FrugalPrompt: Reducing Contextual Overhead in Large Language Models via Token Attribution

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Raiyan, Syed Rifat, Ishmam, Md Farhan, Imran, Abdullah Al, Moni, Mohammad Ali
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915855567159296
author Raiyan, Syed Rifat
Ishmam, Md Farhan
Imran, Abdullah Al
Moni, Mohammad Ali
author_facet Raiyan, Syed Rifat
Ishmam, Md Farhan
Imran, Abdullah Al
Moni, Mohammad Ali
contents Human communication heavily relies on laconism and inferential pragmatics, allowing listeners to successfully reconstruct rich meaning from sparse, telegraphic speech. In contrast, large language models (LLMs) owe much of their stellar performance to expansive input contexts, yet such verbosity inflates monetary costs, carbon footprint, and inference-time latency. This overhead manifests from the redundant low-utility tokens present in typical prompts, as only a fraction of tokens typically carries the majority of the semantic weight. Inspired by the aforementioned cognitive psycholinguistic processes, we address this inefficiency by introducing FrugalPrompt, a novel prompt compression framework for LLMs, which retains only the most semantically significant tokens. Leveraging two state-of-the-art token attribution methods, GlobEnc and DecompX, we assign salience scores to every token in an input sequence, rank them to retain the top-k% tokens, and obtain a sparse frugalized prompt. We establish the theoretical stability of our approach and provide strong empirical results across a suite of four NLP tasks to study the trade-off between the portion of retained tokens and performance. Experimental findings across retention settings reveal asymmetric performance patterns that suggest potential task contamination effects. We posit that our work contributes to a more nuanced understanding of LLM behavior in performance-efficiency trade-offs and delineates the boundary between tasks tolerant of contextual sparsity and those requiring exhaustive context.
format Preprint
id arxiv_https___arxiv_org_abs_2510_16439
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FrugalPrompt: Reducing Contextual Overhead in Large Language Models via Token Attribution
Raiyan, Syed Rifat
Ishmam, Md Farhan
Imran, Abdullah Al
Moni, Mohammad Ali
Computation and Language
Human communication heavily relies on laconism and inferential pragmatics, allowing listeners to successfully reconstruct rich meaning from sparse, telegraphic speech. In contrast, large language models (LLMs) owe much of their stellar performance to expansive input contexts, yet such verbosity inflates monetary costs, carbon footprint, and inference-time latency. This overhead manifests from the redundant low-utility tokens present in typical prompts, as only a fraction of tokens typically carries the majority of the semantic weight. Inspired by the aforementioned cognitive psycholinguistic processes, we address this inefficiency by introducing FrugalPrompt, a novel prompt compression framework for LLMs, which retains only the most semantically significant tokens. Leveraging two state-of-the-art token attribution methods, GlobEnc and DecompX, we assign salience scores to every token in an input sequence, rank them to retain the top-k% tokens, and obtain a sparse frugalized prompt. We establish the theoretical stability of our approach and provide strong empirical results across a suite of four NLP tasks to study the trade-off between the portion of retained tokens and performance. Experimental findings across retention settings reveal asymmetric performance patterns that suggest potential task contamination effects. We posit that our work contributes to a more nuanced understanding of LLM behavior in performance-efficiency trade-offs and delineates the boundary between tasks tolerant of contextual sparsity and those requiring exhaustive context.
title FrugalPrompt: Reducing Contextual Overhead in Large Language Models via Token Attribution
topic Computation and Language
url https://arxiv.org/abs/2510.16439