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

Nov 13

Submitted by

LIQIIIII

SO-Arena: Benchmarking the Capability Spectrum and Reliability of System One Models

NationalUniversityofSingapore

National University of Singapore

Submitted by

jyseo

BirdDreams: Joint Perspective and BEV Video Generation for a Plausible, Reactive Driving World Model

nvidia

1

Submitted by

henry-y1

G-STAR: Group-Scaled Token Capacity for Deeply Compressed Autoregressive Generation

meituan-longcat

Submitted by

jaychempan

Duo: A Reciprocal World Model for Full-Duplex Interaction

inclusionAI

Submitted by

ChenYxyX

PhysLoop: Harnessing LLMs' Physics Awareness for Self-Improving Video Generation Models

DAGroup-PKU

Submitted by

aHapBean

On the Design of Hyper-Connections: Simplification, Stability, and Scalability

dots-studio

Submitted by

wangzx1210

Actor, Judge, or Teacher? Understanding Decision Models in Long-Horizon Agents

Zhejiang University

Submitted by

linjohnss

TTB: Test-time Baking for Efficient Rendering of Decoder-only View Synthesis Models

nvidia

Submitted by

lifuguan

Ropedia Xperience 2: A Gold Standard for Scaling Physical AI Environments and Behaviors

ropedia-ai

Submitted by

yukimasano

Open Vision Banana: A Fully Open Recipe for Generative Visual Perception

FunAILab

Fundamental AI Lab at UTN

Submitted by

Harold328

WorldDraft: If Code Builds the World, Can Video Models Just Render It?

·
10 authors

Submitted by

bhheo

Factored One-sided Adam-Moment for Practical and Scalable SOAP

naver-ai

Submitted by

jaychempan

UniSpark: Exploring Hybrid Attention for Multimodal Understanding and Generation

antgroup

Submitted by

hwbhwbhwb

Learning the Task or Learning the Verifier? Rethinking Criteria Exposure in Embodied Self-Improvement

DAGroup-PKU

Submitted by

wangzx1210

Agent-R^2: Scaling Agents for Realistic Productivity Work via Real-Source Data Synthesis

Zhejiang University

Submitted by

chen-ai-researcher

Evidence-Guided Recovery for AI Agents with Ambiguous Tool Outcomes and Partial Execution

·
4 authors

1

Submitted by

ZefanW

JustRUA: A Controlled Study of GPT-6 Robot-Use Agents

·
12 authors

Submitted by

ismaelR

DeepStress: Stress-Testing Deep Search Agents

Orange

1

Submitted by

pcczc15

Structured Confidence Perturbation via Fisher Geometry for Adversarial Training

Submitted by

XaiverZ

HiVe-Mem: Interleaving Horizontal Memory Expansion and Vertical Evidence Composition for Multimodal Agents

NanyangTechnologicalUniversity

Nanyang Technological University

Submitted by

chen-ai-researcher

Do Multimodal Models Decide Consistently on Equivalent Optical Spectra?

·
3 authors

1

Submitted by

chen-ai-researcher

Humanized but Still True? Auditing Facts and Detection in AI-Text Humanizers

·
3 authors

1

Submitted by

Minh4422

From Large Language Models to Efficient System One Models via Adaptive Computation

·
1 authors

Submitted by

JingjingJiang

A Cross-Scenario Survey and Analysis of Self-Evolving Multimodal Agents

·
1 authors