Papers
Sign in to view your remaining parses.
Tag Filter
Neural Machine Translation by Jointly Learning to Align and Translate
Published:9/2/2014
Neural Machine TranslationEncoder-Decoder ModelSoft Alignment MethodEnglish-to-French TranslationOptimization of Translation Performance
This paper introduces an innovative neural machine translation method that integrates alignment and translation processes, enhancing the encoderdecoder framework to allow soft alignment of relevant source segments, achieving comparable performance to stateoftheart translation
02
SkyNet: Analyzing Alert Flooding from Severe Network Failures in Large Cloud Infrastructures
Published:8/27/2025
Analysis of Severe Failures in Large Cloud InfrastructuresAlert Flooding DetectionNetwork Failure ManagementCloud Computing Security AssessmentHigh-Availability Network Design
SkyNet addresses alert flooding from severe network failures in large cloud infrastructures by integrating multiple monitoring sources and standardizing inputs. It effectively groups alerts, evaluates severity, and filters irrelevant notifications, significantly reducing average
03
CameraCtrl II: Dynamic Scene Exploration via Camera-controlled Video Diffusion Models
Published:3/14/2025
Camera-Controlled Video Diffusion ModelsDynamic Scene ExplorationDynamic Content GenerationWide-Angle Viewpoint GenerationVideo Generation Dataset Construction
CameraCtrl II enables largescale dynamic scene exploration via a cameracontrolled video diffusion model, overcoming limitations in video dynamics and viewpoint range by enhancing individual video clips and allowing userdefined camera trajectories for broader spatial exploratio
02
The Effect of TiO2 Addition on Low-temperature Sintering Behaviors in a SnO2-CoO-CuO System
Published:4/30/2024
Low-Temperature Sintering BehaviorsEffect of TiO2 Addition on SnO2SnO2-Based MaterialsGas Sensor ApplicationsGrain Boundary Diffusion Mechanism
This study investigates the effective lowtemperature (950°C) sintering of a SnO2CoOCuO system by adding TiO2, which significantly enhances densification via grainboundary diffusion, yielding suitable porous microstructures for gas sensor applications.
02
The Effect of Sm2O3 on the Sintering and Grain Growth Behaviors of SnO2-Based Ceramics
Published:6/1/2019
Effect of Samarium Oxide on Sintering of SnO2 CeramicsMicrostructural Study of SnO2-Based CeramicsGrain Growth Behavior of SnO2 CeramicsCo-Precipitation Method for SnO2 CeramicsEffect of Doping on Ceramic Properties
The study examines how Samarium Oxide (Sm2O3) affects the sintering, microstructure, and grain growth in Co and Nbdoped SnO2 ceramics, showing significant grain growth suppression, reducing average size from 2.70μm to 0.887μm due to segregation at grain boundaries.
02
Densification of 0·99SnO2–0·01CuO Mixture: Evidence for Liquid Phase Sintering
Liquid Phase Sintering MechanismTin Oxide-Copper Oxide MixtureHigh Density Material FabricationSintering Temperature and Time OptimizationElectrical Behavior of Copper Ions
The study examines the sintering of a 0.99SnO2–0.01CuO mixture at 1150°C, achieving 98.7% densification. It reveals that liquid phase sintering is the primary densification mechanism, with copper ions dissolving in interstitial positions affecting electrical properties.
02
CameraCtrl: Enabling Camera Control for Text-to-Video Generation
Published:4/3/2024
Video Generation ControlCamera Trajectory ParameterizationDiffusion Model Camera ControlText-to-Video GenerationControllable Video Generation
This paper presents CameraCtrl, a method for precise camera pose control in video generation. By utilizing effective camera trajectory parameterization and a plugandplay control module, CameraCtrl enhances user controllability and creative expression without affecting other bas
02
Color image information transmission in plasma sheath turbulence based on orbital angular momentum mode
Published:4/29/2025
Color Image Transmission in Plasma Sheath TurbulenceOrbital Angular Momentum ModeGaussian Vortex BeamsFree-Space Optical CommunicationPSNR and Bit Error Rate Analysis
This study numerically investigates color image transmission in plasma sheath turbulence using orbital angular momentum modes of Gaussian vortex beams, analyzing the impact of various parameters on image quality, confirming the feasibility of the proposed encoding and decoding sc
02
Robust transmission of pin-like vortex beams in plasma sheath turbulence
Published:7/25/2025
Beam Propagation CharacteristicsPlasma Sheath TurbulencePin-Like Vortex BeamsLaguerre-Gaussian BeamsBit Error Rate (BER)
This study uses the random phasescreen method to analyze the propagation of pinlike vortex beams (PLVBs) in plasma sheath turbulence, finding that PLVBs significantly outperform conventional LaguerreGaussian beams in detection probability, bit error rate, and channel capacity,
022
What Is That Talk About? A Video-to-Text Summarization Dataset for Scientific Presentations
Published:2/12/2025
Video-to-Text Summarization DatasetMultimodal LearningScientific Presentation VideosAI Conference Record ExtractionSummary Quality Evaluation
This paper introduces , a dataset for videototext summarization of scientific presentations, featuring 18,599 AI conference videos and corresponding abstracts. It benchmarks stateoftheart models and applies a planbased framework to enhance summary quality. A notable
02
Diffusion Meets DAgger: Supercharging Eye-in-hand Imitation Learning
Published:7/15/2024
Diffusion Meets DAggerEye-in-Hand Imitation LearningData Synthesis in Imitation LearningOut-of-Distribution Sample GenerationComparison of Behavior Cloning and Diffusion Models
This paper introduces DMD, a method combining diffusion models with DAgger to address compounding execution errors in eyeinhand imitation learning. DMD synthesizes outofdistribution samples, achieving robust performance with fewer data, outperforming traditional behavior clon
02
TileLang: A Composable Tiled Programming Model for AI Systems
Published:4/24/2025
Composable Tiled Programming ModelAI Kernel Programming OptimizationHigh-Performance Computing KernelsDecoupled Scheduling SpaceHardware-Centric Optimization Strategies
TileLang is introduced as a composable tiled programming model for efficient AI kernel development, decoupling scheduling from dataflow through customizable annotations and primitives. Experiments demonstrate its stateoftheart performance, highlighting its power and flexibilit
07
Riemannian Flow Matching Policy for Robot Motion Learning
Published:3/16/2024
Flow Matching PoliciesRobotic Action LearningVisuomotor PoliciesRiemannian Flow Matching PolicyGeometric-Aware Robot Control
The paper presents Riemannian Flow Matching Policies (RFMP), a model for learning robot visuomotor strategies that excels in efficient training and inference. RFMP effectively manages highdimensional, multimodal distributions and incorporates geometric awareness, outperforming e
01
Autoregressive Adversarial Post-Training for Real-Time Interactive Video Generation
Published:6/11/2025
Autoregressive Adversarial Post-TrainingReal-time Video Generationvideo diffusion modelsInteractive Video GenerationLong Video Generation
The paper introduces Autoregressive Adversarial PostTraining (AAPT) to transform a pretrained latent video diffusion model into an efficient realtime interactive video generator. It generates one latent frame per evaluation, streams in real time, and responds to user interacti
03
Fast and Robust Visuomotor Riemannian Flow Matching Policy
Published:12/14/2024
Riemannian Flow Matching PolicyVisuomotor PoliciesStable Riemannian Flow Matching PolicyRobotic Task LearningGeometric Constraints
The paper introduces the Riemannian Flow Matching Policy (RFMP) for visuomotor tasks, offering fast inference and easy training. It incorporates geometric constraints for robustness and outperforms traditional diffusion policies in real and simulated tasks.
02
GentleHumanoid: Learning Upper-body Compliance for Contact-rich Human and Object Interaction
Published:11/7/2025
Upper-body Compliance Learning for Humanoid RobotsSpring-based Impedance ControlContact-rich Human-Robot InteractionSafe Object ManipulationWhole-body Motion Tracking Policy
GentleHumanoid integrates impedance control into a wholebody motion tracking policy for humanoid robots, achieving upperbody compliance. It employs a springbased model to adapt to diverse humanrobot interactions, reducing contact forces while ensuring successful task executio
03
Learning Human-Humanoid Coordination for Collaborative Object Carrying
Published:10/16/2025
Human-Humanoid CollaborationProprioceptive Reinforcement LearningCollaborative Carrying TasksDynamic Object InteractionClosed-Loop Training Environment
The COLA method enables effective humanhumanoid collaboration in complex carrying tasks using proprioceptiononly reinforcement learning. It predicts object motion and human intent, achieving a 24.7% reduction in human effort while maintaining stability, validated across various
02
Humanoid Whole-Body Badminton via Multi-Stage Reinforcement Learning
Published:11/14/2025
Humanoid Whole-Body ControlReinforcement Learning Training PipelineAction Generation in Dynamic EnvironmentsBadminton Motion ControlMultistage Reinforcement Learning
This paper presents a reinforcement learning training pipeline to develop a unified wholebody controller for humanoid badminton, enabling coordinated footwork and striking without reliance on motion priors or expert demonstrations. The training is validated in both simulated and
02
IFEval-Audio: Benchmarking Instruction-Following Capability in Audio-based Large Language Models
Published:5/22/2025
Instruction-Following Capability in Audio LLMsIFEval-Audio Benchmark DatasetMultimodal Model EvaluationAudio Instruction GenerationAudio-Text Instruction Pairing
The study introduces IFEvalAudio, a novel dataset for assessing instructionfollowing capabilities in audiobased large language models, comprising 280 audioinstructionanswer triples across six dimensions, and benchmarks stateoftheart audio LLMs.
01
AHELM: A Holistic Evaluation of Audio-Language Models
Published:8/29/2025
Evaluation of Audio-Language ModelsAHELM BenchmarkPARADE DatasetMultimodal Model Performance AssessmentSpeech Recognition and Language Model Integration
AHELM is a benchmark introduced to holistically assess AudioLanguage Models (ALMs), integrating multiple datasets and introducing PARADE and CoReBench. It covers ten key evaluation aspects and standardizes methods for equitable model comparisons.
01
……