External pipelines · RISE project catalogue
EvoScientist
A self-evolving AI scientist system (arXiv:2603.08127) built on the DeepAgents framework. Six sub-agents (plan, research, code, debug, analyze, write) co-evolve with persistent memory, adaptive per-turn tool selection, dynamic system-prompt rewriting, and installable skill / knowledge packs (EvoSkills). Positioned around "vibe research" — a human-on-the-loop paradigm where the AI acts as a co-evolving research buddy that internalizes scholarly taste.
Where it sits
left: what it builds on · right: what builds on it · pale: exampleContributed by EvoScientist team
How studies reach it
No published study reaches it yet.
Disciplines it reaches
No study reaches it yet.
Solid: published studies. Light: examples.
Computed from the records on this site: what each study, template and specialist names as used, which study extends which, and who contributed what. 0 studies in total.
What it does
Strongest external-benchmark trajectory in the catalog: #1 on DeepResearch Bench, #1 on DeepResearch Bench II, #1 on AstaBench Code & Execution, #1 on AstaBench Data Analysis (all at submission time), plus 6/6 papers accepted at ICAIS 2025 with Best Paper + Appraisal awards. Skills + memory architecture is engineered for *evolution* over sessions, not just task completion.
- Focus
- end-to-end
- Inputs
- research-direction, prior-context
- Outputs
- research-report, code, paper-draft, evolved-skills
- Architecture
- multi-agent, human-in-loop, tool-use, rag-knowledge-base, persistent-memory, iterative-loop, artifact-versioning
- Maintained by
- EvoScientist team
- Started
- 2026
Description
Data model- Discipline
- General
- Method family
- not specified
- Design
- not specified
- Research stage
- Research questionHypothesesLiterature discoveryLiterature synthesisResearch designData analysisCode generationDraftingRevision and editing
- Contributors
- EvoScientist team
- Usage
- not used in published research yet
- Source
- RISE project catalogue · projects/landscape · @4c17bae
- Record
- pipeline:evoscientist · JSON
Solid tags are declared by the source or mapped from its terms; dashed tags are inferred by a published rule. Hover a tag for its provenance.
Bring it into the standard
A pipeline built outside E2ER can meet the standard by describing its steps as a template, attaching the floor of checks and publishing evaluation records. Its authors keep ownership and credit.