The Thieves on Sesame Street are Polyglots - Extracting Multilingual Models from Monolingual APIs (2020.emnlp-main)
Copied to clipboard
| Challenge: | Recent work has demonstrated that deployed NLP models can be stolen by adversaries by querying victim models with gibberish input data that consists of random sequences of words. |
| Approach: | They propose to extract a local copy of a monolingual victim model from an API and query it with gibberish input data paired with the victim's labels. |
| Outcome: | The extracted model learns the task from the monolingual victim, but it generalizes far better than the victim to several other languages. |
Similar Papers
Exploiting Cloze-Questions for Few-Shot Text Classification and Natural Language Inference (2021.eacl-main)
Copied to clipboard
| Challenge: | Existing approaches to learning from examples are limited due to the vast number of languages, domains and tasks. |
| Approach: | They propose a semi-supervised training procedure that reformulates input examples as cloze-style phrases to help language models understand a given task. |
| Outcome: | The proposed approach outperforms supervised training and strong semi-supervised approaches in low-resource settings by a large margin. |
On the Multilingual Capabilities of Very Large-Scale English Language Models (2022.lrec-1)
Copied to clipboard
| Challenge: | Generative Pre-trained Transformers (GPTs) have been scaled to unprecedented sizes in the history of machine learning. |
| Approach: | They investigate the potential and limits of Generative Pre-trained Transformers in three tasks . they find it can be almost as useful for many languages as it is for English . |
| Outcome: | The proposed model can perform tasks in five different languages, and its potential is explored . it can learn from a few examples "via text interaction" and is scalable to many languages . |
Language Contamination Helps Explains the Cross-lingual Capabilities of English Pretrained Models (2022.emnlp-main)
Copied to clipboard
| Challenge: | a large number of pretraining corpora are not publicly available, and it is unclear how much foreign language data exists in monolingual models. |
| Approach: | They propose to use English pretraining corpora to analyze their language composition . they find that even when less than 1% of data is not English, it facilitates cross-lingual transfer . |
| Outcome: | The proposed model is not truly monolingual when pretrained at scale, the authors show . they show that even when less than 1% of data is not English, it facilitates cross-lingual transfer . |
On the Multilingual Ability of Decoder-based Pre-trained Language Models: Finding and Controlling Language-Specific Neurons (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing decoder-based pre-trained language models demonstrate excellent multilingual capabilities, but it is unclear how they handle multilingualism. |
| Approach: | They propose to examine the neuron-level internal behavior of decoder-based PLMs by finding neurons that fire “uniquely for each language” within decoded PLM models. |
| Outcome: | The proposed models fire “uniquely for each language” and show that language-specific neurons are unique, with a slight overlap (5%) between languages. |
Multilingual Chart-based Constituency Parse Extraction from Pre-trained Language Models (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for extracting complete (binary) parses from pre-trained language models are expensive and time-consuming. |
| Approach: | They propose a chart-based method and an effective top-K ensemble technique to extractbinary parses from PLMs. |
| Outcome: | The proposed method can induce non-trivial parses for sentences from nine languages in an integrated and language-agnostic manner, and is robust to cross-lingual transfer. |
Analyzing the Mono- and Cross-Lingual Pretraining Dynamics of Multilingual Language Models (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies on multilingual models have focused on their cross-lingual transfer behavior . a recent study examined multilingual model learning from the multilingual pretraining signal . |
| Approach: | They analyze checkpoints during multilingual pretraining to identify when models acquire in-language and cross-lingual abilities. |
| Outcome: | The proposed model achieves high in-language performance early on, with lower-level linguistic skills acquired before more complex ones. |
Polyglot Contextual Representations Improve Crosslingual Transfer (N19-1)
Copied to clipboard
| Challenge: | Existing methods for crosslingual transfer use multilingual word embeddings, but contextual word representations are not yet available. |
| Approach: | They propose a method to produce multilingual contextual word representations by training a single language model on text from multiple languages. |
| Outcome: | The proposed method compares model models to monolingual and non-contextual variants and shows that polyglot learning can be beneficial for multilingual representations. |
Lost in Activations: A Neuron-level Analysis of Encoders for Cross-Lingual Emotion Detection (2026.eacl-short)
Copied to clipboard
| Challenge: | XLM-R models for multilingual emotion classification are still lacking in understanding of their internal decision-making mechanisms. |
| Approach: | They propose to use neuron-level activation analysis to study the inter-language differences between neurons. |
| Outcome: | The proposed model consistently encodes emotion-related concepts across languages, but others show strong monolingual specialization. |
Can You Tell Me How to Get Past Sesame Street? Sentence-Level Pretraining Beyond Language Modeling (P19-1)
Copied to clipboard
Alex Wang, Jan Hula, Patrick Xia, Raghavendra Pappagari, R. Thomas McCoy, Roma Patel, Najoung Kim, Ian Tenney, Yinghui Huang, Katherin Yu, Shuning Jin, Berlin Chen, Benjamin Van Durme, Edouard Grave, Ellie Pavlick, Samuel R. Bowman
| Challenge: | State-of-the-art models in natural language processing (NLP) often incorporate sentence encoder functions which generate a sequence of vectors intended to represent the in-context meaning of each word in an input text. |
| Approach: | They conduct the first large-scale systematic study of candidate pretraining tasks, comparing 19 different tasks as alternatives and complements to language modeling. |
| Outcome: | The proposed model can be used to train sentences on language modeling tasks. |
Lexicon-Enhanced Self-Supervised Training for Multilingual Dense Retrieval (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Recent multilingual pre-trained models perform poorly on multilingual retrieval tasks due to lack of multilingual training data. |
| Approach: | They propose to mine and generate self-supervised training data based on large-scale unlabeled corpus and introduce query generator to generate more queries in target languages for unlabed passages. |
| Outcome: | The proposed method performs better than baselines on a Mr. TYDI dataset and an industrial dataset from a commercial search engine. |