Search Results for author:
Found 28 papers, 14 papers with code
Transparency at the Source: Evaluating and Interpreting Language Models With Access to the True Distribution
With access to the underlying true source, our results show striking differences and outcomes in learning dynamics between different classes of words.
Identifying and Adapting Transformer-Components Responsible for Gender Bias in an English Language Model
Language models (LMs) exhibit and amplify many types of undesirable biases learned from the training data, including gender bias.
Homophone Disambiguation Reveals Patterns of Context Mixing in Speech Transformers
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.
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.
Feature Interactions Reveal Linguistic Structure in Language Models
We study feature interactions in the context of feature attribution methods for post-hoc interpretability.
Quantifying Context Mixing in Transformers
Self-attention weights and their transformed variants have been the main source of information for analyzing token-to-token interactions in Transformer-based models.
Undesirable Biases in NLP: Addressing Challenges of Measurement
In particular, we will explore two central notions from psychometrics, the construct validity and the reliability of measurement tools, and discuss how they can be applied in the context of measuring model bias.
The Birth of Bias: A case study on the evolution of gender bias in an English language model
With full access to the data and to the model parameters as they change during every step while training, we can map in detail how the representation of gender develops, what patterns in the dataset drive this, and how the model's internal state relates to the bias in a downstream task (semantic textual similarity).
Structural Persistence in Language Models: Priming as a Window into Abstract Language Representations
We investigate the extent to which modern, neural language models are susceptible to structural priming, the phenomenon whereby the structure of a sentence makes the same structure more probable in a follow-up sentence.
DoLFIn: Distributions over Latent Features for Interpretability
Interpreting the inner workings of neural models is a key step in ensuring the robustness and trustworthiness of the models, but work on neural network interpretability typically faces a trade-off: either the models are too constrained to be very useful, or the solutions found by the models are too complex to interpret.
Transferring Inductive Biases through Knowledge Distillation
Having the right inductive biases can be crucial in many tasks or scenarios where data or computing resources are a limiting factor, or where training data is not perfectly representative of the conditions at test time.
Quantifying Attention Flow in Transformers
This makes attention weights unreliable as explanations probes.
Analysing Neural Language Models: Contextual Decomposition Reveals Default Reasoning in Number and Gender Assignment
Extensive research has recently shown that recurrent neural language models are able to process a wide range of grammatical phenomena.
Blackbox Meets Blackbox: Representational Similarity \& Stability Analysis of Neural Language Models and Brains
In this paper, we define and apply representational stability analysis (ReStA), an intuitive way of analyzing neural language models.
Blackbox meets blackbox: Representational Similarity and Stability Analysis of Neural Language Models and Brains
In this paper, we define and apply representational stability analysis (ReStA), an intuitive way of analyzing neural language models.
Siamese recurrent networks learn first-order logic reasoning and exhibit zero-shot compositional generalization
We approach this classic question with current methods, and demonstrate that recurrent neural networks can learn to recognize first order logical entailment relations between expressions.
Formal models of Structure Building in Music, Language and Animal Songs
Human language, music and a variety of animal vocalisations constitute ways of sonic communication that exhibit remarkable structural complexity.
Under the Hood: Using Diagnostic Classifiers to Investigate and Improve how Language Models Track Agreement Information
We show that `diagnostic classifiers', trained to predict number from the internal states of a language model, provide a detailed understanding of how, when, and where this information is represented.
Visualisation and 'diagnostic classifiers' reveal how recurrent and recursive neural networks process hierarchical structure
To develop an understanding of what the recurrent network encodes, visualisation techniques alone do not suffice.
Experiential, Distributional and Dependency-based Word Embeddings have Complementary Roles in Decoding Brain Activity
We evaluate 8 different word embedding models on their usefulness for predicting the neural activation patterns associated with concrete nouns.
Quantifying the vanishing gradient and long distance dependency problem in recursive neural networks and recursive LSTMs
Recursive neural networks (RNN) and their recently proposed extension recursive long short term memory networks (RLSTM) are models that compute representations for sentences, by recursively combining word embeddings according to an externally provided parse tree.
Unsupervised Dependency Parsing: Let's Use Supervised Parsers
We present a self-training approach to unsupervised dependency parsing that reuses existing supervised and unsupervised parsing algorithms.
Ranked #1 on
Unsupervised Dependency Parsing
on Penn Treebank
Compositional Distributional Semantics with Long Short Term Memory
We are proposing an extension of the recursive neural network that makes use of a variant of the long short-term memory architecture.
