Agent Memory Research in 2026: Companion Dataset (papers.yaml, JSON export, README)

Companion dataset for the survey paper "Agent Memory Research in 2026: A Data-Driven Survey and Extended Taxonomy". Contains the curated paper list in YAML format (papers.yaml), a JSON export for web use, and supporting metadata files. This is version 1.1.0 of the dataset, updated July 2026. The dataset now contains 952 papers on agent memory (up from 220 in v1.0.0), classified by the unified 3x3x3 taxonomy with additional Temporal Dynamics, Modality, and Biological Inspiration dimensions. Of these, 498 papers were published since February 2026. Contents: - papers.yaml: structured source of truth (952 entries, 900 unique arXiv IDs) - papers.json: JSON export for the interactive web interface - README.md: auto-generated paper list grouped by taxonomy category Each paper entry includes: title, date (YYYY-MM), URL, taxonomy category (factual/experiential/working), subcategory (token-level/parametric/latent), authors, venue, code URL, project URL, abstract, and tags. Author and abstract metadata is populated for 93% of entries via bulk arXiv API fetching.