WorkFactory: Turning Professional Intents into Executable Work for Scalable Agent Training

Scaling supervised training for working agents requires diverse professional tasks that yield reliable expert demonstrations.

By Zeng, Chen, Deng +10Oct 27, 2026

Score████░░░░░░3.9

Caveats

Preprint; not yet peer reviewed.

VerdictWorth a reader's time today.

Read the original PDF Code

Abstract

Scaling supervised training for working agents requires diverse professional tasks that yield reliable expert demonstrations. However, missing or inconsistent task information can obstruct teacher execution and weaken the resulting supervision. We propose WorkFactory, a scalable framework for constructing coherent and executable professional tasks. Starting from professional intents, information-level planning builds an information support graph to guide task instantiation, explicitly representing the dependencies and locations of information required for the intended outputs. Probe executions then identify support failures and guide repairs while preserving the intended work. We construct 112,686 tasks spanning 612 occupations and use the collected teacher trajectories to fine-tune Qwen models at 4B, 27B, and 35B-A3B scales, yielding the WorkPilot family. All three models improve over their respective base models across four professional-work benchmarks, with WorkPilot-35B-A3B improving from 879 to 1565 Elo on GDPval-AA v2.1. Scaling analyses reveal model-dependent diminishing returns, with execution length first increasing and then decreasing. Controlled component analyses show that planning and repair reduce probe-diagnosed task environment defects and improve downstream performance. The data and models are available at https://github.com/IQuestLab/WorkFactory.

Ziyang Zeng, Tingyang Chen, Jia Deng, Qiyao Wang, Shuo Tang, Yike Yang, Tao Ren, Wayne Xin Zhao, Feng Chang, Jujie He, Chuan Hao, Ran Tao, Bryan Dai

The editor's rubric

Heuristic review

Leverage4 / 5: A general-purpose tool used across several fields (Adam, ResNet, LoRA, next-generation sequencing).; 24% of this desk’s rubric
Magnitude2 / 5: Solid incremental gain on a meaningful problem.; 18% of this desk’s rubric
Evidence3 / 5: Solid: multiple benchmarks or cohorts, ablations, fair baselines, released code or data.; 14% of this desk’s rubric
Novelty2 / 5: A new combination of known ideas.; 16% of this desk’s rubric
Trajectory3 / 5: A clear path to scale.; 18% of this desk’s rubric
Stakes2 / 5: Benefits a professional community (practitioners, clinicians, engineers).; 10% of this desk’s rubric

Dimension	Level	Weight	What that level means
Leverage	████░4/5	24%	A general-purpose tool used across several fields (Adam, ResNet, LoRA, next-generation sequencing).
Magnitude	██░░░2/5	18%	Solid incremental gain on a meaningful problem.
Evidence	███░░3/5	14%	Solid: multiple benchmarks or cohorts, ablations, fair baselines, released code or data.
Novelty	██░░░2/5	16%	A new combination of known ideas.
Trajectory	███░░3/5	18%	A clear path to scale.
Stakes	██░░░2/5	10%	Benefits a professional community (practitioners, clinicians, engineers).

Editor’s rationale

Heuristic triage from title and abstract text only, not a reading of the paper. Cues found: method (we propose, new method); breadth (many tasks); verification (multiple benchmarks, code released); scale (scalable); stakes (general AI).

How the score was computed

rank-2026-10-18

Score████░░░░░░3.9

Score = 10 × (65% × adjusted merit / 10 + 25% × attention + 10% × freshness)

Desks place stronger verdicts first, then use this score to order work within the same verdict.

Merit: 5.6 / 10; Weighted rubric, evidence-gated.
Adjusted merit: 4.7 / 10; Shrunk toward the desk prior by editor confidence (42%).
Attention: 16%; Citations, upvotes, points, mentions.
Freshness: 49%; Half-life decay since publication.

Citations0 (reference 15, via semantic-scholar, Oct 28, 2026, 13:49 UTC)
Influential citations0 (reference 3, via semantic-scholar, Oct 28, 2026, 13:49 UTC)
Hugging Face upvotes5 (reference 25, via hf-daily, Oct 30, 2026, 02:59 UTC)

The record

Reviewed by heuristic-v10 on Oct 28, 2026, 13:49 UTC. Paper type: method.
Categories: cs.AI
BRIEF, No.6 in the Artificial Intelligence edition of October 30, 2026.