Page 14 - Paper Library - AiPaper

Diffusion Policy: Visuomotor Policy Learning via Action Diffusion

Published:3/8/2023

Application of Diffusion Models in Robot Policy LearningVisuomotor Policy LearningDiffusion-based Behavior GenerationStochastic Langevin Dynamics OptimizationMultimodal Action Distribution Handling

The paper introduces Diffusion Policy, modeling visuomotor policies as conditional denoising diffusion processes. It shows a 46.9% performance improvement over existing methods across 12 tasks, effectively handling multimodal actions and ensuring training stability, advancing rob

OpenVLA: An Open-Source Vision-Language-Action Model

Published:6/13/2024

Open-Source Vision-Language-Action ModelRobotic Multi-Task ManipulationLarge-Scale Robot Demonstration DatasetModel Fine-Tuning and AdaptationVision-Language-Action Model

OpenVLA is a 7Bparameter opensource VisionLanguageAction model built on Llama 2. Training on 970k robot demonstrations, it efficiently finetunes for various tasks, outperforming RT2X by 16.5% in absolute success rate across 29 tasks.

A Training-Free Style-Personalization via SVD-Based Feature Decomposition

Published:7/7/2025

SVD-Based Feature DecompositionTraining-Free Style PersonalizationAutoregressive Image GenerationStyle-Guided GenerationStructural Attention Correction

This paper presents a trainingfree framework for stylepersonalized image generation, using a scalewise autoregressive model to maintain semantic consistency and reduce content leakage. It introduces two lightweight modules for precise style modulation and structural stability,

Infinite-Story: A Training-Free Consistent Text-to-Image Generation

Published:11/17/2025

Text-to-Image GenerationTraining-Free Text-to-Image GenerationConsistent Generation FrameworkMulti-Prompt Storytelling ScenariosAutoregressive Modeling

InfiniteStory is a trainingfree framework for consistent texttoimage generation in multiprompt scenarios, addressing identity and style inconsistencies. With key techniques, it achieves stateoftheart performance and is over 6X faster during inference than existing models,

Towards Scalable Semantic Representation for Recommendation

Published:10/12/2024

LLM-based Recommendation SystemsSemantic ID ModelingHigh-Dimensional Embedding Dimensionality ReductionMixture-of-Codes for RecommendationRecommendation System Performance Enhancement

This study introduces the MixtureofCodes (MoC) method to address dimensionality compression when integrating LLM embeddings into recommendation systems. By constructing multiple independent codebooks and incorporating a fusion module, MoC significantly enhances the discriminabi

Training-Free Motion Customization for Distilled Video Generators with Adaptive Test-Time Distillation

Published:6/24/2025

Training-Free Video Generation Based on DiffusionAdaptive Test-Time Distillation FrameworkMotion CustomizationDistilled Video Generation ModelsReference Video-Guided Generation

MotionEcho is a novel trainingfree testtime distillation framework for distilled video generators that addresses motion customization challenges using highquality teacher models to guide fast student models, enhancing motion fidelity and generation quality while maintaining ef

ABO: Dataset and Benchmarks for Real-World 3D Object Understanding

Published:6/1/2022

Amazon Berkeley Objects DatasetBenchmarks for Real-World 3D Object UnderstandingSingle-View 3D ReconstructionMaterial EstimationCross-Domain Multi-View Object Retrieval

The paper presents the Amazon Berkeley Objects (ABO) dataset, bridging the gap between real and virtual 3D worlds, including product images and artistcreated models. It establishes benchmarks to assess stateoftheart methods in singleview 3D reconstruction, material estimatio

ToddlerBot: Open-Source ML-Compatible Humanoid Platform for Loco-Manipulation

Published:2/3/2025

Open-Source Humanoid Robot PlatformRobotics Learning and Data CollectionZero-Shot Policy TransferWhole-Body Loco-ManipulationReproducibility and Maintainability

The paper presents ToddlerBot, a lowcost, opensource humanoid robot platform designed for scalable policy learning and robotics research. It enables highquality data acquisition via zeroshot simtoreal transfer and features a userfriendly teleoperation interface for wholeb

Extended Friction Models for the Physics Simulation of Servo Actuators

Published:10/11/2024

Physics Simulation of Servo ActuatorsExtended Friction ModelsReinforcement Learning Control AlgorithmsDynamic Model Parameter IdentificationIntegration into Physics Engines

The paper proposes extended friction models to enhance the accuracy of servo actuator simulations. By analyzing friction models and identifying parameters from pendulum test data, the study validates these models on four servo actuators, showing significant accuracy improvement o

The Influence of Artificial Intelligence Tools on Learning Outcomes in Computer Programming: A Systematic Review and Meta-Analysis

Published:5/9/2025

AI Tools in Computer ProgrammingEvaluation of AI-Assisted Learning OutcomesLearning Outcomes in Programming CoursesSystematic Review and Meta-AnalysisStudent Perceptions of AI Tools

This systematic review and metaanalysis examine the impact of AI tools like ChatGPT and GitHub Copilot on computer programming learning outcomes. Results show students using AI performed better, but no significant advantage in learning success. Positive student perceptions highl

024

A systematic review and sequential explanatory synthesis: Artificial intelligence in healthcare education, a case of nursing

Artificial Intelligence in Healthcare EducationKnowledge and Skills Development in Nursing StudentsMixed-Methods Systematic ReviewImpact of AI Interventions on Nursing EducationKnowledge Acquisition and Attitudes in Nursing Education

This systematic review assesses the impact of artificial intelligence on nursing students' knowledge acquisition, skills development, and attitudes. Findings indicate positive effects of AI interventions on learning engagement and selfefficacy, highlighting the need for standard

Will the Use of AI Undermine Students Independent Thinking?

Published:5/28/2025

Impact of AI in EducationCultivation of Independent ThinkingCognitive Development and AI ToolsPersonalized Learning EnvironmentsIntegration of Pedagogical Strategies and Technology

The paper examines the impact of AI in education on students' cognitive development, particularly concerning independent and critical thinking. While AI offers personalized learning opportunities, overreliance may diminish cognitive effort and selfanalysis. It analyzes existing

New Era of Artificial Intelligence in Education: Towards a Sustainable Multifaceted Revolution

Published:8/16/2023

Application of AI in EducationIntelligent Tutoring SystemsPersonalized LearningEthics of AI in EducationChallenges of AI in Education

This study reviews the impact of AI in education, focusing on applications, benefits, and challenges, emphasizing the need for sustainable development while advocating for measures to prevent misuse as we embrace technological changes.

The Influence of Artificial Intelligence Tools on Student Performance in e-Learning Environments: Case Study

Published:11/4/2024

Personalized Learning ToolsOnline Educational TechnologyApplication of AI in EducationStudent Engagement and MotivationPre-Service Teacher Training

This study examines the impact of AIdriven personalized learning tools on preservice teachers' academic performance and perceptions in elearning. Findings show that the experimental group using these tools significantly outperformed the control group, highlighting AI's transfo

Synthetic media and computational capitalism: towards a critical theory of artificial intelligence

Published:3/19/2025

Algorithmic Condition in Computational CapitalismAutomimetric Production FrameworkPost-Consciousness TheoryInteraction Analysis of Computational Systems and Cultural FormsNew Computational Turn in Algorithmic Society

The paper develops a critical theory of AI, addressing the collapse of boundaries between human and machine production. It introduces concepts like algorithmic condition and postconsciousness, analyzing automimetric production and its impact on authenticity, emphasizing the need

Hybrid Reinforcement: When Reward Is Sparse, It's Better to Be Dense

Published:10/9/2025

RL Training for Large Language ModelsHybrid Reward OptimizationMath Reasoning BenchmarksReward Model-Based LearningSparse Reward Problem

The HERO framework integrates verifiable rewards with reward models to address the limitations of sparse feedback in large language model reasoning tasks. Using stratified normalization and varianceaware weighting, HERO significantly improves performance on mathematical reasonin

BAPO: Stabilizing Off-Policy Reinforcement Learning for LLMs via Balanced Policy Optimization with Adaptive Clipping

Published:10/21/2025

RL Training for Large Language ModelsBalanced Policy OptimizationAdaptive Clipping MechanismOff-Policy OptimizationEfficient Sample Replay

This paper presents BAPO, a method for stabilizing offpolicy reinforcement learning for large language models by using balanced policy optimization with adaptive clipping, addressing issues of optimization imbalance and improving sample efficiency.

LLMs as Sparse Retrievers:A Framework for First-Stage Product Search

Published:10/21/2025

LLMs Application in Product SearchSparse Retrieval FrameworkVocabulary Mismatch in Product SearchPROSPER FrameworkSemantic Analysis and Retrieval Quality Enhancement

The PROSPER framework utilizes large language models as sparse retrievers for product search, addressing vocabulary mismatch through a literal residual network and lexical focusing window, enhancing keyword weighting and improving retrieval quality amidst hallucinations and initi

HT-Net: Hierarchical Transformer Based Operator Learning Model for Multiscale PDEs

Multiscale PDE ModelsHierarchical Transformer ArchitectureSelf-Attention MechanismEfficient Solver LearningSpectral Bias Mitigation Method

This paper presents HTNet, a hierarchical transformer that efficiently learns solution operators for multiscale PDEs. It features adaptive interaction ranges and hierarchical selfattention to optimize calculations, employing an empirical H1 loss to mitigate spectral bias, outpe

Learning Spatially-Aware Language and Audio Embeddings

Published:9/18/2024

Spatially-Aware Audio and Text Embedding ModelMultimodal Contrastive LearningAudio Event Localization and DetectionOpen Vocabulary Text DescriptionsNon-Spatial Audio and Text Mapping

The paper introduces ELSA, a multimodal contrastive learning model that captures both semantic and spatial features of audio. Using synthetic spatial audio, ELSA demonstrates superior performance in semantic retrieval and 3D localization, improving accuracy over existing models.

261 - 280 / 980

Papers