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 3

Submitted by

Vasily

OCC-RAG: Optimal Cognitive Core for Faithful Question Answering

occ-ai

Submitted by

navvew

From Activation to Causality: Discovery of Causal Visual Representations in the Human Brain

MIT

Massachusetts Institute of Technology

Submitted by

xrxing

Trust Region On-Policy Distillation

SamsungResearch

Samsung Research

Submitted by

taesiri

Humanoid-GPT: Scaling Data and Structure for Zero-Shot Motion Tracking

·
13 authors

Submitted by

pbicho

KVarN: Variance-Normalized KV-Cache Quantization Mitigates Error Accumulation in Reasoning Tasks

huawei-csl

HUAWEI Computing Systems Lab

Submitted by

yl-9

A Local Perturbation Theory for Cross-Domain Interference and Recovery in Multi-Domain RL

·
3 authors

1

Submitted by

YCZhou

World Models Meet Language Models: On the Complementarity of Concrete and Abstract Reasoning

tencent

Submitted by

csjiaya

MIRA: Mid-training Rubric Anchoring for Source-Aware Data Selection

·
12 authors

Submitted by

qth

AutoMedBench: Towards Medical AutoResearch with Agentic AI Models

UCSC

University of California, Santa Cruz

Submitted by

ytz009

TRON: Targeted Rule-Verifiable Online Environments for Visual Reasoning RL

UniversityofGeorgia

University of Georgia

Submitted by

taesiri

Language Models Need Sleep: Learning to Self-Modify and Consolidate Memories

google

Submitted by

Liangqiong-QU

Decoupled Residual Denoising Diffusion Models for Unified and Data Efficient Image-to-Image Translation

hkuhk

The University of Hong Kong

Submitted by

HakHan

Ψ-Bench: Evaluating Persona-Sensitive Influencing in Persuasive Dialogues

·
6 authors

Submitted by

ChengsongHuang

Small RL Controller, Large Language Model: RL-Guided Adaptive Sampling for Test-Time Scaling

·
5 authors

Submitted by

gwkrsrch

Decentralized Instruction Tuning: Conflict-Aware Splitting and Weight Merging

·
2 authors

Submitted by

taesiri

NVIDIA OmniDreams: Real-Time Generative World Model for Closed-Loop Autonomous Vehicle Simulation

nvidia

Submitted by

SteveZeyuZhang

PlatonicNav: Unveiling Semantic Correspondence in Navigation with Platonic Topological Maps

Maincode

Submitted by

mozhu

Diagnosing Harmful Continuation in Answer-Correct Long-CoT Training Traces

·
5 authors

Submitted by

nuuuh

Adaptive Auto-Harness: Sustained Self-Improvement for Agentic System Deployment on Open-Ended Task Streams

·
10 authors

1

Submitted by

ChengCui

PaddleOCR-VL-1.6: Expanding the Frontier of Document Parsing with Under-Optimized Region Refinement and Progressive Post-Training

PaddlePaddle

Submitted by

dorienh

MERIT: Learning Disentangled Music Representations for Audio Similarity

amaai-lab

Submitted by

deqing

Value-Aware Stochastic KV Cache Eviction for Reasoning Models

UniversityofSouthernCalifornia

University of Southern California

Submitted by

EdwinHuang

Benchmarking Visual State Tracking in Multimodal Video Understanding

nyu-visionx

Submitted by

YoadTew

Bootstrap Your Generator: Unpaired Visual Editing with Flow Matching

nvidia

Submitted by

vincentkoc

ClawHub Security Signals: When VirusTotal, Static Analysis, and SkillSpector Disagree

OpenClaw

1

Submitted by

fhlyhv

Domain-Specific Data Synthesis for LLMs via Minimal Sufficient Representation Learning

·
9 authors

Submitted by

josefchen

AURA: Action-Gated Memory for Robot Policies at Constant VRAM

Kaikaku

1

Submitted by

XiangZ

αDepth: Learning Single-Pass Soft Boundary Decomposition for Stereo Conversion

ethz

Submitted by

LeoYML

A Multi-AI-agent Framework Enabling End-to-end Finite Element Analysis for Solid Mechanics Problems

·
6 authors

Submitted by

pajamastao

Prior Availability in Industrial Visual Sim-to-Real: A Review of CAD-Guided and CAD-Unavailable Regimes

·
2 authors

Submitted by

nielsr

Ultralytics YOLO26: Unified Real-Time End-to-End Vision Models

Ultralytics

Submitted by

zhangganlin

BA-T: An Iterative Transformer for Two-View Bundle Adjustment

TUM

Technical University of Munich

Submitted by

sjpark5800

Mitigating Perceptual Judgment Bias in Multimodal LLM-as-a-Judge via Perceptual Perturbation and Reward Modeling

kaist-ai

Submitted by

techsachin

Pressure-Testing Deception Probes in LLMs: Scaling, Robustness, and the Geometry of Deceptive Representations

·
1 authors