Papers by Lingzhi Wang
DPDLLM: A Black-box Framework for Detecting Pre-training Data from Large Language Models (2024.findings-acl)
Copied to clipboard
Baohang Zhou, Zezhong Wang, Lingzhi Wang, Hongru Wang, Ying Zhang, Kehui Song, Xuhui Sui, Kam-Fai Wong
| Challenge: | Existing methods to detect pretraining data from large language models are unrealistic to them. |
| Approach: | They propose to detect pre-training data from LLM in a black-box way by using GPT-2 as reference model and feed it with sequence probabilities to detect whether it was used to train it. |
| Outcome: | The proposed framework outperforms existing methods on the benchmark datasets and shows that it is effective on different popular LLMs. |
Modeling Multi-Dimensional Cognitive States in Large Language Models under Cognitive Crowding (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing Large Language Models (LLMs) mainly address isolated tasks such as emotion analysis or stance detection. |
| Approach: | They propose a large-scale model that combines large-level annotations with hyperbolic space to model human cognitive states. |
| Outcome: | The proposed model outperforms baseline models on cognitive dimensions on single dimension tasks while retaining strong hierarchical structure. |
Learning When and What to Quote: A Quotation Recommender System with Mutual Promotion of Recommendation and Generation (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing quotation recommendation system focuses on what to quote, but ignores whether or when to quote. |
| Approach: | They propose a framework that learns to predict when to quote and what to quote jointly. |
| Outcome: | The proposed framework achieves significantly better performance than baselines on two datasets. |
LLM-REDIAL: A Large-Scale Dataset for Conversational Recommender Systems Created from User Behaviors with LLMs (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing CRS datasets suffer from data inextensibility and semantic inconsistency . |
| Approach: | They introduce the LLM-REDIAL dataset to facilitate the research in CRS by leveraging large language models to generate high-quality dialogues. |
| Outcome: | The proposed dataset is the largest multi-domain CRS dataset which consists of 47.6k multi-turn dialogues with 482.6k utterances across 4 domains. |
RecInDial: A Unified Framework for Conversational Recommendation with Pretrained Language Models (2022.aacl-main)
Copied to clipboard
| Challenge: | Existing generative methods to recommend items are shallowly integrated into the model training and have poor chit-chat ability. |
| Approach: | They propose a framework that integrates recommendation into the dialog generation by introducing a vocabulary pointer. |
| Outcome: | The proposed framework outperforms the state-of-the-art models on a benchmark dataset. |
Coupling Global and Local Context for Unsupervised Aspect Extraction (D19-1)
Copied to clipboard
| Challenge: | Existing studies on aspect extraction focus on sequence tagging models trained on human-annotated data. |
| Approach: | They propose a novel neural model capable of coupling global and local representations to discover aspect words by combining global and locale contexts. |
| Outcome: | The proposed model outperforms state-of-the-art models on laptop and restaurant reviews on two benchmarks. |
IndiVec: An Exploration of Leveraging Large Language Models for Media Bias Detection with Fine-Grained Bias Indicators (2024.findings-eacl)
Copied to clipboard
| Challenge: | Existing studies on social media bias detection focus on fine-tuning models specific to particular datasets and testing them on corresponding test sets. |
| Approach: | They propose a general bias detection framework, IndiVec, built upon large language models and vector databases. |
| Outcome: | The proposed framework outperforms baseline methods on four political bias datasets and provides explicit top-k indicators to interpret bias predictions. |
Investigating Bias in LLM-Based Bias Detection: Disparities between LLMs and Human Perception (2025.coling-main)
Copied to clipboard
| Challenge: | Detecting media bias is critical due to the spread of misinformation and disinformation on social media platforms. |
| Approach: | They investigate the presence and nature of bias within large language models and its consequential impact on media bias detection. |
| Outcome: | The proposed debiasing strategies include prompt engineering and model fine-tuning. |
From Words to Pixels: A Comprehensive Survey on Large Language Models in Visual Segmentation (2026.acl-long)
Copied to clipboard
| Challenge: | Visual segmentation with instruction has been a challenging task for many years . large language models and large multimodal models have spurred a new wave of research . |
| Approach: | They review recent works in LLM-based visual segmentation and analyze their architectural innovations, training strategies, and benchmark performance. |
| Outcome: | The present study reviews the most recent works in LLM-driven visual segmentation . it identifies key challenges and promising future directions . |
Strategize Before Teaching: A Conversational Tutoring System with Pedagogy Self-Distillation (2023.findings-eacl)
Copied to clipboard
| Challenge: | Conversational tutoring systems (CTSs) aim to help students master educational material with natural language interaction in the form of a dialog. |
| Approach: | They propose to jointly predict teaching strategies and generate tutor responses accordingly to help students master educational material through dialog. |
| Outcome: | The proposed framework is based on three dialog tutoring datasets and is more realistic than previous models that generate responses given the strategies as input. |
Continuity of Topic, Interaction, and Query: Learning to Quote in Online Conversations (2020.emnlp-main)
Copied to clipboard
| Challenge: | Quotations are crucial for successful explanations and persuasions in interpersonal communications. |
| Approach: | They propose to use an encoder-decoder neural framework to continue the context with a quotation via language generation to capture latent topics, interactions with the dialogue history, and coherence to the existing contents. |
| Outcome: | The proposed model outperforms state-of-the-art models on two large-scale datasets in English and Chinese and shows that topic, interaction, and query consistency are helpful to learn how to quote in online conversations. |
Cautious Next Token Prediction (2025.findings-acl)
Copied to clipboard
Yizhou Wang, Lingzhi Zhang, Yue Bai, Mang Tik Chiu, Zhengmian Hu, Mingyuan Zhang, Qihua Dong, Yu Yin, Sohrab Amirghodsi, Yun Fu
| Challenge: | Existing methods for decoding autoregressive models are temperature scaling and nucleus sampling to balance diversity and coherence. |
| Approach: | They propose a training-free decoding strategy that uses a model with a low perplexity score to select the trial with the lowest perplexities as the most probable and reliable path. |
| Outcome: | The proposed approach outperforms existing standard decoding strategies consistently by a clear margin. |
Context-Sensitive Generation of Open-Domain Conversational Responses (C18-1)
Copied to clipboard
| Challenge: | Existing studies on single-turn conversation generation focus on coherence and context-sensitive generation of open-domain conversational responses. |
| Approach: | They propose static and dynamic attention based approaches for context-sensitive generation of open-domain conversational responses. |
| Outcome: | The proposed model outperforms all baselines on automatic and human evaluation on two public datasets. |
LLMEdgeRefine: Enhancing Text Clustering with LLM-Based Boundary Point Refinement (2024.emnlp-main)
Copied to clipboard
| Challenge: | LLMEdgeRefine is an iterative clustering method enhanced by large language models . existing clustering methods struggle with domain-specific fine-tuning and outliers . |
| Approach: | They propose an iterative clustering method enhanced by large language models focusing on edge points refinement . authors propose to use LLMs to iterate clusters and iterating to improve semantic coherence . |
| Outcome: | The proposed method outperforms state-of-the-art methods and offers robustness, adaptability, and cost-efficiency for diverse text clustering applications. |
Opportunities and Challenges in Neural Dialog Tutoring (2023.eacl-main)
Copied to clipboard
| Challenge: | Existing approaches to designing dialog tutors have been challenging . current approaches perform poorly in constrained learning scenarios, authors find . |
| Approach: | They analyze dialog tutoring models using automatic and human evaluations to understand the new opportunities brought by dialog tutors. |
| Outcome: | The proposed models perform poorly in less constrained learning scenarios, the authors show . they find large number of model reasoning errors in 45% of conversations . |
FedCSR: A Federated Framework for Multi-Platform Cross-Domain Sequential Recommendation with Dual Contrastive Learning (2025.coling-main)
Copied to clipboard
Dongyi Zheng, Hongyu Zhang, Jianyang Zhai, Lin Zhong, Lingzhi Wang, Jiyuan Feng, Xiangke Liao, Yonghong Tian, Nong Xiao, Qing Liao
| Challenge: | Existing federated frameworks for cross-domain sequential recommendation rely on user alignment, which increases communication costs and privacy risks. |
| Approach: | They propose a federated cross-domain sequential recommendation framework that eliminates the need for user alignment between platforms. |
| Outcome: | The proposed framework eliminates the need for user alignment between platforms. |
KGA: A General Machine Unlearning Framework Based on Knowledge Gap Alignment (2023.acl-long)
Copied to clipboard
| Challenge: | Recent legislation of the "right to be forgotten" has led to the interest in machine unlearning . MU can be used to forget specific training instances as if they have never existed . |
| Approach: | They propose a general unlearning framework called KGA to induce forgetfulness . they propose several unlearning evaluation metrics with pertinence . |
| Outcome: | The proposed framework improves on large-scale datasets and provides insight into unlearning for NLP tasks. |
Quotation Recommendation and Interpretation Based on Transformation from Queries to Quotations (2021.acl-short)
Copied to clipboard
| Challenge: | Existing models of quotation recommendation ignore the relationship between quotations and queries. |
| Approach: | They propose a transformation matrix that directly maps quotations to quotation representations. |
| Outcome: | The proposed model outperforms state-of-the-art models on two datasets in English and Chinese. |
PACAR: Automated Fact-Checking with Planning and Customized Action Reasoning Using Large Language Models (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing studies rely on idealized "gold" evidence for predictions, which is unrealistic due to its limited availability in real-world scenarios. |
| Approach: | They propose a fact-checking framework based on planning and customized action reasoning using LLMs. |
| Outcome: | The proposed framework outperforms baseline methods across three datasets and with varying complexity levels. |
Re-entry Prediction for Online Conversations via Self-Supervised Learning (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Existing work on re-entry prediction ignores conversation thread patterns and repeated engagement of target users. |
| Approach: | They propose to use conversation thread patterns to predict whether a user will come back to a conversation they once participated in to train a model on labels that are automatically derived from the data. |
| Outcome: | The proposed task outperforms the state-of-the-art models on two social media datasets with fewer parameters and faster convergence. |
Cognitive Policy-Driven LLM for Diagnosis and Intervention of Cognitive Distortions in Emotional Support Conversation (2026.acl-long)
Copied to clipboard
| Challenge: | Existing models for ESC ignore cognitive distortions in help-seekers' expressions . current models provide basic emotional comfort, rather than helping help- seekers address psychological distress at a deeper cognitive level. |
| Approach: | They propose a Large Language Model framework to enhance LLMs' ability to diagnose and intervene cognitive distortions in help-seekers. |
| Outcome: | The proposed framework outperforms 15 state-of-the-art baselines in terms of distortion diagnosis accuracy, intervention strategy effectiveness, and safety risk control. |