{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/lost-in-the-middle-how-language-models-use","title":"Lost in the Middle: How Language Models Use Long Contexts","arxiv_id":"2307.03172","date":"2023-07-06","proceeding":null,"authors":["Nelson F. Liu","Kevin Lin","John Hewitt","Ashwin Paranjape","Michele Bevilacqua","Fabio Petroni","Percy Liang"],"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.","url_abs":"https://arxiv.org/abs/2307.03172v3","url_pdf":"https://arxiv.org/pdf/2307.03172v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"lost-in-the-middle-how-language-models-use","repo_url":"https://github.com/nelson-liu/lost-in-the-middle","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"lost-in-the-middle-how-language-models-use","repo_url":"https://github.com/casetext/r-and-r","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"lost-in-the-middle-how-language-models-use","repo_url":"https://github.com/jzhang38/longmamba","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"lost-in-the-middle-how-language-models-use","repo_url":"https://github.com/kbressem/longhealth","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"lost-in-the-middle-how-language-models-use","repo_url":"https://github.com/togethercomputer/llama-2-7b-32k-instruct","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"lost-in-the-middle-how-language-models-use","repo_url":"https://github.com/visual-haystacks/vhs_benchmark","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":null,"task_name":"Position"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2307.03172","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2307.03172"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/togethercomputer/llama-2-7b-32k-instruct","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/nelson-liu/lost-in-the-middle","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kbressem/longhealth","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/visual-haystacks/vhs_benchmark","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jzhang38/longmamba","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/casetext/r-and-r","reach":{"status":"ok"}}],"summary":{"ran_draft_wrong":1,"ran_fixture":1,"unverified":2},"by_repo_kind":{"official":{"samples":2,"ran":0,"repositories":1},"listed":{"samples":2,"ran":2,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":2,"samples":[{"code_sha256_prefix":"870987e39d8888fd","entry":"generate_prompt_landmark","repo":"jzhang38/longmamba","repo_kind":"listed","path":"pass_key.py","file_url":"https://github.com/jzhang38/longmamba/blob/HEAD/pass_key.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"870987e39d8888fd"}},{"code_sha256_prefix":"cff545416361104b","entry":"passkey_retrieval_test","repo":"jzhang38/longmamba","repo_kind":"listed","path":"pass_key.py","file_url":"https://github.com/jzhang38/longmamba/blob/HEAD/pass_key.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"cff545416361104b"}},{"code_sha256_prefix":"91506e645c7a9e50","entry":"get_closedbook_qa_prompt","repo":"nelson-liu/lost-in-the-middle","repo_kind":"official","path":"src/lost_in_the_middle/prompting.py","file_url":"https://github.com/nelson-liu/lost-in-the-middle/blob/HEAD/src/lost_in_the_middle/prompting.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"91506e645c7a9e50"}},{"code_sha256_prefix":"aca69348624f921b","entry":"get_kv_retrieval_prompt","repo":"nelson-liu/lost-in-the-middle","repo_kind":"official","path":"src/lost_in_the_middle/prompting.py","file_url":"https://github.com/nelson-liu/lost-in-the-middle/blob/HEAD/src/lost_in_the_middle/prompting.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"aca69348624f921b"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}