AWhat changed
Integrated sparse expert modeling with long-context training and serving, reporting retrieval and long-document, audio and video evaluations across million-token inputs.
Scientific scope
Integrated sparse expert modeling with long-context training and serving, reporting retrieval and long-document, audio and video evaluations across million-token inputs. The original Pro report is one system contribution, not one record per context length. Synthetic retrieval success does not establish reliable general reasoning over every long input. MoE and long-context predecessors retain credit.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. The original Pro report is one system contribution, not one record per context length. Synthetic retrieval success does not establish reliable general reasoning over every long input. MoE and long-context predecessors retain credit.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
- Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
- Gemini Team: million-token multimodal long-context modeling — official research record
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction