DataPraxis: Scaling Analytics Tasks via Verified Workspace Evolution

arXiv:2610.26924v1 Announce Type: new Abstract: General-purpose agents are increasingly expected to perform professional analytics work, which requires sustained reasoning, iterative investigation, and reliable evidence. Training these capabilities requires diverse and challenging tasks. However, generating tasks directly from raw data often overlooks deeper questions that emerge through analysis and provides limited evidence for verification. To address these limitations, we introduce DataPraxis, an analytics task synthesis framework based on verified workspace evolution. The framework progressively enriches workspaces with validated findings and artifacts, then inherits and combines these workspaces to uncover deeper questions. It constructs tasks from accumulated findings and their dependencies, verifying analytical results both during evolution and after task assembly. Using this framework, we construct DataPraxis-5K, a dataset of 5,000 verified trajectories spanning diverse domains and output formats, to train DataPraxis-27B and DataPraxis-35B-A3B. DataPraxis-27B achieves 55.11 on DABStep and 51.9 on JobBench, matching or surpassing frontier models such as GLM-5.2. DataPraxis-5K is available at huggingface.co/datasets/stargazerhust/DataPraxis-5k

Improves analytics tasks, potentially increasing productivity. No direct societal implications mentioned.

The framework DataPraxis advances AI's ability to perform complex professional analytics through verified task synthesis and workspace evolution. This supports more robust analytical reasoning and verification in AI, which can improve productivity and quality of analytics work. The impact is positive but moderately limited to professional analytic capacities.

This is a technical capability paper on benchmark performance for analytics agents, not a direct claim about societal impact. It implies incremental progress toward automating analyst-level work, a mild labor-relevant trend, but offers no direct evidence of deployment or societal effect.

The paper introduces a framework to generate analytical tasks and datasets for training AI agents, aiming to improve their sustained reasoning and investigative capabilities in professional analytics. This is a technical advancement in AI development without clear immediate positive or negative societal implications, leading to a near-neutral score.

SOURCE     arXiv cs.AI (tier 1)
   ↓
DOCUMENT   325b6ac1-1c9e-472a-b792-972181f80c40
           https://arxiv.org/abs/2610.26924
   ↓
EVIDENCE   1 extracted excerpt
   ↓
MODEL RUN  4 runs, methodology 1.0.1
   ↓
SCORE      -10  (Mixed or uncertain)
   ↓
CONFIDENCE 74%