semi使用者話語解析部分：

1.基於知識共享的大規模多域信念跟蹤 Large-scale Multi-Domain Belief Tracking with Knowledge Sharing(2018)

連結：https://arxiv.org/abs/1807.06517

2.聯合學習特色提取器的封建對話管理Feudal Dialogue Management with Jointly Learned Feature Extractors(2018)

連結：http://aclweb.org/anthology/W18-5038

3.處理物件及其關係:對話實體對話模型: Addressing Objects and Their Relations: The Conversational Entity (2018)

連結http://mi.eng.cam.ac.uk/~sjy/papers/ubcr18.pdf 

4.神經信念跟蹤:資料驅動的對話狀態跟蹤:Neural Belief Tracker: Data-Driven Dialogue State Tracking(2017)

連結:http://mi.eng.cam.ac.uk/~sjy/papers/mowt17.pdf

5.使用簡單的特定於語言的規則微調單詞向量空間: Fine-Tuning Word Vector Spaces with Simple Language-Specific Rules(2017)

連結:http://mi.eng.cam.ac.uk/~sjy/papers/mowt17.pdf

6.使用單語和跨語言約束對分佈詞向量空間進行語義專門化Semantic Specialisation of Distributional Word Vector Spaces using Monolingual and Cross-Lingual Constraints(2017)

連結:http://mi.eng.cam.ac.uk/~sjy/papers/mvol17.pdf

7.潛在意圖對話模式分析:Latent Intention Dialogue Models(2017)

連結:http://mi.eng.cam.ac.uk/~sjy/papers/wmby17.pdf

policy對話政策部分：

1.基於語料庫的口語對話系統優化策略的神經使用者模擬 Neural User Simulation for Corpus-based Policy Optimisation of Spoken Dialogue Systems (2018)

連結：https://arxiv.org/abs/1805.06966

2.基於深度強化學習的不確定度基準估計用於對話政策優化:Benchmarking Uncertainty Estimates with Deep Reinforcement Learning for Dialogue Policy Optimisation(2018)

連結:https://arxiv.org/pdf/1711.11486.pdf

3.多領域的強化學習:“Feudal Reinforcement Learning for Dialogue Management in Large Domains(2018)

連結:https://arxiv.org/pdf/1803.03232.pdf

4.基於強化學習的面向任務的對話管理:A Benchmarking Environment for Reinforcement Learning Based Task Oriented Dialogue Management(2017)

連結:https://arxiv.org/pdf/1711.11023.pdf

5.基於神經網路的對話策略優化的不確定性估計:Uncertainty Estimates for Efficient Neural Network-based Dialogue Policy Optimisation(2017)

連結:https://arxiv.org/pdf/1711.11486.pdf

6.高效的取樣和有監督的強化學習的對話管理:Sample-efficient Actor-Critic Reinforcement Learning with Supervised Data for Dialogue Management(2017)

連結:https://arxiv.org/abs/1707.00130

7.對話管理的子領域建模與分層強化學習:Sub-domain Modelling for Dialogue Management with Hierarchical Reinforcement Learning(2017)

連結:https://arxiv.org/abs/1706.06210

8.對話策略學習領域及使用者滿意度獎勵評估:Domain-independent User Satisfaction Reward Estimation for Dialogue Policy Learning(2017)

連結:http://mi.eng.cam.ac.uk/~sjy/papers/ubcm17.pdf

9.對話管理器領域中使用高斯過程強化學習:Dialogue manager domain adaptation using Gaussian process reinforcement learning(2017)

連結:http://mi.eng.cam.ac.uk/~sjy/papers/gmrs17.pdf

semo對話生成部分：

1.口語對話系統的變域跨域自然語言生成  Variational Cross-domain Natural Language Generation for Spoken Dialogue Systems(2018)

連結：http://mi.eng.cam.ac.uk/~flk24/doc/CVAE_SIGDIAL_2018.pdf

未分類:

1.基於網路的端到端可訓練的面向任務的對話系統：A Network-based End-to-End Trainable Task-oriented Dialogue System(2017)

連結：https://arxiv.org/pdf/1711.11486.pdf

論文連結：http://dialogue.mi.eng.cam.ac.uk/index.php/publications/