knowledge transfer
The process of utilizing knowledge acquired from one task to improve learning or performance on another related task, often seen in transfer learning scenarios.
- A Token is Worth over 1,000 Tokens: Efficient Knowledge Distillation through Low-Rank Clone
- Bit-swapping Oriented Twin-memory Multi-view Clustering in Lifelong Incomplete Scenarios
- CLEAR: Conv-Like Linearization Revs Pre-Trained Diffusion Transformers Up
- Continual Knowledge Adaptation for Reinforcement Learning
- DeepKD: A Deeply Decoupled and Denoised Knowledge Distillation Trainer
- Dual Prototype-Enhanced Contrastive Framework for Class-Imbalanced Graph Domain Adaptation
- Dynamic Siamese Expansion Framework for Improving Robustness in Online Continual Learning
- ECO: Evolving Core Knowledge for Efficient Transfer
- Efficient Multi-bit Quantization Network Training via Weight Bias Correction and Bit-wise Coreset Sampling
- EgoBridge: Domain Adaptation for Generalizable Imitation from Egocentric Human Data
- GRAVER: Generative Graph Vocabularies for Robust Graph Foundation Models Fine-tuning
- Improving Target Sound Extraction via Disentangled Codec Representations with Privileged Knowledge Distillation
- Interaction-Centric Knowledge Infusion and Transfer for Open Vocabulary Scene Graph Generation
- Knowledge Insulating Vision-Language-Action Models: Train Fast, Run Fast, Generalize Better
- Layer as Puzzle Pieces: Compressing Large Language Models through Layer Concatenation
- Learn and Ensemble Bridge Adapters for Multi-domain Task Incremental Learning
- Lessons Learned: A Multi-Agent Framework for Code LLMs to Learn and Improve
- Miss-ReID: Delivering Robust Multi-Modality Object Re-Identification Despite Missing Modalities
- NTKMTL: Mitigating Task Imbalance in Multi-Task Learning from Neural Tangent Kernel Perspective
- PocketSR: The Super-Resolution Expert in Your Pocket Mobiles
- Provably Efficient Multi-Task Meta Bandit Learning via Shared Representations
- Rendering-Aware Reinforcement Learning for Vector Graphics Generation
- SDPGO: Efficient Self-Distillation Training Meets Proximal Gradient Optimization
- SpikingVTG: A Spiking Detection Transformer for Video Temporal Grounding
- Towards A Generalist Code Embedding Model Based On Massive Data Synthesis
- Towards Unsupervised Open-Set Graph Domain Adaptation via Dual Reprogramming
- Training the Untrainable: Introducing Inductive Bias via Representational Alignment
- Transfer Learning for Benign Overfitting in High-Dimensional Linear Regression
- Transforming Gaps into Gains: Bridging Model and Data Heterogeneity in Federated Learning via Knowledge Weak-Aware Zones
- Vicinity-Guided Discriminative Latent Diffusion for Privacy-Preserving Domain Adaptation
- Who You Are Matters: Bridging Interests and Social Roles via LLM-Enhanced Logic Recommendation
- Zebra-Llama: Towards Extremely Efficient Hybrid Models