AI Paper Atlas

Research map / Multi-agent systems and games / LLM agents

$R^3$DAO: Reactive Recovery and Reconstruction for Long-horizon Data Agent Orchestration

Quanxin Liu, Rui Hao, Ruida Xu, Jianwei Zhong, Changhu Chen, Yijun Mo

ICML 2026 · Multi-agent systems and games › LLM agents · approach: Robust and efficient agent execution

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes R³DAO, a reactive recovery and reconstruction framework for orchestrating long-horizon data science agent workflows, aimed at preventing error propagation from early stages.

Problem

End-to-end data science agent workflows consist of tightly coupled sub-processes with strong dynamic dependencies; static chain-like execution plans propagate early errors, causing reasoning chain collapse, task failure, and poor cost-effectiveness.

Approach

Moves away from static chain-like execution plans toward reactive recovery and reconstruction of the workflow when failures occur, so that errors in early stages do not collapse the whole reasoning chain.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add reactive topology reconfiguration (RTR) instead of static retries (ablation)R3DAO · MLE-Bench subset (17 tasks)success rate: 52.9% -> 94.1%Improved
add RTR instead of static retries (ablation)R3DAO · MLE-Bench subset (17 tasks)normalized score: 0.268 -> 0.422Improved
add experience distillation memory (ablation)R3DAO · MLE-Bench subset (17 tasks)average tokens: 215k -> 104kImproved
add experience distillation memory (ablation)R3DAO · MLE-Bench subset (17 tasks)success rate: 82.4% -> 94.1%Improved
add D-HTN hierarchical planning instead of linear plan (ablation)R3DAO · MLE-Bench subset (17 tasks)normalized score: 0.195 -> 0.422Improved
add D-HTN hierarchical planning instead of linear plan (ablation)R3DAO · MLE-Bench subset (17 tasks)success rate: 41.2% -> 94.1%Improved
use R3DAO instead of AIDE under the same Qwen3-Max, <1h budgetR3DAO · MLE-Bench subset (17 tasks) grand averagenormalized score: 0.456 -> 0.646Improved
use R3DAO instead of AIDE on high-complexity tasksR3DAO · MLE-Bench subset (High complexity)normalized score: 0.286 -> 0.495Improved
use R3DAO instead of AIDE on NFL-CDR3DAO · MLE-Bench NFL-CDnormalized score: 0.610 -> 0.268Hurt
use R3DAO (Qwen3-Max) instead of high-resource R&D-AgentR3DAO · MLE-Bench subset (17 tasks)time and tokens: 720 min -> 21.7 min; ~3000k -> 103k tokensImproved
use R3DAO instead of R&D-AgentR3DAO · MLE-Bench subset (17 tasks)normalized score: 0.44 -> 0.42Mixed
raise weight of completion score Scomp in the comprehensive scoreR3DAO vs R&D-Agent · MLE-Bench subset (17 tasks)comprehensive score: 0.422 vs 0.444 at 1.0/0.0; 0.682 vs 0.487 at 0.5/0.5Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy