MHMichael Townsen Hicks

Michael Townsen Hicks

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Michael Townsen Hicks has a total of 19 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • LLMs Bullshit by Design: A Reply to Licon

    by James Humphries, Michael Townsen Hicks, Joe Slater


  • Open Access

    Attention Might Offer Little Benefit for Graph Node Classification

    by Qin Jiang, Chengjia Wang, Michael Lones, Dongdong Chen, Wei Pang

    Attention mechanisms have achieved remarkable success in language models and have since been widely adopted in vision, speech, and multimodal learning. This trend has extended to graph learning, where attention-based models such as Graph Attention Networks (GAT) and Graph Transformers are now prevalent. \textbf{This position paper argues that attention mechanisms may not be as beneficial for graph node classification as commonly believed.} Through systematic ablation studies, we find that attention often provides negligible or even detrimental gains compared to simpler alternatives, with the only notable exception being graphs whose node features are language word embeddings. This suggests that the benefit of attention is largely limited outside language-related applications. We examine attention at three scales: 1-hop (GAT-style), Inception-style, and global mechanisms. We further analyze potential explanations for these results, including the limitations of gradient-based optimization and the fundamental differences between language and graph. Overall, these findings suggest that the prevailing enthusiasm for attention in graph node classification may be overstated, motivating a more critical and evidence-driven re-evaluation of its adoption. The code for all experiments is available at https://github.com/Qin87/ScaleNet/tree/July25.

  • Open Access

    The normative force of natural laws: Humean and non-Humean accounts of nomic normativity

    by Christian Loew, Siegfried Jaag, Michael T. Hicks

    The debate about laws of nature mainly focuses on the laws’ connection to regularities and modal facts. The much-discussed inference problem concerns why ‘it is a law that p’ entails ‘p’. Another problem (the ‘modality problem’) is about the need to explain the laws’ relation to counterfactuals, causation, and dispositions. In this paper, we argue that a third problem, the ‘normativity problem’, should play an equally important role in the debate: A theory of laws needs to explain the laws’ distinctive significance for what agents ought to do and believe. We argue that, just like the other two problems, the normativity problem poses distinct challenges for Humeans and non-Humeans and that it should be taken into account when comparing different views of laws.


  • What Chance Doesn't Know

    by Harjit Bhogal, Michael Hicks

    Humean accounts of chance have a problem with undermining futures: they have to accept that some series of events are physically possible and have a nonzero chance but are inconsistent with the chances being what they are. This contradicts basic platitudes about chances and is inconsistent with plausible constraints on credences. We show how Humeans can avoid these contradictions by drawing on metaphysically impossible worlds that are, nevertheless, scientifically possible. The guiding idea is that chance “doesn't know” at least some of the metaphysical facts. Unlike previous approaches to undermining (e.g., Lewis (1994), Thau (1994), and Hall (1994)), our account is part of a unified solution to a variety of challenges that the Humean faces. And it connects more closely with the way we employ stochastic theories in scientific practice. But perhaps most importantly, the idea that chance doesn’t know metaphysical facts flows from the Humean conception of chance.

  • Open Access

    Correction: ChatGPT is bullshit

    by Michael Townsen Hicks, James Humphries, Joe Slater


  • Top Cited Papers
    Open Access

    ChatGPT is bullshit

    by Michael Townsen Hicks, James Humphries, Joe Slater

    Recently, there has been considerable interest in large language models: machine learning systems which produce human-like text and dialogue. Applications of these systems have been plagued by persistent inaccuracies in their output; these are often called “AI hallucinations”. We argue that these falsehoods, and the overall activity of large language models, is better understood as bullshit in the sense explored by Frankfurt (On Bullshit, Princeton, 2005): the models are in an important way indifferent to the truth of their outputs. We distinguish two ways in which the models can be said to be bullshitters, and argue that they clearly meet at least one of these definitions. We further argue that describing AI misrepresentations as bullshit is both a more useful and more accurate way of predicting and discussing the behaviour of these systems.


  • Open Access

    Third Conference of the GWP (2019)

    by Michael T. Hicks, Andreas Hüttemann, Martin Voggenauer

  • Open Access

    How chance explains

    by Michael Townsen Hicks, Alastair Wilson

    What explains the outcomes of chance processes? We claim that their setups do. Chances, we think, mediate these explanations of outcome by setup but do not feature in them. Facts about chances do feature in explanations of a different kind: higher‐order explanations, which explain how and why setups explain their outcomes. In this paper, we elucidate this 'mediator view' of chancy explanation and defend it from a series of objections. We then show how it changes the playing field in four metaphysical disputes concerning chance. First, it makes it more plausible that even low chances can have explanatory power. Second, it undercuts a circularity objection against reductionist theories of chance. Third, it redirects the debate about a prominent argument against epistemic theories of chance. Finally, it sheds light on potential chancy explanations of the Universe's origin.


  • Open Access

    Breaking the explanatory circle

    by Michael Townsen Hicks

    Humeans are often accused of positing laws which fail to explain or are involved in explanatory circularity. Here, I will argue that these arguments are confused, but not because of anything to do with Humeanism: rather, they rest on false assumptions about causal explanation. I’ll show how these arguments can be neatly sidestepped if one takes on two plausible commitments which are motivated independently of Humeanism: first, that laws don’t directly feature in scientific explanation (a view defended recently by Ruben in R Inst Philos Suppl 27:95–117, 1990, 10.1017/S1358246100005063 and Skow in Reasons why, Oxford University Press, Oxford, 2016) and second, the view that explanation is contrastive. After outlining and motivating these views, I show how they bear on explanation-based arguments against Humeanism.


  • Making Fit Fit

    by Michael Townsen Hicks

    Reductionist accounts of objective chance rely on a notion of fit, which ties the chances at a world to the frequencies at that world. Here, I criticize extant measures of the fit of a chance system and draw on recent literature in epistemic utility theory to propose a new model: chances fit a world insofar as they are accurate at that world. I show how this model of fit does a better job of explaining the normative features of chance, its role in the laws of nature, and its status as an expert function than do previous accounts.