arXiv subjects

Harvey Lederman

Publications and source records attributed to Harvey Lederman.

8 recordsLinked to original sources

arXiv2026

CREATE: Testing LLMs for Associative Creativity

A key component of creativity is associative reasoning: the ability to draw novel yet meaningful connections between concepts. We introduce CREATE, a benchmark designed to evaluate models' capacity for creative associative reasoning. CREATE requires models to generate sets of paths connecting concepts in a model's parametric knowledge. Paths should have high specificity (distinctiveness and closeness of the concept connection) and high diversity (dissimilarity from other paths), and models are scored more highly if they produce a larger set of strong, diverse paths. This task shares demands of real creativity tasks like hypothesis generation, including an extremely large search space, but enables collection of a sizable benchmark with objective answer grading. Evaluation of frontier models shows that the strongest models achieve higher creative utility than others, with the high multiplicity of answers and complexity of the search making benchmark saturation difficult to achieve. Furthermore, our results illustrate that thinking models are not always more effective on our task, even with high token budgets. Recent approaches for creative prompting give some but limited additional improvement. CREATE provides a sandbox for developing new methods to improve models' capacity for associative creativity.

cs.CL↗

arXiv2026

Emergent Introspection in AI is Content-Agnostic

Introspection is a foundational cognitive ability, but its mechanism is not well understood. Recent work has shown that AI models can introspect. We study the mechanism of this introspection. We first extensively replicate Lindsey (2025)'s thought injection detection paradigm in large open-source models. We show that introspection in these models is content-agnostic: models can detect that an anomaly occurred even when they cannot reliably identify its content. The models confabulate injected concepts that are high-frequency and concrete (e.g., "apple"). They also require fewer tokens to detect an injection than to guess the correct concept (with wrong guesses coming earlier). We argue that a content-agnostic introspective mechanism is consistent with leading theories in philosophy and psychology.

cs.AI↗

arXiv2025

Privileged Self-Access Matters for Introspection in AI

Whether AI models can introspect is an increasingly important practical question. But there is no consensus on how introspection is to be defined. Beginning from a recently proposed ''lightweight'' definition, we argue instead for a thicker one. According to our proposal, introspection in AI is any process which yields information about internal states through a process more reliable than one with equal or lower computational cost available to a third party. Using experiments where LLMs reason about their internal temperature parameters, we show they can appear to have lightweight introspection while failing to meaningfully introspect per our proposed definition.

cs.AI↗

arXiv2024

Maximal Social Welfare Relations on Infinite Populations Satisfying Permutation Invariance

We study social welfare relations (SWRs) on an infinite population. Our main result is a new characterization of a utilitarian SWR as the \emph{largest} SWR (in terms of subset when the weak relation is viewed as a set of pairs) which satisfies Strong Pareto, Permutation Invariance (elsewhere called ``Relative Anonymity'' and ``Isomorphism Invariance''), and a further ``Quasi-Independence'' axiom.

econ.GN↗

arXiv2024

Are Language Models More Like Libraries or Like Librarians? Bibliotechnism, the Novel Reference Problem, and the Attitudes of LLMs

Are LLMs cultural technologies like photocopiers or printing presses, which transmit information but cannot create new content? A challenge for this idea, which we call bibliotechnism, is that LLMs generate novel text. We begin with a defense of bibliotechnism, showing how even novel text may inherit its meaning from original human-generated text. We then argue that bibliotechnism faces an independent challenge from examples in which LLMs generate novel reference, using new names to refer to new entities. Such examples could be explained if LLMs were not cultural technologies but had beliefs, desires, and intentions. According to interpretationism in the philosophy of mind, a system has such attitudes if and only if its behavior is well explained by the hypothesis that it does. Interpretationists may hold that LLMs have attitudes, and thus have a simple solution to the novel reference problem. We emphasize, however, that interpretationism is compatible with very simple creatures having attitudes and differs sharply from views that presuppose these attitudes require consciousness, sentience, or intelligence (topics about which we make no claims).

cs.CL↗

arXiv2024

A Dominance Argument Against Incompleteness

This article presents a new argument against many forms of moral and prudential value incompleteness. The argument relies on two central principles: (i) a weak "negative dominance" principle, to the effect that Lottery 1 is better than Lottery 2 only if some possible outcome of Lottery 1 is better than some possible outcome of Lottery 2, and (ii) a weak form of ex ante Pareto, to the effect that, if Lottery 1 gives an unambiguously better (stochastically dominant) prospect to some individuals than Lottery 2, and equally good prospects to everyone else, then Lottery 1 is better than Lottery 2. Given modest auxiliary assumptions, these two principles rule out incompleteness in the prudential ranking of individual lives, and many forms of incompleteness in the moral rankings of outcomes and lotteries.

econ.TH↗

arXiv2023

Incompleteness, Independence, and Negative Dominance

This paper introduces the axiom of Negative Dominance, stating that if a lottery $f$ is strictly preferred to a lottery $g$, then some outcome in the support of $f$ is strictly preferred to some outcome in the support of $g$. It is shown that if preferences are incomplete on a sufficiently rich domain, then this plausible axiom, which holds for complete preferences, is incompatible with an array of otherwise plausible axioms for choice under uncertainty. In particular, in this setting, Negative Dominance conflicts with the standard Independence axiom. A novel theory, which includes Negative Dominance, and rejects Independence, is developed and shown to be consistent.

econ.TH↗

arXiv2016

Standard State Space Models of Unawareness (Extended Abstract)

The impossibility theorem of Dekel, Lipman and Rustichini has been thought to demonstrate that standard state-space models cannot be used to represent unawareness. We first show that Dekel, Lipman and Rustichini do not establish this claim. We then distinguish three notions of awareness, and argue that although one of them may not be adequately modeled using standard state spaces, there is no reason to think that standard state spaces cannot provide models of the other two notions. In fact, standard space models of these forms of awareness are attractively simple. They allow us to prove completeness and decidability results with ease, to carry over standard techniques from decision theory, and to add propositional quantifiers straightforwardly.

cs.AI↗