Lost in the Middle: How Language Models Use Long Contexts
Open accessThe DOI matches its Crossref record; 48 of the 59 references it cites were found in a registry or an index; 15 of the 23 sources its claims cite were checked for support.
Abstract
While recent language models have the ability to take long contexts as input, relatively little is known about how well they use longer context. We analyze the performance of language models on two tasks that require identifying relevant information in their input contexts: multi-document question answering and key-value retrieval. We find that performance can degrade significantly when changing the position of relevant information, indicating that current language models do not robustly make use of information in long input contexts. In particular, we observe that performance is often highest when relevant information occurs at the beginning or end of the input context, and significantly degrades when models must access relevant information in the middle of long contexts, even for explicitly long-context models. Our analysis provides a better understanding of how language models use their input context and provides new evaluation protocols for future long-context language models.
Overview
CoChat Summary
TL;DR
Language models’ performance on tasks requiring access to long input contexts is strongly affected by where relevant information appears, with a U-shaped curve: best at the beginning or end and worst in the middle. Extended-context models are not necessarily better at using context, and reader performance saturates before retriever recall in open-domain QA.
Key findings
- In multi-document QA, performance is highest when the answer-containing document is at the start or end of the context and degrades in the middle; GPT-3.5-Turbo can drop by more than 20%, and in 20- and 30-document settings middle-position performance is lower than its closed-book performance of 56.1% (oracle: 88.3%).
- Extended-context models are not necessarily better at using input context: when the input fits both windows, GPT-3.5-Turbo and GPT-3.5-Turbo(16K) perform nearly identically as a function of relevant-information position.
- In synthetic key-value retrieval, Claude-1.3 and Claude-1.3(100K) do nearly perfectly, while other models struggle especially with 140 or 300 key-value pairs; query-aware contextualization gives GPT-3.5-Turbo(16K) perfect performance at 300 pairs, versus worst-case 45.6% without it.
- In open-domain QA, reader performance saturates long before retriever recall: using 50 instead of 20 retrieved documents only marginally improves performance (∼1.5% for GPT-3.5-Turbo and ∼1% for Claude-1.3).
Methods
- Controlled multi-document QA on NaturalQuestions-Open: 2,655 queries whose annotated long answer is a paragraph; each context has exactly one answer-containing Wikipedia passage plus k−1 Contriever/MS-MARCO distractor passages; the paper varies answer position and context length (10, 20, 30 documents) and measures accuracy.
- Synthetic key-value retrieval: inputs are serialized JSON objects with unique random UUID key-value pairs; models must return the value for a specified key; experiments use 75, 140, and 300 key-value pairs (500 examples each), varying key position and number of pairs, with accuracy as the metric.
- Models include open MPT-30B-Instruct and LongChat-13B(16K), closed GPT-3.5-Turbo, GPT-3.5-Turbo(16K), Claude-1.3, and Claude-1.3(100K), plus additional architecture/instruction experiments with Flan-T5-XXL, Flan-UL2, MPT-30B, Llama-2, and GPT-4; an open-domain QA case study uses a Contriever retriever-reader setup.
Limitations
- Stated: the paper uses greedy decoding and leaves exploration of other decoding methods to future work; the investigations into why models are not robust are described as preliminary; GPT-4 is evaluated only on a subset of 500 multi-document QA examples because full evaluation would cost upwards of $6000.
- Stated: NaturalQuestions-Open has ambiguity, so the authors run an unambiguous subset and also test random distractors and randomized distractor order, finding similar trends.
- Evident: the controlled multi-document QA setting always contains exactly one document that answers the question, unlike open-domain QA; the synthetic key-value task removes natural language semantics; results may not generalize to all tasks, models, or decoding settings.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
7 Valid · 0 Not valid · 5 Not checked · 13 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Language models have become an important and flexible building block in a variety of user-facing language technologies, including conversational interfaces, search and summarization, and collaborative writing (Shuster et al., 2022;Thoppilan et al., 2022;Lee et al., 2022, inter alia).”
§ Lost in the Middle: How Language Models Use Long Contexts > 1 Introduction
What the sources say
- Partly supportedcites Shuster et al., 2022 sourceExperimental — accuracy not yet measured
Section 3.1.1 (p. 4) describes a conversational agent built on a language model that generates internet search queries, supporting conversational interfaces and search as part of dialogue. Section 2 (p. 3) mentions summarization only as related work, and the passages do not establish collaborative writing as an application of this paper’s model. (LLM, conf=0.85)
- Supportedcites Thoppilan et al., 2022 source
The Abstract (p. 1) identifies LaMDA as a family of language models specialized for dialog. In this sentence's list of applications, Thoppilan et al. supports the conversational-interface example; the other cited works need not each cover every example. (LLM, conf=0.92)
- Supportedcites Lee et al., 2022 source
Lee et al. is cited alongside other works, and its contribution to the sentence is the collaborative-writing example. Section 4.2.2 describes writers receiving and selecting model-generated suggestions in a writing interface; Section 2.2.1 also describes language models as applicable to dialogue and summarization. The paper need not establish every example in the jointly cited list. (LLM, conf=0.89)
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“These observations complement prior work, which found that non-instruction fine-tuned language models are biased towards recent tokens (i.e., the end of the input context; Khandelwal et al., 2018;Press et al., 2021).”
§ Lost in the Middle: How Language Models Use Long Contexts > 4.3 Effect of Instruction Fine-Tuning
What the sources say
- Partly supportedcites Khandelwal et al., 2018 sourceExperimental — accuracy not yet measured
The Introduction and Section 5 report that the studied LSTM is much more sensitive to recent word order than to distant word order, supporting a recency effect. The paper studies a standard LSTM on two datasets, however, rather than establishing this tendency across non-instruction-fine-tuned language models generally; it also finds that distant words remain relevant. (LLM, conf=0.82)
- Partly supportedcites Press et al., 2021 sourceExperimental — accuracy not yet measured
Section 3.1 (page not available in our copy) finds that very short contexts hurt performance, while increasing subsequence length from 1,024 to 3,072 tokens gives negligible improvement. That is consistent with limited benefit from more distant context, but the paper does not establish that the models are biased toward recent tokens within an input context. (LLM, conf=0.78)
- Partly supported
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“We instantiate this task with data from NaturalQuestions-Open (Lee et al., 2019;Kwiatkowski et al., 2019), which contains historical queries issued to the Google search engine, coupled with human-annotated answers extracted from Wikipedia.”
§ Lost in the Middle: How Language Models Use Long Contexts > 2.1 Experimental Setup
What the sources say
- Supportedcites Lee et al., 2019 source
Section 7.1 says the paper creates an open version of Natural Questions and uses it for training and evaluation. The Introduction describes the task as answering questions by retrieving evidence from an open corpus, supporting the cited use of NaturalQuestions-Open for this task. (LLM, conf=0.91)
- Not checked — no full text of the sourcecites Kwiatkowski et al., 2019 source
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Following Kandpal et al. (2022) and Mallen et al. (2023), we use accuracy as our primary evaluation metric, judging whether any of the cor- rect answers (as taken from the NaturalQuestions annotations) appear in the predicted output.”
§ Lost in the Middle: How Language Models Use Long Contexts > 2.1 Experimental Setup
What the sources say
- Supportedcites Kandpal et al. (2022) source
Section 3 (p. 4) says the paper evaluates answers using exact match against a ground-truth answer set, and Section 3.1 (p. 4) reports QA accuracies as its evaluation results. This supports citing Kandpal et al. for using accuracy to evaluate question answering. (LLM, conf=0.90)
- Supportedcites Mallen et al. (2023) source
Section 3.1 (‘Focus and Task’) explicitly names accuracy as the metric and says a prediction is correct when it contains an exact match to any gold answer. This supports the evaluation approach attributed to Mallen et al.; the NaturalQuestions annotations describe the citing authors’ answer source. (LLM, conf=0.94)
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“or when language models are augmented with external information (relevant documents from a search engine, database query results, etc.; Petroni et al., 2020;Ram et al., 2023;Shi et al., 2023;Mallen et al., 2023;Schick et al., 2023, inter alia).”
§ Lost in the Middle: How Language Models Use Long Contexts > 1 Introduction
What the sources say
- Partly supportedcites Petroni et al., 2020 sourceExperimental — accuracy not yet measured
The Abstract (p. 1) says the paper integrates information from a retrieval system with a pre-trained language model, and §3.4.2 (p. 5) specifies a retrieved paragraph as context. This supports augmentation with retrieved documents, but our copy does not show language models being augmented with database query results. (LLM, conf=0.84)
- Partly supportedcites Ram et al., 2023 sourceExperimental — accuracy not yet measured
The Introduction describes augmenting language models with relevant documents retrieved from an external knowledge source, and Section 5 describes using a BM25 retriever to find those documents. The passages do not establish database query results as another type of external information used by this paper. (LLM, conf=0.82)
- Partly supportedcites Shi et al., 2023 sourceExperimental — accuracy not yet measured
Section 3.1 (page unavailable) describes retrieving relevant documents from a corpus, and the Conclusion says REPLUG uses those documents to augment a language model. The paper supports the document-retrieval example, but our copy does not describe augmenting models with database query results or documents from a search engine specifically. (LLM, conf=0.82)
- Partly supported
20 more claims · 2 with their cited sources checked
Create a free account to see all 25 claims and every source verdict.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“In vanilla multi-document or NIAH-based benchmarks (Kamradt, 2023; Liu et al., 2024), models perform consistently well when the needle (or gold document) appears at the very beginning or end of the context window, with minimal impact from context length.”
From NoLiMa: Long-Context Evaluation Beyond Literal Matching (2025) · cites this paper as Liu et al., 2024
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Even within a single long sequence, attention effectiveness degrades with distance due to attention dilution , positional encoding limitations, and token interference, leading to the well-known 'lost-in-the-middle' and context-decay phenomena (Liu et al., 2024; Press et al., 2021a).”
From MAGMA: A Multi-Graph based Agentic Memory Architecture for AI Agents (2026) · cites this paper as Liu et al., 2024
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Current large language models (LLMs) exhibit significant deficiencies in episodic memory tasks including encoding, storing, and retrieving specific information from temporally dependent events over a long period of time.”
From ARTEM: Enhancing Large Language Model Agents with Spatial-Temporal Episodic Memory (2026) · cites this paper as Liu et al. 2024
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Recent approaches to handle memory tasks in LLMs, such as in-context learning, retrieval-augmented generation (RAG), and fine-tuning, may resolve the long-term retention issues, but are still inadequate to handle tasks requiring chronological awareness of the stored information.”
From ARTEM: Enhancing Large Language Model Agents with Spatial-Temporal Episodic Memory (2026) · cites this paper as Liu et al. 2024
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“If not ranked at the top, the key information may be neglected by the generative module, resulting in incomplete or inaccurate responses to the user queries24.”
From Leveraging long context in retrieval augmented language models for medical question answering (2025) · cites this paper as [24]
Experimental — accuracy not yet measured
86 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., & Liu, P. J. (2019) · arXiv · doi:10.48550/arxiv.1910.10683 · Cited by 10,929
LLaMA: Open and Efficient Foundation Language Models
Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.-A., Lacroix, T., Rozière, B., Goyal, N., Hambro, E., Azhar, F., et al. (2023) · arXiv · doi:10.48550/arxiv.2302.13971 · Cited by 4,602
Transformer-XL: Attentive language models beyond a fixed-length context
Dai, Z., Yang, Z., Yang, Y., Carbonell, J., Le, Q., & Salakhutdinov, R. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1285 · Cited by 3,169 · cited 2×
“Recent improvements in hardware (e.g., faster GPUs with more memory) and algorithms (Dai et al., 2019; Dao et al., 2022; Poli et al., 2023; Rubin and Berant, 2023, inter alia ) have resulted in language models with larger context windows (e.g., 4096, 32K, and even 100K tokens), but it remains…”
BackgroundLlama 2: Open Foundation and Fine-Tuned Chat Models
Touvron, H., Martin, L., Stone, K., Albert, P., Almahairi, A., Babaei, Y., Bashlykov, N., Batra, S., Bhargava, P., Bhosale, S., et al. (2023) · arXiv · doi:10.48550/arxiv.2307.09288 · Cited by 2,604
Longformer: The Long-Document Transformer
Beltagy, I., Peters, M. E., & Cohan, A. (2020) · arXiv · doi:10.48550/arxiv.2004.05150 · Cited by 2,281
Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering
Izacard, G., & Grave, E. (2020) · Conference of the European Chapter of the Association for Computational Linguistics · doi:10.48550/arxiv.2007.01282 · Cited by 2,263
Natural Questions: A benchmark for question answering research
Kwiatkowski, T., Palomaki, J., Redfield, O., Collins, M., Parikh, A., Alberti, C., Epstein, D., Polosukhin, I., Devlin, J., Lee, K., et al. (2019) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00276 · Cited by 2,069 · cited 1×
“We instantiate this task with data from NaturalQuestions-Open (Lee et al., 2019; Kwiatkowski et al., 2019), which contains historical queries issued to the Google search engine, coupled with human-annotated answers extracted from Wikipedia.”
MethodThe serial position effect of free recall.
Murdock, B. B. (1962) · Journal of Experimental Psychology · doi:10.1037/h0045106 · Cited by 1,717 · cited 1×
“The U-shaped curve we observe in this work has a connection in psychology known as the serial-position effect (Ebbinghaus, 1913; Murdock Jr, 1962), that states that in free-association recall of elements from a list, humans tend to best re-member the first and last elements of the list.”
MethodMemory: A contribution to experimental psychology.
Ebbinghaus, H. (1913) · doi:10.1037/10011-000 · Cited by 1,645 · cited 1×
“The U-shaped curve we observe in this work has a connection in psychology known as the serial-position effect (Ebbinghaus, 1913; Murdock Jr, 1962), that states that in free-association recall of elements from a list, humans tend to best re-member the first and last elements of the list.”
MethodScaling Instruction-Finetuned Language Models
Chung, H. W., Hou, L., Longpre, S., Zoph, B., Tay, Y., Fedus, W., Li, Y., Wang, X., Dehghani, M., Brahma, S., et al. (2022) · arXiv · doi:10.48550/arxiv.2210.11416 · Cited by 1,441 · cited 1×
“To better understand why, we perform some preliminary investigations into the role of model architecture (decoder-only vs. encoder-decoder), query-aware contextualization, and instruction fine-tuning.”
BackgroundBig Bird: Transformers for Longer Sequences
Zaheer, M., Guruganesh, G., Dubey, A., Ainslie, J., Alberti, C., Ontanon, S., Pham, P., Ravula, A., Wang, Q., Yang, L., et al. (2020) · arXiv · doi:10.48550/arxiv.2007.14062 · Cited by 1,349
FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness
Dao, T., Fu, D. Y., Ermon, S., Rudra, A., & Ré, C. (2022) · arXiv · doi:10.48550/arxiv.2205.14135 · Cited by 1,214 · cited 1×
“Dao et al. (2022) instead provide a faster exact attention by a carefully-crafted IO-aware CUDA kernel.”
BackgroundToolformer: Language Models Can Teach Themselves to Use Tools
Schick, T., Dwivedi-Yu, J., Dessì, R., Raileanu, R., Lomeli, M., Zettlemoyer, L., Cancedda, N., & Scialom, T. (2023) · arXiv · doi:10.48550/arxiv.2302.04761 · Cited by 1,150 · cited 1×
“…documents, conversation histories, etc.) or when language models are augmented with exter-nal information (relevant documents from a search engine, database query results, etc.; Petroni et al., 2020; Ram et al., 2023; Shi et al., 2023; Mallen et al., 2023; Schick et al., 2023, inter alia ).”
Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering
Izacard, G., & Grave, E. (2021) · Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume · doi:10.18653/v1/2021.eacl-main.74 · Cited by 1,109
Linformer: Self-Attention with Linear Complexity
Wang, S., Li, B. Z., Khabsa, M., Fang, H., & Ma, H. (2020) · arXiv · doi:10.48550/arxiv.2006.04768 · Cited by 867 · cited 1×
“, 2020), or low-rank approximations (Wang et al., 2020; Peng et al., 2021); see Tay et al.”
BackgroundEfficiently Modeling Long Sequences with Structured State Spaces
Gu, A., Goel, K., & Ré, C. (2021) · arXiv · doi:10.48550/arxiv.2111.00396 · Cited by 805 · cited 1×
“…task-specific since it depends on the marginal value of the added context and the model’s ability to effectively use long input contexts, but we perform a case study with open-domain question answering on NaturalQuestions-Open to better understand this trade-off in existing language models.”
BackgroundLaMDA: Language Models for Dialog Applications
Thoppilan, R., De Freitas, D., Hall, J., Shazeer, N., Kulshreshtha, A., Cheng, H.-T., Jin, A., Bos, T., Baker, L., Du, Y., et al. (2022) · arXiv · doi:10.48550/arxiv.2201.08239 · Cited by 778
Latent retrieval for weakly supervised open domain question answering
Lee, K., Chang, M.-W., & Toutanova, K. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1612 · Cited by 771 · cited 6×
“When retrieving from Wikipedia to answer queries from NaturalQuestions-Open, we find that model performance saturates long before retriever re-call saturates, indicating that current models fail to effectively use additional retrieved doc-uments—using 50 documents instead of 20 retrieved documents only marginally impr…”
MethodInfluentialUL2: Unifying Language Learning ParadigmsPreprint
Tay, Y., Dehghani, M., Tran, V. Q., Garcia, X., Wei, J., Wang, X., Chung, H. W., Shakeri, S., Bahri, D., Schuster, T., et al. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2205.05131 · Cited by 591
In-Context Retrieval-Augmented Language Models
Ram, O., Levine, Y., Dalmedigos, I., Muhlgay, D., Shashua, A., Leyton-Brown, K., & Shoham, Y. (2023) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00605 · Cited by 449
Lee, M., Liang, P., & Yang, Q. (2022) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3491102.3502030 · Cited by 345
Self-citationMallen, A., Asai, A., Zhong, V., Das, R., Khashabi, D., & Hajishirzi, H. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.546 · Cited by 340
CHI Conference on Human Factors in Computing Systems
(2022) · doi:10.1145/3491102 · Cited by 308
Sharp nearby, fuzzy far away: How neural language models use context
Khandelwal, U., He, H., Qi, P., & Jurafsky, D. (2018) · Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p18-1027 · Cited by 303 · cited 3×
“The pioneering work of Khandelwal et al. (2018) showed that small LSTM language models make increasingly coarse use of longer-term context; Sankar et al. (2019) found similar results in dialogue models.”
ResultSharp Nearby, Fuzzy Far Away: How Neural Language Models Use Context
Khandelwal, U., He, H., Qi, P., & Jurafsky, D. (2018) · arXiv · doi:10.48550/arxiv.1805.04623 · Cited by 302
Random Feature AttentionPreprint
Peng, H., Pappas, N., Yogatama, D., Schwartz, R., Smith, N. A., & Kong, L. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2103.02143 · Cited by 291
REPLUG: Retrieval-Augmented Black-Box Language Models
Shi, W., Min, S., Yasunaga, M., Seo, M., James, R., Lewis, M., Zettlemoyer, L., & Yih, W.-T. (2024) · Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers) · doi:10.18653/v1/2024.naacl-long.463 · Cited by 209
Unsupervised Dense Information Retrieval with Contrastive Learning
Izacard, G., Caron, M., Hosseini, L., Riedel, S., Bojanowski, P., Joulin, A., & Grave, E. (2021) · arXiv · doi:10.48550/arxiv.2112.09118 · Cited by 172 · cited 7×
“When retrieving from Wikipedia to answer queries from NaturalQuestions-Open, we find that model performance saturates long before retriever re-call saturates, indicating that current models fail to effectively use additional retrieved doc-uments—using 50 documents instead of 20 retrieved documents only marginally impr…”
MethodInfluentialTrain Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation
Press, O., Smith, N. A., & Lewis, M. (2021) · arXiv · doi:10.48550/arxiv.2108.12409 · Cited by 161
How Context Affects Language Models' Factual Predictions
Petroni, F., Lewis, P., Piktus, A., Rocktäschel, T., Wu, Y., Miller, A. H., & Riedel, S. (2020) · arXiv · doi:10.48550/arxiv.2005.04611 · Cited by 154 · cited 1×
“Petroni et al. (2020) were among the first to demonstrate the potential of combining context from an information retrieval system with a pretrained language models for unsupervised question answering.”
BackgroundSelf-citationDo Neural Dialog Systems Use the Conversation History Effectively? An Empirical Study
Sankar, C., Subramanian, S., Pal, C., Chandar, S., & Bengio, Y. (2019) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/p19-1004 · Cited by 151 · cited 1×
“The pioneering work of Khandelwal et al. (2018) showed that small LSTM language models make increasingly coarse use of longer-term context; Sankar et al. (2019) found similar results in dialogue models.”
ResultLarge Language Models Struggle to Learn Long-Tail Knowledge
Kandpal, N., Deng, H., Roberts, A., Wallace, E., & Raffel, C. (2022) · arXiv · doi:10.48550/arxiv.2211.08411 · Cited by 143
Frustratingly Short Attention Spans in Neural Language ModelingPreprint
Daniluk, M., Rocktäschel, T., Welbl, J., & Riedel, S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1702.04521 · Cited by 136
Hyena Hierarchy: Towards Larger Convolutional Language Models
Poli, M., Massaroli, S., Nguyen, E., Fu, D. Y., Dao, T., Baccus, S., Bengio, Y., Ermon, S., & Ré, C. (2023) · arXiv · doi:10.48550/arxiv.2302.10866 · Cited by 96 · cited 3×
“Finally, a variety of recently-proposed architectures model sequences with millions of tokens, raising the potential of further dramatic increases in language model maximum context length (Gu et al., 2022; Fu et al., 2023; Poli et al., 2023; Yu et al., 2023, inter alia ).”
BackgroundBlenderBot 3: a deployed conversational agent that continually learns to responsibly engage
Shuster, K., Xu, J., Komeili, M., Ju, D., Smith, E. M., Roller, S., Ung, M., Chen, M., Arora, K., Lane, J., et al. (2022) · arXiv · doi:10.48550/arxiv.2208.03188 · Cited by 95
What Context Features Can Transformer Language Models Use?
O’Connor, J., & Andreas, J. (2021) · Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2021.acl-long.70 · Cited by 88 · cited 2×
“For example, Transformer language models may have varying sensitivity to different linguistic features in their input (O’Connor and Andreas, 2021).”
BackgroundAmbigQA: Answering Ambiguous Open-domain Questions
Min, S., Michael, J., Hajishirzi, H., & Zettlemoyer, L. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2020.emnlp-main.466 · Cited by 87 · cited 1×
“We use the ambiguity annotations of Min et al. (2020) to create a subset unambiguous questions.”
MethodWhere to stop reading a ranked list? threshold optimization using truncated score distributions
Arampatzis, A., Kamps, J., & Robertson, S. E. (2009) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/1571941.1572031 · Cited by 73 · cited 1×
“…reranking of retrieved documents (pushing relevant information closer to the start of the input context) or ranked list truncation (retrieving fewer documents when appropriate; Arampatzis et al., 2009) may be promising directions for improving how language-model-based readers use retrieved context.”
BackgroundREPLUG: Retrieval-augmented black-box language models
Shi, W., Min, S., Yasunaga, M., Seo, M., James, R., Lewis, M., Zettlemoyer, L., & Yih, W.-T. (2023) · arXiv · doi:10.48550/arxiv.2301.12652 · Cited by 65
Shortformer: Better language modeling using shorter inputs
Press, O., Smith, N. A., & Lewis, M. (2021) · Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) · doi:10.18653/v1/2021.acl-long.427 · Cited by 55 · cited 1×
“These observations complement prior work, which found that (non-instruction-tuned) language models are biased towards recent tokens (i.e., the end of the input context; Khandelwal et al., 2018; Press et al., 2021).”
ResultRetrieval-Pretrained Transformer: Long-range Language Modeling with Self-retrieval
Rubin, O., & Berant, J. (2024) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00693 · Cited by 44
Do Long-Range Language Models Actually Use Long-Range Context?
Sun, S., Krishna, K., Mattarella-Micke, A., & Iyyer, M. (2021) · Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2021.emnlp-main.62 · Cited by 43 · cited 1×
“Tay et al. (2021) propose the Long Range Arena, which evaluates long-context models on a variety of natural language, visual reasoning, and synthetic tasks.”
BackgroundEfficient long-text understanding with short-text models
Ivgi, M., Shaham, U., & Berant, J. (2023) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00547 · Cited by 42 · cited 1×
“Our experimental setup is similar to the needle-in-a-haystack experiments of Ivgi et al. (2023), who compare question answering performance when the relevant paragraph is placed (i) at the beginning of the input or (ii) a random position within the input.”
MethodRankGen: Improving Text Generation with Large Ranking Models
Krishna, K., Chang, Y., Wieting, J., & Iyyer, M. (2022) · Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2022.emnlp-main.15 · Cited by 30
ZeroSCROLLS: A zero-shot benchmark for long text understanding
Shaham, U., Ivgi, M., Efrat, A., Berant, J., & Levy, O. (2023) · Findings of the Association for Computational Linguistics: EMNLP 2023 · doi:10.18653/v1/2023.findings-emnlp.536 · Cited by 23
ZeroSCROLLS: A Zero-Shot Benchmark for Long Text Understanding
Shaham, U., Ivgi, M., Efrat, A., Berant, J., & Levy, O. (2023) · arXiv · doi:10.48550/arxiv.2305.14196 · Cited by 23
The NLP Task Effectiveness of Long-Range Transformers
Qin, G., Feng, Y., & Van Durme, B. (2023) · Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics · doi:10.18653/v1/2023.eacl-main.273 · Cited by 14 · cited 1×
“Qin et al. (2023) analyze how efficient Transformers perform on a variety of long-context downstream NLP tasks, finding that long-context transformers are recency-biased and do not effectively use long-range context.”
BackgroundHow Context Affects Language Models' Factual Predictions
Fabio Petroni, P. L. (2020) · Automated Knowledge Base Construction (AKBC) · doi:10.24432/c5201w · Cited by 4
Retrieval-Pretrained Transformer: Long-range Language Modeling with Self-retrievalPreprint
Rubin, O., & Berant, J. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2306.13421 · Cited by 3
Research works printed but not linked: 1
- Sharan, 2018, Prediction with a short memory, 10.18653/v1/K18-3013 · DOI
Cited by
Huang, L., Yu, W., Ma, W., Zhong, W., Feng, Z., Wang, H., Chen, Q., Peng, W., Feng, X., Qin, B., et al. (2025) · ACM Transactions on Information Systems · doi:10.1145/3703155 · Cited by 4,688
A Survey of Large Language Models
Zhao, W. X., Zhou, K., Li, J., Tang, T., Dong, Z., Hou, Y., Zhang, B., Min, Y., Zhang, J., Liu, P., et al. (2026) · Frontiers of Computer Science · doi:10.1007/s11704-026-60308-3 · Cited by 2,093
Walter, Y. (2024) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-024-00448-3 · Cited by 1,113
Retrieval-Augmented Generation for Large Language Models: A Survey
Gao, Y., Xiong, Y., Gao, X., Jia, K., Pan, J., Bi, Y., Dai, Y., Sun, J., Wang, M., & Wang, H. (2023) · arXiv · doi:10.48550/arxiv.2312.10997 · Cited by 876
“A notable issue, termed the “lost in the middle” phenomenon, arises when a single retrieval yields redundant content that may dilute or contradict essential information, thereby degrading the generation quality [ Liu et al. , 2023a ] .”
BackgroundEvaluation and mitigation of the limitations of large language models in clinical decision-making
Hager, P., Jungmann, F., Holland, R., Bhagat, K., Hubrecht, I., Knauer, M., Vielhauer, J., Makowski, M., Braren, R., Kaissis, G., et al. (2024) · Nature Medicine · doi:10.1038/s41591-024-03097-1 · Cited by 798
“The sensitivity of LLMs to the order of presented information has been well documented on multiple-choice questions 45,46 and information retrieval 47 .”
Metadata
- Authors
- Liu, N. F., Lin, K., Hewitt, J., Paranjape, A., Bevilacqua, M., Petroni, F., & Liang, P.
- Affiliations
- Stanford University (US)
- University of California, Berkeley (US)
- Samaya AI, UK
- Corresponding author
- Liu, Nelson F.
- Venue
- Transactions of the Association for Computational Linguistics
- Publisher
- MIT Press
- Published
- 2024
- Volume
- 12
- Pages
- 157-173
- ISSN
- 2307-387X (ISSN-L)
- OpenAlex
- W4391876619
- Semantic Scholar
- 1733eb7792f7a43dd21f51f4d1017a1bffd217b5
- Cited by
- 1,612
- References
- 59
- Citations by year
- 2026: 901 · 2025: 516 · 2024: 172
- FWCI
- 233.95
- Type
- journal article
- Language
- en
- Open access
- diamond
- doi.org
- Licence
- Locations
- Transactions of the Association for Computational Linguistics (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
- Keywords
- long-context language models, multi-document question answering, position bias
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-09-27 (prompt extract-v2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.