DiSC-AMC: Token- and Parameter-Efficient Discretized Statistics In-Context Automatic Modulation Classification

2025-09-30

AI Summary

Large language models (LLMs) face significant deployment challenges in open-set automatic modulation classification (AMC) due to excessive context length and parameter count. Method: We propose a lightweight in-context learning framework that (i) discretizes higher-order statistics and cumulants into symbolic tokens; (ii) integrates a k-top neural pre-filtering mechanism coupled with LLM feedback-driven exemplar selection; and (iii) employs a calibration-aware compact prompt template enforcing structured label output. Contribution/Results: Our approach drastically reduces contextual and model-scale requirements: on a ten-class AMC benchmark, a ~5B-parameter LLM achieves 45.5% accuracy—over 8× higher than a 7B baseline—while halving both input/output token count and model parameters. Inference cost is reduced by more than 2×, enabling the first practical, efficient deployment of small-scale LLMs for AMC.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Large Language Models (LLMs) can perform Automatic Modulation Classification (AMC) in an open-set manner without LLM fine-tuning when equipped with carefully designed in-context prompts~cite{rostami2025plug}. Building on this prior work, we target the practical bottlenecks of long prompt contexts and large model sizes that impede in-the-loop deployment. We present Discretized Statistics in-Context Automatic Modulation Classification (DiSC-AMC), a token- and parameter-efficient variant that: (i) discretizes higher-order statistics and cumulants into compact symbolic tokens, (ii) prunes the exemplar list via a lightweight k-top neural prefilter and filters misleading/low-impact features using rationales extracted from prior LLM responses, and (iii) enforces label-only predictions through a calibrated prompt template. Together, these changes reduce both input/output tokens and the model parameter footprint by more than half while maintaining competitive accuracy. On synthetic AMC with ten modulation types under noise, a 7B extit{DeepSeek-R1-Distill-Qwen} baseline achieves 5.2% accuracy, whereas our system, using an approximately 5B-parameter extit{Gemini-2.5-Flash}~cite{comanici2025gemini} model, attains 45.5% accuracy. These results demonstrate that careful discretization and context selection can cut inference cost by over 2x while preserving the advantages of prompt-based AMC and enabling practical in-the-loop use.

Problem

• Reduces long prompt contexts and large model sizes for practical deployment
• Improves token and parameter efficiency in automatic modulation classification
• Maintains competitive accuracy while cutting inference costs significantly

Innovation

• Discretizes statistics into compact symbolic tokens
• Prunes exemplars via lightweight neural prefilter
• Enforces label-only predictions with calibrated prompt template