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

Jun 18

Submitted by

ChrisDing1105

Beyond the Current Observation: Evaluating Multimodal Large Language Models in Controllable Non-Markov Games

internlm

Intern Large Models

Submitted by

shanyou92

Kairos: A Native World Model Stack for Physical AI

·
24 authors

Submitted by

AIcell

Guava: An Effective and Universal Harness for Embodied Manipulation

·
8 authors

4

Submitted by

FuriosaMJLEE

EfficientRollout: System-Aware Self-Speculative Decoding for RL Rollouts

furiosa-ai

Submitted by

jnzhang

MolmoMotion: Forecasting Point Trajectories in 3D with Language Instruction

allenai

Submitted by

adamdad

SAE Interventions are Unreliable: Post-Intervention Recovery of Suppressed Behavior

PolyUHK

The Hong Kong Polytechnic University

Submitted by

jiyatai

Reinforcing Dual-Path Reasoning in Spatial Vision Language Models

HKUCDS

University of Hong Kong

Submitted by

JingyuanHuang

Trust the Right Teacher: Quality-Aware Self-Distillation for GUI Grounding

UGA-AI

University of Georgia

2

Submitted by

FYYDCC

From Trainee to Trainer: LLM-Designed Training Environment for RL with Multi-Agent Reasoning

LARK-Lab

LARK Lab@HKUST (GZ)

Submitted by

harryhsing

Native Active Perception as Reasoning for Omni-Modal Understanding

Qwen

Submitted by

haipeng1

STARE: Surprisal-Guided Token-Level Advantage Reweighting for Policy Entropy Stability

Tencent-Hunyuan

Tencent Hunyuan

Submitted by

indulgeBai

MaineCoon: Pursuing A Real-Time Audio-Visual Social World Model

catnip-ai-tech

Submitted by

taesiri

Sumi: Open Uniform Diffusion Language Model from Scratch

·
6 authors

Submitted by

beiweixiaoxu

Beyond Alignment: Value Diversity as a Collective Property in Multicultural Agent Systems

iNLP-Lab

iNLP Lab @ SUTD

Submitted by

velezbeltran

The Reward Was in Your Data All Along: Correcting Flow Matching with Discriminator-Guided RL

Submitted by

taesiri

PAIWorld: A 3D-Consistent World Foundation Model for Robotic Manipulation

·
28 authors

Submitted by

taesiri

CEO-Bench: Can Agents Play the Long Game?

·
3 authors

Submitted by

xxyQwQ

SciOrch: Learning to Orchestrate Expert LLMs for Solving Frontier Multimodal Scientific Reasoning Tasks

·
9 authors

Submitted by

Krispin

ViT-Up: Faithful Feature Upsampling for Vision Transformers

SJTU

Shanghai Jiao Tong University

Submitted by

RuishanFang

RODS: Reward-Driven Online Data Synthesis for Multi-Turn Tool-Use Agents

inclusionAI

Submitted by

cedzhang

Learning User Simulators with Turing Rewards

MIT

Massachusetts Institute of Technology

Submitted by

taesiri

Externalizing Research Synthesis and Validation in AI Scientists through a Research Harness

·
18 authors

Submitted by

ljang0

MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents

CarnegieMellonU

Carnegie Mellon University

Submitted by

alphadl

IndustryBench-MIPU: Benchmarking Multi-Image Attribute Value Extraction for Industrial Products

alibaba-multimodal-industrial-ai

1688 multimodal & industrial AI

Submitted by

ljang0

iOSWorld: A Benchmark for Personally Intelligent Phone Agents

CarnegieMellonU

Carnegie Mellon University

Submitted by

varun86861993

Bag of Dims: Training-Free Mechanistic Interpretability via Dimension-Level Sign Patterns

·
1 authors

1

Submitted by

taesiri

Physics-IQ Verified

·
7 authors

Submitted by

OedoSoldier

Seeing Before Reasoning: Decoupling Perception and Reasoning for Shortcut-Resilient Multimodal On-Policy Self-Distillation

·
4 authors

Submitted by

dfavenfre-dev

Morpheus: A Morphology-Aware Neural Tokenizer and Word Embedder for Turkish

lonewolflab

Lonewolf Research & Development

Submitted by

hryou1998

HiLo-Token: Input-Adaptive High-Low Frequency Token Compression for Efficient Image Editing

adobe

Submitted by

Tej-a55

A Benchmark and Framework for Evaluating Next Action Predictions in Spreadsheets

microsoft

Submitted by

Henokbfg

LLM-Enabled NWDAF: A Step Toward AI-Native 6G Network Intelligence

Khaledalali

Khalifa university