NeedleChain: Measuring Intact Context Comprehension Capability of Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Moon, Hyeonseok, Lim, Heuiseok
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908742622117888
author Moon, Hyeonseok
Lim, Heuiseok
author_facet Moon, Hyeonseok
Lim, Heuiseok
contents Recent reports suggest that LLMs can handle increasingly long contexts. However, many existing benchmarks for context understanding embed substantial query-irrelevant content, which shifts evaluation toward retrieving relevant snippets rather than fully integrating all provided information. Under this setting, we view that current benchmarks can overestimate true context-understanding ability of LLMs. In particular, we demonstrate that when the context consists entirely of query-relevant text, even advanced models such as GPT-4o fail to reliably integrate inputs as short as 200 tokens. To evaluate this capability more rigorously, we introduce NeedleChain, a benchmark designed to test whether models can faithfully incorporate all given evidence. NeedleChain includes three variants that differ in the required order of comprehension, along with a parallel benchmark based on the needle-in-a-haystack(NIAH) paradigm. By comparing these variants, NeedleChain enables a more comprehensive assessment of context understanding. We further propose a training-free strategy that encourages models to reflect all available information, ROPE contraction, highlighting the importance of full-context integration and pointing to new directions for improving reliable reasoning over context.
format Preprint
id arxiv_https___arxiv_org_abs_2507_22411
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle NeedleChain: Measuring Intact Context Comprehension Capability of Large Language Models
Moon, Hyeonseok
Lim, Heuiseok
Computation and Language
Artificial Intelligence
Recent reports suggest that LLMs can handle increasingly long contexts. However, many existing benchmarks for context understanding embed substantial query-irrelevant content, which shifts evaluation toward retrieving relevant snippets rather than fully integrating all provided information. Under this setting, we view that current benchmarks can overestimate true context-understanding ability of LLMs. In particular, we demonstrate that when the context consists entirely of query-relevant text, even advanced models such as GPT-4o fail to reliably integrate inputs as short as 200 tokens. To evaluate this capability more rigorously, we introduce NeedleChain, a benchmark designed to test whether models can faithfully incorporate all given evidence. NeedleChain includes three variants that differ in the required order of comprehension, along with a parallel benchmark based on the needle-in-a-haystack(NIAH) paradigm. By comparing these variants, NeedleChain enables a more comprehensive assessment of context understanding. We further propose a training-free strategy that encourages models to reflect all available information, ROPE contraction, highlighting the importance of full-context integration and pointing to new directions for improving reliable reasoning over context.
title NeedleChain: Measuring Intact Context Comprehension Capability of Large Language Models
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2507.22411