Geometry of Ordinal Representations in Language Models

• URL: http://arxiv.org/abs/2607.04167v1
• Date: Sun, 05 Jul 2026 08:13:26 GMT
• Title: Geometry of Ordinal Representations in Language Models
• Abstract summary: We show that geometric computation is architecture-dependent, with twisters preserving ordinal order, unlike numeric twisters.<n>We also find that Qwen3-4B shows substantially stronger twisting than Gemma models for indentation, and its twisters preserve ordinal order, unlike its numeric twisters.<n> Activation patching confirms that the identified manifold subspaces concentrate task-relevant information, with manifold-direction ablation causing dramatically larger probe accuracy drops than random-direction controls.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent work showed that language models represent character counts on curved 1D manifolds, with attention heads performing geometric transformations to enable computation. We test whether this generalizes across four ordinal tasks (bracket depth, indentation, table position, numeric magnitude) in Gemma-2-2B, Gemma-2-9B, and Qwen3-4B. We find that 1D manifolds with place-cell feature tiling emerge for tasks where the ordinal variable is locally computable from token identity, while tasks requiring cross-position integration or semantic extraction produce higher-dimensional or incoherent representations. Geometric computation is architecture-dependent: Qwen3-4B shows substantially stronger twisting than Gemma models for indentation, and its twisters preserve ordinal order, unlike its numeric twisters. Activation patching confirms that the identified manifold subspaces concentrate task-relevant information, with manifold-direction ablation causing dramatically larger probe accuracy drops than random-direction controls.

Related papers

• SCOPE: Scale-Consistent One-Pass Estimation of 3D Geometry [69.89196162649091]
  SCOPE (Scale-Consistent One-Pass Estimation of 3D Geometry) is a novel approach for estimating 3D geometry from extended monocular video sequences.<n>Our approach generates affine-invariant 3D point maps with shared parameters across entire sequences, enabling consistent scale-invariant representations.
  arXiv  Detail & Related papers  (2026-06-19T10:23:25Z)
• SpatialEvo: Self-Evolving Spatial Intelligence via Deterministic Geometric Environments [75.60795462502949]
  SpatialEvo is a self-evolving framework for 3D spatial reasoning.<n>It formalizes 16 spatial reasoning task categories under explicit geometric validation rules.<n>It converts unannotated 3D scenes into zero-noise interactive oracles, replacing model consensus with objective physical feedback.
  arXiv  Detail & Related papers  (2026-04-15T17:59:12Z)
• Finding Belief Geometries with Sparse Autoencoders [0.0]
  We introduce a pipeline for discovering candidate simplex-structured subspaces in transformer representations.<n>We validate the pipeline on a transformer trained on a hidden Markov model with known belief-state geometry.<n>Preliminary evidence that genuine belief-like geometry exists in Gemma-2-9B's representation space.
  arXiv  Detail & Related papers  (2026-04-03T03:29:48Z)
• GeoGuide: Hierarchical Geometric Guidance for Open-Vocabulary 3D Semantic Segmentation [38.12499404546835]
  We propose GeoGuide, a framework that integrates hierarchical geometry-semantic consistency for open-vocabulary 3D segmentation.<n> Specifically, we introduce an Uncertainty-based Superpoint Distillation module to fuse geometric and semantic features.<n>Our experiments on ScanNet v2, Matterport3D, and nuScenes demonstrate the superior performance of GeoGuide.
  arXiv  Detail & Related papers  (2026-03-27T10:29:19Z)
• SeqAffordSplat: Scene-level Sequential Affordance Reasoning on 3D Gaussian Splatting [85.87902260102652]
  We introduce the novel task of Sequential 3D Gaussian Affordance Reasoning.<n>We then propose SeqSplatNet, an end-to-end framework that directly maps an instruction to a sequence of 3D affordance masks.<n>Our method sets a new state-of-the-art on our challenging benchmark, effectively advancing affordance reasoning from single-step interactions to complex, sequential tasks at the scene level.
  arXiv  Detail & Related papers  (2025-07-31T17:56:55Z)
• Dens3R: A Foundation Model for 3D Geometry Prediction [44.13431776180547]
  Dens3R is a 3D foundation model designed for joint geometric dense prediction.<n>By integrating image-pair matching features with intrinsic invariance modeling, Dens3R accurately regresses multiple geometric quantities.
  arXiv  Detail & Related papers  (2025-07-22T07:22:30Z)
• Estimating Dataset Dimension via Singular Metrics under the Manifold Hypothesis: Application to Inverse Problems [0.6138671548064356]
  We propose a framework to deal with three key tasks: estimating the intrinsic dimension of the manifold, constructing appropriate local coordinates, and learning mappings between ambient and manifold spaces.<n>We focus on estimating the ID of datasets by analyzing the numerical rank of the VAE decoder pullback metric.<n>The estimated ID guides the construction of an atlas of local charts using a mixture of invertible VAEs, enabling accurate manifold parameterization and efficient inference.
  arXiv  Detail & Related papers  (2025-07-09T21:22:59Z)
• A Recipe for Geometry-Aware 3D Mesh Transformers [2.0992612407358293]
  We study an approach for embedding features at the patch level, accommodating patches with variable node counts. Our research highlights critical insights: 1) the importance of structural and positional embeddings facilitated by heat diffusion in general 3D mesh transformers; 2) the effectiveness of novel components such as geodesic masking and feature interaction via cross-attention in enhancing learning; and 3) the superior performance and efficiency of our proposed methods in challenging segmentation and classification tasks.
  arXiv  Detail & Related papers  (2024-10-31T19:13:31Z)
• Geometry-Contrastive Transformer for Generalized 3D Pose Transfer [95.56457218144983]
  The intuition of this work is to perceive the geometric inconsistency between the given meshes with the powerful self-attention mechanism. We propose a novel geometry-contrastive Transformer that has an efficient 3D structured perceiving ability to the global geometric inconsistencies. We present a latent isometric regularization module together with a novel semi-synthesized dataset for the cross-dataset 3D pose transfer task.
  arXiv  Detail & Related papers  (2021-12-14T13:14:24Z)
• Primal-Dual Mesh Convolutional Neural Networks [62.165239866312334]
  We propose a primal-dual framework drawn from the graph-neural-network literature to triangle meshes. Our method takes features for both edges and faces of a 3D mesh as input and dynamically aggregates them. We provide theoretical insights of our approach using tools from the mesh-simplification literature.
  arXiv  Detail & Related papers  (2020-10-23T14:49:02Z)
• SeqXY2SeqZ: Structure Learning for 3D Shapes by Sequentially Predicting 1D Occupancy Segments From 2D Coordinates [61.04823927283092]
  We propose to represent 3D shapes using 2D functions, where the output of the function at each 2D location is a sequence of line segments inside the shape. We implement this approach using a Seq2Seq model with attention, called SeqXY2SeqZ, which learns the mapping from a sequence of 2D coordinates along two arbitrary axes to a sequence of 1D locations along the third axis. Our experiments show that SeqXY2SeqZ outperforms the state-ofthe-art methods under widely used benchmarks.
  arXiv  Detail & Related papers  (2020-03-12T00:24:36Z)

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.