
电子邮件: linzheng@iie.ac.cn
通信地址: 北京市海淀区树村路19号中科院信工所2号楼
邮政编码: 100085
研究领域
研究方向:自然语言处理、多模态分析、认知计算
具体包括:情感分析、内容生成、大模型轻量化等
招生信息
1. 具备推免资格,计算机相关专业,具有良好的数学、英语、编程、逻辑能力;
2. 本科期间发表过人工智能、自然语言处理等领域论文优先;
3. 希望保研生大四可以来实验室客座或者远程毕设,打好基础,做好科研过渡;
想要更多了解小组情况,可以关注:iienlp.github.io
招生专业
招生方向
工作经历
工作简历
教授课程
专利与奖励
奖励信息
出版信息
Trust-GRS: A Trustworthy Training Framework for Graph Neural Network Based Recommender Systems against Shilling Attacks. AAAI 2025 (CCF A)
Relation Also Knows: Rethinking the Recall and Editing of Factual Associations in Auto-Regressive Transformer Language Models. AAAI 2025 (CCF A)
Multimodal Hypothetical Summary for Retrieval-based Multi-image Question Answering. AAAI 2025 (CCF A)
Sibyl: Empowering Empathetic Dialogue Generation in Large Language Models via Sensible and Visionary Commonsense Inference. COLING, 2025
Towards One-to-Many Visual Question Answering. findings of EMNLP, 2024
Towards Flexible Evaluation for Generative Visual Question Answering. ACM MM, 2024 (CCF A)
Disrupting Diffusion: Token-Level Attention Erasure Attack against Diffusion-based Customization. ACM MM, 2024(CCF A)
Prediction Exposes Your Face: Black-box Model Inversion via Prediction Alignment. ECCV, 2024
Multimodal Table Understanding. ACL, 2024 (CCF A)
Light-PEFT: Lightening Parameter-Efficient Fine-Tuning via Early Pruning. findings of ACL, 2024
Pruning Large Language Models to Intra-module Low-rank Structure with Transitional Activations. findings of ACL, 2024
A Pedestrian is Worth One Prompt: Towards Language Guidance Person Re-Identification. CVPR, 2024 (CCF A)
Object Attribute Matters in Visual Question Answering. AAAI, 2024 (CCF A)
Multi-level Adaptive Contrastive Learning for Knowledge Internalization in Dialogue Generation. EMNLP, 2023
Compressing and Debiasing Vision-Language Pre-Trained Models for Visual Question Answering. EMNLP, 2023
An Empirical Study of Instruction-tuning Large Language Models in Chinese. findings of EMNLP, 2023
Chain-of-Thought Reasoning in Tabular Language Models. findings of EMNLP, 2023
IM-TQA: A Chinese Table Question Answering Dataset with Implicit and Multi-type Table Structures. ACL, 2023 (CCF A)
Combo of Thinking and Observing for Outside-Knowledge VQA. ACL, 2023 (CCF A)
A Gradient Control Method for Backdoor Attacks on Parameter-Efficient Tuning. ACL, 2023 (CCF A)
Question-Interlocutor Scope Realized Graph Modeling over Key Utterances for Dialogue Reading Comprehension. Findings of ACL, 2023
Maximum Entropy Loss, the Silver Bullet Targeting Backdoor Attacks in Pre-trained Language Models. Findings of ACL, 2023
Divide, Conquer, and Combine: Mixture of Semantic-Independent Experts for Zero-Shot Dialogue State Tracking. ACL, 2023 (CCF A)
DiffusEmp: A Diffusion Model-Based Framework with Multi-Grained Control for Empathetic Response Generation. ACL, 2023 (CCF A)
Retrieve-and-Sample: Document-level Event Argument Extraction via Hybrid Retrieval Augmentation. ACL, 2023 (CCF A)
Explainable Hyperbolic Temporal Point Process for User-Item Interaction Sequence Generation. TOIS, 2023 (CCF A)
Multi-Aspect Heterogeneous Graph Augmentation. WWW, 2023 (CCF A)
Towards Polymorphic Adversarial Examples Generation for Short Text. ICASSP, 2023
SERI: SkEtching-Reasoning-Integrating Progressive Workflow for Empathetic Response Generation. ICASSP, 2023
Explainable Hyperbolic Temporal Point Process for User-Item Interaction Sequence Generation. ACM Transactions on Information Systems, 2022 (CCF A)
- A Win-win Deal: Towards Sparse and Robust Pre-trained Language Models. NeurIPS, 2022 (CCF A)
- Detach and Attach: Stylized Image Captioning without Paired Stylized Dataset. ACM MM, 2022 (CCF A)
- Neutral Utterances are Also Causes: Enhancing Conversational Causal Emotion Entailment with Social Commonsense Knowledge. IJCAI, 2022 (CCF A)
- Neural Label Search for Zero-Shot Multi-Lingual Extractive Summarization. ACL, 2022 (CCF A)
- COST-EFF: Collaborative Optimization of Spatial and Temporal Efficiency with Slenderized Multi-exit Language Models. EMNLP, 2022
- Language Prior Is Not the Only Shortcut: A Benchmark for Shortcut Learning in VQA. findings of EMNLP, 2022
- Empathetic Dialogue Generation via Sensitive Emotion Recognition and Sensible Knowledge Selection. findings of EMNLP, 2022
- Towards Robust Visual Question Answering: Making the Most of Biased Samples via Contrastive Learning. findings of EMNLP, 2022
- Target Really Matters: Target-aware Contrastive Learning and Consistency Regularization for Few-shot Stance Detection. COLING, 2022
TAKE: Topic-shift Aware Knowledge sElection for Dialogue Generation. COLING, 2022
Slot Dependency Modeling for Zero-Shot Cross-Domain Dialogue State Tracking. COLING, 2022
CLIO: Role-interactive Multi-event Head Attention Network for Document-level Event Extraction. COLING, 2022
- Learning to Win Lottery Tickets in BERT Transfer via Task-agnostic Mask Training. NAACL, 2022
- Cross-Target Stance Detection via Refined Meta-Learning. ICASSP, 2022
- Connecting Targets via Latent Topics and Contrastive Learning: A Unified Framework for Robust Zero-shot and Few-shot Stance Detection. ICASSP, 2022
- Addressing Extraction and Generation Separately: Keyphrase Prediction with Pre-trained Language Models. IEEE Transactions on Audio, Speech and Language Processing, 2021
- Sarcasm Detection with Commonsense Knowledge. IEEE Transactions on Audio, Speech and Language Processing, 2021
- Past, Present, and Future: Conversational Emotion Recognition through Structural Modeling of Psychological Knowledge. Findings of EMNLP, 2021
- Check It Again: Progressive Visual Question Answering via Visual Entailment. ACL, 2021 (CCF A)
- Marginal Utility Diminishes: Exploring the Minimum Knowledge for BERT Knowledge Distillation. ACL, 2021 (CCF A)
- Enhanching Zero-shot and Few-shot Stance Detection with Commonsense Knowledge Graph. Findings of ACL, 2021 (CCF A)
- ROSITA: Refined BERT Compression with Integrated Technique. AAAI, 2021 (CCF A)
- Learning Class-Transductive Intent Representations for Zero-shot IntentDetection. IJCAI, 2021 (CCF A)
- MACROBERT: Maximizing Certified Region of BERT to Adversarial Word Substitutions. DASFAA, 2021
- Modeling Intra and Inter-modality Incongruity for Multi-Modal Sarcasm Detection. Findings of EMNLP, 2020
- Joint Character-Level Word Embedding and Adversarial Stability Training to Defend Adversarial Text. AAAI, 2020 (CCF A)
- Modeling the Incongruity between Sentence Snippets for Sarcasm Detection. ECAI, 2020
- Generating Paraphrase with Topic as Prior Knowledge. CIKM, 2019
- Ranking and Sampling in Open-Domain Question Answering. EMNLP, 2019
- Learning Sentiment-Specific Word Embedding via Global Sentiment Representation. AAAI, 2018 (CCF A)
- Deep Uniqueness-aware Hashing for Fine-grained Multi-label Image Retrieval. ICASSP, 2018
- Deep Supervised Hashing for Multi-Label and Large-Scale Image Retrieval. ICMR, 2017
- An Unsupervised Cross-Lingual Topic Model Framework for Sentiment Classification, IEEE/ACM Transactions on Audio, Speech and Language Processing, 2016
- A Word Vector and Matrix Factorization Based Method for Opinion Lexicon Extraction. WWW, 2015
- A Cross-Lingual Joint Aspect/Sentiment Model for Sentiment Analysis. CIKM, 2014
- Towards jointly extracting aspects and aspect-specific sentiment knowledge. CIKM, 2012
- Find me opinion sources in blogosphere: a unified framework for opinionated blog feed retrieval. WSDM, 2012
- Effective and efficient?bilingual sentiment lexicon extraction using collocation alignment. CIKM, 2012
- Language-independent sentiment classification using three common words. CIKM, 2011
- 文本后门攻击与防御综述. 计算机研究与发展,2023 (CCF A)
- 智能回复系统研究综述. 信息安全学报,2020
- 文本情绪分析综述. 计算机研究与发展,2018 (CCF A)
- 一种面向大规模社会信息网络的多层社区发现算法. 计算机学报,2016 (CCF A)
发表著作
科研活动
科研项目
毕业生去向
许彦夫 2017级硕士生 腾讯微信
耿 悦 2017级硕士生 国家公务员
吴 行 2017级硕士生 百度搜索
刘 睿 2017级直博生 阿里蚂蚁金服
佀庆一 2019级硕转博 华为天才少年
李江楠 2019级硕转博 腾讯微信模式识别
刘源鑫 2019级硕士生 北京大学
李世鲲 2019级硕转博 香港中文大学
谭宇彤 2020级硕士生 米哈游
温 哲 2020级硕士生 工商银行
梁宇航 2020级硕士生 移动研究院
王 彤 2021级硕士生 华为