Long-context Language Models Fail in Basic Retrieval Tasks Without Sufficient Reasoning Steps

• URL: http://arxiv.org/abs/2410.04422v9
• Date: Tue, 26 Aug 2025 08:55:10 GMT
• Title: Long-context Language Models Fail in Basic Retrieval Tasks Without Sufficient Reasoning Steps
• Authors: Yijiong Yu, Yongfeng Huang, Zhixiao Qi, Wei Wang, Weifeng Liu, Ran Chen, Ji Pei,
• Abstract summary: Long-context language models (LCLMs) characterized by their extensive context window are becoming popular.<n>Despite the fact that they are nearly perfect at standard long-context retrieval tasks, our evaluations demonstrate they fail in some basic cases.<n>We find they can be well addressed with a sufficient number of reasoning steps, guided by specific CoT prompts.
• Score: 21.811552521030137
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Long-context language models (LCLMs), characterized by their extensive context window, are becoming popular. However, despite the fact that they are nearly perfect at standard long-context retrieval tasks, our evaluations demonstrate they fail in some basic cases. Later, we find they can be well addressed with a sufficient number of reasoning steps, guided by specific CoT prompts. This result emphasizes the potential necessity of solving specific long-context tasks using long-CoT methods, while previous long-context benchmarks always ignore the necessity of long reasoning for long-context tasks and treat them as direct QA tasks.

Related papers

• Ref-Long: Benchmarking the Long-context Referencing Capability of Long-context Language Models [36.69535336525585]
  Long-context language models (LCLMs) have exhibited impressive capabilities in long-context understanding tasks.<n>Long-context referencing is a crucial task that requires LCLMs to attribute items of interest to specific parts of long-context data.<n>This paper proposes Ref-Long, a novel benchmark designed to assess the long-context referencing capability of LCLMs.
  arXiv  Detail & Related papers  (2025-07-13T06:17:53Z)
• 100-LongBench: Are de facto Long-Context Benchmarks Literally Evaluating Long-Context Ability? [28.694112253150983]
  Real-task-based long-context evaluation benchmarks have two major shortcomings.<n> benchmarks like LongBench often do not provide proper metrics to separate long-context performance from the model's baseline ability.<n>We introduce a length-controllable long-context benchmark and a novel metric that disentangles baseline knowledge from true long-context capabilities.
  arXiv  Detail & Related papers  (2025-05-25T19:58:31Z)
• LiveLongBench: Tackling Long-Context Understanding for Spoken Texts from Live Streams [4.917265821383127]
  We construct the first spoken long-text dataset, derived from live streams, to reflect the redundancy-rich and conversational nature of real-world scenarios. We evaluate both popular LLMs and specialized methods to assess their ability to understand long-contexts in these tasks. Our findings highlight key limitations of current methods and suggest future directions for improving long-context understanding.
  arXiv  Detail & Related papers  (2025-04-24T08:27:48Z)
• Chain-of-Thought Matters: Improving Long-Context Language Models with Reasoning Path Supervision [40.63870977649693]
  Chain-of-Thought prompting has shown promise for multi-step reasoning, but its effectiveness for long-context scenarios remains underexplored.<n>We propose LongRePS, a process-supervised framework that teaches models to generate high-quality reasoning paths for enhanced long-context performance.<n>Our framework incorporates a self-sampling mechanism to bootstrap reasoning paths and a novel quality assessment protocol specifically designed for long-context scenarios.
  arXiv  Detail & Related papers  (2025-02-28T07:15:12Z)
• WildLong: Synthesizing Realistic Long-Context Instruction Data at Scale [86.25450054683172]
  WildLong extracts meta-information from real user queries to produce scalable data. It supports multi-document reasoning, such as cross-document comparison and aggregation. It surpasses existing open-source long-context-optimized models across benchmarks.
  arXiv  Detail & Related papers  (2025-02-23T18:59:09Z)
• Generalizing From Short to Long: Effective Data Synthesis for Long-Context Instruction Tuning [103.65680870130839]
  We investigate how to design instruction data for the post-training phase of a long context pre-trained model. Our controlled study reveals that models instruction-tuned on short contexts can effectively generalize to longer ones. Based on these findings, we propose context synthesis, a novel data synthesis framework.
  arXiv  Detail & Related papers  (2025-02-21T17:02:40Z)
• Does RAG Really Perform Bad For Long-Context Processing? [15.889864680212147]
  RetroLM is a novel framework for long-context processing. Unlike traditional methods, RetroLM employs KV-level retrieval augmentation. Building on this framework, we develop a specialized retriever for precise retrieval of critical pages.
  arXiv  Detail & Related papers  (2025-02-17T05:02:25Z)
• MM-Embed: Universal Multimodal Retrieval with Multimodal LLMs [78.5013630951288]
  This paper introduces techniques for advancing information retrieval with multimodal large language models (MLLMs) We first study fine-tuning an MLLM as a bi-encoder retriever on 10 datasets with 16 retrieval tasks. We propose modality-aware hard negative mining to mitigate the modality bias exhibited by MLLM retrievers.
  arXiv  Detail & Related papers  (2024-11-04T20:06:34Z)
• What is Wrong with Perplexity for Long-context Language Modeling? [71.34933096461124]
  Long-context inputs are crucial for large language models (LLMs) in tasks such as extended conversations, document summarization, and many-shot in-context learning. Perplexity (PPL) has proven unreliable for assessing long-context capabilities. We propose bfLongPPL, a novel metric that focuses on key tokens by employing a long-short context contrastive method to identify them.
  arXiv  Detail & Related papers  (2024-10-31T09:39:28Z)
• FACT: Examining the Effectiveness of Iterative Context Rewriting for Multi-fact Retrieval [20.217386507637475]
  Large Language Models (LLMs) are proficient at retrieving single facts from extended contexts, yet struggle with tasks requiring the simultaneous retrieval of multiple facts. This paper identifies a novel "lost-in-the-middle" phenomenon, where LLMs progressively lose track of critical information throughout the generation process. We introduce Find All Crucial Texts (FACT), an iterative retrieval method that refines context through successive rounds of rewriting.
  arXiv  Detail & Related papers  (2024-10-28T13:36:41Z)
• FltLM: An Intergrated Long-Context Large Language Model for Effective Context Filtering and Understanding [32.197113821638936]
  We propose a novel integrated Long-Context Large Language Model (FltLM) FltLM incorporates a context filter with a soft mask mechanism, identifying and dynamically excluding irrelevant content to concentrate on pertinent information. Experimental results demonstrate that FltLM significantly outperforms supervised fine-tuning and retrieval-based methods in complex QA scenarios.
  arXiv  Detail & Related papers  (2024-10-09T13:47:50Z)
• ALR$^2$: A Retrieve-then-Reason Framework for Long-context Question Answering [42.146660039671076]
  We develop a retrieve-then-reason framework for large language models (LLMs) We find that modern LLMs struggle to accurately retrieve relevant facts and instead, often hallucinate "retrieved facts" We introduce ALR$2$, a method that augments the long-context reasoning capability of LLMs via an explicit two-stage procedure.
  arXiv  Detail & Related papers  (2024-10-04T08:29:12Z)
• A Controlled Study on Long Context Extension and Generalization in LLMs [85.4758128256142]
  Broad textual understanding and in-context learning require language models that utilize full document contexts. Due to the implementation challenges associated with directly training long-context models, many methods have been proposed for extending models to handle long contexts. We implement a controlled protocol for extension methods with a standardized evaluation, utilizing consistent base models and extension data.
  arXiv  Detail & Related papers  (2024-09-18T17:53:17Z)
• DetectiveQA: Evaluating Long-Context Reasoning on Detective Novels [86.93099925711388]
  We propose textbfDetectiveQA, a dataset specifically designed for narrative reasoning within long contexts.<n>We leverage detective novels, averaging over 100k tokens, to create a dataset containing 1200 human-annotated questions in both Chinese and English.
  arXiv  Detail & Related papers  (2024-09-04T06:28:22Z)
• Stress-Testing Long-Context Language Models with Lifelong ICL and Task Haystack [33.178008350124315]
  We introduce Lifelong ICL, a problem setting that challenges long-context language models (LMs) to learn a sequence of language tasks through in-context learning (ICL)<n>We introduce Task Haystack, an evaluation suite dedicated to assessing and diagnosing how long-context LMs utilize contexts in Lifelong ICL.
  arXiv  Detail & Related papers  (2024-07-23T17:57:41Z)
• NeedleBench: Can LLMs Do Retrieval and Reasoning in 1 Million Context Window? [37.64593022203498]
  NeedleBench is a framework consisting of progressively more challenging tasks for assessing bilingual long-context capabilities. We use the framework to assess how well the leading open-source models can identify key information relevant to the question. We propose the Ancestral Trace Challenge to mimic the complexity of logical reasoning challenges that are likely to be present in real-world long-context tasks.
  arXiv  Detail & Related papers  (2024-07-16T17:59:06Z)
• KV Cache Compression, But What Must We Give in Return? A Comprehensive Benchmark of Long Context Capable Approaches [52.02764371205856]
  Long context capability is a crucial competency for large language models (LLMs) This work provides a taxonomy of current methods and evaluating 10+ state-of-the-art approaches across seven categories of long context tasks.
  arXiv  Detail & Related papers  (2024-07-01T17:59:47Z)
• Leave No Document Behind: Benchmarking Long-Context LLMs with Extended Multi-Doc QA [71.04146366608904]
  Long-context modeling capabilities have garnered widespread attention, leading to the emergence of Large Language Models (LLMs) with ultra-context windows. We propose a novel long-context benchmark, Loong, aligning with realistic scenarios through extended multi-document question answering (QA) Loong introduces four types of tasks with a range of context lengths: Spotlight Locating, Comparison, Clustering, and Chain of Reasoning.
  arXiv  Detail & Related papers  (2024-06-25T09:42:56Z)
• Can Long-Context Language Models Subsume Retrieval, RAG, SQL, and More? [54.667202878390526]
  Long-context language models (LCLMs) have the potential to revolutionize our approach to tasks traditionally reliant on external tools like retrieval systems or databases. We introduce LOFT, a benchmark of real-world tasks requiring context up to millions of tokens designed to evaluate LCLMs' performance on in-context retrieval and reasoning. Our findings reveal LCLMs' surprising ability to rival state-of-the-art retrieval and RAG systems, despite never having been explicitly trained for these tasks.
  arXiv  Detail & Related papers  (2024-06-19T00:28:58Z)
• Are Long-LLMs A Necessity For Long-Context Tasks? [28.54986983107062]
  We argue that the long-LLMs are not a necessity to solve long-context tasks, as common long-context tasks are short-context solvable. We propose a framework called LC-Boost which enables a short-LLM to address the long-context tasks in a bootstrapping manner. By adaptively accessing and utilizing the context based on the presented tasks, LC-Boost can serve as a general framework to handle diversified long-context processing problems.
  arXiv  Detail & Related papers  (2024-05-24T07:59:30Z)
• Long Context Alignment with Short Instructions and Synthesized Positions [56.1267385315404]
  This paper introduces Step-Skipping Alignment (SkipAlign) It is a new technique designed to enhance the long-context capabilities of Large Language Models (LLMs) With a careful selection of the base model and alignment datasets, SkipAlign with only 6B parameters achieves it's best performance and comparable with strong baselines like GPT-3.5-Turbo-16K on LongBench.
  arXiv  Detail & Related papers  (2024-05-07T01:56:22Z)
• NovelQA: Benchmarking Question Answering on Documents Exceeding 200K Tokens [63.7488938083696]
  We introduce NovelQA, a benchmark tailored for evaluating Large Language Models (LLMs) with complex, extended narratives.<n>NovelQA offers a unique blend of complexity, length, and narrative coherence, making it an ideal tool for assessing deep textual understanding.<n>Our evaluation of long-context LLMs on NovelQA reveals significant insights into their strengths and weaknesses.
  arXiv  Detail & Related papers  (2024-03-18T17:32:32Z)
• When does In-context Learning Fall Short and Why? A Study on Specification-Heavy Tasks [54.71034943526973]
  In-context learning (ICL) has become the default method for using large language models (LLMs) We find that ICL falls short of handling specification-heavy tasks, which are tasks with complicated and extensive task specifications. We identify three primary reasons: inability to specifically understand context, misalignment in task schema comprehension with humans, and inadequate long-text understanding ability.
  arXiv  Detail & Related papers  (2023-11-15T14:26:30Z)
• M4LE: A Multi-Ability Multi-Range Multi-Task Multi-Domain Long-Context Evaluation Benchmark for Large Language Models [58.54538318912159]
  M4LE is a benchmark for evaluating the long-sequence capability of large language models (LLMs) M4LE is based on a diverse NLP task pool comprising 36 NLP task types and 12 domains. We conducted a systematic evaluation on 11 well-established LLMs, especially those optimized for long-sequence inputs.
  arXiv  Detail & Related papers  (2023-10-30T03:11:30Z)
• Large Search Model: Redefining Search Stack in the Era of LLMs [63.503320030117145]
  We introduce a novel conceptual framework called large search model, which redefines the conventional search stack by unifying search tasks with one large language model (LLM) All tasks are formulated as autoregressive text generation problems, allowing for the customization of tasks through the use of natural language prompts. This proposed framework capitalizes on the strong language understanding and reasoning capabilities of LLMs, offering the potential to enhance search result quality while simultaneously simplifying the existing cumbersome search stack.
  arXiv  Detail & Related papers  (2023-10-23T05:52:09Z)
• Effective Long-Context Scaling of Foundation Models [90.57254298730923]
  We present a series of long-context LLMs that support effective context windows of up to 32,768 tokens. Our models achieve consistent improvements on most regular tasks and significant improvements on long-context tasks over Llama 2.
  arXiv  Detail & Related papers  (2023-09-27T21:41:49Z)
• LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding [58.20031627237889]
  LongBench is the first bilingual, multi-task benchmark for long context understanding. It comprises 21 datasets across 6 task categories in both English and Chinese, with an average length of 6,711 words (English) and 13,386 characters (Chinese)
  arXiv  Detail & Related papers  (2023-08-28T11:53:40Z)
• Faith and Fate: Limits of Transformers on Compositionality [109.79516190693415]
  We investigate the limits of transformer large language models across three representative compositional tasks. These tasks require breaking problems down into sub-steps and synthesizing these steps into a precise answer. Our empirical findings suggest that transformer LLMs solve compositional tasks by reducing multi-step compositional reasoning into linearized subgraph matching.
  arXiv  Detail & Related papers  (2023-05-29T23:24:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.