OPLD: On-Policy Latent Distillation for Multimodal Reasoning · 相关技术
相关技术
DIffusion Reconstruction ErrorDISDIRECTDIRDEFINEDComposed Object RetrievalCombinatorial OptimizationChain-of-Thought explanation paradigmCalibrated Entropy ScoreCTICOMCAPAutoregressive chain-of-thoughtCoTAugmented Latent IntrinsicsArtificial Language LearningArabic Speech Emotion RecognitionAmodal Mask Prediction and ROI Re-Alignment (AMPR)Adversarial Style OptimizationAdvanced Persistent ThreatsAdaptive Batch ScalingATEANNABI@StateACEAbstractACT