HH-SAE: Discovering and Steering Hierarchical Knowledge of Complex Manifolds

• URL: http://arxiv.org/abs/2605.10536v1
• Date: Mon, 11 May 2026 13:19:45 GMT
• Title: HH-SAE: Discovering and Steering Hierarchical Knowledge of Complex Manifolds
• Abstract summary: We introduce the textbfHybrid (HH-SAE) to resolve rare semantic innovations in high-dimensional, mission-critical domains.<n> HH-SAE prioritizes high-order mechanistic innovation over environmental proxies to enable high-precision discovery in high-stakes environments.
• Score: 12.383053678348281
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Rare semantic innovations in high-dimensional, mission-critical domains are often obscured by dense background contexts, a challenge we define as \textit{feature density conflict}. We introduce the \textbf{Hybrid Hierarchical SAE (HH-SAE)} to resolve this by factorizing manifolds into a nested hierarchy of \textbf{Contextual} ($L_0$), \textbf{Atomic} ($f_1$), and \textbf{Compository} ($f_2$) tiers. Evaluating across disparate manifolds, HH-SAE demonstrates superior resolution by \textbf{``fracturing'' administrative clinical labels into physiological modes} and achieving a peak \textbf{cross-domain zero-shot AUC of 0.9156 in fraud detection}. Path ablation confirms the architecture's structural necessity, revealing a 13.46\% utility collapse when contextual subtraction is removed. Finally, knowledge-steered synthesis achieves a +9.9\% AUPRC lift over state-of-the-art generators, proving that HH-SAE effectively prioritizes high-order mechanistic innovation over environmental proxies to enable high-precision discovery in high-stakes environments.

Related papers

• LoRC: Detecting AI-Generated Images via Low-Rank Collapse in Semantic Residuals [50.157719466558696]
  We show that modern generators exhibit low-rank collapse in the semantic-residual subspace while preserving the dominant semantic direction.<n>We propose textbfLoRC, a framework that decouples semantic dominance to capture the collapsed residual geometry.<n>Our method improves accuracy by an average of 7.0% across multiple benchmarks and achieves 97.0% accuracy on 39 unseen generators.
  arXiv  Detail & Related papers  (2026-08-21T08:58:47Z)
• From Extrinsic to Intrinsic: Geodesic-Guided Representation Learning for 3D Geometric Data [56.41479248637621]
  We introduce a novel 3D representation learning paradigm, namely textbfPRISM, for textbfPre-training.<n>PRISM learns embeddings by textbfRecovering the textbfIntrinsic textbfSurface geodesic textbfMetric.<n> Experiments demonstrate that our approach shows satisfactory accuracy, robustness, and high efficiency in geodesic distance prediction.
  arXiv  Detail & Related papers  (2026-06-01T13:54:26Z)
• Beyond the Aggregation Dilemma: Prior-Retaining Decoupled Learning for Multimodal Graphs [69.56407708794845]
  MultimodalAttributed Graph Learning (MAGL) integrates node attributes with structural topology via graph aggregation.<n>As pretrained encoders evolve into Large Foundation Models (LFMs), the landscape of MAGL fundamentally shifts.<n>Under high-confidence LFM priors, mandatory aggregation introduces topological intrinsic noise that overwhelms discriminative signals.<n>We propose SUPRA (Shared-Unique Prior-Retaining Architecture), a decoupled dual-pathway paradigm.<n> SUPRA processes modality-specific features through topology-agnostics while capturing structural synergy via a lightweight shared GNN, with auxiliary deep supervision counteracting gradient starvation.
  arXiv  Detail & Related papers  (2026-05-23T17:42:25Z)
• High-Dimensional Latents Should Be Diagnosed Through Phase Structure [56.362776482614976]
  We study autoencoder and variational-autoencoder latent spaces through the lens of spin-glass theory.<n>We formalize a latent-space spin-glass dictionary for a fixed decoder.<n>We show that driving the latent system toward the edge-of-stability of the topological trivialization regime has concrete downstream consequences.
  arXiv  Detail & Related papers  (2026-05-23T06:17:23Z)
• 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)
• Robust Deepfake Detection: Mitigating Spatial Attention Drift via Calibrated Complementary Ensembles [8.588817718552418]
  Current deepfake detection models suffer from spatial attention drift under real-world compound degradations.<n>We propose a foundation-driven forensic framework that integrates an extreme compound degradation engine with a structurally constrained, multi-stream architecture.<n>Our approach yields highly stable zero-shot generalization, achieving Fourth Place in the NTIRE 2026 Robust Deepfake Detection Challenge at CVPR.
  arXiv  Detail & Related papers  (2026-04-28T17:32:48Z)
• Hyperbolic Enhanced Representation Learning for Incomplete Multi-view Clustering [57.38215918201251]
  We propose HERL, a Hyperbolic Enhanced Representation Learning framework for Incomplete Multi-View Clustering.<n> operating within the Poincaré ball, HERL constructs a structure-aware latent space to enhance representation learning.<n>We show that HERL consistently outperforms state-of-the-art approaches.
  arXiv  Detail & Related papers  (2026-04-18T10:50:46Z)
• From Intent to Evidence: A Categorical Approach for Structural Evaluation of Deep Research Agents [25.080554451314683]
  Deep research agents (DRAs) have emerged as a promising paradigm for complex information synthesis.<n>We formalize DRA behavior through the lens of category theory, modeling deep research workflow as a composition of structure-preserving maps.<n>We introduce a novel mechanism-aware benchmark with 296 questions designed to stress-test agents along four interpretable axes.
  arXiv  Detail & Related papers  (2026-03-26T11:37:26Z)
• HGNet: Scalable Foundation Model for Automated Knowledge Graph Generation from Scientific Literature [6.001574550157585]
  We propose a two-stage framework for scalable, zero-shot scientific knowledge graph constructions.<n>We release SPHERE, a benchmark for hierarchical relation extraction.<n>In zero-shot settings, gains reach 10.76% for NER and 26.2% for RE.
  arXiv  Detail & Related papers  (2026-03-24T12:35:10Z)
• A Hyperbolic Perspective on Hierarchical Structure in Object-Centric Scene Representations [58.143688187023734]
  We propose a simple pipeline to project Euclidean slot embeddings onto the Lorentz hyperboloid of hyperbolic space.<n>We construct five-level visual hierarchies directly from slot attention masks.<n>We find that hyperbolic projection exposes a consistent scene-level to object-level organisation.
  arXiv  Detail & Related papers  (2026-03-14T16:53:59Z)
• Semantic Level of Detail: Multi-Scale Knowledge Representation via Heat Kernel Diffusion on Hyperbolic Manifolds [0.0]
  We introduce Semantic Level of Detail (SLoD), a framework that defines a continuous zoom operator via heat kernel diffusion on the Poincaré ball.<n>We show that spectral gaps in the graph Laplacian induce emergent scale boundaries.<n>On synthetic hierarchies, our boundary scanner recovers planted levels with ARI up to 1.00, with detection degrading gracefully near the information-theoretic Kesten-Stigum threshold.
  arXiv  Detail & Related papers  (2026-03-09T21:54:08Z)
• Hyper-KGGen: A Skill-Driven Knowledge Extractor for High-Quality Knowledge Hypergraph Generation [63.4604143884703]
  Hyper-KGGen is a skill-driven framework that reformulates extraction as a skill-evolving process.<n>It incorporates an textitadaptive skill acquisition module that actively distills domain expertise into a Global Skill Library.<n>We present textbfHyperDocRED, a rigorously annotated benchmark for document-level knowledge hypergraph extraction.
  arXiv  Detail & Related papers  (2026-02-23T06:32:00Z)
• OMG-Agent: Toward Robust Missing Modality Generation with Decoupled Coarse-to-Fine Agentic Workflows [9.617220633655716]
  We present textbfunderlineOmni-textbfunderlineModality textbfunderlineGeneration Agent (textbfOMG-Agent)
  arXiv  Detail & Related papers  (2026-02-04T02:25:40Z)
• Improving LLM Reasoning with Homophily-aware Structural and Semantic Text-Attributed Graph Compression [55.51959317490934]
  Large language models (LLMs) have demonstrated promising capabilities in Text-Attributed Graph (TAG) understanding.<n>We argue that graphs inherently contain rich structural and semantic information, and that their effective exploitation can unlock potential gains in LLMs reasoning performance.<n>We propose Homophily-aware Structural and Semantic Compression for LLMs (HS2C), a framework centered on exploiting graph homophily.
  arXiv  Detail & Related papers  (2026-01-13T03:35:18Z)
• H3Former: Hypergraph-based Semantic-Aware Aggregation via Hyperbolic Hierarchical Contrastive Loss for Fine-Grained Visual Classification [17.431440244641585]
  H3Former is a novel token-to-region framework for fine-grained visual classification.<n>SAAM exploits multi-scale contextual cues to dynamically construct a weighted hypergraph among tokens.<n>HHCL enforces hierarchical semantic constraints in a non-Euclidean embedding space.
  arXiv  Detail & Related papers  (2025-11-13T12:49:10Z)
• DDL: A Large-Scale Datasets for Deepfake Detection and Localization in Diversified Real-World Scenarios [51.916287988122406]
  We present a novel large-scale deepfake detection and localization (textbfDDL) dataset containing over $textbf1.4M+$ forged samples.<n>Our DDL not only provides a more challenging benchmark for complex real-world forgeries but also offers crucial support for building next-generation deepfake detection, localization, and interpretability methods.
  arXiv  Detail & Related papers  (2025-06-29T15:29:03Z)
• Coupling Global Context and Local Contents for Weakly-Supervised Semantic Segmentation [54.419401869108846]
  We propose a single-stage WeaklySupervised Semantic (WSSS) model with only the image-level class label supervisions. A flexible context aggregation module is proposed to capture the global object context in different granular spaces. A semantically consistent feature fusion module is proposed in a bottom-up parameter-learnable fashion to aggregate the fine-grained local contents.
  arXiv  Detail & Related papers  (2023-04-18T15:29:23Z)

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.