{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/kv-cache-compression-but-what-must-we-give-in","title":"KV Cache Compression, But What Must We Give in Return? A Comprehensive Benchmark of Long Context Capable Approaches","arxiv_id":"2407.01527","date":"2024-07-01","proceeding":null,"authors":["Jiayi Yuan","Hongyi Liu","Shaochen Zhong","Yu-Neng Chuang","Songchen Li","Guanchu Wang","Duy Le","Hongye Jin","Vipin Chaudhary","Zhaozhuo Xu","Zirui Liu","Xia Hu"],"abstract":"Long context capability is a crucial competency for large language models (LLMs) as it mitigates the human struggle to digest long-form texts. This capability enables complex task-solving scenarios such as book summarization, code assistance, and many more tasks that are traditionally manpower-intensive. However, transformer-based LLMs face significant challenges with long context input due to the growing size of the KV cache and the intrinsic complexity of attending to extended inputs; where multiple schools of efficiency-driven approaches - such as KV cache quantization, token dropping, prompt compression, linear-time sequence models, and hybrid architectures - have been proposed to produce efficient yet long context-capable models. Despite these advancements, no existing work has comprehensively benchmarked these methods in a reasonably aligned environment. In this work, we fill this gap by providing a taxonomy of current methods and evaluating 10+ state-of-the-art approaches across seven categories of long context tasks. Our work reveals numerous previously unknown phenomena and offers insights - as well as a friendly workbench - for the future development of long context-capable LLMs. The source code is available at https://github.com/henryzhongsc/longctx_bench.","url_abs":"https://arxiv.org/abs/2407.01527v2","url_pdf":"https://arxiv.org/pdf/2407.01527v2.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":"kv-cache-compression-but-what-must-we-give-in","repo_url":"https://github.com/henryzhongsc/longctx_bench","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":null,"task_name":"Book summarization"},{"task_slug":"quantization","task_name":"Quantization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2407.01527","atlas_url":"https://app.syntology.ai/?focus=2407.01527","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2407.01527"},"calls":[{"tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2407.01527"},"returns":"the samples harvested from this paper, each with its run record; it does not verify the paper"},{"tool":"get_code","arguments_in":"samples[].code_sha256","returns":"one stored sample by its code_sha256; not shown on this site"}]},"developers":"https://syntology.ai/developers","held":true,"reason":"in_graph_layer","checked_against":null,"agent_calls":[{"tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2407.01527"},"call":"get_harvested_code_for_paper(arxiv_id=\"2407.01527\")"},{"tool":"get_code","arguments_in":{"code_sha256":"syntology.samples[].code_sha256"},"call":"get_code(code_sha256=\"…\")"},{"tool":"get_citation_path","arguments":{"paper_1":"2407.01527"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2407.01527\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2407.01527"},"call":"get_concepts_for_paper(arxiv_id=\"2407.01527\")"}],"by_repo":[{"repo":"henryzhongsc/longctx_bench","kind":"official","samples":13,"ran":13,"instrument":5,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code","read_at":"2026-10-05T00:36:11+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","ran_record":{"url":"https://syntology.ai/api/ran/2407.01527.json","schema":"syntology.ran_record/1","counts_crosswalk":{"n_samples":"syntology.counts.lifted","n_ran":"syntology.counts.executed","n_ran_checked":"syntology.counts.checked","n_instrument":"syntology.counts.instrument_failures","n_unverified":"syntology.counts.no_recorded_run"},"note":"live record, measured at request time; may be newer than this page's graph read","badge":"https://syntology.ai/api/ran/2407.01527.svg"},"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/henryzhongsc/longctx_bench","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":5,"ran_honours":3,"ran_draft_wrong":5},"n_samples":13,"n_ran":13,"n_constructed":0,"n_ran_checked":8,"n_instrument":5,"n_unverified":0,"n_instrument_is":"failures of Syntology's instrument, not of the code","every_run_is_an_instrument_failure":false,"key_notes":{"n_ran_checked":"legacy name, kept unchanged so existing readers do not break: it counts the samples that ran with no instrument failure (honoured, violated, and ran with no contract checked); it does not mean a contract was checked, and the pages print it as 'K with no instrument failure', not 'K checked'","n_constructed":"a sub-count of the samples that ran, never subtracted from them and never a failure: an executed sample whose run returned an instance of its own class (fixture_out_type equals the entry name): the run built an object and did not compute a result (Syntology's RAN record, counts.constructed)"},"by_repo_kind":{"official":{"samples":13,"ran":13,"n_ran_checked":8,"n_instrument":5,"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":4,"inline_ok_is":"inline_ok is true only when both licence gates allow Syntology to serve this code's text: this paper's link (inline_ok_link, from the licence of this paper's copy) and the stored copy of the same code that Syntology serves, which may have been first harvested from another repository. withheld_clause is Syntology's own wording of which gate keeps the text a pointer and why, and is null when inline_ok is true. This site shows no code text either way; n_pointer_only_for_licence counts samples with inline_ok false (both gates), n_pointer_only_link_allows those of them whose inline_ok_link is true","n_pointer_only_link_allows":4,"fixture_fields_is":{"signature_and_fixture_out_type":"the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right","level_recorded_at":"when Syntology's graph wrote the sample's verification level, not when the code executed; absent where the level was written before the date was kept, and never filled in with the build date","present_on":"samples that ran only; a key is absent, never empty, where the graph records nothing"},"n_ran_with_level_date":10,"n_ran_without_level_date":3,"n_ran_signature_not_shown":0,"levels_written_between":{"first":"2026-09-24","last":"2026-09-26","is":"the first and last UTC dates on which Syntology's graph wrote a verification level for a sample on this page that ran, not when the code executed"},"code_page_min_papers":5,"code_page_is":"a page for the entry name across every paper's harvested code; the name is a string, not an identity","samples":[{"code_sha256":"3b19a66bec30b1b9","code_sha256_prefix":"3b19a66bec30b1b9","entry":"build_chat","repo":"henryzhongsc/longctx_bench","repo_kind":"official","path":"pipeline/model_utils.py","file_url":"https://github.com/henryzhongsc/longctx_bench/blob/HEAD/pipeline/model_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"tokenizer:MockTokenizer/0/?, prompt:str/0/str, chat_template:str/0/str","fixture_out_type":"str","level_recorded_at":"2026-09-24T15:16:18.137819+00:00","code_page":"/code/build-chat","census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"3b19a66bec30b1b9"}},{"code_sha256":"b349b79d9cc2934b","code_sha256_prefix":"b349b79d9cc2934b","entry":"count_score","repo":"henryzhongsc/longctx_bench","repo_kind":"official","path":"eval/longbench_utils/metrics.py","file_url":"https://github.com/henryzhongsc/longctx_bench/blob/HEAD/eval/longbench_utils/metrics.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"signature":"prediction:str/0/str, ground_truth:int/0/int","fixture_out_type":"float","retrieve_with":{"tool":"get_code","code_sha256":"b349b79d9cc2934b"},"code_page":"/code/count-score","census":null,"licence":"MIT","inline_ok":false,"inline_ok_link":true,"withheld_clause":"this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from borealisai/radar-decoding, and LGPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline","mcp_get_code":{"code_sha256":"b349b79d9cc2934b"}},{"code_sha256":"2207706bbe348d2c","code_sha256_prefix":"2207706bbe348d2c","entry":"count_words","repo":"henryzhongsc/longctx_bench","repo_kind":"official","path":"eval/passkey_utils/passkey_utils.py","file_url":"https://github.com/henryzhongsc/longctx_bench/blob/HEAD/eval/passkey_utils/passkey_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":true,"signature":"input_string:str/0/str","fixture_out_type":"int","level_recorded_at":"2026-09-24T15:16:18.137819+00:00","retrieve_with":{"tool":"get_code","code_sha256":"2207706bbe348d2c"},"code_page":"/code/count-words","census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"2207706bbe348d2c"}},{"code_sha256":"ecac1fc997b729b6","code_sha256_prefix":"ecac1fc997b729b6","entry":"distraction_allowed_partial_match","repo":"henryzhongsc/longctx_bench","repo_kind":"official","path":"eval/passkey_utils/eval_metrics.py","file_url":"https://github.com/henryzhongsc/longctx_bench/blob/HEAD/eval/passkey_utils/eval_metrics.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"list_of_answers:list/1/list, responses:list/1/str","fixture_out_type":"list","level_recorded_at":"2026-09-24T15:16:18.137819+00:00","code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"ecac1fc997b729b6"}},{"code_sha256":"01c26d6cb82784d2","code_sha256_prefix":"01c26d6cb82784d2","entry":"eval_by_metric","repo":"henryzhongsc/longctx_bench","repo_kind":"official","path":"eval/passkey_utils/eval_metrics.py","file_url":"https://github.com/henryzhongsc/longctx_bench/blob/HEAD/eval/passkey_utils/eval_metrics.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"answers:list/1/str, responses:list/1/str, metric:str/0/str","fixture_out_type":"list","level_recorded_at":"2026-09-24T15:16:18.137819+00:00","code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"01c26d6cb82784d2"}},{"code_sha256":"932a134ef3a83a86","code_sha256_prefix":"932a134ef3a83a86","entry":"get_intermediate_values_within_min_max","repo":"henryzhongsc/longctx_bench","repo_kind":"official","path":"eval/passkey_utils/passkey_utils.py","file_url":"https://github.com/henryzhongsc/longctx_bench/blob/HEAD/eval/passkey_utils/passkey_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"value_min:float/0/float, value_max:float/0/float, n_values:int/0/int, desc:str/0/str, int_rounding:bool/0/bool","fixture_out_type":"list","level_recorded_at":"2026-09-24T15:16:18.137819+00:00","code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"932a134ef3a83a86"}},{"code_sha256":"4d685bf704394edd","code_sha256_prefix":"4d685bf704394edd","entry":"load_data","repo":"henryzhongsc/longctx_bench","repo_kind":"official","path":"eval/longbench_utils/eval_long_bench.py","file_url":"https://github.com/henryzhongsc/longctx_bench/blob/HEAD/eval/longbench_utils/eval_long_bench.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"config:dict/0/?","fixture_out_type":"Dataset","level_recorded_at":"2026-09-24T15:16:18.137819+00:00","code_page":"/code/load-data","census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"4d685bf704394edd"}},{"code_sha256":"e7e75981cb464788","code_sha256_prefix":"e7e75981cb464788","entry":"normalize_answer","repo":"henryzhongsc/longctx_bench","repo_kind":"official","path":"eval/longbench_utils/metrics.py","file_url":"https://github.com/henryzhongsc/longctx_bench/blob/HEAD/eval/longbench_utils/metrics.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"signature":"s:str/0/str","fixture_out_type":"str","retrieve_with":{"tool":"get_code","code_sha256":"e7e75981cb464788"},"code_page":"/code/normalize-answer","census":null,"licence":"MIT","inline_ok":false,"inline_ok_link":true,"withheld_clause":"this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from borealisai/radar-decoding, and LGPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline","mcp_get_code":{"code_sha256":"e7e75981cb464788"}},{"code_sha256":"92035fa2c8ef3cbe","code_sha256_prefix":"92035fa2c8ef3cbe","entry":"normalize_answer","repo":"henryzhongsc/longctx_bench","repo_kind":"official","path":"eval/passkey_utils/eval_metrics.py","file_url":"https://github.com/henryzhongsc/longctx_bench/blob/HEAD/eval/passkey_utils/eval_metrics.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"signature":"s:str/0/str","fixture_out_type":"str","level_recorded_at":"2026-09-24T15:16:18.137819+00:00","retrieve_with":{"tool":"get_code","code_sha256":"92035fa2c8ef3cbe"},"code_page":"/code/normalize-answer","census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"92035fa2c8ef3cbe"}},{"code_sha256":"8c5c581f9264c810","code_sha256_prefix":"8c5c581f9264c810","entry":"normalize_zh_answer","repo":"henryzhongsc/longctx_bench","repo_kind":"official","path":"eval/longbench_utils/metrics.py","file_url":"https://github.com/henryzhongsc/longctx_bench/blob/HEAD/eval/longbench_utils/metrics.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"signature":"s:str/0/str","fixture_out_type":"str","retrieve_with":{"tool":"get_code","code_sha256":"8c5c581f9264c810"},"code_page":"/code/normalize-zh-answer","census":null,"licence":"MIT","inline_ok":false,"inline_ok_link":true,"withheld_clause":"this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from borealisai/radar-decoding, and LGPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline","mcp_get_code":{"code_sha256":"8c5c581f9264c810"}},{"code_sha256":"682b08abcc125f52","code_sha256_prefix":"682b08abcc125f52","entry":"register_args_and_configs","repo":"henryzhongsc/longctx_bench","repo_kind":"official","path":"pipeline/main_utils.py","file_url":"https://github.com/henryzhongsc/longctx_bench/blob/HEAD/pipeline/main_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"args:Args/0/?","fixture_out_type":"dict","level_recorded_at":"2026-09-24T15:16:18.137819+00:00","code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"682b08abcc125f52"}},{"code_sha256":"d86a7d51ce10c6e4","code_sha256_prefix":"d86a7d51ce10c6e4","entry":"set_logger","repo":"henryzhongsc/longctx_bench","repo_kind":"official","path":"pipeline/main_utils.py","file_url":"https://github.com/henryzhongsc/longctx_bench/blob/HEAD/pipeline/main_utils.py","link_basis":"plan_row","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"output_folder_dir:str/0/str, args:Args/0/?","fixture_out_type":"Logger","level_recorded_at":"2026-09-26T03:58:20.443611+00:00","code_page":"/code/set-logger","census":null,"licence":"MIT","inline_ok":false,"inline_ok_link":true,"withheld_clause":"this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from amarka8/SOCKET, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"d86a7d51ce10c6e4"}},{"code_sha256":"319e766d2cafe502","code_sha256_prefix":"319e766d2cafe502","entry":"truncate_string_by_word_count","repo":"henryzhongsc/longctx_bench","repo_kind":"official","path":"eval/passkey_utils/passkey_utils.py","file_url":"https://github.com/henryzhongsc/longctx_bench/blob/HEAD/eval/passkey_utils/passkey_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"input_string:str/0/str, word_count_end:int/0/int, word_count_start:int/0/int","fixture_out_type":"str","level_recorded_at":"2026-09-24T15:16:18.137819+00:00","code_page":null,"census":null,"licence":"MIT","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"319e766d2cafe502"}}]},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"c0737650a1b47e267d0dd9193d7ce447d5548d1f9a4fc1390698ec0886fe7bf5","record_changed_at":"2026-10-05","record_changed_at_basis":"changed"}