arXiv subjects

Eamon Duede

Publications and source records attributed to Eamon Duede.

At least 19 recordsLinked to original sources

arXiv2026

The unintended consequences of large language models as a labor-augmenting technology in science

As a labor-augmenting technology, large language models (LLMs) have the potential to accelerate scientific activity across the research pipeline. But even if LLMs perform on par with human experts at selected tasks, their use will bring unintended consequences as they alter the balance of frictions and inducements that steer the allocation of research effort across projects. Here we develop a simple mathematical model to illustrate. In fields where LLMs are useful primarily as tools for discovering promising projects, researchers will become more selective about what they publish; where they facilitate the process of publishing existing data, researchers will become less selective. By allowing scientists to work more quickly, LLMs raise the opportunity cost of researcher time, creating incentives to refine papers less thoroughly before moving on. Enticing as it is to imagine that, by saving us time on mundane tasks, LLMs will provide us with more time to think deeply and develop projects completely, our results temper such hopes.

physics.soc-ph↗

arXiv2026

A correspondence problem for mathematical proof

Mathematical proofs are often said to justify their conclusions by indicating the existence of a corresponding formal derivation. We argue that this widespread view relies on an under-examined notion of correspondence, or what it means for a particular derivation to ''correspond'' to a particular proof. Mere existence of a formalization is not enough, and a substantive account of the required correspondence resolves into two criteria -- adequate representation (of the original theorem) and tracking (of the steps in the original proof). An examination of the actually-existing formalization systems we have today shows the variety of quasi-empirical ways we establish these criteria, and points towards new burdens that may be placed on the future evolution of mathematics itself.

math.HO↗

arXiv2025

Why Slop Matters

AI-generated "slop" is often seen as digital pollution. We argue that this dismissal of the topic risks missing important aspects of AI Slop that deserve rigorous study. AI Slop serves a social function: it offers a supply-side solution to a variety of problems in cultural and economic demand - that, collectively, people want more content than humans can supply. We also argue that AI Slop is not mere digital detritus but has its own aesthetic value. Like other "low" cultural forms initially dismissed by critics, it nonetheless offers a legitimate means of collective sense-making, with the potential to express meaning and identity. We identify three key features of family resemblance for prototypical AI Slop: superficial competence (its veneer of quality is belied by a deeper lack of substance), asymmetry effort (it takes vastly less effort to generate than would be the case without AI), and mass producibility (it is part of a digital ecosystem of widespread generation and consumption). While AI Slop is heterogeneous and depends crucially on its medium, it tends to vary across three dimensions: instrumental utility, personalization, and surrealism. AI Slop will be an increasingly prolific and impactful part of our creative, information, and cultural economies; we should take it seriously as an object of study in its own right.

cs.CY↗

arXiv2025

Critical Confabulation: Can LLMs Hallucinate for Social Good?

LLMs hallucinate, yet some confabulations can have social affordances if carefully bounded. We propose critical confabulation (inspired by critical fabulation from literary and social theory), the use of LLM hallucinations to "fill-in-the-gap" for omissions in archives due to social and political inequality, and reconstruct divergent yet evidence-bound narratives for history's ``hidden figures''. We simulate these gaps with an open-ended narrative cloze task: asking LLMs to generate a masked event in a character-centric timeline sourced from a novel corpus of unpublished texts. We evaluate audited (for data contamination), fully-open models (the OLMo-2 family) and unaudited open-weight and proprietary baselines under a range of prompts designed to elicit controlled and useful hallucinations. Our findings validate LLMs' foundational narrative understanding capabilities to perform critical confabulation, and show how controlled and well-specified hallucinations can support LLM applications for knowledge production without collapsing speculation into a lack of historical accuracy and fidelity.

cs.CL↗

arXiv2025

AI and the Decentering of Disciplinary Creativity

This paper examines the role of artificial intelligence in scientific problem-solving, with a focus on its implications for disciplinary creativity. Drawing on recent work in the philosophy of creativity, I distinguish between creative approaches and creative products, and introduce the concept of disciplinary creativity -the creative application of discipline-specific expertise to a valued problem within that field. Through two cases in mathematics, I show that while computation can extend disciplinary creativity, certain approaches involving AI can serve to displace it. This displacement has the potential to alter (and, perhaps, diminish) the value of scientific pursuit.

cs.AI↗

arXiv2024

The Paradox of Collective Certainty in Science

We explore a paradox of collective action and certainty in science wherein the more scientists research together, the less that work contributes to the value of their collective certainty. When scientists address similar problems and share data, methods, and collaborators, their understanding of and trust in their colleagues' research rises, a quality required for scientific advance. This increases the positive reinforcement scientists receive for shared beliefs as they become more dependent on their colleagues' knowledge, interests, and findings. This collective action increases the potential for scientists to reside in epistemic ''bubbles'' that limit their capacity to make new discoveries or have their discoveries generalize. In short, as scientists grow closer, their experience of scientific validity rises as the likelihood of genuine replication falls, creating a trade-off between certainty and truth.

physics.soc-ph↗

arXiv2024

Confabulation: The Surprising Value of Large Language Model Hallucinations

This paper presents a systematic defense of large language model (LLM) hallucinations or 'confabulations' as a potential resource instead of a categorically negative pitfall. The standard view is that confabulations are inherently problematic and AI research should eliminate this flaw. In this paper, we argue and empirically demonstrate that measurable semantic characteristics of LLM confabulations mirror a human propensity to utilize increased narrativity as a cognitive resource for sense-making and communication. In other words, it has potential value. Specifically, we analyze popular hallucination benchmarks and reveal that hallucinated outputs display increased levels of narrativity and semantic coherence relative to veridical outputs. This finding reveals a tension in our usually dismissive understandings of confabulation. It suggests, counter-intuitively, that the tendency for LLMs to confabulate may be intimately associated with a positive capacity for coherent narrative-text generation.

cs.CL↗

arXiv2024

Oil & Water? Diffusion of AI Within and Across Scientific Fields

This study empirically investigates claims of the increasing ubiquity of artificial intelligence (AI) within roughly 80 million research publications across 20 diverse scientific fields, by examining the change in scholarly engagement with AI from 1985 through 2022. We observe exponential growth, with AI-engaged publications increasing approximately thirteenfold (13x) across all fields, suggesting a dramatic shift from niche to mainstream. Moreover, we provide the first empirical examination of the distribution of AI-engaged publications across publication venues within individual fields, with results that reveal a broadening of AI engagement within disciplines. While this broadening engagement suggests a move toward greater disciplinary integration in every field, increased ubiquity is associated with a semantic tension between AI-engaged research and more traditional disciplinary research. Through an analysis of tens of millions of document embeddings, we observe a complex interplay between AI-engaged and non-AI-engaged research within and across fields, suggesting that increasing ubiquity is something of an oil-and-water phenomenon -- AI-engaged work is spreading out over fields, but not mixing well with non-AI-engaged work.

cs.DL↗

arXiv2024

Apriori Knowledge in an Era of Computational Opacity: The Role of AI in Mathematical Discovery

Can we acquire apriori knowledge of mathematical facts from the outputs of computer programs? People like Burge have argued (correctly in our opinion) that, for example, Appel and Haken acquired apriori knowledge of the Four Color Theorem from their computer program insofar as their program simply automated human forms of mathematical reasoning. However, unlike such programs, we argue that the opacity of modern LLMs and DNNs creates obstacles in obtaining apriori mathematical knowledge from them in similar ways. We claim though that if a proof-checker automating human forms of proof-checking is attached to such machines, then we can obtain apriori mathematical knowledge from them after all, even though the original machines are entirely opaque to us and the proofs they output may not, themselves, be human-surveyable.

cs.AI↗

arXiv2023

The Representational Status of Deep Learning Models

This paper aims to clarify the representational status of Deep Learning Models (DLMs). While commonly referred to as 'representations', what this entails is ambiguous due to a conflation of functional and relational conceptions of representation. This paper argues that while DLMs represent their targets in a relational sense, in general, we have no good reason to believe that DLMs encode locally semantically decomposable representations of their targets. That is, the representational capacity these models have is largely global, rather than decomposable into stable, local subrepresentations. This result has immediate implications for explainable AI (XAI) and directs attention toward exploring the global relational nature of deep learning representations and their relationship both to models more generally to understand their potential role in future scientific inquiry.

cs.AI↗

arXiv2022

Deep Learning Opacity in Scientific Discovery

Philosophers have recently focused on critical, epistemological challenges that arise from the opacity of deep neural networks. One might conclude from this literature that doing good science with opaque models is exceptionally challenging, if not impossible. Yet, this is hard to square with the recent boom in optimism for AI in science alongside a flood of recent scientific breakthroughs driven by AI methods. In this paper, I argue that the disconnect between philosophical pessimism and scientific optimism is driven by a failure to examine how AI is actually used in science. I show that, in order to understand the epistemic justification for AI-powered breakthroughs, philosophers must examine the role played by deep learning as part of a wider process of discovery. The philosophical distinction between the 'context of discovery' and the 'context of justification' is helpful in this regard. I demonstrate the importance of attending to this distinction with two cases drawn from the scientific literature, and show that epistemic opacity need not diminish AI's capacity to lead scientists to significant and justifiable breakthroughs.

cs.AI↗

arXiv2022

The Diminishing Returns of Masked Language Models to Science

Transformer-based masked language models such as BERT, trained on general corpora, have shown impressive performance on downstream tasks. It has also been demonstrated that the downstream task performance of such models can be improved by pretraining larger models for longer on more data. In this work, we empirically evaluate the extent to which these results extend to tasks in science. We use 14 domain-specific transformer-based models (including ScholarBERT, a new 770M-parameter science-focused masked language model pretrained on up to 225B tokens) to evaluate the impact of training data, model size, pretraining and finetuning time on 12 downstream scientific tasks. Interestingly, we find that increasing model sizes, training data, or compute time does not always lead to significant improvements (i.e., >1% F1), if at all, in scientific information extraction tasks and offered possible explanations for the surprising performance differences.

cs.CL↗

arXiv2021

The Social Abduction of Science

The logic of abduction involves a collision between deduction and induction, where empirical surprises violate expectations and scientists innovate to resolve them. Here we reformulate abduction as a social process, occurring not only within individual scientists, but often through conversation between those who understand a particular scientific system and its anomalies (its insiders) and those exposed to alien and disruptive patterns, theories, and findings that could resolve them (its outsiders). These extended conversations between scientists, scholars, and disciplines with divergent backgrounds define a social logic of discovery by catalyzing social syllogisms that yield speculative hypotheses with outsized impact across science. We show how this approach theorizes a number of disparate findings from scientometrics and the science and science, tethering them to powerful interpersonal processes that link diversity with discovery. By reframing abductive reasoning as a social process, we attempt to broker a productive new relationship between science studies' focus on scientific practice and the philosophy of science's concern with propositions in scientific discourse, rendering claims-making through abduction a critical scientific practice, central to scientific change, and increasingly available to science studies through small and large-scale data on scientific communication and deliberation.

physics.soc-ph↗

arXiv2021

Being Together in Place as a Catalyst for Scientific Advance

The COVID-19 pandemic necessitated social distancing at every level of society, including universities and research institutes, raising essential questions concerning the continuing importance of physical proximity for scientific and scholarly advance. Using customized author surveys about the intellectual influence of referenced work on scientists' own papers, combined with precise measures of geographical and semantic distance between focal and referenced works, we find that being at the same institution is strongly associated with intellectual influence on scientists' and scholars' published work. However, this influence increases with intellectual distance: the more different the referenced work done by colleagues at one's institution, the more influential it is on one's own. Universities worldwide constitute places where people doing very different work engage in sustained interactions through departments, committees, seminars, and communities. These interactions come to uniquely influence their published research, suggesting the need to replace rather than displace diverse engagements for sustainable advance.

cs.DL↗

arXiv2020

Status drives how we cite: Evidence from thousands of authors

Researchers cite works for a variety of reasons, including some having nothing to do with acknowledging influence. The distribution of different citation types in the literature, and which papers attract which types, is poorly understood. We investigate high-influence and low-influence citations and the mechanisms producing them using 17,154 ground-truth citation types provided via survey by 9,380 authors systematically sampled across academic fields. Overall, 54% of citations denote little-to-no influence and these citations are concentrated among low status (lightly cited) papers. In contrast, high-influence citations are concentrated among high status (highly cited) papers through a number of steps that resemble a pipeline. Authors discover highly cited papers earlier in their projects, more often through social contacts, and read them more closely. Papers' status, above and beyond any quality differences, directly helps determine their pipeline: experimentally revealing or hiding citation counts during the survey shows that low counts cause lowered perceptions of quality. Accounting for citation types thus reveals a "double status effect": in addition to affecting how often a work is cited, status affects how meaningfully it is cited. Consequently, highly cited papers are even more influential than their raw citation counts suggest.

cs.SI↗

arXiv2017

The Wisdom of Polarized Crowds

As political polarization in the United States continues to rise, the question of whether polarized individuals can fruitfully cooperate becomes pressing. Although diversity of individual perspectives typically leads to superior team performance on complex tasks, strong political perspectives have been associated with conflict, misinformation and a reluctance to engage with people and perspectives beyond one's echo chamber. It is unclear whether self-selected teams of politically diverse individuals will create higher or lower quality outcomes. In this paper, we explore the effect of team political composition on performance through analysis of millions of edits to Wikipedia's Political, Social Issues, and Science articles. We measure editors' political alignments by their contributions to conservative versus liberal articles. A survey of editors validates that those who primarily edit liberal articles identify more strongly with the Democratic party and those who edit conservative ones with the Republican party. Our analysis then reveals that polarized teams---those consisting of a balanced set of politically diverse editors---create articles of higher quality than politically homogeneous teams. The effect appears most strongly in Wikipedia's Political articles, but is also observed in Social Issues and even Science articles. Analysis of article "talk pages" reveals that politically polarized teams engage in longer, more constructive, competitive, and substantively focused but linguistically diverse debates than political moderates. More intense use of Wikipedia policies by politically diverse teams suggests institutional design principles to help unleash the power of politically polarized teams.

cs.SI↗

arXiv2017

Enabling Interactive Analytics of Secure Data using Cloud Kotta

Research, especially in the social sciences and humanities, is increasingly reliant on the application of data science methods to analyze large amounts of (often private) data. Secure data enclaves provide a solution for managing and analyzing private data. However, such enclaves do not readily support discovery science---a form of exploratory or interactive analysis by which researchers execute a range of (sometimes large) analyses in an iterative and collaborative manner. The batch computing model offered by many data enclaves is well suited to executing large compute tasks; however it is far from ideal for day-to-day discovery science. As researchers must submit jobs to queues and wait for results, the high latencies inherent in queue-based, batch computing systems hinder interactive analysis. In this paper we describe how we have augmented the Cloud Kotta secure data enclave to support collaborative and interactive analysis of sensitive data. Our model uses Jupyter notebooks as a flexible analysis environment and Python language constructs to support the execution of arbitrary functions on private data within this secure framework.

cs.DC↗

arXiv2016

Cloud Kotta: Enabling Secure and Scalable Data Analytics in the Cloud

Distributed communities of researchers rely increasingly on valuable, proprietary, or sensitive datasets. Given the growth of such data, especially in fields new to data-driven, computationally intensive research like the social sciences and humanities, coupled with what are often strict and complex data-use agreements, many research communities now require methods that allow secure, scalable and cost-effective storage and analysis. Here we present CLOUD KOTTA: a cloud-based data management and analytics framework. CLOUD KOTTA delivers an end-to-end solution for coordinating secure access to large datasets, and an execution model that provides both automated infrastructure scaling and support for executing analytics near to the data. CLOUD KOTTA implements a fine-grained security model ensuring that only authorized users may access, analyze, and download protected data. It also implements automated methods for acquiring and configuring low-cost storage and compute resources as they are needed. We present the architecture and implementation of CLOUD KOTTA and demonstrate the advantages it provides in terms of increased performance and flexibility. We show that CLOUD KOTTA's elastic provisioning model can reduce costs by up to 16x when compared with statically provisioned models.

cs.DC↗