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

September

Submitted by

benfielding

Sharing is Caring: Efficient LM Post-Training with Collective RL Experience Sharing

Gensyn

Submitted by

wanng

A.S.E: A Repository-Level Benchmark for Evaluating Security in AI-Generated Code

·
21 authors

Submitted by

huangsiteng

VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Model

·
16 authors

Submitted by

henggg

The Landscape of Agentic Reinforcement Learning for LLMs: A Survey

·
25 authors

Submitted by

yangwang825

Drivel-ology: Challenging LLMs with Interpreting Nonsense with Depth

·
7 authors

Submitted by

taesiri

Why Language Models Hallucinate

·
4 authors

Submitted by

iseesaw

A Survey of Reinforcement Learning for Large Reasoning Models

TsinghuaC3I

Submitted by

Andyson

LongLive: Real-time Interactive Long Video Generation

nvidia

Submitted by

taesiri

MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsing

·
61 authors

Submitted by

taesiri

Qwen3-Omni Technical Report

Qwen

Submitted by

taesiri

Reverse-Engineered Reasoning for Open-Ended Generation

·
12 authors

Submitted by

black-yt

A Survey of Scientific Large Language Models: From Data Foundations to Agent Frontiers

InternScience

Submitted by

ztwang

EPO: Entropy-regularized Policy Optimization for LLM Agents Reinforcement Learning

·
9 authors

Submitted by

Hennara

Baseer: A Vision-Language Model for Arabic Document-to-Markdown OCR

Misraj

Submitted by

TianxiangMa

HuMo: Human-Centric Video Generation via Collaborative Multi-Modal Conditioning

·
10 authors

Submitted by

Luo2003

RPG: A Repository Planning Graph for Unified and Scalable Codebase Generation

·
14 authors

Submitted by

lovesnowbest

UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning

ByteDance-Seed

Submitted by

Nothing2Say

VCRL: Variance-based Curriculum Reinforcement Learning for Large Language Models

·
7 authors

2

Submitted by

jt-zhang

SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse-Linear Attention

Tsinghua University

Submitted by

junkang0909

Quantile Advantage Estimation for Entropy-Safe Reasoning

·
6 authors

Submitted by

daixuancheng

FlowRL: Matching Reward Distributions for LLM Reasoning

·
23 authors

Submitted by

callanwu

Scaling Agents via Continual Pre-training

·
22 authors

Submitted by

taesiri

ScaleCUA: Scaling Open-Source Computer Use Agents with Cross-Platform Data

·
21 authors

Submitted by

YannQi

R-4B: Incentivizing General-Purpose Auto-Thinking Capability in MLLMs via Bi-Mode Annealing and Reinforce Learning

·
6 authors

Submitted by

taesiri

WebWeaver: Structuring Web-Scale Evidence with Dynamic Outlines for Open-Ended Deep Research

·
12 authors

Submitted by

taesiri

OmniWorld: A Multi-Domain and Multi-Modal Dataset for 4D World Modeling

·
19 authors

Submitted by

Hanyuezhuohua

Parallel-R1: Towards Parallel Thinking via Reinforcement Learning

tencent

Submitted by

Sicong

MMR1: Enhancing Multimodal Reasoning with Variance-Aware Sampling and Open Resources

MMR1

Submitted by

YangXiao-nlp

LIMI: Less is More for Agency

·
21 authors

Submitted by

taesiri

SciReasoner: Laying the Scientific Reasoning Ground Across Disciplines

·
32 authors

Submitted by

taesiri

Video models are zero-shot learners and reasoners

deepmind

Submitted by

xiaochonglinghu

From Editor to Dense Geometry Estimator

·
9 authors

Submitted by

callanwu

WebSailor-V2: Bridging the Chasm to Proprietary Agents via Synthetic Data and Scalable Reinforcement Learning

·
17 authors

4

Submitted by

xiaochonglinghu

Tree Search for LLM Agent Reinforcement Learning

GD-ML

Submitted by

hammh0a

Hala Technical Report: Building Arabic-Centric Instruction & Translation Models at Scale

IVUL-KAUST

Image and Video Understanding Lab

Submitted by

wujie10

Seedream 4.0: Toward Next-generation Multimodal Image Generation

·
50 authors

Submitted by

taesiri

LLaVA-Critic-R1: Your Critic Model is Secretly a Strong Policy Model

·
7 authors

Submitted by

taesiri

Visual Representation Alignment for Multimodal Large Language Models

·
13 authors

Submitted by

SivilTaram

SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning

·
7 authors

Submitted by

Junteng

WebExplorer: Explore and Evolve for Training Long-Horizon Web Agents

·
15 authors

Submitted by

Haozhan72

SimpleVLA-RL: Scaling VLA Training via Reinforcement Learning

·
21 authors

Submitted by

DongfuJiang

VerlTool: Towards Holistic Agentic Reinforcement Learning with Tool Use

TIGER-Lab

Submitted by

callanwu

ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization

·
14 authors

5

Submitted by

delinqu

EmbodiedOneVision: Interleaved Vision-Text-Action Pretraining for General Robot Control

·
15 authors

Submitted by

XingtaiHF

Towards a Unified View of Large Language Model Post-Training

·
12 authors

Submitted by

lixiaochuan

Droplet3D: Commonsense Priors from Videos Facilitate 3D Generation

·
14 authors

Submitted by

ZiyiXia

Open Data Synthesis For Deep Research

BAAI

Beijing Academy of Artificial Intelligence

Submitted by

taesiri

Towards General Agentic Intelligence via Environment Scaling

·
16 authors

4

Submitted by

taesiri

RewardDance: Reward Scaling in Visual Generation

·
12 authors

Submitted by

P2333

Language Models Can Learn from Verbal Feedback Without Scalar Rewards

sail

Submitted by

P2333

Variational Reasoning for Language Models

sail

Submitted by

taesiri

WebResearcher: Unleashing unbounded reasoning capability in Long-Horizon Agents

·
16 authors

Submitted by

taesiri

Reinforcement Learning on Pre-Training Data

·
36 authors

Submitted by

QbethQ

StableToken: A Noise-Robust Semantic Speech Tokenizer for Resilient SpeechLLMs

·
7 authors

Submitted by

Crayon-Shinchan

OmniInsert: Mask-Free Video Insertion of Any Reference via Diffusion Transformer Models

·
11 authors

Submitted by

HaoyuDong

MachineLearningLM: Continued Pretraining Language Models on Millions of Synthetic Tabular Prediction Tasks Scales In-Context ML

·
5 authors

Submitted by

hyun1905

ReviewScore: Misinformed Peer Review Detection with Large Language Models

kaist-ai

Submitted by

taesiri

MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training Recipe

·
34 authors

Submitted by

fangwu97

Multiplayer Nash Preference Optimization

stanfordnlp

Submitted by

Yoohao

EchoX: Towards Mitigating Acoustic-Semantic Gap via Echo Training for Speech-to-Speech LLMs

·
7 authors

Submitted by

hammh0a

Reasoning Vectors: Transferring Chain-of-Thought Capabilities via Task Arithmetic

IVUL-KAUST

Image and Video Understanding Lab

1

Submitted by

taesiri

Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Search

·
6 authors

Submitted by

taesiri

3D and 4D World Modeling: A Survey

·
23 authors

Submitted by

taesiri

MANZANO: A Simple and Scalable Unified Multimodal Model with a Hybrid Vision Tokenizer

·
27 authors

Submitted by

taesiri

Inverse IFEval: Can LLMs Unlearn Stubborn Training Conventions to Follow Real Instructions?

ByteDance-Seed

DeepResearch Arena: The First Exam of LLMs' Research Abilities via Seminar-Grounded Tasks

·
13 authors

Submitted by

Lingaaaaaaa

Revolutionizing Reinforcement Learning Framework for Diffusion Large Language Models

·
6 authors

Submitted by

taesiri

AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning

·
23 authors

Submitted by

DogNeverSleep

OpenGPT-4o-Image: A Comprehensive Dataset for Advanced Image Generation and Editing

·
12 authors

Submitted by

Yuyang-z

SANA-Video: Efficient Video Generation with Block Linear Diffusion Transformer

nvidia

Submitted by

bltnynk

No Prompt Left Behind: Exploiting Zero-Variance Prompts in LLM Reinforcement Learning via Entropy-Guided Advantage Shaping

kaist-ai

3

Submitted by

itaigat

Set Block Decoding is a Language Model Inference Accelerator

·
9 authors

Submitted by

yaful

Reasoning over Boundaries: Enhancing Specification Alignment via Test-time Delibration

·
7 authors

Submitted by

Wiley085

Robix: A Unified Model for Robot Interaction, Reasoning and Planning

ByteDance-Seed

6

Submitted by

YuanLiuuuuuu

POINTS-Reader: Distillation-Free Adaptation of Vision-Language Models for Document Conversion

tencent

Submitted by

xiangan

LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal Training

·
22 authors

Submitted by

taesiri

EmbeddingGemma: Powerful and Lightweight Text Representations

·
88 authors

3

Submitted by

mhjiang0408

Do You Need Proprioceptive States in Visuomotor Policies?

·
13 authors

Submitted by

xhyandwyy

UI-S1: Advancing GUI Automation via Semi-online Reinforcement Learning

·
11 authors

3

Submitted by

taesiri

Kling-Avatar: Grounding Multimodal Instructions for Cascaded Long-Duration Avatar Animation Synthesis

·
14 authors

Submitted by

yuecao0119

Sequential Diffusion Language Models

OpenGVLab

Submitted by

MasterVito

Beyond the Exploration-Exploitation Trade-off: A Hidden State Approach for LLM Reasoning in RLVR

Tsinghua University

Submitted by

Jarvis1111

Harnessing Uncertainty: Entropy-Modulated Policy Gradients for Long-Horizon LLM Agents

·
10 authors

4

Submitted by

YangyiH

Symbolic Graphics Programming with Large Language Models

·
7 authors

Submitted by

DogNeverSleep

RealUnify: Do Unified Models Truly Benefit from Unification? A Comprehensive Benchmark

·
26 authors

Submitted by

fjxmlzn

Latent Zoning Network: A Unified Principle for Generative Modeling, Representation Learning, and Classification

·
6 authors

Submitted by

taesiri

SAIL-VL2 Technical Report

·
14 authors

Submitted by

taesiri

FLUX-Reason-6M & PRISM-Bench: A Million-Scale Text-to-Image Reasoning Dataset and Comprehensive Benchmark

·
10 authors

Submitted by

taesiri

Paper2Agent: Reimagining Research Papers As Interactive and Reliable AI Agents

·
4 authors

Submitted by

Wiselnn

SIM-CoT: Supervised Implicit Chain-of-Thought

internlm

Intern Large Models

Submitted by

fairyang

Baichuan-M2: Scaling Medical Capability with Large Verifier System

·
34 authors

2

Submitted by

rishiraj

Gated Associative Memory: A Parallel O(N) Architecture for Efficient Sequence Modeling

·
1 authors

Submitted by

taesiri

Hunyuan3D-Omni: A Unified Framework for Controllable Generation of 3D Assets

·
19 authors

Submitted by

taesiri

Hunyuan3D Studio: End-to-End AI Pipeline for Game-Ready 3D Asset Generation

·
100 authors

Submitted by

taesiri

Kwai Keye-VL 1.5 Technical Report

·
60 authors

Submitted by

shgao

Democratizing AI scientists using ToolUniverse

Harvard

Harvard University

Submitted by

sanaka87

Reconstruction Alignment Improves Unified Multimodal Models

·
4 authors

Submitted by

che111

Does DINOv3 Set a New Medical Vision Standard?

·
19 authors

Submitted by

Nicolas-BZRD

When Does Reasoning Matter? A Controlled Study of Reasoning's Contribution to Model Performance

Scale-or-Reason

Scale or Reason ?

3

Submitted by

craigwu

Visual Jigsaw Post-Training Improves MLLMs

·
6 authors

Submitted by

Wiselnn

CapRL: Stimulating Dense Image Caption Capabilities via Reinforcement Learning

internlm

Intern Large Models

Submitted by

taesiri

AToken: A Unified Tokenizer for Vision

·
8 authors

Submitted by

taesiri

ARE: Scaling Up Agent Environments and Evaluations

·
24 authors

Submitted by

zhongwenxu

Single-stream Policy Optimization

tencent

Submitted by

Cloudriver

LongEmotion: Measuring Emotional Intelligence of Large Language Models in Long-Context Interaction

longemotion1