English
Related papers

Related papers: Learning Locality and Isotropy in Dialogue Modelin…

200 papers

Different studies of the embedding space of transformer models suggest that the distribution of contextual representations is highly anisotropic - the embeddings are distributed in a narrow cone. Meanwhile, static word representations…

Computation and Language · Computer Science 2021-09-29 Yue Ding , Karolis Martinkus , Damian Pascual , Simon Clematide , Roger Wattenhofer

Previous work has shown that the representations output by contextual language models are more anisotropic than static type embeddings, and typically display outlier dimensions. This seems to be true for both monolingual and multilingual…

Computation and Language · Computer Science 2023-06-08 Katharina Hämmerl , Alina Fastowski , Jindřich Libovický , Alexander Fraser

Recent dialogue systems rely on turn-based spoken interactions, requiring accurate Automatic Speech Recognition (ASR). Errors in ASR can significantly impact downstream dialogue tasks. To address this, using dialogue context from user and…

Computation and Language · Computer Science 2024-08-13 Wonjun Lee , San Kim , Gary Geunbae Lee

Although transformer-based models have been dominating the field of deep learning, various studies of their embedding space have shown that they suffer from "representation degeneration problem": embeddings tend to be distributed in a…

Machine Learning · Computer Science 2025-01-13 Sergei Kudriashov , Olesya Karpik , Eduard Klyshinsky

Tuning language models for dialogue generation has been a prevalent paradigm for building capable dialogue agents. Yet, traditional tuning narrowly views dialogue generation as resembling other language generation tasks, ignoring the role…

Computation and Language · Computer Science 2024-05-31 Jian Wang , Chak Tou Leong , Jiashuo Wang , Dongding Lin , Wenjie Li , Xiao-Yong Wei

The representation degeneration problem in Contextual Word Representations (CWRs) hurts the expressiveness of the embedding space by forming an anisotropic cone where even unrelated words have excessively positive correlations. Existing…

Computation and Language · Computer Science 2021-06-03 Sara Rajaee , Mohammad Taher Pilehvar

It is widely accepted that fine-tuning pre-trained language models usually brings about performance improvements in downstream tasks. However, there are limited studies on the reasons behind this effectiveness, particularly from the…

Computation and Language · Computer Science 2021-09-13 Sara Rajaee , Mohammad Taher Pilehvar

Typically, a linearly orthogonal transformation mapping is learned by aligning static type-level embeddings to build a shared semantic space. In view of the analysis that contextual embeddings contain richer semantic features, we…

Computation and Language · Computer Science 2021-07-21 Haoran Xu , Philipp Koehn

Self-supervised speech representations can hugely benefit downstream speech technologies, yet the properties that make them useful are still poorly understood. Two candidate properties related to the geometry of the representation space…

Computation and Language · Computer Science 2024-06-14 Mukhtar Mohamed , Oli Danyi Liu , Hao Tang , Sharon Goldwater

Current dialogue summarization systems usually encode the text with a number of general semantic features (e.g., keywords and topics) to gain more powerful dialogue modeling capabilities. However, these features are obtained via open-domain…

Computation and Language · Computer Science 2021-05-31 Xiachong Feng , Xiaocheng Feng , Libo Qin , Bing Qin , Ting Liu

Several studies have explored various advantages of multilingual pre-trained models (such as multilingual BERT) in capturing shared linguistic knowledge. However, less attention has been paid to their limitations. In this paper, we…

Computation and Language · Computer Science 2022-03-18 Sara Rajaee , Mohammad Taher Pilehvar

Contextual word representation models have shown massive improvements on a multitude of NLP tasks, yet their word sense disambiguation capabilities remain poorly explained. To address this gap, we assess whether contextual word…

Computation and Language · Computer Science 2021-04-23 Geetanjali Bihani , Julia Taylor Rayz

Conversational recommender systems (CRSs) capture user preference through textual information in dialogues. However, they suffer from data sparsity on two fronts: the dialogue space is vast and linguistically diverse, while the item space…

Information Retrieval · Computer Science 2025-07-02 Sixiao Zhang , Mingrui Liu , Cheng Long , Wei Yuan , Hongxu Chen , Xiangyu Zhao , Hongzhi Yin

A common approach for sequence tagging tasks based on contextual word representations is to train a machine learning classifier directly on these embedding vectors. This approach has two shortcomings. First, such methods consider single…

Multilingual machine translation has proven immensely useful for both parameter efficiency and overall performance across many language pairs via complete multilingual parameter sharing. However, some language pairs in multilingual models…

Computation and Language · Computer Science 2024-03-27 Neha Verma , Kenton Murray , Kevin Duh

General-purpose text decoding approaches are usually adopted for dialogue response generation. Although the quality of the generated responses can be improved with dialogue-specific encoding methods, conversational decoding methods are…

Computation and Language · Computer Science 2025-02-26 Yuxuan Yao , Han Wu , Qiling Xu , Linqi Song

The representation degeneration problem is a phenomenon that is widely observed among self-supervised learning methods based on Transformers. In NLP, it takes the form of anisotropy, a singular property of hidden representations which makes…

Computation and Language · Computer Science 2024-01-25 Nathan Godey , Éric de la Clergerie , Benoît Sagot

The representation degeneration problem is a phenomenon that is widely observed among self-supervised learning methods based on Transformers. In NLP, it takes the form of anisotropy, a singular property of hidden representations which makes…

Computation and Language · Computer Science 2023-06-14 Nathan Godey , Éric de la Clergerie , Benoît Sagot

The growing number of generative AI-based dialogue systems has made their evaluation a crucial challenge. This paper presents our contribution to this important problem through the Dialogue System Technology Challenge (DSTC-12, Track 1),…

Spoken Dialogue Models (SDMs) have recently attracted significant attention for their ability to generate voice responses directly to users' spoken queries. Despite their increasing popularity, there exists a gap in research focused on…

Computation and Language · Computer Science 2025-10-07 Chengqian Ma , Wei Tao , Yiwen Guo
‹ Prev 1 2 3 10 Next ›