Papers by Jianfeng Gao

How Much Do Language Models Copy From Their Training Data? Evaluating Linguistic Novelty in Text Generation Using RAVEN (2023.tacl-1)

Challenge:	Existing models for machine reading comprehension rely on large amounts of human-annotated in-domain data.
Approach:	They propose an unsupervised domain adaptation framework for Machine Reading Comprehension where the source domain has a large amount of labeled data, while only unlabeled passages are available in the target domain.
Outcome:	The proposed framework can be generalizable to different MRC models and datasets and can be extended to semi-supervised learning.

Copied to clipboard

R. Thomas McCoy, Paul Smolensky, Tal Linzen, Jianfeng Gao, Asli Celikyilmaz

Knowledge-Grounded Dialogue Generation with a Unified Knowledge Representation (2022.naacl-main)

Challenge:	Current language models generate high-quality text, but are they copying it or have they learned generalizable linguistic abstractions?
Approach:	They propose a suite of analyses for assessing the novelty of generated text . they focus on sequential structure (n-grams) and syntactic structure (syntactical structure).
Outcome:	The proposed model-generated text is as novel as the baseline human-generated model- generated text, but it is copied substantially, the authors show .

Copied to clipboard

Yu Li, Baolin Peng, Yelong Shen, Yi Mao, Lars Liden, Zhou Yu, Jianfeng Gao

Pre-training Multi-task Contrastive Learning Models for Scientific Literature Understanding (2023.findings-emnlp)

Challenge:	Existing knowledge-grounded dialogue systems perform poorly on unseen topics due to limited topics covered in training data.
Approach:	They propose a language model that homogenizes different knowledge sources to a unified knowledge representation for knowledge-grounded dialogue generation tasks.
Outcome:	The proposed language model generalizes well across knowledge-grounded dialogue tasks.

Copied to clipboard

Yu Zhang, Hao Cheng, Zhihong Shen, Xiaodong Liu, Ye-Yi Wang, Jianfeng Gao

Understanding the Difficulty of Training Transformers (2020.emnlp-main)

Challenge:	Pre-trained language models (LMs) have shown effectiveness in literature understanding tasks, especially when tuned via contrastive learning.
Approach:	They propose a multi-task contrastive learning framework that enables common knowledge sharing across different scientific literature understanding tasks while preventing task-specific skills from interfering with each other.
Outcome:	The proposed framework outperforms state-of-the-art pre-trained language models on a comprehensive dataset.

Copied to clipboard

Liyuan Liu, Xiaodong Liu, Jianfeng Gao, Weizhu Chen, Jiawei Han

Towards Consistent Natural-Language Explanations via Explanation-Consistency Finetuning (2025.coling-main)

Challenge:	Admin (Adaptive model initialization) is more stable, converges faster, and leads to better performance.
Approach:	They propose a model initialization algorithm to stabilize early training and unleash its full potential in the late stage.
Outcome:	The proposed model initialization method stabilizes early training and unleashes full potential in late stage.

Copied to clipboard

Yanda Chen, Chandan Singh, Xiaodong Liu, Simiao Zuo, Bin Yu, He He, Jianfeng Gao

Explainability & Interpretability in NLP

Challenge:	Large language models generate convincing, fluent explanations, but they often generate inconsistent explanations on different inputs.
Approach:	They propose a method that adapts large language models to generate more consistent explanations on related examples.
Outcome:	The proposed method yields a 10.0% relative explanation consistency improvement across a variety of question-answering datasets and generalizes to 7 out-of-distribution datasets not seen during finetuning (+4.5%)

CodeExp: Explanatory Code Document Generation (2022.findings-emnlp)

Copied to clipboard

Haotian Cui, Chenglong Wang, Junjie Huang, Jeevana Priya Inala, Todd Mytkowicz, Bo Wang, Jianfeng Gao, Nan Duan

Explainability & Interpretability in NLP

Challenge:	Existing code-to-text generation models produce only high-level code summaries that do not capture implementation-level choices essential for these scenarios.
Approach:	They propose a code explanation generation task that uses code docstrings to refine models.
Outcome:	The proposed model can generate well-structured long docstrings comparable to human-written ones.

Z-Code++: A Pre-trained Language Model Optimized for Abstractive Summarization (2023.acl-long)

Copied to clipboard

Pengcheng He, Baolin Peng, Song Wang, Yang Liu, Ruochen Xu, Hany Hassan, Yu Shi, Chenguang Zhu, Wayne Xiong, Michael Zeng, Jianfeng Gao, Xuedong Huang

Adversarial Regularization as Stackelberg Game: An Unrolled Optimization Approach (2021.emnlp-main)

Challenge:	Z-Code++ is a pre-trained language model optimized for abstractive text summarization.
Approach:	They propose a pre-trained language model optimized for abstractive text summarization that uses a two-phase pre-training technique to improve model's performance.
Outcome:	The proposed model outperforms the competing models on low-resource summarization tasks in zero-shot and few-shot settings.

Copied to clipboard

Simiao Zuo, Chen Liang, Haoming Jiang, Xiaodong Liu, Pengcheng He, Jianfeng Gao, Weizhu Chen, Tuo Zhao

AutoMoE: Heterogeneous Mixture-of-Experts with Adaptive Computation for Efficient Neural Machine Translation (2023.findings-acl)

Challenge:	Existing approaches to adversarial regularization treat adversarials and defending players equally, which is undesirable because only the defending player contributes to the generalization performance.
Approach:	They propose a method which formulates adversarial regularization as a Stackelberg game and induces a competition between a leader and a follower.
Outcome:	The proposed method outperforms existing adversarial regularization baselines on a set of machine translation and natural language understanding tasks.

Copied to clipboard

Ganesh Jawahar, Subhabrata Mukherjee, Xiaodong Liu, Young Jin Kim, Muhammad Abdul-Mageed, Laks Lakshmanan, V.S., Ahmed Hassan Awadallah, Sebastien Bubeck, Jianfeng Gao

The Microsoft Toolkit of Multi-Task Deep Neural Networks for Natural Language Understanding (2020.acl-demos)

Challenge:	Existing MoE designs do not consider computational constraints (e.g., FLOPs, latency) Existing works in MoE consider homogeneous design where the same number of experts of the same size are placed uniformly throughout the network.
Approach:	They propose a framework for designing heterogeneous MoEs under computational constraints.
Outcome:	The proposed framework achieves 4x inference speedup and FLOPs reduction over manual models and within 1 BLEU point of MoE SwitchTransformer over benchmark datasets for NMT.

Copied to clipboard

Xiaodong Liu, Yu Wang, Jianshu Ji, Hao Cheng, Xueyun Zhu, Emmanuel Awa, Pengcheng He, Weizhu Chen, Hoifung Poon, Guihong Cao, Jianfeng Gao

Conversing by Reading: Contentful Neural Conversation with On-demand Machine Reading (P19-1)

Challenge:	MT-DNN is an open-source natural language understanding toolkit . it allows researchers and developers to train customized deep learning models .
Approach:	They present MT-DNN, an open-source natural language understanding toolkit . it is designed to facilitate rapid customization for a broad spectrum of NLU tasks . MT supports multi-task knowledge distillation, which can substantially compress a deep neural model without significant performance drop.
Outcome:	The proposed model can significantly compress a large model without significant performance drop.

Copied to clipboard

Lianhui Qin, Michel Galley, Chris Brockett, Xiaodong Liu, Xiang Gao, Bill Dolan, Yejin Choi, Jianfeng Gao

ReEval: Automatic Hallucination Evaluation for Retrieval-Augmented Large Language Models via Transferable Adversarial Attacks (2024.findings-naacl)

Challenge:	a new approach to contentful neural conversation is proposed . end-to-end models are effective in learning fluent responses, but their responses are often vacuous and uninformative.
Approach:	They propose a model that provides the conversation model with relevant text on the fly as a source of external knowledge.
Outcome:	The proposed model improves the informativeness and diversity of generated output compared to previous methods.

Copied to clipboard

Xiaodong Yu, Hao Cheng, Xiaodong Liu, Dan Roth, Jianfeng Gao

Interactive Text Generation (2023.emnlp-main)

Challenge:	Existing static benchmarks do not guarantee that models can use the provided evidence for answering, which is essential to avoid hallucination when the required knowledge is new or private.
Approach:	They propose to automatically perturb existing static one for dynamic evaluation by using a chatGPT framework and a set of open-domain QA datasets.
Outcome:	The proposed framework generates new test cases on two open-domain QA datasets and is human-readable and useful to trigger hallucination in LLMs.

Copied to clipboard

Felix Faltings, Michel Galley, Kianté Brantley, Baolin Peng, Weixin Cai, Yizhe Zhang, Jianfeng Gao, Bill Dolan

Towards Robust Few-Shot Relation Classification: Incorporating Relation Description with Agreement (2025.findings-emnlp)

Challenge:	Advances in generative modeling have made it possible to automatically generate high-quality texts, code, and images, but they can be unsatisfactory in many respects.
Approach:	They propose a task that allows training generation models interactively without the costs of involving real users.
Outcome:	The proposed model trains with Imitation Learning without the cost of involving real users and is superior to non-interactive models.

Copied to clipboard

Mengting Hu, Jianfeng Wu, Ming Jiang, Yalan Xie, Zhunheng Wang, Rui Ying, Xiaoyi Liu, Ruixuan Xu, Hang Gao, Renhong Cheng

Few-Shot Named Entity Recognition: An Empirical Baseline Study (2021.emnlp-main)

Challenge:	Existing approaches to recognize relational relationships with a few support samples are limited for unlimited queries.
Approach:	They propose a simple but effective framework that uses relation descriptions as external knowledge to enhance the model’s comprehension of the relation semantics.
Outcome:	The proposed framework outperforms strong baselines while being robust against various NOTA rates.

Copied to clipboard

Jiaxin Huang, Chunyuan Li, Krishan Subudhi, Damien Jose, Shobana Balakrishnan, Weizhu Chen, Baolin Peng, Jianfeng Gao, Jiawei Han

Microsoft Icecaps: An Open-Source Toolkit for Conversation Modeling (P19-3)

Challenge:	Existing methods to build named entity recognition systems with limited labeled data are lacking.
Approach:	They propose three orthogonal schemes to build named entity recognition systems when labeled data is limited.
Outcome:	The proposed NER systems outperform existing methods on few-shot and training-free settings.

Copied to clipboard

Vighnesh Leonardo Shiv, Chris Quirk, Anshuman Suri, Xiang Gao, Khuram Shahid, Nithya Govindarajan, Yizhe Zhang, Jianfeng Gao, Michel Galley, Chris Brockett, Tulasi Menon, Bill Dolan

Token-wise Curriculum Learning for Neural Machine Translation (2021.findings-emnlp)

Challenge:	upcoming open-source natural language processing repository aims to train conversational agents for multi-turn situations.
Approach:	They present the Intelligent Conversation Engine: Code and Pre-trained Systems (ICECAPS) the framework wraps TensorFlow functionality in a modular component-based architecture.
Outcome:	The Intelligent Conversation Engine: Code and Pre-trained Systems (ICECAPS) is an open-source natural language processing repository.

Copied to clipboard

Chen Liang, Haoming Jiang, Xiaodong Liu, Pengcheng He, Weizhu Chen, Jianfeng Gao, Tuo Zhao

Machine Translation

Challenge:	Existing curriculum learning approaches to Neural Machine Translation (NMT) require sampling sufficient amounts of “easy” samples from training data at the early stage of training.
Approach:	They propose a token-wise curriculum learning approach that creates sufficient amounts of easy samples from training data.
Outcome:	The proposed approach outperforms baselines on five language pairs on low-resource languages.

RADDLE: An Evaluation Benchmark and Analysis Platform for Robust Task-oriented Dialog Systems (2021.acl-long)

Copied to clipboard

Baolin Peng, Chunyuan Li, Zhu Zhang, Chenguang Zhu, Jinchao Li, Jianfeng Gao

NICE: Neural Image Commenting with Empathy (2021.findings-emnlp)

Challenge:	Existing task-oriented dialog systems are less than satisfactory in robustness evaluation . existing systems are weak in robustity evaluation based on pre-training and fine-tuning .
Approach:	They propose to use a set of training examples to evaluate model generalization ability . they propose to include tasks with limited training data to favor models with strong generalization abilities .
Outcome:	The proposed model generalizes well with limited training data and is robust to user input across domains.

Copied to clipboard

Kezhen Chen, Qiuyuan Huang, Daniel McDuff, Xiang Gao, Hamid Palangi, Jianfeng Wang, Kenneth Forbus, Jianfeng Gao

Visual Data in NLP

Challenge:	Emotion and empathy are examples of human qualities lacking in many human-machine interactions.
Approach:	They propose to generate images with human-generated comments with enhanced emotion and empathy while minimizing inappropriate or offensive outputs.
Outcome:	The proposed model generates more human-like and engaging image comments on two images with human-generated comments and human annotations while minimizing inappropriate or offensive outputs.

Iterative Self-Tuning LLMs for Enhanced Jailbreaking Capabilities (2025.naacl-long)

Copied to clipboard

Chung-En Sun, Xiaodong Liu, Weiwei Yang, Tsui-Wei Weng, Hao Cheng, Aidan San, Michel Galley, Jianfeng Gao

Simple but Effective Compound Geometric Operations for Temporal Knowledge Graph Completion (2024.acl-long)

Challenge:	Recent research shows that Large Language Models (LLMs) are vulnerable to automated jailbreak attacks.
Approach:	They propose a framework that crafts adversarial LLMs with enhanced jailbreak ability.
Outcome:	ADV-LLM significantly reduces the computational cost of generating adversarial suffixes while achieving nearly 100% ASR on various open-source LLMs.

Copied to clipboard

Rui Ying, Mengting Hu, Jianfeng Wu, Yalan Xie, Xiaoyi Liu, Zhunheng Wang, Ming Jiang, Hang Gao, Linlin Zhang, Renhong Cheng

Stochastic Answer Networks for Machine Reading Comprehension (P18-1)

Challenge:	Current methods embed factual knowledge into continuous vector space and apply geometric operations to learn potential patterns in temporal knowledge graphs.
Approach:	They propose a temporal knowledge graph completion method that uses two geometric operations to learn missing facts in temporal graphs.
Outcome:	The proposed method significantly outperforms existing temporal knowledge graph embedding models.

Copied to clipboard

Xiaodong Liu, Yelong Shen, Kevin Duh, Jianfeng Gao

Task-Aware Specialization for Efficient and Robust Dense Retrieval for Open-Domain Question Answering (2023.acl-short)

Challenge:	Several recent MRC models employ multi-step reasoning . we show that the use of a stochastic prediction dropout improves robustness .
Approach:	They propose a stochastic answer network that simulates multi-step reasoning in machine reading comprehension.
Outcome:	The proposed model improves robustness and results competitive with state-of-the-art models on the Stanford Question Answering Dataset and Microsoft MAchine Reading COmprehension Dataset.

Copied to clipboard

Hao Cheng, Hao Fang, Xiaodong Liu, Jianfeng Gao

DIALOGPT : Large-Scale Generative Pre-training for Conversational Response Generation (2020.acl-demos)

Challenge:	Existing dense retrieval models are parameter-inefficient and underperform sparse counterparts.
Approach:	They propose a task-aware specialization for dEnse Retrieval architecture that enables parameter sharing by interleaving shared and specialized blocks in a single encoder.
Outcome:	The proposed architecture surpasses BM25 on questions and passages using 60% of the parameters as bi-encoder dense retrievers.

Copied to clipboard

Yizhe Zhang, Siqi Sun, Michel Galley, Yen-Chun Chen, Chris Brockett, Xiang Gao, Jianfeng Gao, Jingjing Liu, Bill Dolan

Deep Dyna-Q: Integrating Planning for Task-Completion Dialogue Policy Learning (P18-1)

Challenge:	DIALOGPT is a large, tunable neural conversational response generation model . trained on 147M conversation-like exchanges extracted from Reddit comment chains .
Approach:	They present a large, tunable neural conversational response generation model, DIALOGPT . the model is trained on 147M conversation-like exchanges extracted from Reddit comment chains .
Outcome:	The proposed model can generate more relevant, contentful and context-consistent responses than baseline systems.

Copied to clipboard

Baolin Peng, Xiujun Li, Jianfeng Gao, Jingjing Liu, Kam-Fai Wong

Optimus: Organizing Sentences via Pre-trained Modeling of a Latent Space (2020.emnlp-main)

Challenge:	Training a task-completion dialogue agent via reinforcement learning (RL) is costly because it requires many interactions with real users.
Approach:	They propose a framework that integrates planning for task-completion dialogue policy learning into a dialogue agent using a world model to mimic real user response and generate simulated experience.
Outcome:	The proposed framework integrates planning for task-completion dialogue policy learning with real user interaction and simulated user behavior.

Copied to clipboard

Chunyuan Li, Xiang Gao, Yuan Li, Baolin Peng, Xiujun Li, Yizhe Zhang, Jianfeng Gao

Conversation Learner - A Machine Teaching Tool for Building Dialog Managers for Task-Oriented Dialog Systems (2020.acl-demos)

Challenge:	Existing models for language understanding and understanding can be trained to provide contextualized representations of words based on text data.
Approach:	They propose a large-scale language VAE model Optimus that is pre-trained on large text corpus and fine-tuned for various language generation and understanding tasks.
Outcome:	The proposed model achieves new state-of-the-art on VAE language modeling benchmarks.

Copied to clipboard

Swadheen Shukla, Lars Liden, Shahin Shayandeh, Eslam Kamal, Jinchao Li, Matt Mazzola, Thomas Park, Baolin Peng, Jianfeng Gao

Reinforcement Learning on Pre-Training Data (2026.acl-long)

Challenge:	a wide variety of tasks have created a need for flexible task-oriented dialog systems . dialog flows are intuitively interpretable but lack the flexibility needed to handle complex dialogs .
Approach:	They propose a machine teaching tool for building dialog managers using familiar tools . they convert the dialog flow into a parametric model and use user-system dialog logs as training data .
Outcome:	The proposed tool combines the best of both approaches to build dialog managers . it converts the dialog flow into a parametric model and improves it over time .

Copied to clipboard

Siheng Li, Kejiao Li, Zenan Xu, Guanhua Huang, Kun Li, Haoyuan Wu, null Wujiajia, Zihao Zheng, Chenchen Zhang, Kun Shi, Xue Gong, Qi Yi, Ruibin Xiong, Tingqiang Xu, Yuhao Jiang, Jianfeng Yan, Yuyuan Zeng, Guanghui Xu, Jinbao Xue, Zhijiang xu, Zheng Fang, Shuai LI, Qibin Liu, Xiaoxue Li, Zhuoyu Li, Yangyu Tao, Fei Gao, Cheng Jiang, Bochao Wang, Kai Liu, Jianchen Zhu, Wai Lam, Bo Zhou, Di Wang

ARCH: Efficient Adversarial Regularized Training with Caching (2021.findings-emnlp)

Challenge:	Recent progress in large language models is driven by scaling of training compute through pre-training with nexttoken prediction (NTP) or post-training (RL) Pre-training using NTP enables models to acquire extensive knowledge and skills from general data, but it suffers from data inefficiency and catastrophic forgetting in continual learning settings.
Approach:	They propose to scale training compute through pre-training with next-token prediction (NTP) or post-training by scaling reinforcement learning (RL) to improve learning from general data.
Outcome:	Experiments on multiple benchmarks and models show that the proposed approach improves continual pre-training and provides a strong foundation for post-training on Qwen3-8B-Base.

Copied to clipboard

Simiao Zuo, Chen Liang, Haoming Jiang, Pengcheng He, Xiaodong Liu, Jianfeng Gao, Weizhu Chen, Tuo Zhao

Grounded Keys-to-Text Generation: Towards Factual Open-Ended Generation (2022.findings-emnlp)

Challenge:	Existing approaches to regularize models require generating a perturbation for each sample in each epoch.
Approach:	They propose an adversarial regularization method where perturbations are generated and cached once every several epochs.
Outcome:	The proposed method significantly eases the computational burden (saves up to 70% of computational time) it produces a notably better (in most of the tasks) or comparable model generalization.

Copied to clipboard

Faeze Brahman, Baolin Peng, Michel Galley, Sudha Rao, Bill Dolan, Snigdha Chaturvedi, Jianfeng Gao

RMM: A Recursive Mental Model for Dialogue Navigation (2020.findings-emnlp)

Challenge:	Large pre-trained language models have enabled open-ended generation frameworks to tackle a variety of tasks beyond data-to-text generation.
Approach:	They propose a new task to generate a factual description about an entity given guiding keys and grounding passages using a dataset.
Outcome:	The proposed model improves factual correctness and recall significantly compared to previous models.

Copied to clipboard

Homero Roman Roman, Yonatan Bisk, Jesse Thomason, Asli Celikyilmaz, Jianfeng Gao

Unsupervised Deep Structured Semantic Models for Commonsense Reasoning (N19-1)

Challenge:	Existing work on language-guided robots focuses on the latter, but little attention is paid to the guiding agent.
Approach:	They propose a two-agent task where one agent navigates and asks questions that a second, guiding agent answers.
Outcome:	The proposed model can be generalized to novel environments.

Copied to clipboard

Shuohang Wang, Sheng Zhang, Yelong Shen, Xiaodong Liu, Jingjing Liu, Jianfeng Gao, Jing Jiang

Neural Approaches to Conversational AI (P18-5)

Challenge:	Existing methods for commonsense reasoning rely on human-crafted features and knowledge bases, but unsupervised learning is not feasible due to the lack of labeled training data or comprehensive knowledge bases.
Approach:	They propose two unsupervised models based on the Deep Structured Semantic Models framework to tackle two commonsense reasoning tasks: Winograd Schema Challenge (WSC) and Pronoun Disambiguation (PDP).
Outcome:	The proposed models capture contextual information in the sentence and co-reference information between pronouns and nouns, and achieve significant improvement over previous state-of-the-art approaches.

Copied to clipboard

Jianfeng Gao, Michel Galley, Lihong Li

Knowledge-Rich Self-Supervision for Biomedical Entity Linking (2022.findings-emnlp)

Challenge:	This tutorial examines neural approaches to conversational AI that have been developed in the last few years.
Approach:	This tutorial presents a review of state-of-the-art neural approaches to conversational AI . they group conversational systems into question answering agents, task-oriented dialogue agents and social bots .
Outcome:	The present tutorial examines state-of-the-art approaches to conversational AI . it draws the connection between neural approaches and traditional symbolic approaches .

Copied to clipboard

Sheng Zhang, Hao Cheng, Shikhar Vashishth, Cliff Wong, Jinfeng Xiao, Xiaodong Liu, Tristan Naumann, Jianfeng Gao, Hoifung Poon

HittER: Hierarchical Transformers for Knowledge Graph Embeddings (2021.emnlp-main)

Challenge:	Entity linking is challenging in high-value domains with myriad entities . standard classification approaches suffer from the annotation bottleneck .
Approach:	They propose a self-supervised approach to learn domain knowledge for biomedical entity linking . it generates self-reported mention examples on unlabeled text and trains contextual encoder .
Outcome:	The proposed method outperforms existing methods by 20 points in accuracy on biomedical datasets.

Copied to clipboard

Sanxing Chen, Xiaodong Liu, Jianfeng Gao, Jian Jiao, Ruofei Zhang, Yangfeng Ji

Few-shot Natural Language Generation for Task-Oriented Dialog (2020.findings-emnlp)

Challenge:	Existing knowledge graph embedding methods to learn representations of knowledge graphs are conceptually simple and can be applied to tasks like factoid question answering (Saxena et al., 2020) and reasoning.
Approach:	They propose a Hierarchical Transformer model to jointly learn Entity-relation composition and Relational contextualization based on a source entity’s neighborhood.
Outcome:	The proposed model achieves state-of-the-art on multiple link prediction datasets and can be integrated into BERT and demonstrate its effectiveness on two Freebase factoid question answering datasets.

Copied to clipboard

Baolin Peng, Chenguang Zhu, Chunyuan Li, Xiujun Li, Jinchao Li, Michael Zeng, Jianfeng Gao

Discriminative Deep Dyna-Q: Robust Planning for Dialogue Policy Learning (D18-1)

Challenge:	Existing methods for NLG depend on heavily annotated data, which is infeasible for new domains.
Approach:	They propose a system that converts a dialog act into a response in natural language . they propose 'nuclear language generation' to simulate a few-shot learning setting .
Outcome:	The proposed model outperforms existing methods on a large set of annotated datasets.

Copied to clipboard

Shang-Yu Su, Xiujun Li, Jianfeng Gao, Jingjing Liu, Yun-Nung Chen

KAT: A Knowledge Augmented Transformer for Vision-and-Language (2022.naacl-main)

Challenge:	Existing approaches to improve the effectiveness and robustness of Deep Dyna-Q (DDQ) are based on a discriminator to control the quality of simulated experiences and to improve learning.
Approach:	They propose to use an RNN-based discriminator to control the quality of simulated experience to improve the effectiveness and robustness of Deep Dyna-Q.
Outcome:	The proposed framework outperforms DDQ by controlling the quality of simulated experience used for planning.

Copied to clipboard

Liangke Gui, Borui Wang, Qiuyuan Huang, Alexander Hauptmann, Yonatan Bisk, Jianfeng Gao

Multi-task Learning with Sample Re-weighting for Machine Reading Comprehension (N19-1)

Challenge:	Existing methods for knowledge retrieval and answer prediction have left open questions about the quality and relevance of the retrieved knowledge and how the reasoning processes over implicit and explicit knowledge should be integrated.
Approach:	They propose a Knowledge Augmented Transformer which integrates both implicit and explicit knowledge in an encoder-decoder architecture while simultaneously reasoning over both knowledge sources during answer generation.
Outcome:	The proposed model achieves a strong state-of-the-art (+6% absolute) on the open-domain multimodal task of OK-VQA.

Copied to clipboard

Yichong Xu, Xiaodong Liu, Yelong Shen, Jingjing Liu, Jianfeng Gao

ConvLab-3: A Flexible Dialogue System Toolkit Based on a Unified Data Format (2023.emnlp-demo)

Challenge:	Existing models for Machine Reading Comprehension (MRC) are small, compared to their size, and there are many studies on using pre-trained word embeddings and back-translation approaches to improve model generalization.
Approach:	They propose a multi-task learning framework to learn a machine reading comprehension model that can be applied to a wide range of MRC tasks in different domains.
Outcome:	The proposed model can be applied to a wide range of MRC tasks in different domains.

Copied to clipboard

Qi Zhu, Christian Geishauser, Hsien-chin Lin, Carel van Niekerk, Baolin Peng, Zheng Zhang, Shutong Feng, Michael Heck, Nurul Lubis, Dazhen Wan, Xiaochen Zhu, Jianfeng Gao, Milica Gasic, Minlie Huang

Chain-of-Skills: A Configurable Model for Open-Domain Question Answering (2023.acl-long)

Challenge:	Existing tools for building TOD systems often lack a user-friendly interface . a toolkit with advanced, easily integrable modules is needed to bridge this gap .
Approach:	They propose a multifaceted dialogue system toolkit that integrates diverse datasets and models with a streamlined training process and in-depth evaluation tools.
Outcome:	The proposed toolkit combines RL and transfer learning to support the rapid development and evaluation of robust dialogue policies.

Copied to clipboard

Kaixin Ma, Hao Cheng, Yu Zhang, Xiaodong Liu, Eric Nyberg, Jianfeng Gao

Information Retrieval

Challenge:	Using customized retrieval models, model transferability and scalability are limited.
Approach:	They propose a modular retrieval model where individual modules correspond to key skills that can be reused across datasets.
Outcome:	The proposed model outperforms self-supervised retrievers in zero-shot evaluations and achieves state-of-the-art fine-tuned retrieval performance on NQ, HotpotQA and OTT-QA.

MindAgent: Emergent Gaming Interaction (2024.findings-naacl)

Copied to clipboard

Ran Gong, Qiuyuan Huang, Xiaojian Ma, Yusuke Noda, Zane Durante, Zilong Zheng, Demetri Terzopoulos, Li Fei-Fei, Jianfeng Gao, Hoi Vo

Logical Transformers: Infusing Logical Structures into Pre-Trained Language Models (2023.findings-acl)

Challenge:	Large foundation models (LFMs) can perform complex scheduling in a multi-agent system and can coordinate agents to complete complex tasks that require extensive collaboration.
Approach:	They propose a gaming-based infrastructure that evaluates LFMs' planning and coordination capabilities in the context of gaming interaction.
Outcome:	The proposed infrastructure can be deployed in a customized VR version of Cuisineworld and adapted in the “Minecraft” domain.

Copied to clipboard

Borui Wang, Qiuyuan Huang, Budhaditya Deb, Aaron Halfaker, Liqun Shao, Daniel McDuff, Ahmed Hassan Awadallah, Dragomir Radev, Jianfeng Gao

Discourse-Aware Neural Rewards for Coherent Text Generation (N18-1)

Challenge:	Existing pre-trained language models that ignore the logical structures underlying natural language text often lack the ability to capture and encode key logical information in the input sequences.
Approach:	They propose to construct logic-aware input embeddings for transformer language models through logic detection, logic mapping and hierarchical logical projections and then develop a new modeling paradigm that can upgrade existing transformer language model into logical transformers to boost their performance.
Outcome:	The proposed model can achieve superior performance on four important and challenging tasks.

Copied to clipboard

Antoine Bosselut, Asli Celikyilmaz, Xiaodong He, Jianfeng Gao, Po-Sen Huang, Yejin Choi

EmailSum: Abstractive Email Thread Summarization (2021.acl-long)

Challenge:	Existing approaches to train text generation models using cross-entropy loss do not always correlate well with achieving high scores on commonly used evaluation measures.
Approach:	They propose to use discourse-aware rewards to model cross-sentence ordering to approximate desired discourse structure to train a model of long, coherent text.
Outcome:	The proposed model produces more coherent and less repetitive text than models trained with cross-entropy or with commonly used scores as rewards.

Copied to clipboard

Shiyue Zhang, Asli Celikyilmaz, Jianfeng Gao, Mohit Bansal

Tree Prompting: Efficient Task Adaptation without Fine-Tuning (2023.emnlp-main)

Challenge:	Recent years have brought about interest in the task of summarizing conversation threads.
Approach:	They develop an email thread summarization dataset that contains human-annotated short and long email threads over a wide variety of topics.
Outcome:	The proposed dataset contains human-annotated short (30 words) and long (100 words) summaries of 2,549 email threads over a wide variety of topics.

Copied to clipboard

Chandan Singh, John Morris, Alexander Rush, Jianfeng Gao, Yuntian Deng

SimulatorArena: Are User Simulators Reliable Proxies for Multi-Turn Evaluation of AI Assistants? (2025.emnlp-main)

Challenge:	Pretrained language models (LMs) are the main interface for applying them to new tasks, but their large size makes them difficult to fine-tune with gradients for specific downstream tasks.
Approach:	They propose to use training data to form a decision tree based on prompt-LM calls, with each prompt determined by the outcomes of previous calls.
Outcome:	The proposed method improves accuracy over competing methods and is competitive with fine-tuning.

Copied to clipboard

Yao Dou, Michel Galley, Baolin Peng, Chris Kedzie, Weixin Cai, Alan Ritter, Chris Quirk, Wei Xu, Jianfeng Gao

UnitedQA: A Hybrid Approach for Open Domain Question Answering (2021.acl-long)

Challenge:	Large language models (LLMs) are increasingly used in interactive applications, and human evaluation remains the gold standard for assessing their performance in multi-turn conversations.
Approach:	They propose to use large language models to simulate users for automatic assistant evaluation.
Outcome:	The proposed model outperforms human evaluations on two interactive tasks and achieves Spearman’s of 0.7 on both tasks.

Copied to clipboard

Hao Cheng, Yelong Shen, Xiaodong Liu, Pengcheng He, Weizhu Chen, Jianfeng Gao

Multi-step Reasoning via Recurrent Dual Attention for Visual Dialog (P19-1)

Challenge:	Recent work on open-domain question answering focuses on either extractive or generative readers exclusively.
Approach:	They propose a hybrid approach to extractive and generative readers that leverages both models.
Outcome:	The proposed approach outperforms state-of-the-art models on NaturalQuestions and TriviaQA respectively.

Copied to clipboard

Zhe Gan, Yu Cheng, Ahmed Kholy, Linjie Li, Jingjing Liu, Jianfeng Gao

Open-domain Question Answering via Chain of Reasoning over Heterogeneous Knowledge (2022.findings-emnlp)

Challenge:	Existing models for visual dialog infer the answer through multiple reasoning steps.
Approach:	They propose a model for visual dialog that uses multi-step reasoning to answer questions about an image.
Outcome:	The proposed model achieves a new state-of-the-art of 64.47% on the VisDial v1.0 dataset .

Copied to clipboard

Kaixin Ma, Hao Cheng, Xiaodong Liu, Eric Nyberg, Jianfeng Gao

Enriching Transformers with Structured Tensor-Product Representations for Abstractive Summarization (2021.naacl-main)

Challenge:	Existing open-domain question answering methods rely on the retriever to gather all evidence in isolation, but our approach uses an intermediary module to perform a chain of reasoning over the retrieved set.
Approach:	They propose a new open-domain question answering framework that integrates an intermediary module into the current retriever-reader pipeline and integrates it into the model.
Outcome:	The proposed framework outperforms the state-of-the-art on two OTT-QA datasets with an exact match score of 47.3 (45% relative gain).

Copied to clipboard

Yichen Jiang, Asli Celikyilmaz, Paul Smolensky, Paul Soulos, Sudha Rao, Hamid Palangi, Roland Fernandez, Caitlin Smith, Mohit Bansal, Jianfeng Gao

Implicit Deep Latent Variable Models for Text Generation (D19-1)

Challenge:	Abstractive summarization is the task of generating a concise summary of input documents . a middle-aged man and a young girl died after they were unable to avoid the plane .
Approach:	They propose a model that enriches the original Transformer with a Tensor Product Representation for abstractive summarization.
Outcome:	The proposed model outperforms the Transformer and the original TP-Transformer significantly on several datasets.

Copied to clipboard

Le Fang, Chunyuan Li, Jianfeng Gao, Wen Dong, Changyou Chen

Model-Generated Pretraining Signals Improves Zero-Shot Generalization of Text-to-Text Transformers (2023.acl-long)

Challenge:	Variational auto-encoders have been used for text generation but their representation power is limited due to two reasons.
Approach:	They advocate sample-based representations of variational distributions for natural language . they further develop an LVM to directly match the aggregated posterior to the prior .
Outcome:	The proposed model can be viewed as a natural extension of VAEs with a regularization of maximizing mutual information, mitigating the "posterior collapse" issue.

Copied to clipboard

Linyuan Gong, Chenyan Xiong, Xiaodong Liu, Payal Bajaj, Yiqing Xie, Alvin Cheung, Jianfeng Gao, Xia Song

Subgoal Discovery for Hierarchical Dialogue Policy Learning (D18-1)

Challenge:	Recent work in NLP has shown that pretrained language models have made notable progress toward generalization to unseen tasks.
Approach:	They propose to pretrain T5 using an auxiliary model to construct more challenging token replacements for the main model to denoise.
Outcome:	The proposed model outperforms similar-sized baseline models on prompted NLP benchmarks and rivals the state-of-the-art model with only 8% of its parameters.

Copied to clipboard

Da Tang, Xiujun Li, Jianfeng Gao, Chong Wang, Lihong Li, Tony Jebara

Self-Checker: Plug-and-Play Modules for Fact-Checking with Large Language Models (2024.findings-naacl)

Challenge:	Existing methods to develop dialogue agents for complex tasks require sparse reward signals.
Approach:	They propose a divide-and-conquer approach that exploits the hidden structure of a task . they use subgoals to divide a goal-oriented task into simpler subgoal sets .
Outcome:	The proposed approach performs competitively against state-of-the-art methods that require human-defined subgoals.

Copied to clipboard

Miaoran Li, Baolin Peng, Michel Galley, Jianfeng Gao, Zhu Zhang

DIONYSUS: A Pre-trained Model for Low-Resource Dialogue Summarization (2023.acl-long)

Challenge:	Existing methods for fact-checking text generated by large language models are expensive and time-consuming.
Approach:	They propose a plug-and-play framework that harnesses large language models for efficient fact-checking in a few-shot manner.
Outcome:	The proposed framework is compared with state-of-the-art models and shows that it can be used to speed up fact-checking in a few-shot manner.

Copied to clipboard

Yu Li, Baolin Peng, Pengcheng He, Michel Galley, Zhou Yu, Jianfeng Gao

Generation-Augmented Retrieval for Open-Domain Question Answering (2021.acl-long)

Challenge:	Existing methods for summarizing dialogues lack in taking into account the structure of dialogues and rely heavily on labeled data.
Approach:	They propose a pre-trained encoder-decoder model for summarizing dialogues in any new domain.
Outcome:	The proposed model outperforms existing methods on six datasets and shows ROUGE scores in zero-shot and few-shot settings.

Copied to clipboard

Yuning Mao, Pengcheng He, Xiaodong Liu, Yelong Shen, Jianfeng Gao, Jiawei Han, Weizhu Chen

PlotMachines: Outline-Conditioned Generation with Dynamic Plot State Tracking (2020.emnlp-main)

Challenge:	Existing approaches to answer open-domain questions use sparse representations and sparsity.
Approach:	They propose a method which augments a query by generating relevant contexts from heuristically discovered contexts without external supervision.
Outcome:	The proposed approach outperforms state-of-the-art dense retrieval methods on natural questions and triviaQA datasets.

Copied to clipboard

Hannah Rashkin, Asli Celikyilmaz, Yejin Choi, Jianfeng Gao

A Hybrid Neural Network Model for Commonsense Reasoning (D19-60)

Challenge:	a novel task is to generate a coherent narrative consistent with an outline . large-scale language models are not sufficient in generating coherent narratives for the given outline despite their impressive generation performance .
Approach:	They propose a task of outline-conditioned story generation that generates a coherent narrative . they propose 'plotmachines' that tracks dynamic plot states and learns different writing styles .
Outcome:	The proposed model can generate a coherent story by tracking the dynamic plot states while conditioning on the input outline while generating the full story.

Copied to clipboard

Pengcheng He, Xiaodong Liu, Weizhu Chen, Jianfeng Gao

GO FIGURE: A Meta Evaluation of Factuality in Summarization (2021.findings-acl)

Challenge:	a hybrid neural network (HNN) model for commonsense reasoning is proposed . it combines language models and semantic similarity models to achieve new state-of-the-art results .
Approach:	They propose a hybrid neural network model for commonsense reasoning . it combines a masked language model and a semantic similarity model .
Outcome:	The proposed model outperforms the WNLI, WSC and PDP60 benchmarks on three commonsense reasoning tasks.

Copied to clipboard

Saadia Gabriel, Asli Celikyilmaz, Rahul Jha, Yejin Choi, Jianfeng Gao

Summarization

Challenge:	Recent advances in text generation systems produce fluent, coherent, relevant, and factually correct text.
Approach:	They propose a metaevaluation framework for evaluating factuality evaluation metrics . they propose five necessary conditions to evaluate factual metrics on diagnostic factuity data .
Outcome:	The proposed framework provides robust evaluation that is extensible to multiple types of factual consistency and standard generation metrics, including QA metrics.

Teaching Language Models to Self-Improve through Interactive Demonstrations (2024.naacl-long)

Copied to clipboard

Xiao Yu, Baolin Peng, Michel Galley, Jianfeng Gao, Zhou Yu

TIGEr: Text-to-Image Grounding for Image Caption Evaluation (D19-1)

Challenge:	Large language models (LLMs) have been shown to improve performance on downstream tasks by prompting them to analyze and revise their outputs.
Approach:	They propose a training algorithm that prompts large language models to analyze and revise their own outputs and uses this feedback to train the small model.
Outcome:	The proposed approach improves LLaMA-7B's performance on math and reasoning tasks by up to 7.13%.

Copied to clipboard

Ming Jiang, Qiuyuan Huang, Lei Zhang, Xin Wang, Pengchuan Zhang, Zhe Gan, Jana Diesner, Jianfeng Gao

Multi-Task Deep Neural Networks for Natural Language Understanding (P19-1)

Challenge:	Existing metrics based on text-level comparisons fail to assess the quality of captions produced by machines.
Approach:	They propose to use a machine-learned text-image grounding model to measure the accuracy of machine-generated captions and their correlation with human judgments.
Outcome:	The proposed metric has higher consistency with human judgments and is more accurate than existing metrics.

Copied to clipboard

Xiaodong Liu, Pengcheng He, Weizhu Chen, Jianfeng Gao

SynthAgent: Adapting Web Agents with Synthetic Supervision (2026.acl-long)

Challenge:	Existing approaches to learning vector-space representations of text are multitask learning and language model pre-training.
Approach:	They propose a multi-task deep neural network (MT-DNN) that leverages cross-task data and incorporates a pre-trained bidirectional transformer language model.
Outcome:	The proposed model achieves state-of-the-art on ten NLU tasks and pushes the GLUE benchmark to 82.7% (2.2% absolute improvement)

Copied to clipboard

Zhaoyang Wang, Yiming Liang, Xuchao Zhang, Qianhui Wu, Siwei Han, Anson Bastos, Rujia Wang, Chetan Bansal, Baolin Peng, Jianfeng Gao, Saravan Rajmohan, Huaxiu Yao

Posterior Differential Regularization with f-divergence for Improving Model Robustness (2021.naacl-main)

Challenge:	Existing studies have focused on synthetic supervision but have encountered data quality issues.
Approach:	They propose a fully synthetic supervision framework that aims at improving data quality via dual refinement of both tasks and trajectories.
Outcome:	The proposed framework outperforms existing methods on standardized benchmarks and shows promising results on a standardized test.

Copied to clipboard

Hao Cheng, Xiaodong Liu, Lis Pereira, Yaoliang Yu, Jianfeng Gao

ConvLab-2: An Open-Source Toolkit for Building, Evaluating, and Diagnosing Dialogue Systems (2020.acl-demos)

Challenge:	Recent studies show that pre-trained models suffer catastrophic degradation in out-of-domain generalization to datasets with domain shift or adversarial scenarios.
Approach:	They propose to regularize the posterior difference between clean and noisy inputs by using a Jacobian regularization framework and a virtual adversarial training framework.
Outcome:	The proposed framework can improve model robustness in fully supervised and semi-supervised settings.

Copied to clipboard

Qi Zhu, Zheng Zhang, Yan Fang, Xiang Li, Ryuichi Takanobu, Jinchao Li, Baolin Peng, Jianfeng Gao, Xiaoyan Zhu, Minlie Huang

Structuring Latent Spaces for Stylized Response Generation (D19-1)

Challenge:	ConvLab-2 inherits Convlab's framework but integrates more powerful dialogue models and supports more datasets.
Approach:	They present ConvLab-2, an open-source toolkit that enables researchers to build task-oriented dialogue systems with state-of-the-art models and perform an end-to-end evaluation.
Outcome:	The new tool inherits ConvLab's framework and extends it by integrating many recently proposed state-of-the-art dialogue models.

Copied to clipboard

Xiang Gao, Yizhe Zhang, Sungjin Lee, Michel Galley, Chris Brockett, Jianfeng Gao, Bill Dolan

Soloist: Building Task Bots at Scale with Transfer Learning and Machine Teaching (2021.tacl-1)

Challenge:	Existing methods for generating responses in a targeted style are limited by the lack of parallel data.
Approach:	They propose a method that bridges conversation modeling and non-parallel style transfer by sharing a structured latent space.
Outcome:	The proposed system generates responses of the targeted style and outperforms baselines without sacrificing appropriateness.

Copied to clipboard

Baolin Peng, Chunyuan Li, Jinchao Li, Shahin Shayandeh, Lars Liden, Jianfeng Gao

Budgeted Policy Learning for Task-Oriented Dialogue Systems (P19-1)

Challenge:	Existing methods for building task-oriented dialog systems are limited to a few tasks and domains.
Approach:	They propose a method that uses transfer learning and machine teaching to build task bots at scale.
Outcome:	The proposed method outperforms existing methods on well-studied task-oriented dialog benchmarks on well studied tasks.

Copied to clipboard

Zhirui Zhang, Xiujun Li, Jianfeng Gao, Enhong Chen

AdaMix: Mixture-of-Adaptations for Parameter-efficient Model Tuning (2022.emnlp-main)

Challenge:	Existing approaches to integrating reinforcement learning into task-oriented dialogue systems require a fixed, small amount of user interactions to learn.
Approach:	They propose a budget-conscious scheduling approach that optimizes a fixed, small amount of user interactions for dialogue agent learning.
Outcome:	The proposed approach improves on a movie-ticket booking task with simulated and real users.

Copied to clipboard

Yaqing Wang, Sahaj Agarwal, Subhabrata Mukherjee, Xiaodong Liu, Jing Gao, Ahmed Hassan Awadallah, Jianfeng Gao

Guided Dialogue Policy Learning without Adversarial Learning in the Loop (2020.findings-emnlp)

Challenge:	Standard fine-tuning of large pre-trained language models requires updating hundreds of millions to billions of parameters and storing a large copy of the PLM weights for every task.
Approach:	They propose a parameter-efficient fine-tuning technique where small trainable components are injected into the PLM and updated during fine-uning.
Outcome:	The proposed method outperforms SOTA parameter-efficient fine-tuning and full model fine-uning on GLUE development set with RoBERTa-large encoder.

Copied to clipboard

Ziming Li, Sungjin Lee, Baolin Peng, Jinchao Li, Julia Kiseleva, Maarten de Rijke, Shahin Shayandeh, Jianfeng Gao

Toward Compositional Behavior in Neural Models: A Survey of Current Views (2024.emnlp-main)

Challenge:	Reinforcement learning methods suffer from sparse and unstable reward signals . alternating training of dialogue agent and reward model can get stuck in local optima .
Approach:	They propose to decompose adversarial training into two steps to improve dialogue policy learning.
Outcome:	The proposed method achieves remarkable task success rate using both on-policy and off-poly reinforcement learning methods.

Copied to clipboard

Kate McCurdy, Paul Soulos, Paul Smolensky, Roland Fernandez, Jianfeng Gao

Challenge:	Compositionality is a core property of natural language, and it is regarded as a key goal for modern NLP systems.
Approach:	They propose a conceptual framework to address compositionality in NLP . they propose to use this framework to survey researchers active in this area .
Outcome:	The proposed framework finds consensus on key points and suggests that scale alone is unlikely to achieve the desired behavior.

MIND: A Large-scale Dataset for News Recommendation (2020.acl-main)

Copied to clipboard

Fangzhao Wu, Ying Qiao, Jiun-Hung Chen, Chuhan Wu, Tao Qi, Jianxun Lian, Danyang Liu, Xing Xie, Jianfeng Gao, Winnie Wu, Ming Zhou

SciEx: Benchmarking Large Language Models on Scientific Exams with Human Expert Grading and Automatic Grading (2024.emnlp-main)

Challenge:	Personalized news recommendation is an important technique for personalized news service.
Approach:	They propose to build a large-scale news recommendation dataset from Microsoft News . they demonstrate that news recommendation relies on the quality of news content understanding .
Outcome:	The proposed dataset contains 1 million users and more than 160k English news articles, each of which has rich textual content such as title, abstract and body.

Copied to clipboard

Tu Dinh, Carlos Mullov, Leonard Bärmann, Zhaolin Li, Danni Liu, Simon Reiß, Jueun Lee, Nathan Lerzer, Jianfeng Gao, Fabian Peller-Konrad, Tobias Röddiger, Alexander Waibel, Tamim Asfour, Michael Beigl, Rainer Stiefelhagen, Carsten Dachsbacher, Klemens Böhm, Jan Niehues

Open Domain Question Answering with A Unified Knowledge Interface (2022.acl-long)

Challenge:	Large Language Models (LLMs) are rapidly developing and are becoming more and more useful in scientific tasks.
Approach:	They propose to use LLM-as-a-judge to grade LLMs on SciEx to assess their ability on scientific tasks.
Outcome:	The proposed benchmarks show that the LLMs perform decently on free-form exams, achieving 0.948 Pearson correlation with expert grading.

Copied to clipboard

Kaixin Ma, Hao Cheng, Xiaodong Liu, Eric Nyberg, Jianfeng Gao

Structured Data in NLP

Challenge:	a retriever-reader framework is popular for open domain question answering . however, accessing heterogeneous knowledge sources through a unified interface remains unknown .
Approach:	They propose a retriever-reader framework that uses explicit knowledge to access heterogeneous knowledge sources through a unified interface.
Outcome:	The proposed framework can benefit from the expanded knowledge index, the authors show . their approach sets the single-model state-of-the-art on Natural Questions .

REO-Relevance, Extraness, Omission: A Fine-grained Evaluation for Image Captioning (D19-1)

Copied to clipboard

Ming Jiang, Junjie Hu, Qiuyuan Huang, Lei Zhang, Jana Diesner, Jianfeng Gao

Targeted Adversarial Training for Natural Language Understanding (2021.naacl-main)

Challenge:	Existing metrics for image captioning evaluation provide an overall quality score, which is difficult to infer specific description errors.
Approach:	They propose a fine-grained evaluation method REO for automatically measuring the performance of image captioning systems.
Outcome:	The proposed method achieves higher consistency with human judgments and provides more intuitive evaluation results than other metrics.

Copied to clipboard

Lis Pereira, Xiaodong Liu, Hao Cheng, Hoifung Poon, Jianfeng Gao, Ichiro Kobayashi

Robust Navigation with Language Pretraining and Stochastic Sampling (D19-1)

Challenge:	Existing adversarial training approaches focus on making adversarials less expensive or regularizing rather than replacing the standard training objective.
Approach:	They propose an algorithm to introspect current mistakes and prioritize adversarial training steps to where the model errs the most.
Outcome:	The proposed algorithm improves adversarial training for natural language understanding by introspecting mistakes and prioritizing training steps to where the model errs the most.

Copied to clipboard

Xiujun Li, Chunyuan Li, Qiaolin Xia, Yonatan Bisk, Asli Celikyilmaz, Jianfeng Gao, Noah A. Smith, Yejin Choi

Diversifying the Expert Knowledge for Task-Agnostic Pruning in Sparse Mixture-of-Experts (2025.findings-acl)

Challenge:	Existing methods to learn visual representations and action decoding schemes are limited to previously unseen instructions and environments.
Approach:	They propose a stochastic sampling scheme to reduce the gap between the expert actions in training and sampled actions in test to correct its own mistakes.
Outcome:	The proposed methods achieve 6% absolute gain over the previous best results on the Room-to-Room benchmark.

Copied to clipboard

Zeliang Zhang, Xiaodong Liu, Hao Cheng, Chenliang Xu, Jianfeng Gao

Challenge:	Large Language Models (LLMs) have outstanding performance by learning a large number of model parameters on large amounts of data.
Approach:	They propose a method of grouping and pruning similar experts to improve the model’s parameter efficiency by a range of natural language tasks.
Outcome:	The proposed method outperforms other model pruning methods on a range of natural language tasks.

Text Editing by Command (2021.naacl-main)

Copied to clipboard

Felix Faltings, Michel Galley, Gerold Hintz, Chris Brockett, Chris Quirk, Jianfeng Gao, Bill Dolan

LiST: Lite Prompted Self-training Makes Parameter-efficient Few-shot Learners (2022.findings-naacl)

Challenge:	Recent work has focused on making such models more controllable and factually grounded.
Approach:	They propose a novel interactive text generation setting in which the user interacts with the system by issuing commands to edit existing text.
Outcome:	The proposed model outperforms baseline models and obtains positive results in automatic and human evaluations.

Copied to clipboard

Yaqing Wang, Subhabrata Mukherjee, Xiaodong Liu, Jing Gao, Ahmed Awadallah, Jianfeng Gao

ConvLab: Multi-Domain End-to-End Dialog System Platform (P19-3)

Challenge:	LiST is an efficient method for fine-tuning large pre-trained language models in few-shot learning settings.
Approach:	They propose a method for efficient fine-tuning of large pre-trained language models in few-shot settings using self-training and meta-learning.
Outcome:	The proposed method outperforms GPT-3 in-context learning by 33% on few-shot tasks.

Copied to clipboard

Sungjin Lee, Qi Zhu, Ryuichi Takanobu, Zheng Zhang, Yaoqin Zhang, Xiang Li, Jinchao Li, Baolin Peng, Xiujun Li, Minlie Huang, Jianfeng Gao

Towards Generating Long and Coherent Text with Multi-Level Latent Variable Models (P19-1)

Challenge:	ConvLab is an open-source multi-domain end-to-end dialog system platform . it allows researchers to quickly set up experiments with reusable components and compare a large set of different approaches in common environments.
Approach:	They propose to use an open-source multi-domain end-to-end dialog system platform to train and evaluate dialog bots in common environments.
Outcome:	The proposed system enables researchers to quickly set up experiments with reusable components and compare a large set of different approaches in common environments.

Copied to clipboard

Dinghan Shen, Asli Celikyilmaz, Yizhe Zhang, Liqun Chen, Xin Wang, Jianfeng Gao, Lawrence Carin

Cyclical Annealing Schedule: A Simple Approach to Mitigating KL Vanishing (N19-1)

Challenge:	Variational autoencoders (VAEs) have received much attention as an end-to-end architecture for text generation with latent variables.
Approach:	They propose to leverage several multi-level structures to learn a variational autoencoder model for generating long, and coherent text.
Outcome:	The proposed model produces more coherent and less repetitive long text compared to baselines and mitigates posterior collapse issue.

Copied to clipboard

Hao Fu, Chunyuan Li, Xiaodong Liu, Jianfeng Gao, Asli Celikyilmaz, Lawrence Carin

Jointly Optimizing Diversity and Relevance in Neural Response Generation (N19-1)

Challenge:	Variational autoencoders (VAEs) with an auto-regressive decoder have been applied for many natural language processing tasks.
Approach:	They propose a cyclical annealing schedule which repeats the process of increasing multiple times to learn more meaningful latent codes progressively by leveraging previous learning cycles as warm re-restart.
Outcome:	The proposed method improves on a broad range of NLP tasks, including language modeling, dialog response generation and semi-supervised text classification.

Copied to clipboard

Xiang Gao, Sungjin Lee, Yizhe Zhang, Chris Brockett, Michel Galley, Jianfeng Gao, Bill Dolan

SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization (2020.acl-main)

Challenge:	Recent neural conversation models often generate bland and generic responses . however, the improvement often comes at the cost of decreased relevance .
Approach:	They propose a spacefusion model to jointly optimize diversity and relevance that fuses the latent space of a sequence-to-sequence model and that of an autoencoder model by leveraging novel regularization terms.
Outcome:	The proposed model improves diversity and relevance compared to baselines in both diversity and diversity.

Copied to clipboard

Haoming Jiang, Pengcheng He, Weizhu Chen, Xiaodong Liu, Jianfeng Gao, Tuo Zhao

Contrastive Multi-document Question Generation (2021.eacl-main)

Challenge:	Existing methods for fine-tuning pre-trained models fail to generalize to unseen data.
Approach:	They propose a framework for robust and efficient fine-tuning for pre-trained models . proposed framework achieves new state-of-the-art performance on a number of NLP tasks .
Outcome:	The proposed framework outperforms the state-of-the-art T5 model on GLUE, SNLI, SciTail and ANLI.

Copied to clipboard

Woon Sang Cho, Yizhe Zhang, Sudha Rao, Asli Celikyilmaz, Chenyan Xiong, Jianfeng Gao, Mengdi Wang, Bill Dolan

ECoK: Emotional Commonsense Knowledge Graph for Mining Emotional Gold (2024.findings-acl)

Challenge:	Multi-document question generation focuses on generating a question that covers the common aspect of multiple documents, but a naive model trained only using the targeted document set may generate too generic questions that cover a larger scope than delineated by the document set.
Approach:	They propose a contrastive learning strategy where given ‘positive’ and ‘negative’ sets of documents, generate a question that is closely related to the ‘positive' set but far away from the ‘negative' set.
Outcome:	The proposed model significantly outperforms several strong baselines, as measured by automatic metrics and human evaluation.

Copied to clipboard

Zhunheng Wang, Xiaoyi Liu, Mengting Hu, Rui Ying, Ming Jiang, Jianfeng Wu, Yalan Xie, Hang Gao, Renhong Cheng