Babbling Suppression: Making LLMs Greener One Token at a Time

TL;DR

This paper introduces Babbling Suppression, a method that reduces unnecessary token generation in large language models during code generation by integrating test execution, leading to significant energy savings without sacrificing accuracy.

Contribution

It proposes a practical, model-agnostic approach to minimize babbling in LLMs by terminating generation after passing tests, improving efficiency and sustainability.

Findings

Babbling occurs across all tested models, more in Java than Python.

Babbling Suppression reduces energy consumption by up to 65%.

Token generation decreases in most cases, with minimal overhead.

Abstract

Context: Large Language Models (LLMs) are increasingly used in modern software development, aiding in code generation, code completion, and refactoring through AI-powered assistants. While they accelerate development workflows, they often produce extraneous output, referred to as "babbling", which incurs additional cognitive, economic, and energy costs. Objective: This work investigates the problem of babbling in LLM-based code generation and proposes a practical, model-agnostic approach to reduce unnecessary output without compromising solution accuracy. Method: We introduce Babbling Suppression (BS), a method that integrates test execution into the LLM generation process by evaluating intermediate outputs and terminating generation once a solution passes all tests. This prevents excessive token generation while having no impact on model accuracy. An empirical study was conducted across two Python and two Java benchmarks, targeting four 3-4B parameter models and six 6-7B parameter models. Results: Our findings show that babbling occurs across all tested models, with higher frequency in Java than in Python. Applying BS significantly reduces energy consumption by up to 65% for Python and 62% for Java in models prone to babbling. Across 40 model-benchmark pairs, 29 showed reduced mean energy consumption, with reductions exceeding 20% in 22 cases. Generated token count decreased in 35 pairs, while the GPU energy-per-token overhead of BS remained below 10% for 26 pairs, decreased for 2, and reached a maximum of 24%, yielding net energy savings in most cases. Implications: BS can make AI-assisted programming more efficient and sustainable by reducing energy consumption and minimizing cognitive effort by developers. Its model-agnostic design allows easy integration, suggesting broad applicability.