Papers by Yufeng Zhang

A Learning-Exploring Method to Generate Diverse Paraphrases with Multi-Objective Deep Reinforcement Learning (2020.coling-main)

Challenge:	Existing methods to answer subjective questions about products are often imbalanced across product domains.
Approach:	They propose a domain-adaptive model that integrates multiple viewpoints into a good answer by integrating these heterogeneous and inconsistent viewpoints.
Outcome:	The proposed model integrates multiple viewpoints into a single answer span and is able to integrate them into the answer.

Copied to clipboard

Mingtong Liu, Erguang Yang, Deyi Xiong, Yujie Zhang, Yao Meng, Changjian Hu, Jinan Xu, Yufeng Chen

Can Multi-agent Help Disambiguation in Multi-domain Translation? (2026.findings-acl)

Challenge:	Paraphrase generation is of great importance for many downstream tasks in natural language processing.
Approach:	They propose a method to generate sentences as learning objectives from the learned data distribution and employ reinforcement learning to combine these new learning objectives for model training.
Outcome:	The proposed method gains significant diversity and improves generation quality over state-of-the-art datasets.

Copied to clipboard

Zhibo Man, Shaoyang Xu, Yujie Zhang, Yi Feng, Yuanmeng Chen, Yufeng Chen, Xu Jinan, Wenxuan Zhang

Machine Translation

Challenge:	Existing multi-agent systems have shown strong potential for machine translation (MT) but their performance in multidomain translation remains unsatisfactory due to cross-domain word ambiguity .
Approach:	They propose a multi-agent collaborative disambiguation framework for MDT that leverages the collaborative capabilities of LLMs for disambiguations.
Outcome:	The proposed framework improves translation performance across multiple domains and improves disambiguation accuracy.

Agent-in-the-Loop: A Data Flywheel for Continuous Improvement in LLM-based Customer Support (2025.emnlp-industry)

Copied to clipboard

Cen Zhao, Tiantian Zhang, Hanchen Su, Yufeng Zhang, Shaowei Su, Mingzhi Xu, Yu Liu, Wei Han, Jeremy Werner, Claire Na Cheng, Yashar Mehdad

Challenge:	Existing offline approaches to improve an LLM-based customer support system rely on batch annotations.
Approach:	They propose an agent-in-the-loop framework that integrates four key types of annotations directly into live customer operations: (1) pairwise response preferences, (2) agent adoption and rationales, (3) knowledge relevance checks, and (4) identification of missing knowledge.
Outcome:	The proposed framework reduces retraining cycles from months to weeks by integrating four key types of annotations directly into live customer operations.

Target-oriented Fine-tuning for Zero-Resource Named Entity Recognition (2021.findings-acl)

Copied to clipboard

Ying Zhang, Fandong Meng, Yufeng Chen, Jinan Xu, Jie Zhou

CM-Align: Consistency-based Multilingual Alignment for Large Language Models (2025.findings-emnlp)

Challenge:	Named entity recognition (NER) is one of the fundamental tasks in natural language processing.
Approach:	They propose four practical guidelines to guide knowledge transfer and task finetuning . they propose a framework to exploit data from three aspects in a unified training manner .
Outcome:	The proposed framework improves on six benchmarks and shows that it is state-of-the-art in five languages.

Copied to clipboard

Xue Zhang, Yunlong Liang, Fandong Meng, Songming Zhang, Yufeng Chen, Jinan Xu, Jie Zhou

Original Semantics-Oriented Attention and Deep Fusion Network for Sentence Matching (D19-1)

Challenge:	Current large language models (LLMs) show a significant performance gap in alignment between English and other languages.
Approach:	They propose a consistency-based method to construct high-quality multilingual preference data for improving multilingual alignment.
Outcome:	The proposed method is based on three LLMs and three common tasks and shows that it performs better than current methods.

Copied to clipboard

Mingtong Liu, Yujie Zhang, Jinan Xu, Yufeng Chen

Challenge:	Sentence matching is a key issue in natural language inference and paraphrase identification.
Approach:	They propose a semantics-oriented attention and deep fusion network (OSOA-DFN) that is oriented to the original semantic representation of another sentence and propagates attention information at each matching layer.
Outcome:	The proposed model can model sentence matching more precisely on three sentence matching benchmark datasets.

Structure and Label Constrained Data Augmentation for Cross-domain Few-shot NER (2023.findings-emnlp)

Copied to clipboard

Jingyi Zhang, Ying Zhang, Yufeng Chen, Jinan Xu

Low-Resource NLP

Challenge:	Named entity recognition (NER) tasks require large datasets with accurate annotations that are labor-intensive and time-consuming.
Approach:	They propose a method to leverage domain gaps to model cross-domain few-shot named entity recognition (NER) NER is a natural language processing task to detect entity mentions and classify them into predefined labels .
Outcome:	The proposed method achieves state-of-the-art or competitive results on standard datasets.

Generating Authentic Adversarial Examples beyond Meaning-preserving with Doubly Round-trip Translation (2022.naacl-main)

Copied to clipboard

Siyu Lai, Zhen Yang, Fandong Meng, Xue Zhang, Yufeng Chen, Jinan Xu, Jie Zhou

Robustness in NLP

Challenge:	Existing approaches to generate adversarial examples for NMT use the meaning-preserving restriction.
Approach:	They propose a new definition for adversarial examples based on the Doubly Round-Trip Translation (DRTT) they introduce masked language models to construct bilingual adversarials based upon DRTT .
Outcome:	The proposed approach significantly improves the robustness of the NMT model on clean and noisy test sets.

Language-Coupled Reinforcement Learning for Multilingual Retrieval-Augmented Generation (2026.findings-acl)

Copied to clipboard

Rui Qi, Fengran Mo, Yufeng Chen, Xue Zhang, Shuo Wang, Hongliang Li, Xu Jinan, Meng Jiang, Jian-Yun Nie, Kaiyu Huang

Multilinguality

Challenge:	Existing approaches to multilingual retrieval-augmented generation (MRAG) use a single-turn retrieval and subsequent optimization to acquire and integrate beneficial external knowledge from multilingual collections.
Approach:	They propose a multilingual search-augmented reinforcement learning framework that integrates a language-coupled Group Relative Policy Optimization into the policy and reward models.
Outcome:	The proposed framework achieves competitive performance and is appropriate for various practical scenarios such as constrained training data and retrieval over collections encompassing a large number of languages.

Multilingual Knowledge Editing with Language-Agnostic Factual Neurons (2025.coling-main)

Copied to clipboard

Xue Zhang, Yunlong Liang, Fandong Meng, Songming Zhang, Yufeng Chen, Jinan Xu, Jie Zhou

Knowledge Representation

Challenge:	Existing methods to update factual knowledge overlook connections of same knowledge between different languages, resulting in knowledge conflicts and limited edit performance.
Approach:	They propose a method to edit multilingual knowledge simultaneously that avoids knowledge conflicts and improves edit performance.
Outcome:	The proposed method avoids knowledge conflicts and improves edit performance on bi-ZsRE and MzsRE benchmarks.

A Quality-based Syntactic Template Retriever for Syntactically-Controlled Paraphrase Generation (2023.emnlp-main)

Copied to clipboard

Xue Zhang, Songming Zhang, Yunlong Liang, Yufeng Chen, Jian Liu, Wenjuan Han, Jinan Xu

Iterative Constrained Back-Translation for Unsupervised Domain Adaptation of Machine Translation (2022.coling-1)

Challenge:	Existing syntactically-controlled paraphrase generation models perform well with human-annotated or well-chosen syntaktic templates.
Approach:	They propose a quality-based Syntactic Template Retriever to retrieve templates based on the quality of the to-be-generated paraphrases.
Outcome:	The proposed algorithm can generate high-quality paraphrases without sacrificing quality.

Copied to clipboard

Hongxiao Zhang, Hui Huang, Jiale Gao, Yufeng Chen, Jinan Xu, Jian Liu

Think Natively: Unlocking Multilingual Reasoning with Consistency-Enhanced Reinforcement Learning (2026.acl-long)

Challenge:	Existing back-translation methods focus on in-domain lexical knowledge, which may lead to poor translation of unseen in- domain words.
Approach:	They propose an iterative constrained back-translation method to incorporate in-domain lexical knowledge into synthetic parallel data from BT.
Outcome:	The proposed method improves the BLEU score by up to 3.08 on four domains.

Copied to clipboard

Xue Zhang, Yunlong Liang, Fandong Meng, Songming Zhang, Kaiyu Huang, Yufeng Chen, Xu Jinan, Jie Zhou

Syntactically-Informed Unsupervised Paraphrasing with Non-Parallel Data (2021.emnlp-main)

Challenge:	Current Large Reasoning Models exhibit two critical limitations when processing non-English languages: (1) They struggle to maintain input-output language consistency; (2) They generally perform poorly with wrong reasoning paths and lower answer accuracy compared to English.
Approach:	They propose a language-consistency reward and a cross-lingual thinking alignment reward to improve the model's interpretability and accuracy.
Outcome:	The proposed model achieves nearly 100% language consistency and superior performance on two multilingual benchmarks (MMATH and PolyMath).

Copied to clipboard

Erguang Yang, Mingtong Liu, Deyi Xiong, Yujie Zhang, Yao Meng, Changjian Hu, Jinan Xu, Yufeng Chen

CM-Net: A Novel Collaborative Memory Network for Spoken Language Understanding (D19-1)

Challenge:	Existing studies on syntactically controlled paraphrase generation rely on large-scale parallel data.
Approach:	They propose a syntactically-informed unsupervised paraphrasing model based on conditional variational auto-encoder which can generate texts in a specified syntastic structure.
Outcome:	The proposed model can generate diverse paraphrases with specified syntactic structure using non-parallel data.

Copied to clipboard

Yijin Liu, Fandong Meng, Jinchao Zhang, Jie Zhou, Yufeng Chen, Jinan Xu

Challenge:	Existing models for slot filling and intent detection fail to fully utilize cooccurrence relations between slots and intents, which restricts their potential performance.
Approach:	They propose a novel Collaborative Memory Network (CM-Net) that captures slot-specific and intent-specific features in a collaborative manner.
Outcome:	The proposed network outperforms existing models on two benchmarks and a self-collected corpus.

SWE-Mutation: Can LLMs Generate Reliable Test Suites in Software Engineering? (2026.findings-acl)

Copied to clipboard

Yuxuan Sun, Yuze Zhao, Yufeng Wang, Yao Du, Zhiyuan Ma, Jinbo Wang, Mengdi Zhang, Kai Zhang, Zhenya Huang

Multimodality

Challenge:	Evaluating software engineering capabilities is a core component of large language models (LLMs).
Approach:	They propose a benchmark to evaluate LLM-generated test suites that introduces mutated solutions that attempt to "fool" them.
Outcome:	The proposed test suites are based on 2,636 mutated variants derived from 800 original instances and include a multilingual subset spanning nine programming languages.

A Reinforcement Learning Approach to Improve Low-Resource Machine Translation Leveraging Domain Monolingual Data (2024.lrec-main)

Copied to clipboard

Hongxiao Zhang, Mingtong Liu, Chunyou Li, Yufeng Chen, Jinan Xu, Ming Zhou

Less, but Better: Efficient Multilingual Expansion for LLMs via Layer-wise Mixture-of-Experts (2025.acl-long)

Challenge:	Existing methods for fine-tuning domain adaptation have overfitting problem in low-resource domains . lack of parallel data makes it difficult for model to learn domain-specific knowledge .
Approach:	They propose a Reinforcement Learning Domain Adaptation method for Neural Machine Translation that uses in-domain source monolingual data to make up for the lack of parallel data.
Outcome:	The proposed method can alleviate overfitting and reinforce the model to learn domain-specific knowledge.

Copied to clipboard

Xue Zhang, Yunlong Liang, Fandong Meng, Songming Zhang, Yufeng Chen, Jinan Xu, Jie Zhou

Green & Sustainable NLP

Challenge:	Existing large language models (LLMs) have remarkable ability in high-resource languages, but their performance in multilingual scenarios is still limited.
Approach:	They propose a layer-wise expert allocation algorithm to determine the appropriate number of new experts for each layer.
Outcome:	The proposed method outperforms the previous state-of-the-art baseline with 60% fewer experts in the single-expansion setting and 33.3% fewer in the lifelong-expanding setting.

ICL: Iterative Continual Learning for Multi-domain Neural Machine Translation (2024.findings-emnlp)

Copied to clipboard

Zhibo Man, Kaiyu Huang, Yujie Zhang, Yuanmeng Chen, Yufeng Chen, Jinan Xu

Machine Translation

Challenge:	Existing studies have focused on learning domain knowledge from multiple domains, but task-specific parameters hinder mutual transfer of knowledge between new domains.
Approach:	They propose an iterative Continual learning framework for multi-domain neural machine translation that leverages previously acquired domain knowledge.
Outcome:	The proposed model outperforms baseline models on UM-Corpus and OPUS datasets.

C3D: Enhancing LLM Reasoning via Confidence-Guided Contrastive Decoding (2026.findings-acl)

Copied to clipboard

Yufeng Zhang, Xuepeng Wang, Lingxiang Wu, Jinqiao Wang

Reasoning

Challenge:	Large language models are prone to distraction by contextual information during reasoning tasks.
Approach:	They propose a decoding method that uses predicted logits to estimate the model's confidence.
Outcome:	The proposed method reveals how the model dynamically activates and adjusts its consideration of each premise as reasoning progresses.

AlignDistil: Token-Level Language Model Alignment as Adaptive Policy Distillation (2025.acl-long)

Copied to clipboard

Songming Zhang, Xue Zhang, Tong Zhang, Bojie Hu, Yufeng Chen, Jinan Xu

Semantic Text Processing

Challenge:	Existing methods for LLM alignment optimize tokens using a sparse, response-level reward or preference annotation.
Approach:	They propose an RLHF-equivalent distillation method for token-level reward optimization that incorporates the reward learned by DPO into the RLHG objective and builds a token-based teacher distribution.
Outcome:	The proposed method bridges the accuracy gap between the reward from the DPO model and the pure reward model by building a contrastive DPO reward with a normal and a reverse DPO.

MT2: Towards a Multi-Task Machine Translation Model with Translation-Specific In-Context Learning (2023.emnlp-main)

Copied to clipboard

Chunyou Li, Mingtong Liu, Hongxiao Zhang, Yufeng Chen, Jinan Xu, Ming Zhou

Low-Resource NLP

Challenge:	Sentence-level translation, document-level and terminology constrained translations are important in machine translation.
Approach:	They propose a multi-task machine translation model that integrates translation memory sentences . they propose 'in-context learning' paradigm that allows translation-specific context learning .
Outcome:	The proposed model improves translation memory, document-level translation, and document-constrained translation tasks.

A Multi-modal Debiasing Model with Dynamical Constraint for Robust Visual Question Answering (2023.findings-acl)

Copied to clipboard

Yu Li, Bojie Hu, Fengshuo Zhang, Yahan Yu, Jian Liu, Yufeng Chen, Jinan Xu

Adversarially Improving NMT Robustness to ASR Errors with Confusion Sets (2022.aacl-short)

Challenge:	Recent studies have shown that many well-developed Visual Question Answering systems suffer from bias problem.
Approach:	They propose a way to mitigate bias problem by subtracting bias score from standard VQA base score.
Outcome:	The proposed method improves on the VQA v2.0 and VQA-CP V2,0 datasets.

Copied to clipboard

Shuaibo Wang, Yufeng Chen, Songming Zhang, Deyi Xiong, Jinan Xu

An Iterative Multi-Knowledge Transfer Network for Aspect-Based Sentiment Analysis (2021.findings-emnlp)

Challenge:	Existing methods for robustness against homophone errors are limited to homophones . substitution errors are the most common errors in NMT models .
Approach:	They propose an adversarial example generation method based on confusion sets that contain words easily confusable with a target word by ASR to conduct adversarially training for NMT models.
Outcome:	The proposed method improves on the clean test set and can be used in real-world scenarios.

Copied to clipboard

Yunlong Liang, Fandong Meng, Jinchao Zhang, Yufeng Chen, Jinan Xu, Jie Zhou

Sentiment Analysis

Challenge:	Existing approaches to Aspect-based sentiment analysis do not exploit the interactive relations among subtasks and do not utilize document-level labeled domain/sentiment knowledge, which restricts their performance.
Approach:	They propose an iterative multi-knowledge transfer network for end-to-end ABSA that leverages the inter-task interaction between subtasks.
Outcome:	The proposed approach improves on three benchmark datasets.

Long Text Generation with Topic-aware Discrete Latent Variable Model (2022.emnlp-main)

Copied to clipboard

Erguang Yang, Mingtong Liu, Deyi Xiong, Yujie Zhang, Yufeng Chen, Jinan Xu

GCDT: A Global Context Enhanced Deep Transition Architecture for Sequence Labeling (P19-1)

Challenge:	Recent work focuses on the modeling of discourse relation, resulting in discrete codes learning shallow semantics.
Approach:	They propose a topic-aware latent code-guided text generation model that encourages discrete codes to model information about topics.
Outcome:	The proposed model generates more topic-relevant and coherent texts.

Copied to clipboard

Yijin Liu, Fandong Meng, Jinchao Zhang, Jinan Xu, Yufeng Chen, Jie Zhou

Chunking

Challenge:	Existing systems for sequence labeling are limited by shallow connections between consecutive hidden states and insufficient modeling of global information.
Approach:	They propose a global context enhanced deep transition architecture for sequence labeling . they deepen the state transition path at each position in a sentence and assign tokens with global representations .
Outcome:	The proposed architecture outperforms the best reported results on two standard sequence labeling tasks.

Exploration-Exploitation Reshaping towards Efficient Reasoning for Large Language Models (2026.findings-acl)

Copied to clipboard

Yufeng Shi, Weilin Luo, Yuxiang Zhang, Zongmeng Zhang, Haoyang Liu, Yubing Wang, Bin Wang, Wengang Zhou, Houqiang Li

Every Document Owns Its Structure: Inductive Text Classification via Graph Neural Networks (2020.acl-main)

Challenge:	Large Reasoning Models (LRMs) are constrained by the overthinking issue.
Approach:	They propose a policy optimization framework that reshapes the exploration and exploitation through two core components: self-imitation and self-guidance exploration.
Outcome:	The proposed model achieves superior reasoning efficiency without compromising overall accuracy.

Copied to clipboard

Yufeng Zhang, Xueli Yu, Zeyu Cui, Shu Wu, Zhongzhen Wen, Liang Wang

Information Extraction & Text Mining

Challenge:	Existing graph-based methods for text classification cannot capture contextual word relationships within each document nor can they produce inductive learning of new words.
Approach:	They propose to use Graph Neural Networks to learn the local word representations and then aggregate the word nodes as the document embeddings.
Outcome:	The proposed method outperforms state-of-the-art methods on four benchmark datasets.

DavIR: Data Selection via Implicit Reward for Large Language Models (2025.acl-long)

Copied to clipboard

Haotian Zhou, Tingkai Liu, Qianli Ma, Yufeng Zhang, Jianbo Yuan, Pengfei Liu, Yang You, Hongxia Yang

Dual-Space Knowledge Distillation for Large Language Models (2024.emnlp-main)

Challenge:	6% of Alpaca dataset selected with DavIR can steer both LLaMA and Gemma models to produce superior performance compared to the same models trained on the full 52K dataset.
Approach:	They propose a model-based data selection method for post-training Large Language Models . they generalize Reducible Holdout Loss to core-set selection problem of causal language modeling .
Outcome:	The proposed method can steer both LLaMA and Gemma models to superior performance compared to the same models trained on the full 52K dataset.

Copied to clipboard

Songming Zhang, Xue Zhang, Zengkui Sun, Yufeng Chen, Jinan Xu

Learning Structural Information for Syntax-Controlled Paraphrase Generation (2022.findings-naacl)

Challenge:	Existing large language models (LLMs) have strong generalization abilities due to their huge model capacities.
Approach:	They propose a dual-space knowledge distillation framework that unifies the output spaces of the two models for KD.
Outcome:	The proposed framework outperforms existing white-box KD frameworks on task-agnostic instruction-following benchmarks and can automatically align representations of two models with different vocabularies.

Copied to clipboard

Erguang Yang, Chenglin Bai, Deyi Xiong, Yujie Zhang, Yao Meng, Jinan Xu, Yufeng Chen