In Wyoming, improving the ai-human collaboration in the development of IEP

Key points:

The University of Wyoming is using its new Coiep initiative, a system of multiple agents fed by large language models (LLM), to significantly improve artificial intelligence and human collaboration to address the challenges in education (special).

The first complex challenge that the UW College of Education will use the collaboration of ai-Human to address is the joint and coevalúa creation of individualized education programs (IEP) for students with disabilities in Wyoming and beyond.

Ordered by federal persons of the Education Law (idea) of people with disabilities, IEP are legally binding documents that describe the interconnected components (eg, annual educational objectives and specially designed instruction) to ensure that students with Disabilities have access to the General Education curriculum and high quality learning. experiences. Serving as the cornerstone of special education (Yell et al., 2016), IEP play a fundamental role in adapting educational experiences to meet the unique needs of students with disabilities.

The UW research team has developed a medium fidelity prototype of COIEP, specially designed to optimize the complex IEP development process when the process step by step of the creation of central components of IEPs. Preliminary evaluations suggest that COIEP has the potential to support educators in the creation and evaluation of Central IEP components and provide individualized instruction for students with disabilities (Zhang et al., In review).

The development of an IEP is a complex process that requires significant time and experience to analyze the data of students from various sources, identify the strengths and needs of students to develop objectives, design evidence -based instructional practices and address others Support (for example, modifications, adaptations) that can support students to achieve the objectives and make significant progress in the General Education curriculum. The process can take a long time and discouraging for special education teachers, particularly pre-service teachers and rookies who are recently introduced into the field, as well as experienced educators who can have a burden of up to 50 students at the same time, But it lacks sufficient resources and time (HOGUE AND TAYLOR, 2020).

In addition, many school districts in the United States, including Wyoming, have faced problems related to IEP requirements due to the complex development of the IEP. These compliance problems can negatively affect the workforce of special education teachers, which, in turn, affects access to high quality instruction for students with disabilities. To address these challenges, COIEP takes advantage of a team of agents fed by LLM to support educators in the creation of three interconnected components required of an IEP, including 1) the current level of academic performance and declaration of functional performance, 2) Objectives of IEP and 3) individualized services and supports in a sequential order.

A system of multiple agents is an advanced artificial intelligence system in which multiple agents fueled by LLM collaborate to perform sub-tareas of a complex task (Guo et al., 2024). An advantage of such systems lies in taking advantage of the capacity of multiple agents, instead of a single agent, to complete specific parts of a complex task. In addition, the different agents can be equipped with different tools (for example, data analysis tool, calculation tool, databases for evidence -based instructional practices) that allow them to complete the subtartea more efficiently and precisely. Compared to other general purpose -purpose systems (for example, chatgpt), this specialized and collaborative approach prevents ai agents from deviating from their specific tasks, produce more reliable results and reduce the risk of hallucination (that is, information information inaccurate).

In Coiep, the different agents are designed and fast engineering, techniques to instruct the ai ​​agents that generate the desired exit, to demonstrate the step -by -step creation process and the specific elements of each component of an IEP. In addition, this system admits the Keeping-Humans-in-the-Loop (Hitl) function, allowing educators to provide comments on each component generated by an agent before the task passes to the next agent. These collaborative approaches step by step and ai-Humanos further increase the transparency of the agent's workflow and guarantee the quality of the components of the IEP generated.

The research team plans to carry out a series of pilot studies in Wyoming schools this year to prove the effectiveness, usability and user experience of the COIEP's average fidelity prototype. Prior special and service teachers and directors in Wyoming and other states will be recruited to participate in these studies.

It is important to emphasize that the development of IEP implies collaboration efforts of IEP team members (for example, educators, families, students, school psychologists and school administrators). Therefore, Coiep serves as a collaboration tool that admits the IEP team through the creation and coevaluation of the Core IEP components. Instead of spending valuable time in paperwork, the team can focus on critically reviewing the generated content, participating in the significant resolution of problems and discussing effective strategies to implement the IEP.

The team can further refine the components during the IEP meeting, incorporating data and knowledge of updated students of all members to guarantee an integral, individualized and high quality IEP to better meet the needs of the students. Coiep's highly flexible workflow, where each agent operates independently, makes it a versatile tool that can meet different purposes based on the needs of its end users. For example, COIEP can act as a professional learning tool demonstrating the step -by -step process of creating IEP documents, helping educators to develop their experience. It can also function as an evaluation tool to assess whether an IEP aligns with the requirements of ideas, ensuring compliance and quality.

Coiep will make a significant contribution to research on ai in special education, particularly when providing a promising tool to reduce excessive workloads of special education teachers in Wyoming and improve the quality of IEP for students with disabilities. From an economic perspective, our cost analysis suggested that COIEP could be a profitable professional learning tool compared to traditional professional development or coaching on the site, which are often high cost and intensive in resources (Kraft et al. , 2018).

More importantly, Coiep offers a personalized approach to support educators in the development of high quality IEP, which has great potential to improve the results for SWDs. After being fully developed and implemented, educators can access COIEP at any time, anywhere in Wyoming. Given the important scarcity of teachers in special education and the high associated cost of the teacher's wear (Billingsley and Bettini, 2019), Coiep has the potential to address historical and continuous challenges in special education.