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

Jul 7

Submitted by

chengtan9907

OmniOpt: Taxonomy, Geometry, and Benchmarking of Modern Optimizers

opendatalab

Submitted by

BroAlanTaps

UI-MOPD: Multi-Platform On-Policy Distillation for Continual GUI Agent Learning

·
11 authors

Submitted by

taesiri

PixWorld: Unifying 3D Scene Generation and Reconstruction in Pixel Space

NanyangTechnologicalUniversity

Nanyang Technological University

Submitted by

yangyu90

ResearchStudio-Reel: Automate the Last Mile of Research from Paper to Poster, Video, and Blog

microsoft

Submitted by

fistyyyy

ResearchStudio-Idea: An Evidence-Grounded Research-Ideation Skill Suite from ML Conference Outcomes

microsoft

Submitted by

Jeff-Wang

GigaWorld-1: A Roadmap to Build World Models for Robot Policy Evaluation

open-gigaai

Submitted by

cherubicxn

Vision Pretraining for Dense Spatial Perception

robbyant

Submitted by

MatanAvitan

MANCE: Manifold Aware Concept Erasure

barilan

Bar-Ilan University

Submitted by

LjHuang

EVA-Client: A Unified Data Collection, Inference, and Deployment Framework for Embodied Policies on Real Robots

·
11 authors

Submitted by

lhhuang

Wan-Streamer v0.2: Higher Resolution, Same Latency

Wan-AI

Submitted by

taesiri

InternVLA-A1.5: Unifying Understanding, Latent Foresight, and Action for Compositional Generalization

·
29 authors

1

Submitted by

JunhaJung

Do All Visual Tokens Matter Equally? Object-Evidence Preserving Token Merging for Vision-Language Retrieval

·
4 authors

Submitted by

sirluk

KVpop -- Key-Value Cache Compression with Predictive Online Pruning

NX-AI

Submitted by

LIQIIIII

dOPSD: On-Policy Self-Distillation for Diffusion Language Models

·
3 authors

Submitted by

taesiri

Multiplayer Interactive World Models with Representation Autoencoders

·
27 authors

Submitted by

simingfu

Perceptual Flow Matching for Few-Step Generative Modeling

jingdong1

Submitted by

mtilyjason

Multi-Turn Agentic Scientific Literature Search via Workflow Induction

illinois

University of Illinois at Urbana-Champaign

Submitted by

taesiri

EdgeBench: Unveiling Scaling Laws of Learning from Real-World Environments

ByteDance-Seed

Submitted by

taesiri

LLM-as-a-Verifier: A General-Purpose Verification Framework

·
9 authors

Submitted by

galfiebelman

MV-Forcing: Long Multi-View Video Generation via 4D-Grounded Spatio-Temporal Self-Forcing

·
3 authors

1

Submitted by

nielsr

Unified Audio Intelligence Without Regressing on Text Intelligence

nvidia

Submitted by

lzq2021

PraMem: Practice-derived Experiential Memory for Long-horizon Behavior Prediction

·
14 authors

Submitted by

Elfsong

Mastermind: Strategy-grounded Learning for Repository-Scale Vulnerability Reproduction

·
7 authors

1

Submitted by

AmirhosseinAbaskohi

SeKV: Resolution-Adaptive KV Cache with Hierarchical Semantic Memory for Long-Context LLM Inference

microsoft

Submitted by

Yunhao-Feng

Safety Testing LLM Agents at Scale: From Risk Discovery to Evidence-Grounded Verification

antgroup

Submitted by

alessiotoniolo

GORGO: Online Tuning for Cross-Region Network-Aware LLM Serving

·
3 authors

Submitted by

gulluk

Transition-Aware best-of-N sampling for Longitudinal Chest X-ray Reports

gevaertlab

Submitted by

huzican

Bridging Interleaved Multi-Modal Reasoning as a Unified Decision Process

tencent

Submitted by

anishapat

Speaker-Aware Temporal Aggregation Strategies on Segment Representations for Depression Detection in Dyadic Interaction: A Benchmark Study

·
4 authors

1

Submitted by

mxvp

CONFLUX: A Latent Diusion Model for 3D Chest-CT Synthesis with RL Post-Training

·
4 authors

3

Submitted by

paulengstler

SynCity 3000: Bootstrapping Scene-Scale 3D Diffusion

Oxford

University of Oxford

Submitted by

lihongyu0807

Deform360: A Massive Multi-view Visuotactile Dataset for Deformable World Models

brownu

Brown University

Submitted by

huzican

Look Before You Leap: Distilling Tree Search into Action Evaluation for Frozen VLA Models

NJU

Nanjing University

Submitted by

matteospanio

Taste-aware music retrieval from audio embeddings

csc-unipd

Centro di Sonologia Computazionale

Submitted by

gawon1224

ACID: Action Consistency via Inverse Dynamics for Planning with World Models

POSTECH

Pohang University of Science and Technology

1

Submitted by

MatteoFasulo

AI Wizards at EXIST 2026: Hierarchical Soft-Label Learning for Multimodal Sexism Identification in Memes

AIWizards

Submitted by

ryota-komatsu

Speaker-Disentangled Chunk-Wise Regression for Syllabic Tokenization

·
4 authors

Submitted by

psychofict

PixCon: Clean-Positive Contrastive Learning for Foundation-Model Semi-Supervised Segmentation

ebenworks

Ebenworks Systems

Submitted by

zixinding

Learning to Trigger: Reinforcement Learning at the Large Hadron Collider

Uchicago

University of Chicago