Beyond Instruction Optimization: Multi-Agent Error-Driven Class Description Refinement for LLM-Based Classification

Hamvir Dev, Shivam Ratnakant Mhaskar, Sasanka Vutla, Anup Pattnaik

Keywords: Prompt Optimization, Multi Agent System, LLM Classification, Error Driven Class Description Refinement, Contact Center, Agentic AI

TL;DR: LLM classification accuracy hinges not just on how you prompt but on how you define your classes — this paper introduces a multi-agent framework that iteratively refines class descriptions from misclassification errors, yielding up to 20.71% gains.

Abstract:

Large Language Models (LLMs) have demonstrated considerable efficacy in
classification tasks, yet their performance depends on two critical prompt components: Task Instructions (HOW to classify) and Class Descriptions (WHAT defines each class). While prompt engineering research has extensively explored instruction optimization, class descriptions have received comparatively less attention, often being treated as fixed inputs or simple label names. This represents a critical gap for real-world classification tasks, particularly in contact center domains, where labels often suffer from ambiguous boundaries, overlapping definitions, and incomplete coverage of possible cases—substantially limiting accuracy regardless of instruction
quality.

We propose a multi-agent framework for iteratively refining class descriptions based on classification errors. By analyzing misclassified
instances, language agents automatically generate improved descriptions that better capture class distinctions and resolve ambiguities. Empirical evaluation across contact center and public benchmark datasets demonstrates upto 20.71% accuracy improvements over static class descriptions, addressing an orthogonal dimension to existing instruction optimization techniques.

Submission Type: Deployed

Submission Number: 476