Papers by Nikola Mrkšić
Training Neural Response Selection for Task-Oriented Dialogue Systems (P19-1)
Copied to clipboard
Matthew Henderson, Ivan Vulić, Daniela Gerz, Iñigo Casanueva, Paweł Budzianowski, Sam Coope, Georgios Spithourakis, Tsung-Hsien Wen, Nikola Mrkšić, Pei-Hao Su
| Challenge: | Despite their popularity, retrieval-based models have had modest impact on task-oriented dialogue systems . main obstacle to their application is the low-data regime of most task-orientated dialogue tasks . e-commerce, banking, and other domains are applications of retrieval models . |
| Approach: | They propose a method which pretrains a retrieval-based model on large general-domain conversational corpora and fine-tunes it for the target dialogue domain. |
| Outcome: | The proposed method is evaluated on five diverse domains, ranging from e-commerce to banking. |
Deep Learning for Conversational AI (N18-6)
Copied to clipboard
| Challenge: | Spoken Dialogue Systems (SDS) have great commercial potential . the advent of deep learning has led to significant advances in this area of NLP research . |
| Approach: | This tutorial will introduce researchers to the pipeline framework for modelling goal-oriented dialogue systems. |
| Outcome: | This tutorial will familiarise researchers with the latest advances in spoken dialogue systems . the aim of the course is to encourage dialogue research in the NLP community . |
Post-Specialisation: Retrofitting Vectors of Words Unseen in Lexical Resources (N18-1)
Copied to clipboard
| Challenge: | Word vector specialisation is a portable, light-weight approach to fine-tuning distributional word vector spaces by injecting external knowledge from rich lexical resources such as WordNet. |
| Approach: | They propose a constraint-driven vector space specialisation method that embeds external knowledge into lexical resources into a deep neural network to specialise unseen words. |
| Outcome: | The proposed method preserves useful linguistic knowledge for seen words while propagating external signal to unseen words to improve their vector representations. |
Multilingual and Cross-Lingual Intent Detection from Spoken Data (2021.emnlp-main)
Copied to clipboard
Daniela Gerz, Pei-Hao Su, Razvan Kusztos, Avishek Mondal, Michał Lis, Eshan Singhal, Nikola Mrkšić, Tsung-Hsien Wen, Ivan Vulić
| Challenge: | a systematic study on multilingual and cross-lingual intent detection from spoken data is presented . current work on intent detection is limited to English, and standard benchmarks exist only in English. |
| Approach: | They present a systematic study on multilingual and cross-lingual intent detection from spoken data. |
| Outcome: | The proposed resource is called MInDS-14, and it provides strong intent detection in most target languages. |
Fully Statistical Neural Belief Tracking (P18-2)
Copied to clipboard
| Challenge: | Existing framework for a dialogue state tracking model requires an expensive manual retuning step . |
| Approach: | They propose to improve existing NBT model by removing a manual retuning step . they propose two different statistical update mechanisms to improve model performance . |
| Outcome: | The proposed model achieves competitive performance and provides a robust framework for building resource-light DST models. |
Adversarial Propagation and Zero-Shot Cross-Lingual Transfer of Word Vector Specialization (D18-1)
Copied to clipboard
| Challenge: | Semantic specialization is a process of fine-tuning pre-trained distributional word vectors using external lexical knowledge to accentuate a particular semantic relation in the specialized vector space. |
| Approach: | They propose a method for specializing distributional word vectors using external lexical knowledge. |
| Outcome: | The proposed method improves on word similarity, dialog state tracking, and lexical simplification across three languages and on three tasks. |
ConveRT: Efficient and Accurate Conversational Representations from Transformers (2020.findings-emnlp)
Copied to clipboard
| Challenge: | ConveRT is a pretraining framework for conversational AI that is computationally heavy, slow, and expensive to train. |
| Approach: | They propose a pretraining framework for conversational tasks that is efficient, lightweight, and inexpensive. |
| Outcome: | The proposed model achieves state-of-the-art performance across widely established responses . it trains substantially faster than existing state- of-the art models . |
PolyResponse: A Rank-based Approach to Task-Oriented Dialogue with Application in Restaurant Search and Booking (D19-3)
Copied to clipboard
Matthew Henderson, Ivan Vulić, Iñigo Casanueva, Paweł Budzianowski, Daniela Gerz, Sam Coope, Georgios Spithourakis, Tsung-Hsien Wen, Nikola Mrkšić, Pei-Hao Su
| Challenge: | a task-oriented dialogue system is based on task-specific ontologies that constrain slots to specific values . we present a conversational search engine that can be used to search for restaurant reservations . |
| Approach: | They propose a conversational search engine that supports task-oriented dialogue . the polyresponse engine is trained on hundreds of millions of examples extracted from real conversations . |
| Outcome: | The proposed system is available in 8 different languages. |
Specialising Word Vectors for Lexical Entailment (N18-1)
Copied to clipboard
| Challenge: | Existing word representation learning methods rely on the distributional hypothesis to learn meaningful word representations. |
| Approach: | They propose a method that emphasises the asymmetric relation of lexical entailment by injecting external linguistic constraints into the input word vector space. |
| Outcome: | The proposed method achieves state-of-the-art in the tasks of hypernymy directionality, hypernomia detection, and graded lexical entailment. |
ConvFiT: Conversational Fine-Tuning of Pretrained Language Models (2021.emnlp-main)
Copied to clipboard
Ivan Vulić, Pei-Hao Su, Samuel Coope, Daniela Gerz, Paweł Budzianowski, Iñigo Casanueva, Nikola Mrkšić, Tsung-Hsien Wen
| Challenge: | Existing Transformer-based language models (LMs) are not effective as sentence encoders when used off-the-shelf. |
| Approach: | They propose a method which turns a pretrained LM into a universal conversational encoder and task-specialised sentence encoder. |
| Outcome: | The proposed framework achieves state-of-the-art ID performance across the board with particular gains in the most challenging, few-shot setups. |