Papers by Zornitsa Kozareva
Efficient Large Scale Language Modeling with Mixtures of Experts (2022.emnlp-main)
Copied to clipboard
Mikel Artetxe, Shruti Bhosale, Naman Goyal, Todor Mihaylov, Myle Ott, Sam Shleifer, Xi Victoria Lin, Jingfei Du, Srinivasan Iyer, Ramakanth Pasunuru, Giridharan Anantharaman, Xian Li, Shuohui Chen, Halil Akin, Mandeep Baines, Louis Martin, Xing Zhou, Punit Singh Koura, Brian O’Horo, Jeffrey Wang, Luke Zettlemoyer, Mona Diab, Zornitsa Kozareva, Veselin Stoyanov
| Challenge: | Mixture of Experts layers (MoEs) enable efficient scaling of language models . large autoregressive language models such as GPT-3 can be adapted to a wide range of tasks . |
| Approach: | They propose to use Mixture of Experts layers to enable efficient scaling of language models . they find that MoEs are substantially more compute efficient than dense models compared to MoE models - but only when they are more modestly trained . |
| Outcome: | The proposed model outperforms dense models in a wide range of tasks and domains. |
Few-shot Learning with Multilingual Generative Language Models (2022.emnlp-main)
Copied to clipboard
Xi Victoria Lin, Todor Mihaylov, Mikel Artetxe, Tianlu Wang, Shuohui Chen, Daniel Simig, Myle Ott, Naman Goyal, Shruti Bhosale, Jingfei Du, Ramakanth Pasunuru, Sam Shleifer, Punit Singh Koura, Vishrav Chaudhary, Brian O’Horo, Jeff Wang, Luke Zettlemoyer, Zornitsa Kozareva, Mona Diab, Veselin Stoyanov, Xian Li
| Challenge: | Large-scale generative language models such as GPT-3 are competitive few-shot learners. |
| Approach: | They train multilingual generative language models on a corpus covering a diverse set of languages and study their few- and zero-shot learning capabilities. |
| Outcome: | The proposed model outperforms GPT-3 on 171 out of 182 directions with 32 training examples and surpasses the official supervised baseline in 45 directions. |
Improving In-Context Few-Shot Learning via Self-Supervised Training (2022.naacl-main)
Copied to clipboard
Mingda Chen, Jingfei Du, Ramakanth Pasunuru, Todor Mihaylov, Srini Iyer, Veselin Stoyanov, Zornitsa Kozareva
| Challenge: | Existing approaches to improve in-context few-shot learning are pretraining and downstream fewshot evaluation. |
| Approach: | They propose to use self-supervision as an intermediate training stage between pretraining and downstream fewshot usage to train models to perform in-context few shot learning. |
| Outcome: | The proposed model outperforms baseline models on two benchmarks. |
On-device Structured and Context Partitioned Projection Networks (P19-1)
Copied to clipboard
| Challenge: | A challenge in on-device text classification is to build highly accurate models that fit in small memory footprint and have low latency. |
| Approach: | They propose an on-device neural network which learns compact projection vectors from raw text using structured and context-dependent partition projections. |
| Outcome: | The proposed model outperforms baseline models and surpasses RNN, CNN and BiLSTM models on dialog act and intent prediction. |
Self-Governing Neural Networks for On-Device Short Text Classification (D18-1)
Copied to clipboard
| Challenge: | Existing deep neural networks have a tiny memory footprint and low computational capacity compared to high performance computing systems such as CPUs, GPUs and TPUs on the cloud. |
| Approach: | They propose on-device self-governing neural networks which learn compact projection vectors with local sensitive hashing. |
| Outcome: | The proposed models perform better on dialog act classification tasks while maintaining high accuracy. |
Transferable Neural Projection Representations (N19-1)
Copied to clipboard
| Challenge: | Neural word embeddings require lookup and a large memory footprint making it hard to deploy on-device. |
| Approach: | They propose a skip-gram based architecture coupled with Locality-Sensitive Hashing projections to learn efficient dynamically computable representations. |
| Outcome: | The proposed model performs better than previous models on multiple NLP tasks. |
Methods for Measuring, Updating, and Visualizing Factual Beliefs in Language Models (2023.eacl-main)
Copied to clipboard
Peter Hase, Mona Diab, Asli Celikyilmaz, Xian Li, Zornitsa Kozareva, Veselin Stoyanov, Mohit Bansal, Srinivasan Iyer
| Challenge: | Pretrained language models store a large amount of factual information that can be elicited by prompting or finetuning. |
| Approach: | They propose methods to measure model factual beliefs and update incorrect beliefs in models . they propose a new visualization tool that shows relationships between stored model beliefs . |
| Outcome: | The proposed methods improve models' consistency and accuracy, the authors show . their methods outperform existing methods in more difficult settings, the paper shows . |
ProFormer: Towards On-Device LSH Projection Based Transformers (2021.eacl-main)
Copied to clipboard
| Challenge: | ProFormer is a projection based transformer architecture that is faster and lighter making it suitable to deploy to memory constraint devices such as mobile phones, watches and IoT. |
| Approach: | They propose a projection based transformer architecture that generates word representations on-the-fly without embedding lookup tables and a local projection attention layer that transforms the input sequence of N LSH word projections into a sequence of K representations. |
| Outcome: | The proposed architecture reduces memory footprint from 92.16 MB to 1.7 KB and requires 16x less computation overhead making it suitable to deploy to memory constraint devices and preserve user privacy. |
On-Device Text Representations Robust To Misspellings via Projections (2021.eacl-main)
Copied to clipboard
| Challenge: | Recent advances in Locality-Sensitive Hashing (LSH)-based projection networks have demonstrated state-of-the-art performance in various classification tasks without explicit word embedding lookup tables by computing on-the fly text representations. |
| Approach: | They propose to use locality-sensitive hashing to compute on-the-fly text representations without explicit word embedding tables. |
| Outcome: | The proposed classifiers are more robust to common misspellings and perturbations of the input text compared to biLSTMs and fine-tuned BERT based methods. |
PRADO: Projection Attention Networks for Document Classification On-Device (D19-1)
Copied to clipboard
| Challenge: | Recent advances in deep learning have improved the performance of on-device neural networks for long text classification. |
| Approach: | They propose a projection attention neural network PRADO that combines trainable projections with attention and convolutions to train tiny neural networks that achieve high performance on multiple long document classification tasks. |
| Outcome: | The proposed model achieves high performance on multiple long document classification tasks while maintaining compact size. |
Self-Governing Neural Networks for On-Device Short Text Classification (D18-1)
Copied to clipboard
| Challenge: | Existing deep neural networks have a tiny memory footprint and low computational capacity compared to high performance computing systems such as CPUs, GPUs and TPUs on the cloud. |
| Approach: | They propose on-device self-governing neural networks which learn compact projection vectors with local sensitive hashing. |
| Outcome: | The proposed models perform better on dialog act classification tasks while maintaining high accuracy. |
ProSeqo: Projection Sequence Networks for On-Device Text Classification (D19-1)
Copied to clipboard
| Challenge: | ProSeqo is a novel on-device sequence model for text classification . it uses dynamic recurrent projections without the need to store or look up pre-trained embeddings. |
| Approach: | They propose a novel on-device sequence model for text classification using recurrent projections that uses dynamic recursion projections without the need to store or look up any pre-trained embeddings. |
| Outcome: | The proposed model outperforms state-of-the-art neural and on-device approaches for short and long text classification tasks while maintaining low memory footprint and high accuracy. |
ToKen: Task Decomposition and Knowledge Infusion for Few-Shot Hate Speech Detection (2022.emnlp-main)
Copied to clipboard
Badr AlKhamissi, Faisal Ladhak, Srinivasan Iyer, Veselin Stoyanov, Zornitsa Kozareva, Xian Li, Pascale Fung, Lambert Mathias, Asli Celikyilmaz, Mona Diab
| Challenge: | Hate speech detection is complex and requires commonsense reasoning and social nuance . prior work has shown that even humans cannot achieve a high agreement on whether a post constitutes HS . |
| Approach: | They frame a few-shot learning task to decompose a hate speech detection task into its "constituent" parts. they show that infusing commonsense knowledge from reasoning datasets improves the performance even further. |
| Outcome: | The proposed method outperforms baseline methods in the 16-shot case. |