Structured Code Representations Enable Data-Efficient Adaptation of Code Language Models

Learning to Decode Collaboratively with Multiple Language Models

CHAMP: A Competition-level Dataset for Fine-Grained Analyses of LLMs' Mathematical Reasoning Capabilities

Diversity Measurement and Subset Selection for Instruction Tuning Datasets

Improving Black-box Robustness with In-Context Rewriting

Improving Dialogue Agents by Decomposing One Global Explicit Annotation with Local Implicit Multimodal Feedback

Mitigating the Impact of Outlier Channels for Language Model Quantization with Activation Regularization

Reuse Your Rewards: Reward Model Transfer for Zero-Shot Cross-Lingual Alignment

MDCR: A Dataset for Multi-Document Conditional Reasoning

Design and Application of 2d Goldene Heterostructures with Transition Metal Dichalcogenide for Efficient Overall Water Splitting