English
Related papers

Related papers: DENIAHL: In-Context Features Influence LLM Needle-…

200 papers

Recent large language models (LLMs) have demonstrated versatile capabilities in long-context scenarios. Although some recent benchmarks have been developed to evaluate the long-context capabilities of LLMs, there is a lack of benchmarks…

Computation and Language · Computer Science 2024-10-08 Lei Wang , Shan Dong , Yuhui Xu , Hanze Dong , Yalu Wang , Amrita Saha , Ee-Peng Lim , Caiming Xiong , Doyen Sahoo

Video understanding is a crucial next step for multimodal large language models (MLLMs). Various benchmarks are introduced for better evaluating the MLLMs. Nevertheless, current video benchmarks are still inefficient for evaluating video…

Computer Vision and Pattern Recognition · Computer Science 2025-03-10 Zijia Zhao , Haoyu Lu , Yuqi Huo , Yifan Du , Tongtian Yue , Longteng Guo , Bingning Wang , Weipeng Chen , Jing Liu

Recent studies have shown that Large Language Models (LLMs) struggle to accurately retrieve information and maintain reasoning capabilities when processing long-context inputs. To address these limitations, we propose a finetuning approach…

Machine Learning · Computer Science 2024-10-15 Zheyang Xiong , Vasilis Papageorgiou , Kangwook Lee , Dimitris Papailiopoulos

Recent advances in long-context language models (LMs) have enabled million-token inputs, expanding their capabilities across complex tasks like computer-use agents. Yet, the safety implications of these extended contexts remain unclear. To…

Cryptography and Security · Computer Science 2025-11-10 Rishi Rajesh Shah , Chen Henry Wu , Shashwat Saxena , Ziqian Zhong , Alexander Robey , Aditi Raghunathan

As language models support larger and larger context sizes, evaluating their ability to make effective use of that context becomes increasingly important. We analyze the ability of several code generation models to handle long range…

Computation and Language · Computer Science 2025-06-26 Yannick Assogba , Donghao Ren

Video sequences offer valuable temporal information, but existing large multimodal models (LMMs) fall short in understanding extremely long videos. Many works address this by reducing the number of visual tokens using visual resamplers.…

Computer Vision and Pattern Recognition · Computer Science 2024-07-02 Peiyuan Zhang , Kaichen Zhang , Bo Li , Guangtao Zeng , Jingkang Yang , Yuanhan Zhang , Ziyue Wang , Haoran Tan , Chunyuan Li , Ziwei Liu

Many benchmarks exist for evaluating long-context language models (LCLMs), yet developers often rely on synthetic tasks such as needle-in-a-haystack (NIAH) or an arbitrary subset of tasks. However, it remains unclear whether these…

Computation and Language · Computer Science 2025-03-07 Howard Yen , Tianyu Gao , Minmin Hou , Ke Ding , Daniel Fleischer , Peter Izsak , Moshe Wasserblat , Danqi Chen

LLMs and RAG systems are now capable of handling millions of input tokens or more. However, evaluating the output quality of such systems on long-context tasks remains challenging, as tasks like Needle-in-a-Haystack lack complexity. In this…

Computation and Language · Computer Science 2024-07-02 Philippe Laban , Alexander R. Fabbri , Caiming Xiong , Chien-Sheng Wu

Current benchmarks like Needle-in-a-Haystack (NIAH), Ruler, and Needlebench focus on models' ability to understand long-context input sequences but fail to capture a critical dimension: the generation of high-quality long-form text.…

Computation and Language · Computer Science 2025-01-24 Yuhao Wu , Ming Shan Hee , Zhiqing Hu , Roy Ka-Wei Lee

As the context limits of Large Language Models (LLMs) increase, the range of possible applications and downstream functions broadens. In many real-world tasks, decisions depend on details scattered across collections of often disparate…

Computation and Language · Computer Science 2025-04-24 Jonathan Roberts , Kai Han , Samuel Albanie

Existing multilingual long-context benchmarks, often based on the popular needle-in-a-haystack test, primarily evaluate a model's ability to locate specific information buried within irrelevant texts. However, such a retrieval-centric…

Computation and Language · Computer Science 2025-04-18 Amey Hengle , Prasoon Bajpai , Soham Dan , Tanmoy Chakraborty

The ability to recognize patterns from examples and apply them to new ones is a primal ability for general intelligence, and is widely studied by psychology and AI researchers. Many benchmarks have been proposed to measure such ability for…

Artificial Intelligence · Computer Science 2025-10-24 Kai Yan , Zhan Ling , Kang Liu , Yifan Yang , Ting-Han Fan , Lingfeng Shen , Zhengyin Du , Jiecao Chen

Large language models (LLMs) have demonstrated strong capabilities in using external tools to address user inquiries. However, most existing evaluations assume tool use in short contexts, offering limited insight into model behavior during…

Computation and Language · Computer Science 2025-11-24 Beong-woo Kwak , Minju Kim , Dongha Lim , Hyungjoo Chae , Dongjin Kang , Sunghwan Kim , Dongil Yang , Jinyoung Yeo

Recent advances have been improving the context windows of Large Language Models (LLMs). To quantify the real long-context capabilities of LLMs, evaluators such as the popular Needle in a Haystack have been developed to test LLMs over a…

Software Engineering · Computer Science 2024-06-11 Jiawei Liu , Jia Le Tian , Vijay Daita , Yuxiang Wei , Yifeng Ding , Yuhan Katherine Wang , Jun Yang , Lingming Zhang

Recently proposed evaluation benchmarks aim to characterize the effective context length and the forgetting tendencies of large language models (LLMs). However, these benchmarks often rely on simplistic 'needle in a haystack' retrieval or…

Computation and Language · Computer Science 2025-10-07 Raquib Bin Yousuf , Aadyant Khatri , Shengzhe Xu , Mandar Sharma , Naren Ramakrishnan

Recent research in Needle-in-a-Haystack (NIAH) benchmarks has explored the capabilities of Large Language Models (LLMs) in retrieving contextual information from large text documents. However, as LLMs become increasingly integrated into…

Artificial Intelligence · Computer Science 2024-06-24 Hokyung Lee , Sumanyu Sharma , Bing Hu

The ability of large language models (LLMs) to recall and retrieve information from long contexts is critical for many real-world applications. Prior work (Liu et al., 2023) reported that LLMs suffer significant drops in retrieval accuracy…

Information Retrieval · Computer Science 2025-11-11 Max McKinnon

The capability of large language models to handle long-context information is crucial across various real-world applications. Existing evaluation methods often rely either on real-world long texts, making it difficult to exclude the…

Computation and Language · Computer Science 2025-09-18 Mo Li , Songyang Zhang , Taolin Zhang , Haodong Duan , Yunxin Liu , Kai Chen

Large language models (LLMs) increasingly support very long input contexts. Yet it remains unclear how reliably they extract and infer information at scale. Performance varies with context length and strongly interacts with how information…

Computation and Language · Computer Science 2026-01-06 Amirali Ebrahimzadeh , Seyyed M. Salili

Existing frameworks for evaluating long-context language models (LCLM) can be broadly categorized into real-world applications (e.g, document summarization) and synthetic tasks (e.g, needle-in-a-haystack). Despite their utility, both…

Computation and Language · Computer Science 2025-10-21 Yijun Yang , Zeyu Huang , Wenhao Zhu , Zihan Qiu , Fei Yuan , Jeff Z. Pan , Ivan Titov