arXiv subjects

Nathaniel Imel

Publications and source records attributed to Nathaniel Imel.

4 recordsLinked to original sources

arXiv2026

Agent-based imitation dynamics can yield efficiently compressed population-level vocabularies

Natural languages have been argued to evolve under pressure to efficiently compress meanings into words by optimizing the Information Bottleneck (IB) complexity-accuracy tradeoff. However, the underlying social dynamics that could drive the optimization of a language's vocabulary towards efficiency remain largely unknown. In parallel, evolutionary game theory has been invoked to explain the emergence of language from rudimentary agent-level dynamics, but it has not yet been tested whether such an approach can lead to efficient compression in the IB sense. Here, we provide a unified model integrating evolutionary game theory with the IB framework and show how near-optimal compression can arise in a population through an independently motivated dynamic of imprecise strategy imitation in signaling games. We find that key parameters of the model -- namely, those that regulate precision in these games, as well as players' tendency to confuse similar states -- lead to constrained variation of the tradeoffs achieved by emergent vocabularies. Our results suggest that evolutionary game dynamics could potentially provide a mechanistic basis for the evolution of vocabularies with information-theoretically optimal and empirically attested properties.

cs.CL↗

arXiv2026

On convexity and efficiency in semantic systems

There are two widely held characterizations of human semantic category systems: (1) they form convex partitions of conceptual spaces, and (2) they are efficient for communication. While prior work observed that convexity and efficiency co-occur in color naming, the analytical relation between them and why they co-occur have not been well understood. We address this gap by combining analytical and empirical analyses that build on the Information Bottleneck (IB) framework for semantic efficiency. First, we show that convexity and efficiency are distinct in the sense that neither entails the other: there are convex systems which are inefficient, and optimally-efficient systems that are non-convex. Crucially, however, the IB-optimal systems are mostly convex in the domain of color naming, explaining the main empirical basis for the convexity approach. Second, we show that efficiency is a stronger predictor for discriminating attested color naming systems from hypothetical variants, with convexity adding negligible improvement on top of that. Finally, we discuss a range of empirical phenomena that convexity cannot account for but efficiency can. Taken together, our work suggests that while convexity and efficiency can yield similar structural observations, they are fundamentally distinct, with efficiency providing a more comprehensive account of semantic typology.

cs.CL↗

arXiv2025

Evolution and compression in LLMs: On the emergence of human-aligned categorization

Converging evidence suggests that human systems of semantic categories achieve near-optimal compression via the Information Bottleneck (IB) complexity-accuracy tradeoff. Large language models (LLMs) are not trained for this objective, which raises the question: are LLMs capable of evolving efficient human-aligned semantic systems? To address this question, we focus on color categorization -- a key testbed of cognitive theories of categorization with uniquely rich human data -- and replicate with LLMs two influential human studies. First, we conduct an English color-naming study, showing that LLMs vary widely in their complexity and English-alignment, with larger instruction-tuned models achieving better alignment and IB-efficiency. Second, to test whether these LLMs simply mimic patterns in their training data or actually exhibit a human-like inductive bias toward IB-efficiency, we simulate cultural evolution of pseudo color-naming systems in LLMs via a method we refer to as Iterated in-Context Language Learning (IICLL). We find that akin to humans, LLMs iteratively restructure initially random systems towards greater IB-efficiency. However, only a model with strongest in-context capabilities (Gemini 2.0) is able to recapitulate the wide range of near-optimal IB-tradeoffs observed in humans, while other state-of-the-art models converge to low-complexity solutions. These findings demonstrate how human-aligned semantic categories can emerge in LLMs via the same fundamental principle that underlies semantic efficiency in humans.

cs.CL↗

arXiv2025

Density, asymmetry and citation dynamics in scientific literature

Scientific behavior is often characterized by a tension between building upon established knowledge and introducing novel ideas. Here, we investigate whether this tension is reflected in the relationship between the similarity of a scientific paper to previous research and its eventual citation rate. To operationalize similarity to previous research, we introduce two complementary metrics to characterize the local geometry of a publication's semantic neighborhood: (1) \emph{density} ($\rho$), defined as the ratio between a fixed number of previously-published papers and the minimum distance enclosing those papers in a semantic embedding space, and (2) asymmetry ($\alpha$), defined as the average directional difference between a paper and its nearest neighbors. We tested the predictive relationship between these two metrics and its subsequent citation rate using a Bayesian hierarchical regression approach, surveying $\sim 53,000$ publications across nine academic disciplines and five different document embeddings. While the individual effects of $\rho$ on citation count are small and variable, incorporating density-based predictors consistently improves out-of-sample prediction when added to baseline models. These results suggest that the density of a paper's surrounding scientific literature may carry modest but informative signals about its eventual impact. Meanwhile, we find no evidence that publication asymmetry improves model predictions of citation rates. Our work provides a scalable framework for linking document embeddings to scientometric outcomes and highlights new questions regarding the role that semantic similarity plays in shaping the dynamics of scientific reward.

cs.DL↗