Daily Arxiv

전 세계에서 발간되는 인공지능 관련 논문을 정리하는 페이지 입니다.
본 페이지는 Google Gemini를 활용해 요약 정리하며, 비영리로 운영 됩니다.
논문에 대한 저작권은 저자 및 해당 기관에 있으며, 공유 시 출처만 명기하면 됩니다.

OregairuChar: A Benchmark Dataset for Character Appearance Frequency Analysis in My Teen Romantic Comedy SNAFU

A Gate-Based Quantum Genetic Algorithm for Real-Valued Global Optimization

Accurate online action and gesture recognition system using detectors and Deep SPD Siamese Networks

4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos

No One-Model-Fits-All: Uncovering Spatio-Temporal Forecasting Trade-offs with Graph Neural Networks and Foundation Models

Model Merging Improves Zero-Shot Generalization in Bioacoustic Foundation Models

Generating Software Architecture Description from Source Code using Reverse Engineering and Large Language Model

SmartSecChain-SDN: A Blockchain-Integrated Intelligent Framework for Secure and Efficient Software-Defined Networks

From Linear Probing to Joint-Weighted Token Hierarchy: A Foundation Model Bridging Global and Cellular Representations in Biomarker Detection

DL101 Neural Network Outputs and Loss Functions

Deep learning models are vulnerable, but adversarial examples are even more vulnerable

No Pose Estimation? No Problem: Pose-Agnostic and Instance-Aware Test-Time Adaptation for Monocular Depth Estimation

Accelerating HDC-CNN Hybrid Models Using Custom Instructions on RISC-V GPUs

UA-Code-Bench: A Competitive Programming Benchmark for Evaluating LLM Code Generation in Ukrainian

PECL: A Heterogeneous Parallel Multi-Domain Network for Radar-Based Human Activity Recognition

Dynamic Residual Encoding with Slide-Level Contrastive Learning for End-to-End Whole Slide Image Representation

OvA-LP: A Simple and Efficient Framework for Federated Learning on Non-IID Data

8bit-GPT: Exploring Human-AI Interaction on Obsolete Macintosh Operating Systems

Pluralistic Behavior Suite: Stress-Testing Multi-Turn Adherence to Custom Behavioral Policies

Multi-agent Coordination via Flow Matching

Query Generation Pipeline with Enhanced Answerability Assessment for Financial Information Retrieval

BiPETE: A Bi-Positional Embedding Transformer Encoder for Risk Assessment of Alcohol and Substance Use Disorder with Electronic Health Records

Enhancing Public Speaking Skills in Engineering Students Through AI

Learning Fourier shapes to probe the geometric world of deep neural networks

Pattern-Aware Diffusion Synthesis of fMRI/dMRI with Tissue and Microstructural Refinement

Too Good to be Bad: On the Failure of LLMs to Role-Play Villains

DeepForgeSeal: Latent Space-Driven Semi-Fragile Watermarking for Deepfake Detection Using Multi-Agent Adversarial Reinforcement Learning

A benchmark multimodal oro-dental dataset for large vision-language models

Search Is Not Retrieval: Decoupling Semantic Matching from Contextual Assembly in RAG

BudgetMem: Learning Selective Memory Policies for Cost-Efficient Long-Context Processing in Language Models

MERaLiON-SER: Robust Speech Emotion Recognition Model for English and SEA Languages

A Dual Perspective on Decision-Focused Learning: Scalable Training via Dual-Guided Surrogates

You Need Reasoning to Learn Reasoning: The Limitations of Label-Free RL in Weak Base Models

Beta Distribution Learning for Reliable Roadway Crash Risk Assessment

Minimal and Mechanistic Conditions for Behavioral Self-Awareness in LLMs

Software Defined Vehicle Code Generation: A Few-Shot Prompting Approach

Prompt-Based Safety Guidance Is Ineffective for Unlearned Text-to-Image Diffusion Models

Isaac Lab: A GPU-Accelerated Simulation Framework for Multi-Modal Robot Learning

A Standardized Benchmark for Multilabel Antimicrobial Peptide Classification

Unified Multimodal Diffusion Forcing for Forceful Manipulation

An Active Learning Pipeline for Biomedical Image Instance Segmentation with Minimal Human Intervention

PuzzleMoE: Efficient Compression of Large Mixture-of-Experts Models via Sparse Expert Merging and Bit-packed inference

Data Efficiency and Transfer Robustness in Biomedical Image Segmentation: A Study of Redundancy and Forgetting with Cellpose

MDM: Manhattan Distance Mapping of DNN Weights for Parasitic-Resistance-Resilient Memristive Crossbars

Causal Structure and Representation Learning with Biomedical Applications

ScheduleStream: Temporal Planning with Samplers for GPU-Accelerated Multi-Arm Task and Motion Planning & Scheduling

CPO: Condition Preference Optimization for Controllable Image Generation

Knowledge-based anomaly detection for identifying network-induced shape artifacts

Trustworthiness Calibration Framework for Phishing Email Detection Using Large Language Models

IndicVisionBench: Benchmarking Cultural and Multilingual Understanding in VLMs

Temporal convolutional and fusional transformer model with Bi-LSTM encoder-decoder for multi-time-window remaining useful life prediction

Learning to reason about rare diseases through retrieval-augmented agents

Ada-FCN: Adaptive Frequency-Coupled Network for fMRI-Based Brain Disorder Classification

P-MIA: A Profiled-Based Membership Inference Attack on Cognitive Diagnosis Models

First is Not Really Better Than Last: Evaluating Layer Choice and Aggregation Strategies in Language Model Data Influence Estimation

SWAP: Towards Copyright Auditing of Soft Prompts via Sequential Watermarking

Jailbreaking in the Haystack

Prioritize Economy or Climate Action? Investigating ChatGPT Response Differences Based on Inferred Political Orientation

POLIS-Bench: Towards Multi-Dimensional Evaluation of LLMs for Bilingual Policy Tasks in Governmental Scenarios

Measuring what Matters: Construct Validity in Large Language Model Benchmarks

Separate the Wheat from the Chaff: Winnowing Down Divergent Views in Retrieval Augmented Generation

multiMentalRoBERTa: A Fine-tuned Multiclass Classifier for Mental Health Disorder

Simulating Misinformation Vulnerabilities With Agent Personas

EncouRAGe: Evaluating RAG Local, Fast, and Reliable

Reasoning Up the Instruction Ladder for Controllable Language Models

A Penny for Your Thoughts: Decoding Speech from Inexpensive Brain Signals

Adaptive Testing for LLM Evaluation: A Psychometric Alternative to Static Benchmarks

Stateful KV Cache Management for LLMs: Balancing Space, Time, Accuracy, and Positional Fidelity

AI-Powered Citation Auditing: A Zero-Assumption Protocol for Systematic Reference Verification in Academic Research

Efficient Deployment of CNN Models on Multiple In-Memory Computing Units

Reasoning Is All You Need for Urban Planning AI

Cleaning Maintenance Logs with LLM Agents for Improved Predictive Maintenance

Autonomous generation of different courses of action in mechanized combat operations

ORCHID: Orchestrated Retrieval-Augmented Classification with Human-in-the-Loop Intelligent Decision-Making for High-Risk Property

Real-Time Reasoning Agents in Evolving Environments

DMA: Online RAG Alignment with Human Feedback

Epistemic Reject Option Prediction

A hybrid solution approach for the Integrated Healthcare Timetabling Competition 2024

A Theoretical Framework for Environmental Similarity and Vessel Mobility as Coupled Predictors of Marine Invasive Species Pathways

CareMedEval dataset: Evaluating Critical Appraisal and Reasoning in the Biomedical Field

Node-Based Editing for Multimodal Generation of Text, Audio, Image, and Video

Control Barrier Function for Aligning Large Language Models

A Criminology of Machines

FATE: A Formal Benchmark Series for Frontier Algebra of Multiple Difficulty Levels

PoCo: Agentic Proof-of-Concept Exploit Generation for Smart Contracts

Causal Graph Neural Networks for Healthcare

LA-MARRVEL: A Knowledge-Grounded and Language-Aware LLM Reranker for AI-MARRVEL in Rare Disease Diagnosis

OceanAI: A Conversational Platform for Accurate, Transparent, Near-Real-Time Oceanographic Insights

Pay for The Second-Best Service: A Game-Theoretic Approach Against Dishonest LLM Providers

Artificial Intelligence in Elementary STEM Education: A Systematic Review of Current Applications and Future Challenges

Non-Convex Over-the-Air Heterogeneous Federated Learning: A Bias-Variance Trade-off

TraceTrans: Translation and Spatial Tracing for Surgical Prediction

TowerVision: Understanding and Improving Multilinguality in Vision-Language Models

Gestura: A LVLM-Powered System Bridging Motion and Semantics for Real-Time Free-Form Gesture Understanding

ADPO: Anchored Direct Preference Optimization

Rewarding the Journey, Not Just the Destination: A Composite Path and Answer Self-Scoring Reward Mechanism for Test-Time Reinforcement Learning

RealDPO: Real or Not Real, that is the Preference

Deep Edge Filter: Return of the Human-Crafted Layer in Deep Learning

Mathematics with large language models as provers and verifiers

Integrating Sequential and Relational Modeling for User Events: Datasets and Prediction Tasks

Uncertainty Propagation in XAI: A Comparison of Analytical and Empirical Estimators

Created by

Haebom

저자

Teodor Chiaburu, Felix Bie{\ss}mann, Frank Hau{\ss}er

개요

본 논문은 설명 가능한 AI(XAI)에서 불확실성을 정량화하고 해석하기 위한 통합 프레임워크를 제시합니다. 입력 데이터와 모델 파라미터의 변동으로부터 불확실성 전파를 포착하는 일반적인 설명 함수 $e_{\theta}(x, f)$를 정의하여, 분석적 및 경험적 설명 분산 추정치를 사용하여 불확실성이 설명에 미치는 영향을 체계적으로 평가합니다. 1차 불확실성 전파를 분석적 추정기로 사용하여 접근 방식을 설명하고, 다양한 데이터 세트에 대한 종합적인 평가를 통해 분석적 및 경험적 불확실성 전파 추정치를 비교하고 강건성을 평가합니다. 설명의 불일치에 대한 이전 연구를 확장하여, 불확실성을 안정적으로 포착하고 전파하지 않는 XAI 방법을 실험적으로 확인합니다. 고위험 응용 프로그램에서 불확실성 인식 설명의 중요성을 강조하고 현재 XAI 방법의 한계에 대한 새로운 통찰력을 제공합니다. 실험 코드는 https://github.com/TeodorChiaburu/UXAI 에서 확인할 수 있습니다.

GitHub - TeodorChiaburu/UXAI: Sources and propagation of uncertainty in XAI

Sources and propagation of uncertainty in XAI. Contribute to TeodorChiaburu/UXAI development by creating an account on GitHub.

시사점, 한계점

•

시사점:

◦

XAI에서 불확실성을 정량화하고 해석하기 위한 통합 프레임워크 제공

◦

분석적 및 경험적 방법을 통한 불확실성 전파 평가 및 비교

◦

불확실성을 제대로 반영하지 못하는 XAI 방법 식별

◦

고위험 응용 프로그램에서 불확실성 인식 설명의 중요성 강조

•

한계점:

◦

제시된 프레임워크의 일반성 및 적용 범위에 대한 추가 연구 필요

◦

1차 불확실성 전파만 고려하여 고차 불확실성 효과는 반영하지 못할 수 있음

◦

사용된 데이터셋의 다양성 및 대표성에 대한 검토 필요

◦

특정 XAI 방법에 대한 편향 가능성 존재

Made with Slashpage