image Generation with Discrete Style Space

Kwai-Kolors

Kolors Team, Kuaishou Technology

Submitted by

YiyangAiLab

When Visualizing is the First Step to Reasoning: MIRA, a Benchmark for Visual Chain-of-Thought

ByteDance-Seed

Submitted by

ambud26

What Does It Take to Be a Good AI Research Agent? Studying the Role of Ideation Diversity

3

Submitted by

prayerdan

GroupRank: A Groupwise Reranking Paradigm Driven by Reinforcement Learning

AQ-MedAI

Submitted by

giantPanda0906

Step-Audio-R1 Technical Report

stepfun-ai

Submitted by

SoloWayG

MADD: Multi-Agent Drug Discovery Orchestra

ITMO-NSS

Submitted by

Zihao1

Too Good to be Bad: On the Failure of LLMs to Role-Play Villains

tencent

Submitted by

taesiri

V-ReasonBench: Toward Unified Reasoning Benchmark Suite for Video Generation Models

·
10 authors

Submitted by

taesiri

Time-to-Move: Training-Free Motion Controlled Video Generation via Dual-Clock Denoising

Technion

Technion Israel institute of technology

Submitted by

SivanSX

MedSAM3: Delving into Segment Anything with Medical Concepts

Submitted by

zli12321

First Frame Is the Place to Go for Video Content Customization

UMCP

University of Maryland College Park

Submitted by

KevinQHLin

Computer-Use Agents as Judges for Generative User Interface

showlab

Submitted by

wangsssssss

UniAVGen: Unified Audio and Video Generation with Asymmetric Cross-Modal Interactions

NJU

Nanjing University

Submitted by

taesiri

PhysX-Anything: Simulation-Ready Physical 3D Assets from Single Image

·
5 authors

Submitted by

xxzcc

DRIVE: Data Curation Best Practices for Reinforcement Learning with Verifiable Reward in Competitive Code Generation

tencent

Submitted by

taesiri

Visual Spatial Tuning

ByteDance-Seed

Submitted by

ytz20

Black-Box On-Policy Distillation of Large Language Models

MicrosoftResearch

Microsoft Research

Submitted by

alexiglad

EBT-Policy: Energy Unlocks Emergent Physical Reasoning Capabilities

·
8 authors

Submitted by

taesiri

Inferix: A Block-Diffusion based Next-Generation Inference Engine for World Simulation

Alibaba-DAMO-Academy

Submitted by

JiaaqiLiu

Agent0-VL: Exploring Self-Evolving Agent for Tool-Integrated Vision-Language Reasoning

UNC-ChapelHill

University of North Carolina at Chapel Hill

2

Submitted by

taesiri

LEGO-Eval: Towards Fine-Grained Evaluation on Synthesizing 3D Embodied Environments with Tool Augmentation

·
6 authors

Submitted by

caizhongang

Scaling Spatial Intelligence with Multimodal Foundation Models

sensenova

Submitted by

zgzaacm

Video Generation Models Are Good Latent Reward Models

Tencent-Hunyuan

Tencent Hunyuan

Submitted by

weijiawu

WEAVE: Unleashing and Benchmarking the In-context Interleaved Comprehension and Generation

·
13 authors

Submitted by

taesiri

DeepEyesV2: Toward Agentic Multimodal Model

rednote-hilab

Submitted by

taesiri

World Simulation with Video Foundation Models for Physical AI

·
89 authors

Submitted by

hankaixyz

ROOT: Robust Orthogonalized Optimizer for Neural Network Training

huawei-noah

HUAWEI Noah's Ark Lab

Submitted by

taesiri

VisPlay: Self-Evolving Vision-Language Models from Images

UIUC-CS

University of Illinois at Urbana-Champaign

Submitted by

zhangzixin02

TiViBench: Benchmarking Think-in-Video Reasoning for Video Generative Models

·
13 authors

Submitted by

ShuaiBai623

Soft Adaptive Policy Optimization

Qwen

Submitted by

jiamingZ

SteadyDancer: Harmonized and Coherent Human Image Animation with First-Frame Preservation

MCG-NJU

Multimedia Computing Group-Nanjing University

Submitted by

starsuzi

Adaptive Multi-Agent Response Refinement in Conversational Systems

amazon

Submitted by

FairyFali

Generalizing Test-time Compute-optimal Scaling as an Optimizable Graph

Pennsylvania State University

Submitted by

Diegooor

Large Language Models Meet Extreme Multi-label Classification: Scaling and Multi-modal Framework

nielseniq

Submitted by

eliebak

Motif 2 12.7B technical report

Motif-Technologies

Motif Technologies

Submitted by

dongdongunique

Evolve the Method, Not the Prompts: Evolutionary Synthesis of Jailbreak Attacks on LLMs

·
8 authors

Submitted by

taesiri

Virtual Width Networks

ByteDance-Seed

Submitted by

scofield7419

UniVA: Universal Video Agent towards Open-Source Next-Generation Video Generalist

UniVA-Agent

Submitted by

taesiri

Cambrian-S: Towards Spatial Supersensing in Video

·
15 authors

Submitted by

CodeGoat24

UniREditBench: A Unified Reasoning-based Image Editing Benchmark

Fudan-University

Fudan University

Submitted by

Owen777

UltraFlux: Data-Model Co-Design for High-quality Native 4K Text-to-Image Generation across Diverse Aspect Ratios

Submitted by

JacobYuan

UniLumos: Fast and Unified Image and Video Relighting with Physics-Plausible Feedback

Alibaba-DAMO-Academy

Submitted by

taesiri

Can World Simulators Reason? Gen-ViRe: A Generative Visual Reasoning Benchmark

·
5 authors

Submitted by

taesiri

The Station: An Open-World Environment for AI-Driven Discovery

·
2 authors

Submitted by

shkim0116

KLASS: KL-Guided Fast Inference in Masked Diffusion Models

kaist-ai

Submitted by

ydalva

Canvas-to-Image: Compositional Image Generation with Multimodal Controls

·
8 authors

6

Submitted by

AnnieFeng

VeriCoT: Neuro-symbolic Chain-of-Thought Validation via Logical Consistency Checks

AWS

Amazon Web Services

Submitted by

hanqing666

The Path Not Taken: RLVR Provably Learns Off the Principals

Submitted by

luoyingfeng

Beyond English: Toward Inclusive and Scalable Multilingual Machine Translation with LLMs

NiuTrans

Submitted by

JustinTX

Budget-Aware Tool-Use Enables Effective Agent Scaling

google

Submitted by

wenyi

UI2Code^N: A Visual Language Model for Test-Time Scalable Interactive UI-to-Code Generation

·
8 authors

Submitted by

Wayne-King

Does Understanding Inform Generation in Unified Multimodal Models? From Analysis to Path Forward

PekingUniversity

Peking University

Submitted by

thomagram

STARFlow-V: End-to-End Video Generative Modeling with Normalizing Flow

apple

Submitted by

taesiri

GGBench: A Geometric Generative Reasoning Benchmark for Unified Multimodal Models

·
10 authors

Submitted by

AviadDahan

LiteAttention: A Temporal Sparse Attention for Diffusion Transformers

MoonMath-ai

Submitted by

taesiri

iMontage: Unified, Versatile, Highly Dynamic Many-to-many Image Generation

stepfun-ai

Submitted by

horseee

In-Video Instructions: Visual Signals as Generative Control

NationalUniversityofSingapore

National University of Singapore

Submitted by

Hennara

Wasm: A Pipeline for Constructing Structured Arabic Interleaved Multimodal Corpora

·
7 authors

2