Papers by Zhengyan Zhang

28 papers
RiOT: Efficient Prompt Refinement with Residual Optimization Tree (2025.acl-long)

Copied to clipboard

Challenge: Existing methods for automatic prompt optimization face two challenges: lack of diversity and semantic drift.
Approach: They propose a framework for automatic prompt optimization that iteratively refines prompts through text gradients and selects the best prompt using perplexity.
Outcome: The proposed framework outperforms existing prompt optimization methods and manual prompting on commonsense, mathematical, logical, temporal, and semantic reasoning benchmarks.
Automatic Label Sequence Generation for Prompting Sequence-to-sequence Models (2022.coling-1)

Copied to clipboard

Challenge: Prompting has shown to be sample efficient compared to fine-tuning with pre-trained models.
Approach: They propose a fully automatic prompting method that uses natural language prompts on sequence-to-sequence models and a beam search method to generate a large amount of label sequence candidates.
Outcome: The proposed method significantly outperforms other no-manual-design methods on single label words and generates large amount of label sequence candidates.
Knowledge Inheritance for Pre-trained Language Models (2022.naacl-main)

Copied to clipboard

Challenge: Existing large-scale pre-trained language models are mainly trained from scratch individually, ignoring that many well-taught PLMs are available.
Approach: They propose a pre-training framework called knowledge inheritance and propose auxiliary supervision to efficiently learn larger PLMs.
Outcome: The proposed framework can be used to train large-scale language models with huge parameters and a large dataset can be adapted to domain adaptation and knowledge transfer.
EventRAG: Enhancing LLM Generation with Event Knowledge Graphs (2025.acl-long)

Copied to clipboard

Challenge: Existing approaches to text generation often neglect event structures that shape real-world narratives.
Approach: They propose a framework that integrates structured event semantics with iterative retrieval and inference to enhance text generation.
Outcome: Experiments on UltraDomain and MultiHopRAG show that the proposed framework outperforms baseline RAG systems in generation effectiveness, logical consistency, and multi-hop reasoning accuracy.
KEPLER: A Unified Model for Knowledge Embedding and Pre-trained Language Representation (2021.tacl-1)

Copied to clipboard

Challenge: Existing language representation models (PLMs) cannot capture factual knowledge from text.
Approach: They propose a unified model for Knowledge Embedding and Pre-trained LanguagERepresentation which integrates factual knowledge into PLMs and produces effective text-enhanced KE with the strong PLM.
Outcome: The proposed model improves on existing pre-trained language representation models and improves their performance on various NLP tasks.
Prompt Tuning for Discriminative Pre-trained Language Models (2022.findings-acl)

Copied to clipboard

Challenge: Recent studies have shown promising results of prompt tuning in stimulating pre-trained language models (PLMs) for natural language processing tasks.
Approach: They propose a prompt tuning framework that reformulates NLP tasks into a discriminative language modeling problem.
Outcome: The proposed framework improves on text classification and question answering tasks and prevents unstable tuning problems in low-resource settings.
Sub-Character Tokenization for Chinese Pretrained Language Models (2023.tacl-1)

Copied to clipboard

Challenge: Existing tokenization methods for Chinese PLMs treat each character as an indivisible token, but ignore the unique feature of the writing system where additional linguistic information exists below the character level.
Approach: They propose to encode Chinese characters into short sequences and construct Chinese vocabulary based on the encoded text.
Outcome: The proposed tokenizers can tokenize inputs into much shorter sequences, improving computational efficiency.
READIN: A Chinese Multi-Task Benchmark with Realistic and Diverse Input Noises (2023.acl-long)

Copied to clipboard

Challenge: Existing benchmarks for Chinese inputs often lack a realistic representation of real-world noises.
Approach: They construct a Chinese multi-task benchmark with REalistic and Diverse input noises . they use pinyin input and speech input to recruit speakers from diverse dialects based on their inputs - a feature that is important for Chinese NLP benchmarks if it is implemented in real-world applications.
Outcome: The proposed benchmarks are based on four different tasks and are designed to maximize diversity.
Emergent Modularity in Pre-trained Transformers (2023.findings-acl)

Copied to clipboard

Challenge: Existing studies on pre-trained Transformers show that they learn fine-grained neuron functions.
Approach: They examine the presence of modularity in pre-trained Transformers . they focus on Mixture-of-Experts, a promising candidate for modularity .
Outcome: The proposed structure stabilizes at the early stage, which is faster than neuron stabilization.
BMInf: An Efficient Toolkit for Big Model Inference and Tuning (2022.acl-demo)

Copied to clipboard

Challenge: Recent years, pre-trained language models (PLMs) have achieved promising results on various NLP tasks.
Approach: They propose an open-source toolkit for big model inference and tuning which can support big model tuning at extremely low computation cost.
Outcome: The proposed toolkit can support big model inference and tuning at extremely low computation cost.
Train No Evil: Selective Masking for Task-Guided Pre-Training (2020.emnlp-main)

Copied to clipboard

Challenge: Pre-trained language models can't capture domain-specific and task-specific patterns because of the task-agnostic pre-training stage.
Approach: They propose a task-guided pre-training stage with selective masking between general pre-train and fine-tuning to learn domain-specific patterns.
Outcome: The proposed method can achieve comparable or even better performance with less than 50% of computation cost.
UniSpeaker: A Unified Approach for Multimodality-driven Speaker Generation (2025.findings-emnlp)

Copied to clipboard

Challenge: a new framework for speaker generation is proposed to enable multimodal speaker generation . multimodal cues such as visual appearance, textual descriptions, and other biometric signals are still in its early stages.
Approach: a new framework is proposed to enable multimodal speaker generation . the framework uses self-distillation to apply speaker disentanglement to speech generation a model is developed .
Outcome: The proposed framework is the first to support unified voice generation from arbitrary modality combinations.
MoEfication: Transformer Feed-forward Layers are Mixtures of Experts (2022.findings-acl)

Copied to clipboard

Challenge: Recent work has shown that feed-forward networks (FFNs) in pre-trained Transformers are a key component, storing various linguistic and factual knowledge.
Approach: They propose to convert a model into its MoE version with the same parameters and build expert routers to decide which experts will be used for each input.
Outcome: The proposed model can use 10% to 30% of FFN parameters while maintaining over 95% original performance.
Robust and Scalable Model Editing for Large Language Models (2024.lrec-main)

Copied to clipboard

Challenge: Existing methods that ignore contextual knowledge fail to reliably fall back to parametric knowledge when presented with irrelevant context.
Approach: They propose to use contextual knowledge to update and correct LLMs' knowledge by in-context editing instead of retraining.
Outcome: The proposed method outperforms current state-of-the-art methods by a large margin on a dataset that contains irrelevant questions.
Plug-and-Play Knowledge Injection for Pre-trained Language Models (2023.acl-long)

Copied to clipboard

Challenge: Existing knowledge injection methods are not suitable for enhancing pre-trained language models with external knowledge bases.
Approach: They propose a plug-and-play knowledge injection method where knowledge bases are injected into frozen existing downstream models by a knowledge plugin.
Outcome: The proposed method improves the performance of knowledge injection on knowledge-driven tasks while keeping model parameters frozen.
BMCook: A Task-agnostic Compression Toolkit for Big Models (2022.emnlp-demos)

Copied to clipboard

Challenge: Existing efforts to compress medium-sized models for specific tasks have limited results.
Approach: They propose a task-agnostic compression toolkit for big models that implements quantization, pruning, distillation and MoEfication methods.
Outcome: The proposed tool improves performance on a model with 3 billion parameters by 12x . it also outperforms the original model on three typical NLP benchmarks.
Finding Skill Neurons in Pre-trained Transformer-based Language Models (2022.emnlp-main)

Copied to clipboard

Challenge: Pre-trained language models have demonstrated superior performance on various natural language processing tasks.
Approach: They find that after prompt tuning, some neurons encode task-specific skills . they also show that skill neurons are most likely generated in pre-training .
Outcome: The neurons are highly predictive of task labels after prompt tuning for specific tasks.
Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Trigger (2021.acl-long)

Copied to clipboard

Challenge: Existing methods for textual backdoor attacks insert additional contents into normal samples as triggers, causing detection and blocking of backdoors.
Approach: They propose to use syntactic structure as trigger in textual backdoor attacks . they propose to achieve similar attack performance but have higher invisibility .
Outcome: The proposed method achieves almost 100% success rate but has higher invisibility and stronger resistance to defenses than the insertion-based methods.
LLM-as-a-Coauthor: Can Mixed Human-Written and Machine-Generated Text Be Detected? (2024.findings-naacl)

Copied to clipboard

Challenge: Current research focuses on purely MGT detection without adequately addressing mixed scenarios including AI-revised Human-Written Text (HWT) and human-revealed MGT.
Approach: They define mixtext, a form of mixed text involving both AI and human-generated content, and then use a MixSet dataset to assess their effectiveness.
Outcome: The proposed detectors struggle to identify mixtext, particularly in dealing with subtle modifications and style adaptability.
Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention (2025.acl-long)

Copied to clipboard

Challenge: Long-context modeling is crucial for next-generation language models, but high computational cost of standard attention mechanisms poses significant computational challenges.
Approach: They propose a natively trained Sparse Attention mechanism that integrates algorithms with hardware-aligned optimizations to achieve efficient long-context modeling.
Outcome: The proposed model maintains or exceeds Full Attention models across general benchmarks, long-context tasks, and instruction-based reasoning.
Variator: Accelerating Pre-trained Models with Plug-and-Play Compression Modules (2023.findings-emnlp)

Copied to clipboard

Challenge: Large language models (LLMs) have been successful on NLP tasks but require huge parameter sizes and computational resources.
Approach: They propose a parameter-efficient acceleration method that enhances computational efficiency through plug-and-play compression plugins.
Outcome: The proposed method saves 53% computational costs using only 0.9% additional parameters with a performance drop of less than 2%.
Learning Implicit Sentiment in Aspect-based Sentiment Analysis with Supervised Contrastive Pre-Training (2021.emnlp-main)

Copied to clipboard

Challenge: Recent studies have focused on identifying the sentiment polarity of aspects in product reviews.
Approach: They propose to use supervised Contrastive Pre-Training to learn implicit sentiment . they propose to train large-scale sentiment-annotated corpora from in-domain language resources .
Outcome: The proposed model achieves state-of-the-art performance on SemEval2014 benchmarks and comprehensively validates its effectiveness on learning implicit sentiment.
Uncertainty-Aware Label Refinement for Sequence Labeling (2020.emnlp-main)

Copied to clipboard

Challenge: Conditional random fields (CRF) for label decoding have been a problem for many tasks.
Approach: They propose a two-stage label decoding framework that model long-term label dependencies while being much more computationally efficient.
Outcome: The proposed method outperforms the CRF-based methods and greatly accelerates the inference process.
Better Robustness by More Coverage: Adversarial and Mixup Data Augmentation for Robust Finetuning (2021.findings-acl)

Copied to clipboard

Challenge: Pretrained language models perform poorly under adversarial attacks due to the large search space.
Approach: They propose a method to cover a much larger proportion of the attack search space by adding textual adversarial examples during training.
Outcome: The proposed method covers a much larger proportion of the attack search space.
Plug-and-Play Document Modules for Pre-trained Models (2023.acl-long)

Copied to clipboard

Challenge: Large-scale pre-trained models have been widely adopted for document-oriented NLP tasks, such as question answering.
Approach: They propose to decouple document encoding from downstream tasks by introducing a document plugin into the backbone of a PTM.
Outcome: The proposed model can encode documents once and for all across different scenarios.
ERNIE: Enhanced Language Representation with Informative Entities (P19-1)

Copied to clipboard

Challenge: Existing pre-trained language models rarely consider incorporating knowledge graphs (KGs) Existing models capture rich semantic patterns from plain text and can be fine-tuned to improve performance of NLP tasks.
Approach: They propose to incorporate knowledge graphs into pre-trained language models to enhance language representation with external knowledge.
Outcome: The proposed model can take full advantage of lexical, syntactic, and knowledge information simultaneously.
TextFlint: Unified Multilingual Robustness Evaluation Toolkit for Natural Language Processing (2021.acl-demo)

Copied to clipboard

Challenge: Existing approaches to textual robustness evaluation focus on slightly modifying the input data, which maintains the original meaning and results in a different prediction.
Approach: They propose a multilingual robustness evaluation toolkit for NLP that integrates universal text transformations, task-specific transformations and adversarial attack.
Outcome: The toolkit includes universal text transformation, task-specific transformation, adversarial attack, subpopulation, and their combinations to provide comprehensive robustness analyses.
ProSparse: Introducing and Enhancing Intrinsic Activation Sparsity within Large Language Models (2025.coling-main)

Copied to clipboard

Challenge: Activation sparsity is a promising paradigm for accelerating model inference . few large language models achieve high activation spar and comparable performance .
Approach: They propose a method to achieve activation sparsity and acceleration in large language models . they introduce ReLU activation and adopt progressive sparse regularization .
Outcome: The proposed method achieves high activation sparsity and comparable model performance.

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