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

October

Submitted by

janchorowski

The Dragon Hatchling: The Missing Link between the Transformer and Models of the Brain

pathwaycom

Submitted by

AlexiaJM

Less is More: Recursive Reasoning with Tiny Networks

SamsungSAILMontreal

Samsung AI Lab (SAIL) Montreal

Submitted by

taesiri

Agent Learning via Early Experience

metaresearch

Submitted by

taesiri

Scaling Latent Reasoning via Looped Language Models

ByteDance-Seed

9

Submitted by

AaronHuangWei

QeRL: Beyond Efficiency -- Quantization-enhanced Reinforcement Learning for LLMs

nvidia

Submitted by

Gofinge

Concerto: Joint 2D-3D Self-Supervised Learning Emerges Spatial Representations

Pointcept

Submitted by

Jakumetsu

MCPMark: A Benchmark for Stress-Testing Realistic and Comprehensive MCP Use

NationalUniversityofSingapore

National University of Singapore

Submitted by

tsbpp

Diffusion Transformers with Representation Autoencoders

nyu-visionx

Submitted by

WNJXYK

A Theoretical Study on Bridging Internal Probability and Self-Consistency for LLM Reasoning

LAMDA-NeSy

Submitted by

Wenxuan123

Spatial Forcing: Implicit Spatial Representation Alignment for Vision-language-action Model

Submitted by

lastdefiance20

D2E: Scaling Vision-Action Pretraining on Desktop Data for Transfer to Embodied AI

·
10 authors

Submitted by

fangwu97

DeepSearch: Overcome the Bottleneck of Reinforcement Learning with Verifiable Rewards via Monte Carlo Tree Search

stanfordnlp

Submitted by

taesiri

Vision-Zero: Scalable VLM Self-Improvement via Strategic Gamified Self-Play

·
9 authors

Submitted by

fracapuano

Robot Learning: A Tutorial

lerobot

Submitted by

taesiri

Kimi Linear: An Expressive, Efficient Attention Architecture

moonshotai

Submitted by

taesiri

Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models

·
13 authors

Submitted by

taesiri

PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Model

PaddlePaddle

8

Submitted by

KangLiao

Thinking with Camera: A Unified Multimodal Model for Camera-Centric Understanding and Generation

mmlab-ntu

Submitted by

amant555

Apriel-1.5-15b-Thinker

ServiceNow-AI

5

Submitted by

MoshiQAQ

ReCode: Unify Plan and Action for Universal Granularity Control

·
13 authors

Submitted by

GindaChen

Efficient Long-context Language Model Training by Core Attention Disaggregation

·
9 authors

5

Submitted by

GMFTBY

The End of Manual Decoding: Towards Truly End-to-End Language Models

tencent

Submitted by

KevinQHLin

Paper2Video: Automatic Video Generation from Scientific Papers

showlab

Submitted by

xinlongwang

Emu3.5: Native Multimodal Models are World Learners

BAAI

Beijing Academy of Artificial Intelligence

Submitted by

taesiri

Every Attention Matters: An Efficient Hybrid Architecture for Long-Context Reasoning

antgroup

Submitted by

Vasily

When Models Lie, We Learn: Multilingual Span-Level Hallucination Detection with PsiloQA

·
9 authors

Submitted by

Ningyu

LightMem: Lightweight and Efficient Memory-Augmented Generation

Zhejiang University

Submitted by

xiaochonglinghu

Advancing End-to-End Pixel Space Generative Modeling via Self-supervised Pre-training

GD-ML

Submitted by

zhangshaolei

DeepAnalyze: Agentic Large Language Models for Autonomous Data Science

RUC-DataLab

Submitted by

ZhuofengLi

In-the-Flow Agentic System Optimization for Effective Planning and Tool Use

Stanford

Submitted by

PhoenixZ

MM-HELIX: Boosting Multimodal Long-Chain Reflective Reasoning with Holistic Platform and Adaptive Hybrid Policy Optimization

·
14 authors

Submitted by

dongguanting

Agentic Entropy-Balanced Policy Optimization

RUC

Renmin University of China

Submitted by

gowitheflow

Scaling Language-Centric Omnimodal Representation Learning

LCO-Embedding

Submitted by

YerbaPage

LongCodeZip: Compress Long Context for Code Language Models

Stanford-University

Stanford University

Submitted by

taesiri

Tongyi DeepResearch Technical Report

AlibabaTongyiLab

4

Submitted by

lixiaoxi45

DeepAgent: A General Reasoning Agent with Scalable Toolsets

·
11 authors

Submitted by

didiforhugface

InteractComp: Evaluating Search Agents With Ambiguous Queries

·
25 authors

Submitted by

QiushiSun

JanusCoder: Towards a Foundational Visual-Programmatic Interface for Code Intelligence

internlm

Intern Large Models

Submitted by

fuvty

Cache-to-Cache: Direct Semantic Communication Between Large Language Models

nics-efc

Tsinghua-NICS-EFC

Submitted by

cuijiaxing

Self-Forcing++: Towards Minute-Scale High-Quality Video Generation

ByteDance-Seed

Submitted by

Everything-is-Ok

DITING: A Multi-Agent Evaluation Framework for Benchmarking Web Novel Translation

NextGenWhu

Submitted by

eliebak

DeepSeek-OCR: Contexts Optical Compression

deepseek-ai

Submitted by

taesiri

OmniVinci: Enhancing Architecture and Data for Omni-Modal Understanding LLM

nvidia

Submitted by

taesiri

GEM: A Gym for Agentic LLMs

sail

Submitted by

lixiaoxi45

Video-Thinker: Sparking "Thinking with Videos" via Reinforcement Learning

·
9 authors

Submitted by

mjkmain

KORMo: Korean Open Reasoning Model for Everyone

KORMo-Team

Submitted by

taesiri

WithAnyone: Towards Controllable and ID Consistent Image Generation

stepfun-ai

Submitted by

WooooDyy

BAPO: Stabilizing Off-Policy Reinforcement Learning for LLMs via Balanced Policy Optimization with Adaptive Clipping

nex-agi

Submitted by

Rbin

RAG-Anything: All-in-One RAG Framework

hkuds

Data Intelligence Lab@HKU

Submitted by

rzzhan

ExGRPO: Learning to Reason from Experience

·
8 authors

4

Submitted by

taesiri

FineVision: Open Data Is All You Need

huggingface

Submitted by

taesiri

UniVideo: Unified Understanding, Generation, and Editing for Videos

KlingTeam

Submitted by

xytian1008

More Thought, Less Accuracy? On the Dual Nature of Reasoning in Vision-Language Models

·
8 authors

Submitted by

Ogkunal

Fathom-DeepResearch: Unlocking Long Horizon Information Retrieval and Synthesis for SLMs

FractalAIResearch

Fractal AI Research

Submitted by

pbicho

SINQ: Sinkhorn-Normalized Quantization for Calibration-Free Low-Precision LLM Weights

huawei-csl

HUAWEI Computing Systems Lab

Submitted by

jienengchen

World-in-World: World Models in a Closed-Loop World

·
17 authors

Submitted by

forde450

Ming-UniVision: Joint Image Understanding and Generation with a Unified Continuous Tokenizer

inclusionAI

Submitted by

zichenwen

AI for Service: Proactive Assistance with AI Glasses

SJTU

Shanghai Jiao Tong University

2

Submitted by

Carlanlarkk

Low-probability Tokens Sustain Exploration in Reinforcement Learning with Verifiable Reward

tencent

Submitted by

imlixinyang

FlashWorld: High-quality 3D Scene Generation within Seconds

·
6 authors

Submitted by

binxia

DreamOmni2: Multimodal Instruction-based Editing and Generation

·
13 authors

Submitted by

Blue-Giant

MemMamba: Rethinking Memory Patterns in State Space Model

·
5 authors

Submitted by

callanwu

AgentFold: Long-Horizon Web Agents with Proactive Context Management

AlibabaTongyiLab

Submitted by

taesiri

Every Step Evolves: Scaling Reinforcement Learning for Trillion-Scale Thinking Model

inclusionAI

Submitted by

Mqleet

Human-Agent Collaborative Paper-to-Page Crafting for Under $0.1

Submitted by

tmencatt

Language Models are Injective and Hence Invertible

gladia

Gladia Research Group

Submitted by

Paranioar

From Pixels to Words -- Towards Native Vision-Language Primitives at Scale

SenseTime

Submitted by

CCCCCC

Glyph: Scaling Context Windows via Visual-Text Compression

·
14 authors

Submitted by

taesiri

UniGenBench++: A Unified Semantic Evaluation Benchmark for Text-to-Image Generation

·
11 authors

Submitted by

taesiri

VLA-RFT: Vision-Language-Action Reinforcement Fine-tuning with Verified Rewards in World Simulators

·
11 authors

Submitted by

Jessamine

Winning the Pruning Gamble: A Unified Approach to Joint Sample and Token Pruning for Efficient Supervised Fine-Tuning

alibabagroup

Submitted by

derrickzhu

A Survey of Data Agents: Emerging Paradigm or Overstated Hype?

·
25 authors

Submitted by

jiaruz2

TaTToo: Tool-Grounded Thinking PRM for Test-Time Scaling in Tabular Reasoning

amazon

3

Submitted by

Andrew613

PICABench: How Far Are We from Physically Realistic Image Editing?

·
13 authors

Submitted by

yejunliang23

NANO3D: A Training-Free Approach for Efficient 3D Editing Without Masks

·
8 authors

Submitted by

foggyforest

UniMoE-Audio: Unified Speech and Music Generation with Dynamic-Capacity MoE

HIT-TMG

Submitted by

taesiri

VideoCanvas: Unified Video Completion from Arbitrary Spatiotemporal Patches via In-Context Conditioning

KlingTeam

Submitted by

sayakpaul

The Principles of Diffusion Models

·
5 authors

Submitted by

lynazhang

LoongRL:Reinforcement Learning for Advanced Reasoning over Long Contexts

MicrosoftResearch

Microsoft Research

5

Submitted by

buaahsh

BitNet Distillation

MicrosoftResearch

Microsoft Research

Submitted by

sinwang

RoboOmni: Proactive Robot Manipulation in Omni-modal Context

OpenMOSS-Team

Submitted by

yuezhouhu

AdaSPEC: Selective Knowledge Distillation for Efficient Speculative Decoders

GeorgiaTech

Georgia Institute of Technology

Submitted by

yulunliu

StealthAttack: Robust 3D Gaussian Splatting Poisoning via Density-Guided Illusions

NYCU

National Yang Ming Chiao Tung University

Submitted by

menghao22

Bee: A High-Quality Corpus and Full-Stack Suite to Unlock Advanced Fully Open MLLMs

Open-Bee

Submitted by

wujie10

FARMER: Flow AutoRegressive Transformer over Pixels

·
9 authors

1

Submitted by

yangcole

Attention Illuminates LLM Reasoning: The Preplan-and-Anchor Rhythm Enables Fine-Grained Policy Optimization

alibabagroup

2

Submitted by

WuChengyue

Fast-dLLM v2: Efficient Block-Diffusion LLM

nvidia

Submitted by

taesiri

Game-TARS: Pretrained Foundation Models for Scalable Generalist Multimodal Game Agents

ByteDance

Submitted by

xiaochonglinghu

ImagerySearch: Adaptive Test-Time Search for Video Generation Beyond Semantic Dependency Constraints

GD-ML

Submitted by

taesiri

StockBench: Can LLM Agents Trade Stocks Profitably In Real-world Markets?

·
7 authors

Submitted by

yjyjyj98

Meta-Awareness Enhances Reasoning Models: Self-Alignment Reinforcement Learning

kaist-ai

Submitted by

taesiri

Open-o3 Video: Grounded Video Reasoning with Explicit Spatio-Temporal Evidence

ByteDance

Submitted by

weizhepei

TruthRL: Incentivizing Truthful LLMs via Reinforcement Learning

Submitted by

taesiri

Lumina-DiMOO: An Omni Diffusion Large Language Model for Multi-Modal Generation and Understanding

Alpha-VLLM

Submitted by

Jeff-Wang

GigaBrain-0: A World Model-Powered Vision-Language-Action Model

open-gigaai

5

Submitted by

weidawang

Chem-R: Learning to Reason as a Chemist

Submitted by

taesiri

Detect Anything via Next Point Prediction

IDEA-Research

Submitted by

Monta3Pt

Latent Refinement Decoding: Enhancing Diffusion-Based Language Models by Refining Belief States

KCL

King's College London

2

Submitted by

taesiri

StreamingVLM: Real-Time Understanding for Infinite Video Streams

·
7 authors

Submitted by

LightChen2333

AutoPR: Let's Automate Your Academic Promotion!

·
13 authors

Submitted by

QingyanBai

Scaling Instruction-Based Video Editing with a High-Quality Synthetic Dataset

·
13 authors

Submitted by

taesiri

Video-LMM Post-Training: A Deep Dive into Video Reasoning with Large Multimodal Models

·
27 authors

Submitted by

BianYx

Video-As-Prompt: Unified Semantic Control for Video Generation

ByteDance

Submitted by

MingleiShi

Latent Diffusion Model without Variational Autoencoder

KlingTeam

Submitted by

GuoxinChen

ReForm: Reflective Autoformalization with Prospective Bounded Sequence Optimization

·
9 authors