arXiv subjects

Willem Zuidema

Publications and source records attributed to Willem Zuidema.

At least 19 recordsLinked to original sources

arXiv2026

Transformer See, Transformer Do: Copying as an Intermediate Step in Learning Analogical Reasoning

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 intelligence systems capable of robust human-like analogical reasoning has proven difficult. In this work, we train transformers using Meta-Learning for Compositionality (MLC) on an analogical reasoning task (letter-string analogies) and assess their generalization capabilities. We find that letter-string analogies become learnable when guiding the models to attend to the most informative problem elements, induced by including copy tasks in the training data. Furthermore, generalization to new alphabets improves when models are trained with more heterogeneous datasets. For the best training run, our 3-layer encoder-decoder model performs on par with frontier models on our letter-string analogy datasets. The MLC approach also enables some generalization to compositions of trained transformations, but not to completely novel transformations. To understand how the model solves the analogies, we identify an algorithm for one analogy task that approximates the model's computations. We verify this using interpretability analyses and show that head-level attention weights causally affect model output, allowing steering between different analogy tasks. We discuss the implications of our findings for generalization capabilities of larger models and the parallels to human analogical reasoning.

cs.LG↗

arXiv2026

Transformers Struggle to Use Their Emergent World Models: Revisiting the Tower of Hanoi, and the Illusion of Thinking

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 puzzle, but still struggle with the flat-to-flat variant (where initial and goal states are not restricted to have all rings on a single peg). This paper presents an in-depth study of how both small, in-house Transformers and large, third-party LRMs solve this task. To understand the failures mechanistically, we first train small Transformers from scratch on precomputed solution traces. Using a variety of interpretability techniques, we show that these Transformers develop an emergent world model: a linearly decodable, geometrically faithful representation of the puzzle's state space (the Sierpinski triangle), that is causally involved in solving the puzzles. Second, we return to the large LLMs and apply our techniques to two frontier reasoning models, Qwen3.6-27B and DeepSeek-R1-Distill-Qwen-32B, that attempt to solve the task through extended chain-of-thought. Surprisingly, we find that both models encode the Sierpinski world model near-perfectly at the end of the prompt, and yet fail at the majority of tasks when there are more than 3 rings. We locate the source of this failure in the decaying representation of the world model. We probe for the representation at different stages during planning, and establish causality by showing that performance can be improved by injecting the prompt-time representation at inference. The failure of the models is thus one of maintenance of the required representations, not their absence, and performance is at least partially recoverable. These results thus reframe the reported collapse in performance from prior work: current Large Reasoning Models build a world model, and then lose it.

cs.AI↗

arXiv2026

Melodic contour does not cluster: Reconsidering contour typology

How to describe the shape of a melodic phrase? Scholars have often relied on typologies with a small set of contour types. We question their adequacy: we find no evidence that phrase contours cluster into discrete types, neither in German or Chinese folksongs, nor in Gregorian chant. The test for clustering we propose applies the dist-dip test of multimodality after a UMAP dimensionality reduction. The test correctly identifies clustering in a synthetic dataset, but not in actual phrase contours. These results raise problems for discrete typologies. In particular, type frequencies may be unreliable, as we see with Huron's typology. We also show how a recent finding of four contour shapes may be an artefact of the analysis. Our findings suggest that melodic contour is best seen as a continuous phenomenon.

cs.SD↗

arXiv2026

In-Context Learning in Speech Language Models: Analyzing the Role of Acoustic Features, Linguistic Structure, and Induction Heads

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 acoustic features affect ICL in Speech Language Models. We focus on the Text-to-Speech (TTS) task, which allows us to analyze ICL from two angles: (1) how accurately the model infers the task from the demonstrations (i.e., generating the correct spoken content), and (2) to what extent the model mimics the acoustic characteristics of the demonstration speech in its output. We find that speaking rate strongly affects ICL performance and is also mimicked in the output, whereas pitch range and intensity have little impact on performance and are not consistently reproduced. Finally, we investigate the role of induction heads in speech-based ICL and show that these heads play a causal role: ablating the top-k induction heads completely removes the model's ICL ability, mirroring findings from text-based ICL.

cs.CL↗

arXiv2026

Tracking the emergence of linguistic structure in self-supervised models learning from speech

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 structure emerge in model training? We study the encoding of a wide range of linguistic structures, across layers and intermediate checkpoints of six Wav2Vec2 and HuBERT models trained on spoken Dutch. We find that different levels of linguistic structure show notably distinct layerwise patterns as well as learning trajectories, which can partially be explained by differences in their degree of abstraction from the acoustic signal and the timescale at which information from the input is integrated. Moreover, we find that the level at which pre-training objectives are defined strongly affects both the layerwise organization and the learning trajectories of linguistic structures, with greater parallelism induced by higher-order prediction tasks (i.e. iteratively refined pseudo-labels).

cs.CL↗

arXiv2026

Black Big Boxes: Tracing Adjective Order Preferences in Large Language Models

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 test case for assessing how language models (LMs) acquire graded and context-sensitive word order preferences. We ask to what extent adjective order preferences in LMs can be explained by distributional learning alone, and where models exhibit behaviour that goes beyond surface co-occurrence patterns. We find that LM predictions are largely explained by training data frequencies: simple n-gram statistics account for much of their behaviour and closely mirror the preferences learned during training. However, by analysing learning dynamics we reveal that models also generalize robustly to unseen adjective combinations, indicating that their behaviour cannot be reduced to memorization of observed orders alone. Moreover, we show how LMs leverage word order cues from sentence context, demonstrating with feature attribution methods that contextual cues are an additional driver of adjective order in LM output.

cs.CL↗

arXiv2025

Linguists should learn to love speech-based deep learning models

Futrell and Mahowald present a useful framework bridging technology-oriented deep learning systems and explanation-oriented linguistic theories. Unfortunately, the target article's focus on generative text-based LLMs fundamentally limits fruitful interactions with linguistics, as many interesting questions on human language fall outside what is captured by written text. We argue that audio-based deep learning models can and should play a crucial role.

cs.CL↗

arXiv2025

Interpretability for Time Series Transformers using A Concept Bottleneck Framework

Mechanistic interpretability focuses on reverse engineering the internal mechanisms learned by neural networks. We extend our focus and propose to mechanistically forward engineer using our framework based on Concept Bottleneck Models. In the context of long-term time series forecasting, we modify the training objective to encourage a model to develop representations which are similar to predefined, interpretable concepts using Centered Kernel Alignment. This steers the bottleneck components to learn the predefined concepts, while allowing other components to learn other, undefined concepts. We apply the framework to the Vanilla Transformer, Autoformer and FEDformer, and present an in-depth analysis on synthetic data and on a variety of benchmark datasets. We find that the model performance remains mostly unaffected, while the model shows much improved interpretability. Additionally, we verify the interpretation of the bottleneck components with an intervention experiment using activation patching.

cs.LG↗

arXiv2025

A Linguistically Motivated Analysis of Intonational Phrasing in Text-to-Speech Systems: Revealing Gaps in Syntactic Sensitivity

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 phrase boundaries, which can often be predicted by identifying syntactic boundaries within a sentence. We find that TTS systems struggle to accurately generate intonational phrase boundaries in sentences where syntactic boundaries are ambiguous (e.g., garden path sentences or sentences with attachment ambiguity). In these cases, systems need superficial cues such as commas to place boundaries at the correct positions. In contrast, for sentences with simpler syntactic structures, we find that systems do incorporate syntactic cues beyond surface markers. Finally, we finetune models on sentences without commas at the syntactic boundary positions, encouraging them to focus on more subtle linguistic cues. Our findings indicate that this leads to more distinct intonation patterns that better reflect the underlying structure.

cs.CL↗

arXiv2025

The Curious Case of Visual Grounding: Different Effects for Speech- and Text-based Language Encoders

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-internal representations of words, and find substantially different effects in speech- vs. text-based language encoders. Firstly, global representational comparisons reveal that visual grounding increases alignment between representations of spoken and written language, but this effect seems mainly driven by enhanced encoding of word identity rather than meaning. We then apply targeted clustering analyses to probe for phonetic vs. semantic discriminability in model representations. Speech-based representations remain phonetically dominated with visual grounding, but in contrast to text-based representations, visual grounding does not improve semantic discriminability. Our findings could usefully inform the development of more efficient methods to enrich speech-based models with visually-informed semantics.

cs.CL↗

arXiv2025

What do self-supervised speech models know about Dutch? Analyzing advantages of language-specific pre-training

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-to-end models trained only on speech recordings. However, it's less clear to what extent pre-training on specific languages improves language-specific linguistic information. Here we test the encoding of Dutch phonetic and lexical information in internal representations of self-supervised Wav2Vec2 models. Pre-training exclusively on Dutch improves the representation of Dutch linguistic features as compared to pre-training on similar amounts of English or larger amounts of multilingual data. This language-specific advantage is well-detected by trained clustering or classification probes, and partially observable using zero-shot metrics. Furthermore, the language-specific benefit on linguistic feature encoding aligns with downstream performance on Automatic Speech Recognition.

cs.CL↗

arXiv2025

Propositional Logic for Probing Generalization in Neural Networks

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 capabilities of large language models, as well as their often ill-understood failures on a wide range of reasoning tasks. In this paper, in contrast, we investigate the generalization behavior of three key neural architectures (Transformers, Graph Convolution Networks and LSTMs) in a controlled task rooted in propositional logic. The task requires models to generate satisfying assignments for logical formulas, making it a structured and interpretable setting for studying compositionality. We introduce a balanced extension of an existing dataset to eliminate superficial patterns and enable testing on unseen operator combinations. Using this dataset, we evaluate the ability of the three architectures to generalize beyond the training distribution. While all models perform well in-distribution, we find that generalization to unseen patterns, particularly those involving negation, remains a significant challenge. Transformers fail to apply negation compositionally, unless structural biases are introduced. Our findings highlight persistent limitations in the ability of standard architectures to learn systematic representations of logical operators, suggesting the need for stronger inductive biases to support robust rule-based reasoning.

cs.LG↗

arXiv2025

PolyPythias: Stability and Outliers across Fifty Language Model Pre-Training Runs

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 different results in response to slight variations in initial conditions, e.g., the random seed. Crucially, the research community still lacks sufficient resources and tools to systematically investigate pre-training stability, particularly for decoder-only language models. We introduce the PolyPythias, a set of 45 new training runs for the Pythia model suite: 9 new seeds across 5 model sizes, from 14M to 410M parameters, resulting in about 7k new checkpoints that we release. Using these new 45 training runs, in addition to the 5 already available, we study the effects of different initial conditions determined by the seed -- i.e., parameters' initialisation and data order -- on (i) downstream performance, (ii) learned linguistic representations, and (iii) emergence of training phases. In addition to common scaling behaviours, our analyses generally reveal highly consistent training dynamics across both model sizes and initial conditions. Further, the new seeds for each model allow us to identify outlier training runs and delineate their characteristics. Our findings show the potential of using these methods to predict training stability.

cs.CL↗

arXiv2024

Disentangling Textual and Acoustic Features of Neural Speech Representations

Neural speech models build deeply entangled internal representations, which capture a variety of features (e.g., fundamental frequency, loudness, syntactic category, or semantic content of a word) in a distributed encoding. This complexity makes it difficult to track the extent to which such representations rely on textual and acoustic information, or to suppress the encoding of acoustic features that may pose privacy risks (e.g., gender or speaker identity) in critical, real-world applications. In this paper, we build upon the Information Bottleneck principle to propose a disentanglement framework that separates complex speech representations into two distinct components: one encoding content (i.e., what can be transcribed as text) and the other encoding acoustic features relevant to a given downstream task. We apply and evaluate our framework to emotion recognition and speaker identification downstream tasks, quantifying the contribution of textual and acoustic features at each model layer. Additionally, we explore the application of our disentanglement framework as an attribution method to identify the most salient speech frame representations from both the textual and acoustic perspectives.

cs.CL↗

arXiv2024

Do Language Models Exhibit Human-like Structural Priming Effects?

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 reflective of results found in humans and human corpora (Gries and Kootstra, 2017). We make use of the structural priming paradigm, where recent exposure to a structure facilitates processing of the same structure. We don't only investigate whether, but also where priming effects occur, and what factors predict them. We show that these effects can be explained via the inverse frequency effect, known in human priming, where rarer elements within a prime increase priming effects, as well as lexical dependence between prime and target. Our results provide an important piece in the puzzle of understanding how properties within their context affect structural prediction in language models.

cs.CL↗

arXiv2024

Human-like Linguistic Biases in Neural Speech Models: Phonetic Categorization and Phonotactic Constraints in Wav2Vec2.0

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 interactions between units. Inspired by classic experiments on human speech perception, we study how Wav2Vec2 resolves phonotactic constraints. We synthesize sounds on an acoustic continuum between /l/ and /r/ and embed them in controlled contexts where only /l/, only /r/, or neither occur in English. Like humans, Wav2Vec2 models show a bias towards the phonotactically admissable category in processing such ambiguous sounds. Using simple measures to analyze model internals on the level of individual stimuli, we find that this bias emerges in early layers of the model's Transformer module. This effect is amplified by ASR finetuning but also present in fully self-supervised models. Our approach demonstrates how controlled stimulus designs can help localize specific linguistic knowledge in neural speech models.

cs.CL↗

arXiv2024

Perception of Phonological Assimilation by Neural Speech Recognition Models

Human listeners effortlessly compensate for phonological changes during speech perception, often unconsciously inferring the intended sounds. For example, listeners infer the underlying /n/ when hearing an utterance such as "clea[m] pan", where [m] arises from place assimilation to the following labial [p]. This article explores how the neural speech recognition model Wav2Vec2 perceives assimilated sounds, and identifies the linguistic knowledge that is implemented by the model to compensate for assimilation during Automatic Speech Recognition (ASR). Using psycholinguistic stimuli, we systematically analyze how various linguistic context cues influence compensation patterns in the model's output. Complementing these behavioral experiments, our probing experiments indicate that the model shifts its interpretation of assimilated sounds from their acoustic form to their underlying form in its final layers. Finally, our causal intervention experiments suggest that the model relies on minimal phonological context cues to accomplish this shift. These findings represent a step towards better understanding the similarities and differences in phonological processing between neural ASR models and humans.

cs.CL↗

arXiv2024

DecoderLens: Layerwise Interpretation of Encoder-Decoder Transformers

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, to analyze encoder-decoder Transformers, we propose a simple, new method: DecoderLens. Inspired by the LogitLens (for decoder-only Transformers), this method involves allowing the decoder to cross-attend representations of intermediate encoder layers instead of using the final encoder output, as is normally done in encoder-decoder models. The method thus maps previously uninterpretable vector representations to human-interpretable sequences of words or symbols. We report results from the DecoderLens applied to models trained on question answering, logical reasoning, speech recognition and machine translation. The DecoderLens reveals several specific subtasks that are solved at low or intermediate layers, shedding new light on the information flow inside the encoder component of this important class of models.

cs.CL↗