Analysis and Explainability of LLMs Via Evolutionary Methods

• URL: http://arxiv.org/abs/2605.02930v1
• Date: Mon, 27 Apr 2026 18:07:58 GMT
• Title: Analysis and Explainability of LLMs Via Evolutionary Methods
• Abstract summary: We show how relating weights to genotypes and output text to phenotypes can improve our understanding of model lineage.<n>We generate an unsupervised evolutionary tree of black-box foundation models.
• Score: 0.5387709967970504
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Evolutionary methods have long been useful for analysis and explanation in genetics, biology, ecology, and related fields. In this work, we extend these methods to neural networks, specifically large language models (LLMs), to better analyze and explain relationships among models. We show how relating weights to genotypes and output text to phenotypes can improve our understanding of model lineage, important datasets, the roles of different model layers, and visualization of model relationships. We demonstrate this in a controlled experiment, where our estimated evolutionary trees reliably recover the topology of the ground-truth training tree. We further identify the most important weight layers according to weight differences and show through phenotypic experiments that one training dataset appears to contribute more useful information than the others. Finally, we generate an unsupervised evolutionary tree of black-box foundation models. Throughout, we provide visualizations that support a clearer understanding of evolutionary relationships among LLMs.

Related papers

• Interpretable and Fair Generalized Additive Neural Networks via Multi-objective Learning [52.66717986145448]
  Interpretability and fairness are two of the most emphasized dimensions in trustworthy artificial intelligence (AI)<n>This paper focuses on neural network (NN)-based generalized additive models (GAMs), a class of self-interpretable models.
  arXiv  Detail & Related papers  (2026-09-05T07:29:42Z)
• Deep-learning-based pan-phenomic data reveals the explosive evolution of avian visual disparity [8.871465190908003]
  This study uses deep learning techniques to explore avian morphological evolution.<n>We show that the high-dimensional embedding space encodes phenotypic convergence.<n>We also demonstrate that hierarchical semantic structures emerged in the high-dimensional embedding space despite being trained on flat labels.
  arXiv  Detail & Related papers  (2026-02-03T18:32:15Z)
• KAN-Matrix: Visualizing Nonlinear Pairwise and Multivariate Contributions for Physical Insight [0.0]
  We introduce a novel application of Kolmogorov-Arnold Networks (KANs) to enhance interpretability and parsimony.<n>We present two interpretable, color-coded visualization tools: the Pairwise KAN Matrix (PKAN) and the Multivariable KAN Contribution Matrix (MKAN)
  arXiv  Detail & Related papers  (2025-12-12T02:04:53Z)
• I Predict Therefore I Am: Is Next Token Prediction Enough to Learn Human-Interpretable Concepts from Data? [76.15163242945813]
  Large language models (LLMs) have led many to conclude that they exhibit a form of intelligence.<n>We introduce a novel generative model that generates tokens on the basis of human-interpretable concepts represented as latent discrete variables.
  arXiv  Detail & Related papers  (2025-03-12T01:21:17Z)
• PhyloVAE: Unsupervised Learning of Phylogenetic Trees via Variational Autoencoders [5.505257238864315]
  PhyloVAE is an unsupervised learning framework designed for representation learning and generative modeling of tree topologies.<n>We develop a deep latent-variable generative model that facilitates fast, parallelized topology generation.<n>Experiments demonstrate PhyloVAE's robust representation learning capabilities and fast generation of phylogenetic tree topologies.
  arXiv  Detail & Related papers  (2025-02-07T07:58:47Z)
• Cross-Entropy Is All You Need To Invert the Data Generating Process [29.94396019742267]
  Empirical phenomena suggest that supervised models can learn interpretable factors of variation in a linear fashion.<n>Recent advances in self-supervised learning have shown that these methods can recover latent structures by inverting the data generating process.<n>We prove that even in standard classification tasks, models learn representations of ground-truth factors of variation up to a linear transformation.
  arXiv  Detail & Related papers  (2024-10-29T09:03:57Z)
• Learning Discrete Concepts in Latent Hierarchical Models [73.01229236386148]
  Learning concepts from natural high-dimensional data holds potential in building human-aligned and interpretable machine learning models.<n>We formalize concepts as discrete latent causal variables that are related via a hierarchical causal model.<n>We substantiate our theoretical claims with synthetic data experiments.
  arXiv  Detail & Related papers  (2024-06-01T18:01:03Z)
• Beyond DAGs: A Latent Partial Causal Model for Multimodal Learning [80.44084021062105]
  We propose a novel latent partial causal model for multimodal data, featuring two latent coupled variables, connected by an undirected edge, to represent the transfer of knowledge across modalities.<n>Under specific statistical assumptions, we establish an identifiability result, demonstrating that representations learned by multimodal contrastive learning correspond to the latent coupled variables up to a trivial transformation.<n>Experiments on a pre-trained CLIP model embodies disentangled representations, enabling few-shot learning and improving domain generalization across diverse real-world datasets.
  arXiv  Detail & Related papers  (2024-02-09T07:18:06Z)
• Neural Echos: Depthwise Convolutional Filters Replicate Biological Receptive Fields [56.69755544814834]
  We present evidence suggesting that depthwise convolutional kernels are effectively replicating the biological receptive fields observed in the mammalian retina. We propose a scheme that draws inspiration from the biological receptive fields.
  arXiv  Detail & Related papers  (2024-01-18T18:06:22Z)
• Tertiary Lymphoid Structures Generation through Graph-based Diffusion [54.37503714313661]
  In this work, we leverage state-of-the-art graph-based diffusion models to generate biologically meaningful cell-graphs. We show that the adopted graph diffusion model is able to accurately learn the distribution of cells in terms of their tertiary lymphoid structures (TLS) content.
  arXiv  Detail & Related papers  (2023-10-10T14:37:17Z)
• Unsupervised ensemble-based phenotyping helps enhance the discoverability of genes related to heart morphology [57.25098075813054]
  We propose a new framework for gene discovery entitled Un Phenotype Ensembles. It builds a redundant yet highly expressive representation by pooling a set of phenotypes learned in an unsupervised manner. These phenotypes are then analyzed via (GWAS), retaining only highly confident and stable associations.
  arXiv  Detail & Related papers  (2023-01-07T18:36:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.