跳到主要內容

臺灣博碩士論文加值系統

(44.192.48.196) 您好!臺灣時間:2024/06/14 17:21
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:郭蕙綺
研究生(外文):Hui-Chi Kuo
論文名稱:利用常識推理於對話系統中隱含意圖預測與任務推薦
論文名稱(外文):Implicit Intent Prediction and Recommendation with Commonsense Reasoning for Conversational Systems
指導教授:陳縕儂
指導教授(外文):Yun-Nung Chen
口試委員:李宏毅蔡宗翰曹昱
口試委員(外文):Hung-Yi LeeTzong-Han TsaiYu Tsao
口試日期:2023-01-19
學位類別:碩士
校院名稱:國立臺灣大學
系所名稱:資訊工程學系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2023
畢業學年度:111
論文頁數:35
中文關鍵詞:常識推理零樣本模板隱含意圖推薦對話系統
外文關鍵詞:Commonsense ReasoningZero-shot PromptingImplicit IntentRecommendationConversational Systems
DOI:10.6342/NTU202300366
相關次數:
  • 被引用被引用:0
  • 點閱點閱:176
  • 評分評分:
  • 下載下載:33
  • 收藏至我的研究室書目清單書目收藏:0
現有的虛擬智慧助理用以執行用戶明確提出的任務或服務,因此,跨多個領域的任務需要透過具有多個明確提出的意圖的對話來完成。然而,一般的人類助理能夠經由常識來推理出語句裡隱含的多個意圖,進而減少互動的次數,增加實用性。
本篇論文提出了一個多領域對話系統框架,先根據用戶的語句推斷隱含意圖,接著對大型預訓練語言模型進行零樣本提示,用以觸發適當的任務導向機器人。
提出的框架展現了對於隱含意圖理解的能力,並且可以經由零樣本的方式推薦相關的機器人。
Intelligent virtual assistants are currently designed to perform tasks or services explicitly mentioned by users, so multiple related domains or tasks need to be performed one by one through a long conversation with many explicit intents.
Instead, human assistants are capable of reasoning (multiple) implicit intents based on user utterances via commonsense knowledge, reducing complex interactions and improving practicality.
Therefore, this paper proposes a framework of multi-domain dialogue systems, which can automatically infer implicit intents based on user utterances and then perform zero-shot prompting using a large pre-trained language model to trigger suitable single task-oriented bots.
The proposed framework is demonstrated effective to realize implicit intents and recommend associated bots in a zero-shot manner.
口試委員審定書 i
致謝 iii
摘要 v
Abstract vii
Contents ix
List of Figures xiii
List of Tables xv
Chapter 1 Introduction 1
1.1 Motivation 1
1.2 Main Contribution 1
1.3 Thesis Structure 2
Chapter 2 Background 5
2.1 Deep Learning Models 5
2.1.1 Transformer 5
2.1.2 BART 6
2.1.3 GPT-J 7
2.1.4 GPT-3 8
Chapter 3 Related Work 9
3.1 Multi-Domain Realization 9
3.2 Commonsense Reasoning 9
3.3 Zero-Shot Prompting 10
Chapter 4 Framework 11
4.1 Overview 11
4.2 Commonsense-Inferred Intent Generation 11
4.2.1 Relation Trigger Selection 12
4.2.2 Commonsense Knowledge Generation 13
4.3 Zero-Shot Bot Recommendation 14
4.3.1 Pre-trained Language Model 14
4.3.2 Prompting for Bot Recommendation 14
Chapter 5 Experiments 17
5.1 Baseline 17
5.2 Strong Baseline 18
5.3 Automatic Evaluation Results 18
5.4 Human Evaluation Results 20
5.5 Qualitative Analysis 23
5.6 Extended Study 23
5.6.1 Framework Overview 23
5.6.2 Commonsense Inference 24
5.6.3 Paraphrase to Query 25
5.6.4 Qualitative Analysis 26
5.7 Reproducibility 26
Chapter 6 Discussion 29
Chapter 7 Conclusion 31
References 33
M. Sun, Y.-N. Chen, Z. Hua, Y. Tamres-Rudnicky, A. Dash, and A. Rudnicky, “App- dialogue: Multi-app dialogues for intelligent assistants,” in Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC’16), 2016, pp. 3127–3132.
A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin, “Attention is all you need,” Advances in neural information processing systems, vol. 30, 2017.
M. Lewis, Y. Liu, N. Goyal, M. Ghazvininejad, A. Mohamed, O. Levy, V. Stoyanov, and L. Zettlemoyer, “BART: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension,” in Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Online: Association for Computational Linguistics, Jul. 2020, pp. 7871–7880. [Online]. Available: https://aclanthology.org/2020.acl-main.703
B. Wang, “Mesh-Transformer-JAX: Model-Parallel Implementation of Transformer Language Model with JAX,” https://github.com/kingoflolz/mesh-transformer-jax, May 2021.
A. Radford, J. Wu, R. Child, D. Luan, D. Amodei, and I. Sutskever, “Language models are unsupervised multitask learners,” 2019.
R. Child, S. Gray, A. Radford, and I. Sutskever, “Generating long sequences with sparse transformers,” arXiv preprint arXiv:1904.10509, 2019.
M. Sun, Y.-N. Chen, and A. I. Rudnicky, “Helpr: A framework to break the barrier across domains in spoken dialog systems,” in Dialogues with social robots. Springer, 2017, pp. 257–269.
A. Bosselut, H. Rashkin, M. Sap, C. Malaviya, A. Celikyilmaz, and Y. Choi, “COMET: Commonsense transformers for automatic knowledge graph construction,” in Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. Florence, Italy: Association for Computational Linguistics, Jul. 2019, pp. 4762–4779. [Online]. Available: https://aclanthology.org/P19-1470
M. Sap, R. Le Bras, E. Allaway, C. Bhagavatula, N. Lourie, H. Rashkin, B. Roof, N. A. Smith, and Y. Choi, “Atomic: An atlas of machine commonsense for if-then reasoning,” in Proceedings of the AAAI conference on artificial intelligence, vol. 33, no. 01, 2019, pp. 3027–3035.
H. Zhang, D. Khashabi, Y. Song, and D. Roth, “Transomcs: From linguistic graphs to commonsense knowledge,” arXiv preprint arXiv:2005.00206, 2020.
J. D. Hwang, C. Bhagavatula, R. Le Bras, J. Da, K. Sakaguchi, A. Bosselut, and Y. Choi, “(comet-) atomic 2020: On symbolic and neural commonsense knowledge graphs,” in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, no. 7, 2021, pp. 6384–6392.
T. Brown, B. Mann, N. Ryder, M. Subbiah, J. D. Kaplan, P. Dhariwal, A. Neelakantan, P. Shyam, G. Sastry, A. Askell et al., “Language models are few-shot learners,” Advances in neural information processing systems, vol. 33, pp. 1877 1901, 2020.
Z. Zhao, E. Wallace, S. Feng, D. Klein, and S. Singh, “Calibrate before use: Improving few-shot performance of language models,” in International Conference on Machine Learning. PMLR, 2021, pp. 12697–12706.
T. Schick and H. Schütze, “Exploiting cloze questions for few shot text classification and natural language inference,” arXiv preprint arXiv:2001.07676, 2020.
T. Shin, Y. Razeghi, R. L. Logan IV, E. Wallace, and S. Singh, “Autoprompt: Eliciting knowledge from language models with automatically generated prompts,” arXiv preprint arXiv:2010.15980, 2020.
M. Sun, Y.-N. Chen, Z. Hua, Y. Tamres-Rudnicky, A. Dash, and A. Rudnicky, “AppDialogue: Multi-app dialogues for intelligent assistants,” in Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC’16). Portorož, Slovenia: European Language Resources Association (ELRA), May 2016, pp. 3127–3132. [Online]. Available: https://aclanthology.org/L16-1499
連結至畢業學校之論文網頁點我開啟連結
註: 此連結為研究生畢業學校所提供,不一定有電子全文可供下載,若連結有誤,請點選上方之〝勘誤回報〞功能,我們會盡快修正,謝謝!
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top