Author name cluster

Ivan Vulić

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

7 papers

1 author row

NeurIPS Conference 2025 Conference Paper

Universal Cross-Tokenizer Distillation via Approximate Likelihood Matching

Benjamin Minixhofer
Ivan Vulić
Edoardo Maria Ponti

Distillation has shown remarkable success in transferring knowledge from a Large Language Model (LLM) teacher to a student LLM. However, current distillation methods require similar tokenizers between the teacher and the student, restricting their applicability to only a small subset of teacher--student pairs. In this work, we develop a principled cross-tokenizer distillation method to solve this crucial deficiency. Our method is the first to enable effective distillation across fundamentally different tokenizers, while also substantially outperforming prior methods in all other cases. We verify the efficacy of our method on three distinct use cases. First, we show that viewing tokenizer transfer as self-distillation enables unprecedentedly effective transfer across tokenizers, including rapid transfer of subword models to the byte-level. Transferring different models to the same tokenizer also enables ensembling to boost performance. Secondly, we distil a large maths-specialised LLM into a small general-purpose model with a different tokenizer, achieving competitive maths problem-solving performance. Thirdly, we use our method to train state-of-the-art embedding prediction hypernetworks for training-free tokenizer transfer. Our results unlock an expanded range of teacher--student pairs for distillation, enabling new ways to adapt and enhance interaction between LLMs.

PDF Details

NeurIPS Conference 2024 Conference Paper

Zero-Shot Tokenizer Transfer

Benjamin Minixhofer
Edoardo M. Ponti
Ivan Vulić

Language models (LMs) are bound to their tokenizer, which maps raw text to a sequence of vocabulary items (tokens). This restricts their flexibility: for example, LMs trained primarily on English may still perform well in other natural and programming languages, but have vastly decreased efficiency due to their English-centric tokenizer. To mitigate this, we should be able to swap the original LM tokenizer with an arbitrary one, on the fly, without degrading performance. Hence, in this work we define a new problem: Zero-Shot Tokenizer Transfer (ZeTT). The challenge at the core of ZeTT is finding embeddings for the tokens in the vocabulary of the new tokenizer. Since prior heuristics for initializing embeddings often perform at chance level in a ZeTT setting, we propose a new solution: we train a hypernetwork taking a tokenizer as input and predicting the corresponding embeddings. We empirically demonstrate that the hypernetwork generalizes to new tokenizers both with encoder (e. g. , XLM-R) and decoder LLMs (e. g. , Mistral-7B). Our method comes close to the original models' performance in cross-lingual and coding tasks while markedly reducing the length of the tokenized sequence. We also find that the remaining gap can be quickly closed by continued training on less than 1B tokens. Finally, we show that a ZeTT hypernetwork trained for a base (L)LM can also be applied to fine-tuned variants without extra training. Overall, our results make substantial strides toward detaching LMs from their tokenizer.

PDF Details DOI

TMLR Journal 2023 Journal Article

Modular Deep Learning

Jonas Pfeiffer
Sebastian Ruder
Ivan Vulić
Edoardo Ponti

Transfer learning has recently become the dominant paradigm of machine learning. Pre-trained models fine-tuned for downstream tasks achieve better performance with fewer labelled examples. Nonetheless, it remains unclear how to develop models that specialise towards multiple tasks without incurring negative interference and that generalise systematically to non-identically distributed tasks. Modular deep learning has emerged as a promising solution to these challenges. In this framework, units of computation are often implemented as autonomous parameter-efficient modules. Information is conditionally routed to a subset of modules and subsequently aggregated. These properties enable positive transfer and systematic generalisation by separating computation from routing and updating modules locally. We offer a survey of modular architectures, providing a unified view over several threads of research that evolved independently in the scientific literature. Moreover, we explore various additional purposes of modularity, including scaling language models, causal inference and discovery, programme simulation, and hierarchical reinforcement learning. Finally, we report various concrete applications where modularity has been successfully deployed such as cross-lingual and cross-modal knowledge transfer.

PDF Details

AAAI Conference 2021 Conference Paper

Analogy Training Multilingual Encoders

Nicolas Garneau
Mareike Hartmann
Anders Sandholm
Sebastian Ruder
Ivan Vulić
Anders Søgaard

Language encoders encode words and phrases in ways that capture their local semantic relatedness, but are known to be globally inconsistent. Global inconsistency can seemingly be corrected for, in part, by leveraging signals from knowledge bases, but previous results are partial and limited to monolingual English encoders. We extract a large-scale multilingual, multi-word analogy dataset from Wikidata for diagnosing and correcting for global inconsistencies and implement a four-way Siamese BERT architecture for grounding multilingual BERT (mBERT) in Wikidata through analogy training. We show that analogy training not only improves the global consistency of mBERT, as well as the isomorphism of language-specific subspaces, but also leads to significant gains on downstream tasks such as bilingual dictionary induction and sentence retrieval.

PDF Details

AAAI Conference 2020 Conference Paper

A General Framework for Implicit and Explicit Debiasing of Distributional Word Vector Spaces

Anne Lauscher
Goran Glavaš
Simone Paolo Ponzetto
Ivan Vulić

Distributional word vectors have recently been shown to encode many of the human biases, most notably gender and racial biases, and models for attenuating such biases have consequently been proposed. However, existing models and studies (1) operate on under-speciﬁed and mutually differing bias deﬁnitions, (2) are tailored for a particular bias (e. g. , gender bias) and (3) have been evaluated inconsistently and non-rigorously. In this work, we introduce a general framework for debiasing word embeddings. We operationalize the deﬁnition of a bias by discerning two types of bias speciﬁcation: explicit and implicit. We then propose three debiasing models that operate on explicit or implicit bias speciﬁcations and that can be composed towards more robust debiasing. Finally, we devise a full-ﬂedged evaluation framework in which we couple existing bias metrics with newly proposed ones. Experimental ﬁndings across three embedding methods suggest that the proposed debiasing models are robust and widely applicable: they often completely remove the bias both implicitly and explicitly without degradation of semantic information encoded in any of the input distributional spaces. Moreover, we successfully transfer debiasing models, by means of crosslingual embedding spaces, and remove or attenuate biases in distributional word vector spaces of languages that lack readily available bias speciﬁcations.

PDF Details

JAIR Journal 2019 Journal Article

A Survey of Cross-lingual Word Embedding Models

Sebastian Ruder
Ivan Vulić
Anders Søgaard

Cross-lingual representations of words enable us to reason about word meaning in multilingual contexts and are a key facilitator of cross-lingual transfer when developing natural language processing models for low-resource languages. In this survey, we provide a comprehensive typology of cross-lingual word embedding models. We compare their data requirements and objective functions. The recurring theme of the survey is that many of the models presented in the literature optimize for the same objectives, and that seemingly different models are often equivalent, modulo optimization strategies, hyper-parameters, and such. We also discuss the different ways cross-lingual word embeddings are evaluated, as well as future challenges and research horizons.

PDF Details DOI

JAIR Journal 2016 Journal Article

Bilingual Distributed Word Representations from Document-Aligned Comparable Data

Ivan Vulić
Marie-Francine Moens

We propose a new model for learning bilingual word representations from non-parallel document-aligned data. Following the recent advances in word representation learning, our model learns dense real-valued word vectors, that is, bilingual word embeddings (BWEs). Unlike prior work on inducing BWEs which heavily relied on parallel sentence-aligned corpora and/or readily available translation resources such as dictionaries, the article reveals that BWEs may be learned solely on the basis of document-aligned comparable data without any additional lexical resources nor syntactic information. We present a comparison of our approach with previous state-of-the-art models for learning bilingual word representations from comparable data that rely on the framework of multilingual probabilistic topic modeling (MuPTM), as well as with distributional local context-counting models. We demonstrate the utility of the induced BWEs in two semantic tasks: (1) bilingual lexicon extraction, (2) suggesting word translations in context for polysemous words. Our simple yet effective BWE-based models significantly outperform the MuPTM-based and context-counting representation models from comparable data as well as prior BWE-based models, and acquire the best reported results on both tasks for all three tested language pairs.

PDF Details DOI