@inproceedings{jayalath-etal-2025-prism,
title = "{PRISM}: Efficient Long-Range Reasoning With Short-Context {LLM}s",
author = "Jayalath, Dulhan and
Wendt, James Bradley and
Monath, Nicholas and
Tata, Sandeep and
Gunel, Beliz",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.emnlp-main.517/",
pages = "10207--10229",
ISBN = "979-8-89176-332-6",
abstract = "Long-range tasks demand reasoning over long inputs. However, existing solutions are limited, e.g., long-context models require large compute budgets, parameter-efficient fine-tuning (PEFT) needs training data, and retrieval-augmented generation (RAG) entails complex task-specific designs. Though in-context approaches overcome many of these issues, methods with short-context LLMs are inefficient, trading context for processing more tokens. We introduce **PRISM**, a highly token-efficient in-context method based on structured schemas that outperforms baselines on diverse tasks with **4x shorter contexts**. This approach produces concise outputs and efficiently leverages key-value (KV) caches to **reduce costs by up to 54{\%}**. PRISM scales down to tiny contexts without increasing costs or sacrificing quality, and generalizes to new tasks with minimal effort by generating schemas from task descriptions."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="jayalath-etal-2025-prism">
<titleInfo>
<title>PRISM: Efficient Long-Range Reasoning With Short-Context LLMs</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dulhan</namePart>
<namePart type="family">Jayalath</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">James</namePart>
<namePart type="given">Bradley</namePart>
<namePart type="family">Wendt</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nicholas</namePart>
<namePart type="family">Monath</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sandeep</namePart>
<namePart type="family">Tata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Beliz</namePart>
<namePart type="family">Gunel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-332-6</identifier>
</relatedItem>
<abstract>Long-range tasks demand reasoning over long inputs. However, existing solutions are limited, e.g., long-context models require large compute budgets, parameter-efficient fine-tuning (PEFT) needs training data, and retrieval-augmented generation (RAG) entails complex task-specific designs. Though in-context approaches overcome many of these issues, methods with short-context LLMs are inefficient, trading context for processing more tokens. We introduce **PRISM**, a highly token-efficient in-context method based on structured schemas that outperforms baselines on diverse tasks with **4x shorter contexts**. This approach produces concise outputs and efficiently leverages key-value (KV) caches to **reduce costs by up to 54%**. PRISM scales down to tiny contexts without increasing costs or sacrificing quality, and generalizes to new tasks with minimal effort by generating schemas from task descriptions.</abstract>
<identifier type="citekey">jayalath-etal-2025-prism</identifier>
<location>
<url>https://aclanthology.org/2025.emnlp-main.517/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>10207</start>
<end>10229</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T PRISM: Efficient Long-Range Reasoning With Short-Context LLMs
%A Jayalath, Dulhan
%A Wendt, James Bradley
%A Monath, Nicholas
%A Tata, Sandeep
%A Gunel, Beliz
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-332-6
%F jayalath-etal-2025-prism
%X Long-range tasks demand reasoning over long inputs. However, existing solutions are limited, e.g., long-context models require large compute budgets, parameter-efficient fine-tuning (PEFT) needs training data, and retrieval-augmented generation (RAG) entails complex task-specific designs. Though in-context approaches overcome many of these issues, methods with short-context LLMs are inefficient, trading context for processing more tokens. We introduce **PRISM**, a highly token-efficient in-context method based on structured schemas that outperforms baselines on diverse tasks with **4x shorter contexts**. This approach produces concise outputs and efficiently leverages key-value (KV) caches to **reduce costs by up to 54%**. PRISM scales down to tiny contexts without increasing costs or sacrificing quality, and generalizes to new tasks with minimal effort by generating schemas from task descriptions.
%U https://aclanthology.org/2025.emnlp-main.517/
%P 10207-10229
Markdown (Informal)
[PRISM: Efficient Long-Range Reasoning With Short-Context LLMs](https://aclanthology.org/2025.emnlp-main.517/) (Jayalath et al., EMNLP 2025)
ACL
- Dulhan Jayalath, James Bradley Wendt, Nicholas Monath, Sandeep Tata, and Beliz Gunel. 2025. PRISM: Efficient Long-Range Reasoning With Short-Context LLMs. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 10207–10229, Suzhou, China. Association for Computational Linguistics.