Knowledge catalogue
AD2-Bench diagnoses multimodal reasoning via a Chain of Evidence, showing failures from spatial ambiguity and semantic uncertainty in adverse urban scenes.
Atlassian says Rovo Chat's live Long Horizon engine uses Claude Opus 4.x-class models and can swap reasoning models.
SAG retrieves structured multi-hop evidence via event-entity hyperedges and query-time joins, without a maintained global knowledge graph.
A practical method for converting Figma theme evidence into an LLM handoff that fits an existing application and preserves accessibility.
Apple paper: keep specs, schemas, tool config, and constraints; drop session traces. Selective shared memory beat both no memory and full history.
Atlassian uses Claude for complex Rovo workflows and Claude Code internally; its multimodel platform also uses Google Cloud.