Page 8 - Paper Library - AiPaper

This paper introduces a novel distributed realtime database indexing algorithm that integrates B Tree and consistent hashing. It efficiently determines storage locations for TAG points and organizes historical data, addressing highfrequency data storage challenges.

Single-cell whole-genome analyses by Linear Amplification via Transposon Insertion (LIANTI)

Published:4/13/2017

Single-Cell Whole-Genome AmplificationLinear Amplification via Transposon Insertion (LIANTI)Copy-Number Variation DetectionSingle Nucleotide Variation AnalysisObservation of DNA Replication Origins

The study presents LIANTI, an enhanced singlecell wholegenome amplification method that improves copynumber variation detection with kilobase resolution, enabling the observation of DNA replication origin activation differences among cells, and identifying SNV spectra postUV

DiffuRec: A Diffusion Model for Sequential Recommendation

Published:4/3/2023

Diffusion Model for Sequential RecommendationUncertainty InjectionDistribution-Based Item RepresentationSequential Recommender SystemsGenerative Recommendation Systems

This paper introduces DiffuRec, the first diffusion model for sequential recommendation, representing item embeddings as distributions to better reflect multiple user interests and diverse item features. The method leverages noise addition for uncertainty injection and reconstruc

Least Angle Regression

Published:6/23/2004

Least Angle RegressionModel Selection AlgorithmsLasso RegressionForward Stagewise Linear RegressionEfficient Prediction Model

Least Angle Regression (LARS) is an efficient model selection algorithm that offers advantages over traditional forward selection methods. It enables fast computation of Lasso estimates and effectively implements Forward Stagewise regression, improving model parsimony and predict

A Comprehensive Survey on World Models for Embodied AI

Published:10/19/2025

World Models for Embodied AITemporal Modeling and InferenceSpatial Representation and RenderingRobotics and Autonomous Driving DatasetsLong-Horizon Consistency Modeling

This survey presents a unified framework for world models in embodied AI, categorizing functionality, temporal modeling, and spatial representation. It systematizes data resources and metrics across robotics and autonomous driving, compares stateoftheart models, and identifies

LLVM: A Compilation Framework for Lifelong Program Analysis & Transformation

Published:6/10/2004

LLVM Compiler FrameworkLifelong Program Analysis and TransformationStatic Single Assignment (SSA) formLanguage-Independent Type SystemCompiler Performance Evaluation

LLVM is a compiler framework that enables transparent, lifelong program analysis and transformation by providing highlevel information for optimizations at compile, link, run, and idle times. It features a languageindependent type system and is evaluated for its efficiency and

Advanced Skills through Multiple Adversarial Motion Priors in Reinforcement Learning

Published:3/23/2022

Adversarial Motion PriorsRobotic Locomotion ControlImitation LearningDiverse Skill LearningReinforcement Learning

This study presents MultiAMP, an enhanced adversarial motion prior method that simplifies reward function tuning in reinforcement learning for robotic motion control. Experiments demonstrate simultaneous learning of multiple styles and skills without significant performance diff

CosyVoice 3: Towards In-the-wild Speech Generation via Scaling-up and Post-training

Published:5/23/2025

CosyVoice 3 Speech Synthesis ModelLLM-based Speech SynthesisZero-Shot Multilingual Speech SynthesisPost-Training TechniquesSpeech Naturalness and Similarity

CosyVoice 3 is a zeroshot multilingual speech synthesis model designed for realworld applications, featuring an innovative speech tokenizer and differentiable reward optimization, significantly enhancing content consistency, speaker similarity, and prosody naturalness over its

肉身的殖民: 身体、空间与资本主义劳动地理

Capitalist Labor GeographyInteraction of Body and SpaceRegulation and Control of Labor BodiesTheory of Bodily ColonialismSpatial Syntax and Capital Expansion

This paper introduces the concept of 'corporeal colonialism', examining how capitalism regulates laboring bodies through spatial syntax to ensure profit growth. It constructs a framework that reveals how labor spaces discipline and reduce workers to mere tools, diminishing their

023

HunyuanVideo: A Systematic Framework For Large Video Generative Models

Published:12/4/2024

Open Source Video Generation ModelsLarge-Scale Video Generation FrameworkVideo Generation Technology EvaluationText-Video AlignmentVideo Generation Benchmark Comparison

HunyuanVideo is an opensource video generative model designed to bridge the performance gap between proprietary models and public ones, featuring a comprehensive framework and achieving over 13 billion parameters, outperforming industry leaders.

Empathic neural responses are modulated by the perceived fairness of others

Published:1/1/2006

Modulation of Empathic Neural ResponsesEconomic Models of Social PreferencesFunctional Magnetic Resonance ImagingNeural Activity During Pain ObservationFairness Impact on Empathy

The study reveals that empathic neural responses are modulated by the perceived fairness of others, as shown through an economic game and fMRI. Male participants displayed increased brain activation when observing fair players in pain, but this empathy was reduced for unfair play

RecGPT-V2 Technical Report

Published:12/16/2025

RecGPT-V2 Recommender SystemLLM-based Recommendation SystemsHierarchical Multi-Agent SystemHybrid Representation InferenceConstrained Reinforcement Learning

RecGPTV2 introduces four innovations to enhance intent reasoning and efficiency, reducing GPU consumption by 60% and improving generalization and evaluation consistency. Online tests show significant performance gains in metrics like CTR and IPV, indicating its industrial applic

Video Prediction Policy: A Generalist Robot Policy with Predictive Visual Representations

Published:12/19/2024

video diffusion modelsRobotic Action LearningVideo Prediction PolicyDynamic Visual RepresentationsComplex Manipulation Tasks

The Video Prediction Policy (VPP) utilizes Video Diffusion Models (VDMs) to generate visual representations that incorporate both current static and predicted dynamic information, enhancing robot action learning and achieving a 31.6% increase in success rates for complex tasks.

Real-World Reinforcement Learning of Active Perception Behaviors

Published:12/1/2025

Reinforcement Learning for Active Perception BehaviorsAsymmetric Advantage Weighted Regression (AAWR)Robot Learning under Partial ObservabilityPrivileged Value Function EstimationRobot Manipulation Task Evaluation

The paper introduces Asymmetric Advantage Weighted Regression (AAWR) to train active perception policies for robots facing partial observability. Utilizing privileged sensors allows for highquality value function training, significantly enhancing task performance across various

A Learned Cache Eviction Framework with Minimal Overhead

Published:1/27/2023

Machine Learning Cache Eviction FrameworkIntegration of Traditional Cache Systems with Machine LearningEfficient Caching AlgorithmsProduction Workload EvaluationLow-Overhead Cache Decision Making

The MAT framework reduces the number of ML predictions for cache eviction from 63 to 2 by using a heuristic as a filter, maintaining low miss ratios similar to stateoftheart ML systems, which enhances practicality for highthroughput environments.

WorldPlay: Towards Long-Term Geometric Consistency for Real-Time Interactive World Modeling

Published:12/17/2025

Real-Time Interactive World ModelingLong-Term Geometric Consistencyvideo diffusion modelsMemory-Aware ModelingDynamic Context Reconstruction

This paper introduces WorldPlay, a video diffusion model for realtime interactive world modeling with longterm geometric consistency, achieved through three innovations: Dual Action Representation, Reconstituted Context Memory, and Context Forcing.

ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation

Published:4/13/2023

Text-to-Image GenerationHuman Preference Reward ModelReward Feedback LearningDiffusion Model OptimizationExpert Comparison Ratings

This study introduces ImageReward, a generalpurpose human preference reward model for texttoimage generation, trained on a systematic annotation process with 137,000 expert comparisons. It outperforms existing models and proposes Reward Feedback Learning (ReFL) for optimizing

A Biologically Plausible Parser

Published:8/5/2021

Biologically Plausible ParserAssembly CalculusLanguage ParsingComputational Framework for Cognitive FunctionsEnglish Sentence Parsing

The paper presents a biologically plausible parser using Assembly Calculus, demonstrating that simple neural mechanisms can effectively parse complex sentences in English and Russian, highlighting the potential of biological models in advanced language processing.

Detailed balance in large language model-driven agents

Published:12/11/2025

LLM Generative DynamicsApplication of Least Action PrincipleTransition Probability Statistical AnalysisMacroscopic Dynamics TheoryComplex AI Systems

This paper introduces a method based on the least action principle to uncover detailed balance in LLMdriven agents, highlighting that their generative processes depend on potential functions rather than generic rules, marking a significant theoretical advance in AI dynamics.

A-LAMP: Agentic LLM-Based Framework for Automated MDP Modeling and Policy Generation

Published:12/12/2025

RL Training for Large Language ModelsMarkov Decision Process ModelingAutomated Policy GenerationVerifiable Stage-wise ModelingAdvanced Reinforcement Learning Applications

The ALAMP framework automates the transition from natural language task descriptions to MDP modeling and policy generation. By decomposing modeling, coding, and training into verifiable stages, ALAMP enhances policy generation capabilities, outpacing traditional large language

141 - 160 / 972

Papers