Tracking Representation Dynamics in Large Language Models with Persistent Homology

• URL: http://arxiv.org/abs/2606.19542v1
• Date: Wed, 17 Jun 2026 19:35:10 GMT
• Title: Tracking Representation Dynamics in Large Language Models with Persistent Homology
• Abstract summary: We study alignment dynamics using persistent homology.<n>We find that the majority of topological reorganization occurs during the earliest stages of training.<n>Different alignment objectives induce distinguishable topological trajectories, while instruction-tuned and pretrained models exhibit qualitatively different patterns of evolution.
• Score: 5.665477411284105
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models are commonly aligned through supervised fine-tuning, yet little is known about how their internal representations evolve during this process. We study alignment dynamics using persistent homology by tracking the topology of activation spaces throughout fine-tuning. Across four transformer language models ranging from 1B to 7B parameters and three alignment objectives corresponding to helpful, harmless, and mixed training data, we find that the majority of topological reorganization occurs during the earliest stages of training. A dense checkpoint analysis reveals a transient peak in topological activity followed by rapid stabilization. We further show that different alignment objectives induce distinguishable topological trajectories, while instruction-tuned and pretrained models exhibit qualitatively different patterns of evolution. Our results suggest that persistent homology provides a complementary perspective on alignment, revealing representation-level changes that are not apparent from behavioral metrics alone.

Related papers

• DyG$^2$T: Modeling Object Dynamics with 3D Gaussian Temporal-Spatial Particle Graph Transformer [86.4618122245946]
  DyG$2$T is a dynamics modeling framework that infers object motion trajectories by spatially completing and temporally discriminating Key Point representations.<n>Experiments on both synthetic and real-world datasets demonstrate that DyG$2$T achieves accurate dynamics modeling and reasoning.
  arXiv  Detail & Related papers  (2026-08-19T03:44:28Z)
• Interpreting learning dynamics of autoencoders: Transient scaling and emerging concepts of the Ising model [0.0]
  unsupervised autoencoders trained on microscopic spin configurations from the Ising model learn macroscopic, theory-relevant variables.<n>We quantify learning across multiple spatial scales and reveal two distinct dynamical regimes.<n>A dynamical viewpoint of learning is established in which intrinsic properties expose the effects of forced changes in representation.
  arXiv  Detail & Related papers  (2026-07-11T12:31:34Z)
• Self-Evolving Spatial Reasoning in Vision Language Models via Geometric Logic Consistency [53.28076739140119]
  We propose a framework that enforces logical consistency in Vision-Language Models (VLMs) through geometric and linguistic duality operations.<n>SAGE is model-agnostic, data-efficient compared to prior GRPO methods, and can be applied as a lightweight post-training stage to any existing VLM.<n> Experiments on video and spatial reasoning benchmarks demonstrate consistent improvements over strong baselines.
  arXiv  Detail & Related papers  (2026-05-18T10:05:21Z)
• A Mechanistic Analysis of Looped Reasoning Language Models [41.640597313586525]
  We conduct a mechanistic analysis of the latent states in looped language models.<n>We show that for many of the studied models each layer in the cycle converges to a distinct fixed point.<n>We discover that recurrent blocks learn stages of inference that closely mirror those of feedforward models.
  arXiv  Detail & Related papers  (2026-04-13T17:55:36Z)
• UniG2U-Bench: Do Unified Models Advance Multimodal Understanding? [50.92401586025528]
  Unified multimodal models have recently demonstrated strong generative capabilities, yet whether and when generation improves understanding remains unclear.<n>We introduce UniG2U-Bench, a comprehensive benchmark categorizing generation-to-understanding (G2U) evaluation into 7 regimes and 30 subtasks.
  arXiv  Detail & Related papers  (2026-03-03T18:36:16Z)
• Mechanistic Analysis of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning [0.0]
  Large language models exhibit remarkable performance across diverse tasks through pre-training and fine-tuning paradigms.<n>Continual fine-tuning on sequential tasks induces catastrophic forgetting, where newly acquired knowledge interferes with previously learned capabilities.<n>We identify three primary mechanisms driving forgetting: gradient interference in attention weights, representational drift in intermediate layers, and loss landscape flattening.
  arXiv  Detail & Related papers  (2026-01-26T17:15:10Z)
• Dynamical Systems Analysis Reveals Functional Regimes in Large Language Models [0.8694591156258423]
  Large language models perform text generation through high-dimensional internal dynamics.<n>Most interpretability approaches emphasise static representations or causal interventions, leaving temporal structure largely unexplored.<n>We discuss a composite dynamical metric, computed from activation time-series during autoregressive generation.
  arXiv  Detail & Related papers  (2026-01-11T21:57:52Z)
• Empirical Investigation of Latent Representational Dynamics in Large Language Models: A Manifold Evolution Perspective [4.935224714809964]
  This paper introduces the Dynamical Manifold Evolution Theory (DMET), a conceptual framework that models large language model (LLM) generation as a continuous trajectory evolving on a low-dimensional semantic manifold.<n>The theory characterizes latent dynamics through three interpretable metrics-state continuity ($C$), attractor compactness ($Q$), and topological persistence ($P$)-which jointly capture the smoothness, stability, and structure of representation evolution.
  arXiv  Detail & Related papers  (2025-05-24T14:17:50Z)
• Bridging Critical Gaps in Convergent Learning: How Representational Alignment Evolves Across Layers, Training, and Distribution Shifts [1.9458156037869137]
  convergent learning is the degree to which neural systems arrive at similar internal representations.<n>We present a large-scale audit of convergent learning spanning dozens of vision models and thousands of layer-pair comparisons.<n>Findings fill critical gaps in our understanding of representational convergence, with implications for neuroscience and AI.
  arXiv  Detail & Related papers  (2025-02-26T00:04:24Z)
• Relative Representations: Topological and Geometric Perspectives [50.85040046976025]
  Relative representations are an established approach to zero-shot model stitching.<n>We introduce a normalization procedure in the relative transformation, resulting in invariance to non-isotropic rescalings and permutations.<n>Second, we propose to deploy topological densification when fine-tuning relative representations, a topological regularization loss encouraging clustering within classes.
  arXiv  Detail & Related papers  (2024-09-17T08:09:22Z)
• Inverse Dynamics Pretraining Learns Good Representations for Multitask Imitation [66.86987509942607]
  We evaluate how such a paradigm should be done in imitation learning. We consider a setting where the pretraining corpus consists of multitask demonstrations. We argue that inverse dynamics modeling is well-suited to this setting.
  arXiv  Detail & Related papers  (2023-05-26T14:40:46Z)
• Latent Traversals in Generative Models as Potential Flows [113.4232528843775]
  We propose to model latent structures with a learned dynamic potential landscape. Inspired by physics, optimal transport, and neuroscience, these potential landscapes are learned as physically realistic partial differential equations. Our method achieves both more qualitatively and quantitatively disentangled trajectories than state-of-the-art baselines.
  arXiv  Detail & Related papers  (2023-04-25T15:53:45Z)

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.