| The Ubuntu Dialogue Corpus: A Large Dataset for Research in Unstructured Multi-Turn Dialogue Systems |
SIGDIAL 2015 |
1. Introduces the Ubuntu Dialogue Corpus, a dataset containing almost 1 million multi-turn dialogues, with a total of over 7 million utterances and 100 million words. 2. Describe two neural learning architectures suitable for analyzing this dataset, and provide benchmark performance on the task of selecting the best next response. |
| On the Evaluation of Dialogue Systems with Next Utterance Classification |
Arxiv 2016 |
1. Investigate the performance of humans on this task to validate the relevance of Next Utterance Classification (NUC) as a method of evaluation. 2. Results show three main findings: (1) humans are able to correctly classify responses at a rate much better than chance, thus confirming that the task is feasible, (2) human performance levels vary across task domains (we consider 3 datasets) and expertise levels (novice vs experts), thus showing that a range of performance is possible on this type of task, (3) automated dialogue systems built using state-of-the-art machine learning methods have similar performance to the human novices, but worse than the experts, thus confirming the utility of this class of tasks for driving further research in automated dialogue systems. |
| Sequential Matching Network - A New Architecture for Multi-turn Response Selection in Retrieval-Based Chatbots |
ACL 2017 |
1. Propose a sequential matching network (SMN) to address both problems. SMN first matches a response with each utterance in the context on multiple levels of granularity, and distills important matching information from each pair as a vector with convolution and pooling operations. The vectors are then accumulated in a chronological order through a recurrent neural network (RNN) which models relationships among utterances. The final matching score is calculated with the hidden states of the RNN. |
| Modeling Multi-turn Conversation with Deep Utterance Aggregation |
COLING 2018 |
1. Formulate previous utterances into context using a proposed deep utterance aggregation model to form a fine-grained context representation, where a self-matching attention is first introduced to route the vital information in each utterance. Then the model matches a response with each refined utterance and the final matching score is obtained after attentive turns aggregation. |
| Multi-Turn Response Selection for Chatbots with Deep Attention Matching Network |
ACL 2018 |
1. Investigate matching a response with its multi-turn context using dependency information based entirely on attention. 2. First, construct representations of text segments at different granularities solely with stacked self-attention. Second, extract the truly matched segment pairs with attention across the context and response. We jointly introduce those two kinds of attention in one uniform neural network. |
| Improving Response Selection in Multi-Turn Dialogue Systems by Incorporating Domain Knowledge |
CoNLL 2018 |
Proposes a novel neural network architecture for response selection in an end-to-end multi-turn conversational dialogue setting. The architecture applies context level attention and incorporates additional external knowledge provided by descriptions of domain-specific words. |
| Learning a Matching Model with Co-teaching for Multi-turn Response Selection in Retrieval-based Dialogue Systems |
ACL 2019 |
1. Propose a general co-teaching framework with three specific teaching strategies that cover both teaching with loss functions and teaching with data curriculum. |
| Dually Interactive Matching Network for Personalized Response Selection in Retrieval-Based Chatbots |
EMNLP 2019 |
1. Proposes a dually interactive matching network (DIM) for presenting the personalities of dialogue agents in retrieval-based chatbots. 2. The DIM model adopts a dual matching architecture, which performs interactive matching between responses and contexts and between responses and personas respectively for ranking response candidates. |
| Multi-hop Selector Network for Multi-turn Response Selection in Retrieval-based Chatbots |
EMNLP 2019 |
1. Analyze the side effect of using too many context utterances and propose a multi-hop selector network (MSN) to alleviate the problem. 2. MSN firstly utilizes a multi-hop selector to select the relevant utterances as context. Then, the model matches the filtered context with the candidate response and obtains a matching score. |
| Sampling Matters! An Empirical Study of Negative Sampling Strategies |
EMNLP 2019 |
1. Study how to sample negative examples to automatically construct a training set for effective model learning in retrieval-based dialogue systems. 2. Consider four strategies including minimum sampling, maximum sampling, semi-hard sampling, and decay-hard sampling. 3. Empirical studies on two benchmarks with three matching models indicate that compared with the widely used random sampling strategy, although the first two strategies lead to performance drop, the latter two ones can bring consistent improvement to the performance of all the models on both benchmarks. |
| Deep Retrieval-Based Dialogue Systems: A Short Review |
Arxiv 2019 |
1. Present the latest and most relevant retrieval-based dialogue systems and the available datasets used to build and evaluate them. 2. Discuss their limitations and provide insights and guidelines for future work. |
| Poly-encoders - Transformer Architectures and Pre-training Strategies for Fast and Accurate Multi-sentence Scoring |
ICLR 2020 |
1. Develop a new transformer architecture, the Poly-encoder, that learns global rather than token level self-attention features. 2. Perform a detailed comparison of all three approaches, including what pre-training and fine-tuning strategies work best. |