DataEnvGym: Data Generation Agents in Teacher Environments with Student Feedback
Zaid Khan, Elias Stengel-Eskin, Jaemin Cho, Mohit Bansal

TL;DR
DataEnvGym is a new testbed framework that simulates feedback-driven data generation environments, enabling the development and testing of autonomous data creation agents to improve machine learning models across various domains.
Contribution
Introduces DataEnvGym, a versatile testbed for training and evaluating autonomous data generation agents in feedback-driven, structured environments across multiple domains.
Findings
Agents can iteratively improve student performance.
Environments support multiple domains and skill levels.
Structured environments enhance interpretability and control.
Abstract
The process of creating training data to teach models is currently driven by humans, who manually analyze model weaknesses and plan how to create data that improves a student model. Approaches using LLMs as annotators reduce human effort, but still require humans to interpret feedback from evaluations and control the LLM to produce data the student needs. Automating this labor-intensive process by creating autonomous data generation agents - or teachers - is desirable, but requires environments that can simulate the feedback-driven, iterative, closed loop of data creation. To enable rapid, scalable testing for such agents and their modules, we introduce DataEnvGym, a testbed of teacher environments for data generation agents. DataEnvGym frames data generation as a sequential decision-making task, involving an agent consisting of a data generation policy (which generates a plan for…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
Taxonomy
TopicsIntelligent Tutoring Systems and Adaptive Learning · Online Learning and Analytics
