Papers by Zhen-Hua Ling
UniVocal: Unified Speech-Singing Code-Switching Synthesis (2026.acl-long)
Copied to clipboard
| Challenge: | Existing systems cannot automatically determine when to switch between modes based on text content. |
| Approach: | They propose a unified framework that implicitly infers vocal modes from text context to pioneer SCS Synthesis. |
| Outcome: | The proposed framework infers vocal modes solely from text context to pioneer SCS Synthesis. |
Constraining Sequential Model Editing with Editing Anchor Compression (2025.findings-naacl)
Copied to clipboard
| Challenge: | Large language models (LLMs) exhibit hallucinations due to incorrect or outdated knowledge embedded in their parameters. |
| Approach: | They propose a framework to constrain the deviation of the parameter matrix during sequential editing by selecting editing anchors that are important in encoding new relations without deviating too much from the original matrix. |
| Outcome: | The proposed framework minimizes deviations caused by model editing while retaining over 70% of the general abilities. |
Retrieving, Rethinking and Revising: The Chain-of-Verification Can Improve Retrieval Augmented Generation (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Recent Retrieval Augmented Generation (RAG) aims to enhance Large Language Models . however, such approach can generate inconsistent answer with external references . |
| Approach: | They propose to integrate the verification module into the RAG to improve external retrieval correctness and internal generation consistency. |
| Outcome: | The proposed model can significantly surpass the state-of-the-art baselines using different LLM backbones. |
Multi-Level Matching and Aggregation Network for Few-Shot Relation Classification (P19-1)
Copied to clipboard
| Challenge: | Existing methods for few-shot relation classification use supervised training, but lack of large-scale manually labeled data. |
| Approach: | They propose a multi-level matching and aggregation network (MLMAN) for few-shot relation classification. |
| Outcome: | The proposed model achieves state-of-the-art performance on the FewRel dataset. |
RPO: Retrieval Preference Optimization for Robust Retrieval-Augmented Generation (2025.acl-long)
Copied to clipboard
| Challenge: | Large language models struggle to evaluate the correctness of non-parametric knowledge when it differs from internal memorization, leading to knowledge conflicts during response generation. |
| Approach: | They propose a lightweight alignment method to leverage multi-source knowledge based on retrieval relevance. |
| Outcome: | Experiments on four datasets show that the proposed method outperforms RAG by 4-10% in accuracy without any extra component. |
X-ACE: Explainable and Multi-factor Audio Captioning Evaluation (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing evaluation metrics for automated audio captioning only provide an overall score . current evaluation checklists are inadequate to characterize the nuanced differences . |
| Approach: | They propose an explainable and multi-factor audio captioning evaluation paradigm . they define sound event, source, attribute and relation as four factors tailored for the audio description . |
| Outcome: | The proposed evaluation paradigm improves the quality of audio captions . it can detect mismatches and align with human perception, the authors show . |
Distant Supervision Relation Extraction with Intra-Bag and Inter-Bag Attentions (N19-1)
Copied to clipboard
| Challenge: | Existing methods to extract relational data generated by distant supervision generate noisy training data. |
| Approach: | They propose a neural relation extraction method to deal with noisy training data generated by distant supervision. |
| Outcome: | Experimental results show that the proposed method is more accurate than state-of-the-art methods on the New York Times dataset. |
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. |
Select to Know: An Internal-External Knowledge Self-Selection Framework for Domain-Specific Question Answering (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) perform well in general QA but often struggle in domain-specific scenarios. |
| Approach: | They propose a framework that internalizes domain knowledge through internal-external knowledge self-selection and selective supervised fine-tuning. |
| Outcome: | The proposed framework outperforms existing methods and matches domain-pretrained LLMs with significantly lower cost. |
GenesisFunc: Multi-Agent Data Generation for Accurate and Generalizable Function-Calling (2026.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) extend their capabilities through function-calling (FC) however, obtaining and annotating real function-called data is challenging, and synthetic data from existing pipelines suffers from unreliable APIs, limited tool scalability, insufficient diversity, and weak quality control. |
| Approach: | They propose a pipeline for generating FC training data using reliable tools and a multi-agent framework that supports a dialogue generation system that produces conversations spanning diverse scenarios. |
| Outcome: | The proposed pipeline outperforms open-source models in in-domain FC performance and out-of-domain generalization while reaching FC capabilities comparable to some of the latest API-based models. |
HeterMPC: A Heterogeneous Graph Neural Network for Response Generation in Multi-Party Conversations (2022.acl-long)
Copied to clipboard
| Challenge: | Experimental results show that HeterMPC outperforms various baseline models for response generation in multi-party conversations. |
| Approach: | They propose a heterogeneous graph-based neural network for response generation in multi-party conversations which models the semantics of utterances and interlocutors simultaneously with two types of nodes in a graph. |
| Outcome: | The proposed model outperforms baseline models on the Ubuntu Internet Relay Chat (IRC) channel. |
TegTok: Augmenting Text Generation via Task-specific and Open-world Knowledge (2022.findings-acl)
Copied to clipboard
| Challenge: | Generating natural and informative texts has been a long-standing problem in NLP. |
| Approach: | They propose to augment TExt Generation via Task-specific and Open-world Knowledge in a unified framework. |
| Outcome: | The proposed model can learn what and how to generate on two text generation tasks. |
MADNet: Maximizing Addressee Deduction Expectation for Multi-Party Conversation Generation (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for multi-party conversations rely on addressee labels and can only be applied to an ideal setting where addresses are missing. |
| Approach: | They propose a method that maximizes addressee deduction expectation in heterogeneous graph neural networks for MPC generation. |
| Outcome: | The proposed method outperforms baseline models on Ubuntu IRC channel benchmarks on the task of MPC generation under a common and challenging setting where addressee labels are missing. |
Model Editing Harms General Abilities of Large Language Models: Regularization to the Rescue (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods that edit large language models with updated knowledge can cause side effects on the general abilities of LLMs such as reasoning, natural language inference, and question answering. |
| Approach: | They propose to regularize the edit update weights by imposing constraints on their complexity based on the RElative Change in weighT. |
| Outcome: | The proposed method can significantly mitigate the side effects while maintaining over 94% editing performance. |
Hybrid semi-Markov CRF for Neural Sequence Labeling (P18-2)
Copied to clipboard
| Challenge: | Existing conditional random fields (CRFs) use hand-crafted features to perform sequence labeling tasks. |
| Approach: | They propose to use semi-Markov conditional random fields for neural sequence labeling in natural language processing to extract features from segments instead of words. |
| Outcome: | The proposed model achieves state-of-the-art when no external knowledge is used. |
RISE: Reasoning Enhancement via Iterative Self-Exploration in Multi-hop Question Answering (2025.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) excel in many areas but face challenges with complex reasoning tasks, such as Multi-Hop Question Answering (MHQA). |
| Approach: | They propose a framework to enhance models’ reasoning capability through iterative self-exploration that addresses key errors in MHQA tasks such as Evidence Aggregation and Reasoning Decomposition. |
| Outcome: | Extensive experiments on multiple MHQA benchmarks show that the proposed framework significantly improves reasoning accuracy and task performance. |
Symbolization, Prompt, and Classification: A Framework for Implicit Speaker Identification in Novels (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for speaker identification in novel dialogues are limited to handling explicit narrative patterns and complex cases. |
| Approach: | They propose a framework which identifies implicit speakers in novels via symbolization, prompt, and classification. |
| Outcome: | The proposed framework outperforms existing methods by 4.8% accuracy on the web novel collection, which reduces 47% of speaker identification errors, and outperfies the emerging ChatGPT. |
Neural Natural Language Inference Models Enhanced with External Knowledge (P18-1)
Copied to clipboard
| Challenge: | Existing datasets that allow for complex models to be trained are limited . if data is not available, can machines learn all knowledge needed to perform natural language inference? |
| Approach: | They propose to enrich neural natural language inference models with external knowledge . they propose to use this knowledge to build NLI models to leverage it . |
| Outcome: | The proposed models improve on the SNLI and MultiNLI datasets. |
Enhancing Sentence Embedding with Generalized Pooling (C18-1)
Copied to clipboard
| Challenge: | Existing methods for learning sentence embedding are limited, but still need to be improved. |
| Approach: | They propose a vector-based multi-head attention model that uses special cases of max pooling, mean pooling and scalar self-attention. |
| Outcome: | The proposed model improves on natural language inference, author profiling, and sentiment classification tasks. |
Is ChatGPT a Good Multi-Party Conversation Solver? (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are powerful tools for multi-party conversations, but their capacity to handle multi-parties remains unexplored. |
| Approach: | They propose to evaluate ChatGPT and GPT-4's zero-shot learning capabilities within the context of multi-party conversations (MPCs) they also propose to incorporate MPC structures, encompassing both speaker and addressee architecture. |
| Outcome: | The proposed models perform poorly on a number of MPC tasks while GPT-4 performs well on speaker and addressee architecture. |
Dually Interactive Matching Network for Personalized Response Selection in Retrieval-Based Chatbots (D19-1)
Copied to clipboard
| Challenge: | Existing models for personalized dialogues rank responses according to their semantic relevance with the given context. |
| Approach: | They propose a dually interactive matching network (DIM) for presenting personalities of dialogue agents in retrieval-based chatbots. |
| Outcome: | The proposed model outperforms the existing model by 14.5% and 27.7% on a PERSONA-CHAT dataset. |