Article

Geometric Representations of Knowledge Inside Biological Large Language Models: an Empirical Analysis

2026-07-22

Abstract excerpt

<title>Abstract</title> <p>In large language models, factual and relational knowledge is often carried by a strikingly regular geometry: concepts correspond to linear directions, antonyms and analogies to parallel offsets, and taxonomies to nested, near-orthogonal subspaces. Single-cell foundation models (SCFMs)—“biological large language models” trained with the same masked-token recipe on transcriptomes—are now...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: d29b42e4-00bb-5ad5-b6b5-79575929ba39
DOI: 10.21203/rs.3.rs-10434032/v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Geometric Representations of Knowledge Inside Biological Large Language Models: an Empirical AnalysisDOI 10.21203/rs.3.rs-10434032/v1Read

Tokenizing single-cell transcriptomes as a native language for large language modelsSemantic proximity Read

CLOP-DiT: Structured-Metadata-Conditioned Single-Cell Latent Generation via Contrastive Language-Omics Pretraining and Diffusion TransformersSemantic proximity Read

CellFM: a large-scale foundation model pre-trained on transcriptomics of 100 million human cellsSemantic proximity Read

scGenAI: A generative AI platform with biological context embedding of multimodal features enhances single cell state classificationSemantic proximity Read

Language may be all omics needs: Harmonizing multimodal data for omics understanding with CellHermesSemantic proximity Read

CellTosg2Sequence: A Unified Text-Omics-Signaling-Graph Large Language Model for Single-Cell AnalysisSemantic proximity Read

Scalable Non-negative Matrix Factorization of the Human Cell Census Reveals Interpretable Transcriptional ProgramsSemantic proximity Read

Coding agents author interpretable single-cell embedding models from the literatureSemantic proximity Read

Scaling Dense Representations for Single Cell with Transcriptome-Scale ContextSemantic proximity Read

Gene-Family Encoding Boosts Domain-Adapted Single-Cell Language ModelsSemantic proximity Read

scKEPLM: Knowledge enhanced large-scale pre-trained language model for single-cell transcriptomicsSemantic proximity Read

Hierarchical Interpretation of Out-of-Distribution Cells Using Bottlenecked TransformerSemantic proximity Read