3 citations · 6 across the 17 of their papers we have counts for
18 papers
Transformers Struggle to Use Their Emergent World Models: Revisiting the Tower of Hanoi, and the Illusion of Thinking
Devin Pereira, Willem Zuidema
The Tower of Hanoi is a simple planning puzzle that in prior work has proven challenging for large reasoning models (LRMs). Current models solve the standard formulation of the puz…
In-Context Learning in Speech Language Models: Analyzing the Role of Acoustic Features, Linguistic Structure, and Induction Heads
Charlotte Pouw, Hosein Mohebbi, Afra Alishahi +1
In-Context Learning (ICL) has been extensively studied in text-only Language Models, but remains largely unexplored in the speech domain. Here, we investigate how linguistic and ac…
Tracking the emergence of linguistic structure in self-supervised models learning from speech
Marianne de Heer Kloots, Martijn Bentum, Hosein Mohebbi +3
Self-supervised speech models learn effective representations of spoken language, which have been shown to reflect various aspects of linguistic structure. But when does such struc…
Transformer See, Transformer Do: Copying as an Intermediate Step in Learning Analogical Reasoning
Philipp Hellwig, Willem Zuidema, Claire E. Stevenson +1
Analogical reasoning is a hallmark of human intelligence, enabling us to solve new problems by transferring knowledge from one situation to another. Yet, developing artificial inte…
The Curious Case of Visual Grounding: Different Effects for Speech- and Text-based Language Encoders
Adrian Sauter, Willem Zuidema, Marianne de Heer Kloots
How does visual information included in training affect language processing in audio- and text-based deep learning models? We explore how such visual grounding affects model-intern…
What do self-supervised speech models know about Dutch? Analyzing advantages of language-specific pre-training
Marianne de Heer Kloots, Hosein Mohebbi, Charlotte Pouw +3
How language-specific are speech representations learned by self-supervised models? Existing work has shown that a range of linguistic features can be successfully decoded from end…
Propositional Logic for Probing Generalization in Neural Networks
Anna Langedijk, Jaap Jumelet, Willem Zuidema
The extent to which neural networks are able to acquire and represent symbolic rules remains a key topic of research and debate. Much current work focuses on the impressive capabil…
A Linguistically Motivated Analysis of Intonational Phrasing in Text-to-Speech Systems: Revealing Gaps in Syntactic Sensitivity
Charlotte Pouw, Afra Alishahi, Willem Zuidema
We analyze the syntactic sensitivity of Text-to-Speech (TTS) systems using methods inspired by psycholinguistic research. Specifically, we focus on the generation of intonational p…
PolyPythias: Stability and Outliers across Fifty Language Model Pre-Training Runs
Oskar van der Wal, Pietro Lesci, Max Muller-Eberstein +4
The stability of language model pre-training and its effects on downstream performance are still understudied. Prior work shows that the training process can yield significantly di…
Post-hoc Disentanglement of Textual and Acoustic Features in Pre-trained Speech Encoders
Hosein Mohebbi, Grzegorz Chrupała, Willem Zuidema +2
Pre-trained speech models build entangled internal representations, which capture a variety of features (e.g., pitch, loudness, syntax, or semantics of an utterance) in a distribut…
Human-like Linguistic Biases in Neural Speech Models: Phonetic Categorization and Phonotactic Constraints in Wav2Vec2.0
Marianne de Heer Kloots, Willem Zuidema
What do deep neural speech models know about phonology? Existing work has examined the encoding of individual linguistic units such as phonemes in these models. Here we investigate…
Black Big Boxes: Tracing Adjective Order Preferences in Large Language Models
Jaap Jumelet, Lisa Bylinina, Willem Zuidema +1
In English and other languages, multiple adjectives in noun phrases follow intricate ordering patterns. These patterns have been widely studied in linguistics and provide a useful…
Perception of Phonological Assimilation by Neural Speech Recognition Models
Charlotte Pouw, Marianne de Heer Kloots, Afra Alishahi +1
Human listeners effortlessly compensate for phonological changes during speech perception, often unconsciously inferring the intended sounds. For example, listeners infer the under…
Do Language Models Exhibit Human-like Structural Priming Effects?
Jaap Jumelet, Willem Zuidema, Arabella Sinclair
We explore which linguistic factors -- at the sentence and token level -- play an important role in influencing language model predictions, and investigate whether these are reflec…
Transparency at the Source: Evaluating and Interpreting Language Models With Access to the True Distribution
Jaap Jumelet, Willem Zuidema
We present a setup for training, evaluating and interpreting neural language models, that uses artificial, language-like data. The data is generated using a massive probabilistic g…
Identifying and Adapting Transformer-Components Responsible for Gender Bias in an English Language Model
Abhijith Chintam, Rahel Beloch, Willem Zuidema +2
Language models (LMs) exhibit and amplify many types of undesirable biases learned from the training data, including gender bias. However, we lack tools for effectively and efficie…
Homophone Disambiguation Reveals Patterns of Context Mixing in Speech Transformers
Hosein Mohebbi, Grzegorz Chrupała, Willem Zuidema +1
Transformers have become a key architecture in speech processing, but our understanding of how they build up representations of acoustic and linguistic structure is limited. In thi…
DecoderLens: Layerwise Interpretation of Encoder-Decoder Transformers
Anna Langedijk, Hosein Mohebbi, Gabriele Sarti +2
In recent years, many interpretability methods have been proposed to help interpret the internal states of Transformer-models, at different levels of precision and complexity. Here…