Liang Ding

Senior Staff Researcher@Alibaba Group·中国·OpenReview

研究方向

Language Agent · Large Language Model · Efficient Transfer Learning · Self-Supervised Learning · Foundation Models · Natural Language Processing · Neural Machine Translation · Statistical Machine Translation

Liang Ding

Memory Efficient Fine-Tuning of LLMs via Forward-Only Hessian-Free Coordinate Descent

Better, Faster: Harnessing Self-Improvement in Large Reasoning Models

TeZO: Empowering the Low-Rankness on the Temporal Dimension in the Zeroth-Order Optimization for Fine-tuning LLMs

Retrieval-Augmented Perception: High-resolution Image Perception Meets Visual RAG

The Energy Loss Phenomenon in RLHF: A New Perspective on Mitigating Reward Hacking

Self-Evolving Pseudo-Rehearsal for Catastrophic Forgetting with Task Similarity in LLMs

Short-length Adversarial Training Helps LLMs Defend Long-length Jailbreak Attacks: Theoretical and Empirical Evidence

Layer as Puzzle Pieces: Compressing Large Language Models through Layer Concatenation

MMA: Benchmarking Multi-Modal Large Language Model in Ambiguity Contexts

DynamicKV: Task-Aware Adaptive KV Cache Compression for Long Context LLMs

Memory Efficient Block Coordinate Descent Method for Forward-Only Second-Order Finetuning of LLM Models

Towards Efficient Mixture of Experts: A Holistic Study of Compression Techniques

Simultaneous Computation and Memory Efficient Zeroth-Order Optimizer for Fine-Tuning Large Language Models

Memory-Efficient Block Coordinate Descent for Hessian-Informed Zeroth-Order Optimizer