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 2

Submitted by

Thu-zqc

R^2Orch: Recursive Self-Improvement for Agent Orchestration via Self-Play within a Skill-Internalization Loop

CUHKSZ

Chinese University of Hong Kong, Shenzhen

1

Submitted by

CoreloneH

FrameForge: Harnessing Priors of Video Foundation Models for Unified Image Generation

FrameXAI

Submitted by

yunqu

Does Scaling Toward AGI Depend on What System-1 Lets System-2 Execute?

Tsinghua University

1

Submitted by

amant555

Apriel Voice 27B Preview: An Open Model for Speech, Reasoning, and Tool Use

ServiceNow-AI

1

Submitted by

rooty2020

Latent Reasoning in Continuous Space for Unified Multimodal Models

kaist-ai

Submitted by

onepiece1999

Wizard: Manipulate Any World in Real Time via Agentic Control

tencent

1

Submitted by

Gwzzz

OSWAM: An Effective World Action Model with One-Step Action Generation

PolyUHK

The Hong Kong Polytechnic University

Submitted by

quhongyu123

HiFOR: Hierarchical Foresight with Action-Guided World Grounding for Vision-Language-Action Model

·
7 authors

Submitted by

assafsinger

ReLift: Dense 3D Tracking as Lifted Trajectory Repair

Technion

Technion Israel institute of technology

Submitted by

jonghakim

CEPO: A General Framework for Improving Prompt Optimization in Multi-Agent Systems

·
5 authors

Submitted by

qcf568

Venus-DefakerOne2: A Vision-Forensics Foundation Model with Broader Coverage

inclusionAI

Submitted by

C0notSilly

CaPtain: A Pipeline for Post-Training Code-as-Policy Robot Agents

CUOHK

Chinese University of Hong Kong

Submitted by

yinbq

Probing Spatial State Representations in Visual Foundation Models

mmlab-ntu

1

Submitted by

An998

WeFair: Fidelity-Aware Alignment from Human Feedback for Personalized Query Rewriting in Short-Video Search

tencent

Submitted by

IntJudge

REA: A Routed Embodied Agent Runtime for Asynchronous Brain-Cerebellum Control

·
8 authors

Submitted by

jcenaa

WorldBridge: Bridging Video and Action for Task Generalization in World Action Models

·
12 authors

Submitted by

YuvalH9

GLOW-FM: Efficient Generative Low-Light Image Enhancement with One-Step Latent Flow Matching

Arm

Submitted by

walkeralan123789

When Requirements Interact: Scaling Long-Horizon Coding via Interdependent Specifications

XiaomiMiMo

Submitted by

CedPei

HALO: Human-Grounded Granular Evaluation of Proactive Agents

2077AIDataFoundation

Submitted by

yuli02gwu

On the Origin and Evolution of Training-Inference Mismatch in LLM Reinforcement Learning

George-Washington-University

George Washington University

Submitted by

HU22333

Supervising the Visible State: Separating Representation from View Selection in Streaming Video

LINs-lab

Submitted by

tunatone

No Channels Left Behind: Understanding Diffusion Acceleration through Residual Gates

·
5 authors

1

Submitted by

danielwusg

RoboArchitect: Automating the Design of Robotic Agents with Persistent Memory

·
6 authors

Submitted by

Skywalker0410

MemSteer: Turning User Histories into Supervision for Agent Memory Optimization

MicrosoftResearch

Microsoft Research

1

Submitted by

yujiazheng

CausalWAM: Learning the Causes of Actions in World Action Models

Causal Intelligence

Submitted by

pratyushmaini

Follow the Gradient: Tracing Safety Degradation Back to Training Data

DatologyAI

Submitted by

Lixiaoming

SciGame: Let Agents Play the Scientific Discovery Games in the Real and Counterfactual World

·
4 authors

Submitted by

ReaganWZY

TernScope: Model-Hardware Co-Design for Efficient Ternary Endoscopic Segmentation

Monash University

Submitted by

ArtKanke

rTaste: Do Research Agents Have Taste? Evaluating Early Stage Research Judgment

·
1 authors

Submitted by

GiuliannoV

Think Twice: The Metis-1.6 Technical Report

Lernex

Submitted by

SinjaeKang

Learning Quality-Controllable Robot Policies with Iterative Hindsight Relabeling

kaist-ai

1

Submitted by

arbazsiddiqui

VoxLock: Target-Speaker Extraction for Voice Agents Without an Enrollment Step

·
1 authors

Submitted by

vinthony

SketchDirector: Complex Scene Sketching via Agentic Layout Reasoning and Detail Refinement

GVCLab

GVC Lab at Great Bay University

Submitted by

tugrulbayrak

TurnTurk-GPT: Text-Based Turn-Taking Prediction for Turkish

·
4 authors