Skip to content
Demonstrator · items marked Example are invented · what exists today
E2ER

External pipelines · RISE project catalogue

ARIS (Auto-Research-In-Sleep)

An open-source research harness for autonomous ML research (arXiv:2605.03042) built around *cross-model adversarial collaboration*: an executor model drives forward progress while a reviewer from a different model family critiques intermediate artifacts and requests revisions. Ships as 74+ Markdown-only `SKILL.md` files plus an optional standalone CLI (ARIS-Code), with the explicit design stance that the harness is a *methodology*, not a platform, portable across Claude Code, Codex CLI, Cursor, Trae, Antigravity, Copilot CLI, OpenClaw, Windsurf, and custom agents.

Indexed in RISE · activeConformance with the standard plannedProject site

What it does

Treats *long-running plausible-unsupported-success* as the central failure mode of autonomous research and addresses it by mandating cross-model review as the default (executor and reviewer from different model families, e.g., Claude Code + Codex MCP at GPT-5.4 xhigh, or MiniMax-M2.7 + GLM-5). Three-stage evidence verification, five-pass scientific editing pipeline, mathematical-proof checks, visual PDF inspection, and a persistent *Research Wiki* (papers / ideas / experiments / claims + relationship graph). Notably most- starred (9.8k+) Claude-Code-native research harness in the catalog.

Focus
end-to-end
Inputs
research-direction, prior-paper, base-repo, reviewer-comments
Outputs
paper, code, figures, rebuttal, slides, research-wiki
Architecture
multi-agent, debate-consensus, human-in-loop, tool-use, rag-knowledge-base, persistent-memory, iterative-loop, artifact-versioning
Maintained by
Ruofeng Yang; Yongcan Li; Shuai Li
Started
2026

Description

Data model
Discipline
Computer science
Method family
not specified
Design
not specified
Research stage
Research questionHypothesesLiterature discoveryLiterature synthesisResearch designData analysisCode generationDraftingRevision and editingReviewDissemination
Contributors
Ruofeng Yang; Yongcan Li; Shuai Li
Usage
not used in published research yet
Source
RISE project catalogue · projects/landscape · @4c17bae
Record
pipeline:aris · 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.

Other pipelines in Computer science