data efficiency
A measure of how effectively an algorithm can achieve good performance with limited data. In AI, improving data efficiency is crucial for models to be trained effectively with less labeled data or under diverse conditions.
- 3D Equivariant Visuomotor Policy Learning via Spherical Projection
- Asymmetric REINFORCE for off-Policy Reinforcement Learning: Balancing positive and negative rewards
- Bohdi: Heterogeneous LLM Fusion with Automatic Data Exploration
- BridgeVLA: Input-Output Alignment for Efficient 3D Manipulation Learning with Vision-Language Models
- Computational Algebra with Attention: Transformer Oracles for Border Basis Algorithms
- Compute-Optimal Scaling for Value-Based Deep RL
- Data Efficient Adaptation in Large Language Models via Continuous Low-Rank Fine-Tuning
- Data Selection Matters: Towards Robust Instruction Tuning of Large Multimodal Models
- From Pixels to Views: Learning Angular-Aware and Physics-Consistent Representations for Light Field Microscopy
- GRIT: Teaching MLLMs to Think with Images
- IOSTOM: Offline Imitation Learning from Observations via State Transition Occupancy Matching
- Improving Data Efficiency for LLM Reinforcement Fine-tuning Through Difficulty-targeted Online Data Selection and Rollout Replay
- Incremental Sequence Classification with Temporal Consistency
- Learning from Reward-Free Offline Data: A Case for Planning with Latent Dynamics Models
- Less is More: Improving LLM Alignment via Preference Data Selection
- LoMix: Learnable Weighted Multi-Scale Logits Mixing for Medical Image Segmentation
- Pretraining a Shared Q-Network for Data-Efficient Offline Reinforcement Learning
- Scaling Data-Driven Probabilistic Robustness Analysis for Semantic Segmentation Neural Networks
- Systematic Reward Gap Optimization for Mitigating VLM Hallucinations
- Training a Scientific Reasoning Model for Chemistry