Yang Deng

RAVENEA: A Benchmark for Multimodal Retrieval-Augmented Visual Culture Understanding

DualEdit: Mitigating Safety Fallback in LLM Backdoor Editing via Affirmation-Refusal Regulation

Bounds of Chain-of-Thought Robustness: Reasoning Steps, Embed Norms, and Beyond

InComeS: Integrating Compression and Selection Mechanisms into LLMs for Efficient Model Editing

Learning-to-Context Slope: Evaluating In-Context Learning Effectiveness Beyond Performance Illusions

Contrastive Weak-to-Strong Generalization

E2EDev: Benchmarking Large Language Models in End-to-End Software Development Task

When Less Language is More: Language-Reasoning Disentanglement Makes LLMs Better Multilingual Reasoners

The Rise of Parameter Specialization for Knowledge Storage in Large Language Models

Lens: Rethinking Multilingual Enhancement for Large Language Models

PEARL: Towards Permutation-Resilient LLMs

Towards Comprehensive and Efficient Post Safety Alignment of Large Language Models via Safety Patching