From Topology to Retrieval: Decoding Embedding Spaces with Unified Signatures

arXiv:2511.22150 · cs.LG, cs.CL, cs.IR · Submitted 2025-11-27 · Read on arXiv

cs.LG, cs.CL, cs.IR

Submitted: 2025-11-27

Updated: 2026-10-12

Code: https://github.com/Boehringer-Ingelheim/unified-topological-signatures

Terminology

• Embeddings — 2× in this paper · explained in 2 episodes
• Interpretability — 1× in this paper · explained in 15 episodes
• Redundancy — 1× in this paper · explained in 2 episodes
• Unified Topological Signatures — 1× in this paper

Sources

• Representation Topology Divergence: A Method for Comparing Neural Network Representations
• Beyond Benchmarks: Evaluating Embedding Model Similarity for Retrieval Augmented Generation Systems
• SPECTER: Document-level Representation Learning using Citation-informed Transformers
• Reliability of CKA as a Similarity Measure in Deep Learning
• How Contextual are Contextualized Word Representations? Comparing the Geometry of BERT, ELMo, and GPT-2 Embeddings
• The Vendi Score: A Diversity Evaluation Metric for Machine Learning
• Persistent Topological Features in Large Language Models
• Large Language Models Are Zero-Shot Time Series Forecasters
• Geometric Interpretation of Layer Normalization and a Comparative Analysis with RMSNorm
• The Platonic Representation Hypothesis
• Manifold Approximation leads to Robust Kernel Alignment
• A Comparative Study of Learning Paradigms in Large Language Models via Intrinsic Dimension
• Measuring Intrinsic Dimension of Token Embeddings
• Shared Global and Local Geometry of Language Model Embeddings
• Gemini Embedding: Generalizable Embeddings from Gemini
• Geometric Signatures of Compositionality Across a Language Model's Lifetime
• Towards General Text Embeddings with Multi-stage Contrastive Learning
• Spectral Insights into Data-Oblivious Critical Layers in Large Language Models
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
• Arctic-Embed: Scalable, Efficient, and Accurate Text Embedding Models

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks