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

Aug 7

Submitted by

zzzzhw

AgentOPSD: Recursive Self-Distillation for Agentic Reinforcement Learning

Tsinghua University

Submitted by

QiushiSun

OSReward: Instituting Standardized Evaluation for Cross-Platform Computer-Use Reward Models

hkunlp

NLP Group of The University of Hong Kong

Submitted by

taesiri

WorldClaw: Agentic 3D Open-World Generation at Scale

Tencent-Hunyuan

Tencent Hunyuan

Submitted by

isemenkov

Interpretable MEG Decoding of Perceived Speech: Cortical Sources and the Stimulus Features That Drive Retrieval

·
5 authors

Submitted by

taesiri

GST-Bench: Can VLMs Develop Global Spatial Awareness from Video?

ByteDance-Seed

Submitted by

xuzishan

EnvACE: Internalizing Environment Dynamics via World Rehearsal for Agentic Reinforcement Learning

tencent

Submitted by

Kaichengalex

Learning from Failures: Retrieval-Centric CoT via Hard Negatives for Unified Multimodal Retrieval

DeepGlint-AI

Submitted by

SivanSX

ChronoVision: Temporal Reasoning via Latent State Reconstruction

PediaMedAI

1

Submitted by

Benyou

From Economic Agents to Agentic Economies: A Systems Blueprint for Economic World Models

Submitted by

bhheo

On-Policy Delta Distillation for Multilingual Math Reasoning

naver-ai

1

Submitted by

taesiri

HarnessOpt-Bench: Evaluating LLMs at Harness Optimization

ScaleAI

Submitted by

BugMaker-Boyan

DataSpace: Benchmarking Data Agents for Verifiable Analytics over Heterogeneous Workspaces

HKUSTDial

Submitted by

ayoubkirouane

Teaching Nemotron Greek: Mining a Corpus, Adapting Retrieval, and Grounding Generation for Modern Greek across Specialist Domains

KIEFERSA

1

Submitted by

taesiri

DyPES-VLA: Learning Shared Dynamics Priors and Embodiment-Specific Control for Cross-Embodiment Manipulation

·
15 authors

Submitted by

HarrisonPENG

World-to-Wrist: Task-Conditioned Future Wrist Modeling for Fine-Grained Robot Manipulation

·
11 authors

Submitted by

Phoebux

EffectLearner: World-Aware Object-Effect Reasoning for Real-World Video Object Removal

·
11 authors

Submitted by

GuoxinChen

CalibForge: Adversarial Solver Calibration for Scaling Learnable Terminal Tasks

AweAI-Team

Submitted by

Yunqiu

SmartMage: Dynamic Modality Orchestration for 3D Scene Understanding

·
5 authors

Submitted by

Uri-ka

MameLoshnLM: Yiddish Language Model and Evaluation Benchmark

Submitted by

Longin-Yu

PaDoc: Layout-Grounded Parallel Decoding for Document Parsing

tsinghua-uni

Tsinghua University

Submitted by

rtyasdf

KVAE: Family of Tokenizers for Multimodal Generative Models

kandinskylab

Submitted by

Nos7

Activity Frames: Deterministic Screen-Activity Compilation for Agent Memory and Replay

·
1 authors

Submitted by

XuGuo699

ContextMaster: Interactive Multi-Shot Video Creation via Fixed-Budget Sparse Context Routing

KlingTeam

1

Submitted by

stojnvla

Invisible Shortcuts: Why Vision Encoders Know Your Camera

·
5 authors

2

Submitted by

GhostCai

MASS: Multiplayer World Models with Authoritative Shared State

AlayaLab

Submitted by

zd21

Continual Learning in Transition

CASIA

Chinese Academic of Science Institute of Automation

Submitted by

Tirth021

Task-Conditional Flow Matching for Balanced Multilingual Text Embedding Adaptation

LingoIITGN

Lingo Research Group

1