@inproceedings{yu-etal-2025-sequential,
title = "Sequential-{NIAH}: A Needle-In-A-Haystack Benchmark for Extracting Sequential Needles from Long Contexts",
author = "Yu, Yifei and
Zhang, Qian-Wen and
Qiao, Lingfeng and
Yin, Di and
Li, Fang and
Wang, Jie and
Xi, Chen Zeng and
Zheng, Suncong and
Liang, Xiaolong and
Sun, Xing",
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.1497/",
doi = "10.18653/v1/2025.emnlp-main.1497",
pages = "29450--29468",
ISBN = "979-8-89176-332-6",
abstract = "Evaluating the ability of large language models (LLMs) to process lengthy contexts is critical, especially for retrieving query-relevant information embedded within them. We introduce Sequential-NIAH, a benchmark specifically designed to evaluate the capability of LLMs to extract sequential information items (known as \textit{needles}) from long contexts. The benchmark includes three needle generation pipelines: synthetic-temporal, real-temporal, and real-logical orders, with context lengths ranging from 8K to 128K, which comprises 14,000 samples (2,000 for testing). To facilitate the evaluation of this benchmark, we trained an evaluation model that assesses the correctness of LLM responses by comparing their completeness and sequential consistency against the ground truth, which provides a more reliable evaluation metric than GPT-4 or Claude. We conducted experiments on six well-known LLMs, revealing that even the best-performing model achieved a maximum accuracy of only 63.50{\%} on test set of this benchmark. Further analysis highlights the growing challenges posed by increasing the context length or the number of needles, underscoring substantial room for improvement of LLMs. Additionally, noise analysis validates the reliability and challenge of the benchmark, making Sequential-NIAH an important reference for advancing research on long text information extraction capabilities of LLMs."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yu-etal-2025-sequential">
<titleInfo>
<title>Sequential-NIAH: A Needle-In-A-Haystack Benchmark for Extracting Sequential Needles from Long Contexts</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yifei</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qian-Wen</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lingfeng</namePart>
<namePart type="family">Qiao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Di</namePart>
<namePart type="family">Yin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fang</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jie</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chen</namePart>
<namePart type="given">Zeng</namePart>
<namePart type="family">Xi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Suncong</namePart>
<namePart type="family">Zheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaolong</namePart>
<namePart type="family">Liang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xing</namePart>
<namePart type="family">Sun</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>Evaluating the ability of large language models (LLMs) to process lengthy contexts is critical, especially for retrieving query-relevant information embedded within them. We introduce Sequential-NIAH, a benchmark specifically designed to evaluate the capability of LLMs to extract sequential information items (known as needles) from long contexts. The benchmark includes three needle generation pipelines: synthetic-temporal, real-temporal, and real-logical orders, with context lengths ranging from 8K to 128K, which comprises 14,000 samples (2,000 for testing). To facilitate the evaluation of this benchmark, we trained an evaluation model that assesses the correctness of LLM responses by comparing their completeness and sequential consistency against the ground truth, which provides a more reliable evaluation metric than GPT-4 or Claude. We conducted experiments on six well-known LLMs, revealing that even the best-performing model achieved a maximum accuracy of only 63.50% on test set of this benchmark. Further analysis highlights the growing challenges posed by increasing the context length or the number of needles, underscoring substantial room for improvement of LLMs. Additionally, noise analysis validates the reliability and challenge of the benchmark, making Sequential-NIAH an important reference for advancing research on long text information extraction capabilities of LLMs.</abstract>
<identifier type="citekey">yu-etal-2025-sequential</identifier>
<identifier type="doi">10.18653/v1/2025.emnlp-main.1497</identifier>
<location>
<url>https://aclanthology.org/2025.emnlp-main.1497/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>29450</start>
<end>29468</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Sequential-NIAH: A Needle-In-A-Haystack Benchmark for Extracting Sequential Needles from Long Contexts
%A Yu, Yifei
%A Zhang, Qian-Wen
%A Qiao, Lingfeng
%A Yin, Di
%A Li, Fang
%A Wang, Jie
%A Xi, Chen Zeng
%A Zheng, Suncong
%A Liang, Xiaolong
%A Sun, Xing
%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 yu-etal-2025-sequential
%X Evaluating the ability of large language models (LLMs) to process lengthy contexts is critical, especially for retrieving query-relevant information embedded within them. We introduce Sequential-NIAH, a benchmark specifically designed to evaluate the capability of LLMs to extract sequential information items (known as needles) from long contexts. The benchmark includes three needle generation pipelines: synthetic-temporal, real-temporal, and real-logical orders, with context lengths ranging from 8K to 128K, which comprises 14,000 samples (2,000 for testing). To facilitate the evaluation of this benchmark, we trained an evaluation model that assesses the correctness of LLM responses by comparing their completeness and sequential consistency against the ground truth, which provides a more reliable evaluation metric than GPT-4 or Claude. We conducted experiments on six well-known LLMs, revealing that even the best-performing model achieved a maximum accuracy of only 63.50% on test set of this benchmark. Further analysis highlights the growing challenges posed by increasing the context length or the number of needles, underscoring substantial room for improvement of LLMs. Additionally, noise analysis validates the reliability and challenge of the benchmark, making Sequential-NIAH an important reference for advancing research on long text information extraction capabilities of LLMs.
%R 10.18653/v1/2025.emnlp-main.1497
%U https://aclanthology.org/2025.emnlp-main.1497/
%U https://doi.org/10.18653/v1/2025.emnlp-main.1497
%P 29450-29468
Markdown (Informal)
[Sequential-NIAH: A Needle-In-A-Haystack Benchmark for Extracting Sequential Needles from Long Contexts](https://aclanthology.org/2025.emnlp-main.1497/) (Yu et al., EMNLP 2025)
ACL
- Yifei Yu, Qian-Wen Zhang, Lingfeng Qiao, Di Yin, Fang Li, Jie Wang, Chen Zeng Xi, Suncong Zheng, Xiaolong Liang, and Xing Sun. 2025. Sequential-NIAH: A Needle-In-A-Haystack Benchmark for Extracting Sequential Needles from Long Contexts. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 29450–29468, Suzhou, China. Association for Computational Linguistics.