Semantic Sections: An Atlas-Native Feature Ontology for Obstructed Representation Spaces

• URL: http://arxiv.org/abs/2603.20867v1
• Date: Sat, 21 Mar 2026 15:59:29 GMT
• Title: Semantic Sections: An Atlas-Native Feature Ontology for Obstructed Representation Spaces
• Abstract summary: We introduce an atlas-native replacement object, the semantic section.<n>We show that cycle consistency is the key criterion for genuine globalization.<n>We find nontrivial populations of semantic sections across layer-16 atlases for Llama 3.2 3B Instruct, Qwen 2.5 3B Instruct, and Gemma 2 2B IT.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent interpretability work often treats a feature as a single global direction, dictionary atom, or latent coordinate shared across contexts. We argue that this ontology can fail in obstructed representation spaces, where locally coherent meanings need not assemble into one globally consistent feature. We introduce an atlas-native replacement object, the semantic section: a transport-compatible family of local feature representatives defined over a context atlas. We formalize semantic sections, prove that tree-supported propagation is always pathwise realizable, and show that cycle consistency is the key criterion for genuine globalization. This yields a distinction between tree-local, globalizable, and twisted sections, with twisted sections capturing locally coherent but holonomy-obstructed meanings. We then develop a discovery-and-certification pipeline based on seeded propagation, synchronization across overlaps, defect-based pruning, cycle-aware taxonomy, and deduplication. Across layer-16 atlases for Llama 3.2 3B Instruct, Qwen 2.5 3B Instruct, and Gemma 2 2B IT, we find nontrivial populations of semantic sections, including cycle-supported globalizable and twisted regimes after deduplication. Most importantly, semantic identity is not recovered by raw global-vector similarity. Even certified globalizable sections show low cross-chart signed cosine similarity, and raw similarity baselines recover only a small fraction of true within-section pairs, often collapsing at moderate thresholds. By contrast, section-based identity recovery is perfect on certified supports. These results support semantic sections as a better feature ontology in obstructed regimes.

Related papers

• A Dynamic-Semantics Framework for Grounding Human Referring Expressions in Visual Perceptual Data [0.0]
  We present a framework that externalizes pact state into three explicit, inspectable sets of referent-object bindings.<n>We evaluate the framework on the Stanford Repeated Reference Game corpus.
  arXiv  Detail & Related papers  (2026-08-09T12:14:10Z)
• C$^{2}$R: Cross-sample Consistency Regularization Mitigates Feature Splitting and Absorption in Sparse Autoencoders [56.67585330425431]
  Sparse Autoencoders (SAEs) are widely used to interpret large language models by decomposing activations into sparse, human-understandable features.<n>These issues stem from inconsistent latent assignment across samples.<n>We introduce C$2$R, which penalizes the co-activation of directionally similar latents.
  arXiv  Detail & Related papers  (2026-06-29T17:45:31Z)
• SAM-NER: Semantic Archetype Mediation for Zero-Shot Named Entity Recognition [50.82878172248818]
  emphSemantic Archetype Mediation stabilizes cross-domain transfer through an intermediate, domain-invariant archetype space.<n> SAM-NER consistently outperforms strong prior ZS-NER baselines in cross-domain settings.
  arXiv  Detail & Related papers  (2026-05-05T12:54:17Z)
• Hierarchical Context Alignment with Disentangled Geometric and Temporal Modeling for Semantic Occupancy Prediction [61.484280369655536]
  Camera-based 3D Semantic Occupancy Prediction (SOP) is crucial for understanding complex 3D scenes from limited 2D image observations.<n>Existing SOP methods typically aggregate contextual features to assist the occupancy representation learning.<n>We introduce a new Hierarchical context alignment paradigm for a more accurate SOP (Hi-SOP)
  arXiv  Detail & Related papers  (2024-12-11T09:53:10Z)
• Spatial Semantic Recurrent Mining for Referring Image Segmentation [63.34997546393106]
  We propose Stextsuperscript2RM to achieve high-quality cross-modality fusion. It follows a working strategy of trilogy: distributing language feature, spatial semantic recurrent coparsing, and parsed-semantic balancing. Our proposed method performs favorably against other state-of-the-art algorithms.
  arXiv  Detail & Related papers  (2024-05-15T00:17:48Z)
• Semantic Connectivity-Driven Pseudo-labeling for Cross-domain Segmentation [89.41179071022121]
  Self-training is a prevailing approach in cross-domain semantic segmentation. We propose a novel approach called Semantic Connectivity-driven pseudo-labeling. This approach formulates pseudo-labels at the connectivity level and thus can facilitate learning structured and low-noise semantics.
  arXiv  Detail & Related papers  (2023-12-11T12:29:51Z)
• Leveraging Hidden Positives for Unsupervised Semantic Segmentation [5.937673383513695]
  We leverage contrastive learning by excavating hidden positives to learn rich semantic relationships. We introduce a gradient propagation strategy to learn semantic consistency between adjacent patches. Our proposed method achieves new state-of-the-art (SOTA) results in COCO-stuff, Cityscapes, and Potsdam-3 datasets.
  arXiv  Detail & Related papers  (2023-03-27T08:57:28Z)
• Understanding Imbalanced Semantic Segmentation Through Neural Collapse [81.89121711426951]
  We show that semantic segmentation naturally brings contextual correlation and imbalanced distribution among classes. We introduce a regularizer on feature centers to encourage the network to learn features closer to the appealing structure. Our method ranks 1st and sets a new record on the ScanNet200 test leaderboard.
  arXiv  Detail & Related papers  (2023-01-03T13:51:51Z)
• Finding the global semantic representation in GAN through Frechet Mean [4.688163910878411]
  We propose an unsupervised method for finding this global semantic basis in the intermediate latent space in GANs. The proposed global basis, called Fr'echet basis, is derived by introducing Fr'echet mean to the local semantic perturbations in a latent space. Experimental results demonstrate that Fr'echet basis provides better semantic factorization and robustness compared to the previous methods.
  arXiv  Detail & Related papers  (2022-10-11T15:01:25Z)
• Latent Topology Induction for Understanding Contextualized Representations [84.7918739062235]
  We study the representation space of contextualized embeddings and gain insight into the hidden topology of large language models. We show there exists a network of latent states that summarize linguistic properties of contextualized representations.
  arXiv  Detail & Related papers  (2022-06-03T11:22:48Z)
• Human Correspondence Consensus for 3D Object Semantic Understanding [56.34297279246823]
  In this paper, we introduce a new dataset named CorresPondenceNet. Based on this dataset, we are able to learn dense semantic embeddings with a novel geodesic consistency loss. We show that CorresPondenceNet could not only boost fine-grained understanding of heterogeneous objects but also cross-object registration and partial object matching.
  arXiv  Detail & Related papers  (2019-12-29T04:24:22Z)

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.