Papers by Zhang Dongjie

5 papers
PyramidInfer: Pyramid KV Cache Compression for High-throughput LLM Inference (2024.findings-acl)

Copied to clipboard

Challenge: Existing methods to reduce memory usage for large language models neglect inter-layer dependency between layers and huge memory consumption in pre-computation.
Approach: They propose a method that compresses the KV cache by layer-wise retaining crucial context.
Outcome: The proposed method reduces memory usage by layer-wise retaining crucial context . it can improve 2.2x throughput compared to Accelerate with over 54% memory reduction .
Blind Spot Navigation in Large Language Model Reasoning with Thought Space Explorer (2026.findings-eacl)

Copied to clipboard

Challenge: Existing studies show that large language models have strong reasoning capabilities through chain-structured methods.
Approach: They propose a framework for navigating and expanding thought structures to overcome blind spots in LLM reasoning.
Outcome: The proposed framework overcomes blind spots in large language models by expanding thought structures . the proposed framework improves accuracy of the final answer and intermediate reasoning steps .
Multimodal Knowledge Learning for Named Entity Disambiguation (2022.findings-emnlp)

Copied to clipboard

Challenge: Existing attempts to model multimodal information at the knowledge level are lacking multimodal annotation data against the large-scale unlabeled corpus.
Approach: They propose to use multimodal knowledge learning to link ambiguous mentions with textual and visual contexts to a predefined knowledge graph.
Outcome: The proposed method achieves improvements over the state-of-the-art methods on two public MNED datasets.
Learning Better Masking for Better Language Model Pre-training (2023.acl-long)

Copied to clipboard

Challenge: Existing PrLMs adopt a Random-Token Masking strategy with a fixed masking ratio and different contents are masked by an equal probability throughout the training.
Approach: They propose two scheduled masking approaches that adaptively tune masking ratio and masked content in different training stages, which improves pre-training efficiency and effectiveness.
Outcome: The proposed methods improve the pre-training efficiency and effectiveness on the downstream tasks.
Rectifying the Emotional Flow: Aligning Priors and Dynamic Guidance for High-Arousal Text-to-Speech (2026.acl-long)

Copied to clipboard

Challenge: Existing systems suffer from linguistic collapse when pursuing high intensity or fail to meet target emotional levels.
Approach: They propose an inference framework that introduces a neutral prosody bias and a uniform Classifier-Free Guidance that distorts the acoustic manifold, leading to artifacts.
Outcome: The proposed framework achieves superior linguistic accuracy and expressiveness without model retraining.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations