{"id":7359,"date":"2025-01-31T23:56:40","date_gmt":"2025-01-31T23:56:40","guid":{"rendered":"https:\/\/megagon.ai\/publications\/holistic-reasoning-with-long-context-lms-a-benchmark-for-database-operations-on-massive-textual-data\/"},"modified":"2025-10-02T22:59:34","modified_gmt":"2025-10-02T22:59:34","slug":"holistic-reasoning-with-long-context-lms","status":"publish","type":"publications","link":"https:\/\/megagon.ai\/jp\/publications\/holistic-reasoning-with-long-context-lms\/","title":{"rendered":"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual Data"},"template":"","publications-tags":[324,192],"conference-year":[186],"conference":[188],"class_list":["post-7359","publications","type-publications","status-publish","has-post-thumbnail","hentry","publications-tags-data-ai-symbiosis","publications-tags-llm-nlp","conference-year-186","conference-iclr"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.1 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual Data - Megagon<\/title>\n<meta name=\"description\" content=\"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual... Megagon Labs paper. Presented at ICLR 2025.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/megagon.ai\/jp\/publications\/holistic-reasoning-with-long-context-lms\/\" \/>\n<meta property=\"og:locale\" content=\"ja_JP\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual Data\" \/>\n<meta property=\"og:description\" content=\"LCLM performance is more sensitive to how much information is packed into the context than to the length of that context. Moreover, tasks requiring aggregation of multiple facts across the input lead to noticeable performance drops, especially as complexity increases.These insights reveal a critical bottleneck in current LCLMs and point to where future work must focus: not just expanding context windows, but improving how models reason within them.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/megagon.ai\/jp\/publications\/holistic-reasoning-with-long-context-lms\/\" \/>\n<meta property=\"og:site_name\" content=\"Megagon\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/megagonlabs\/\" \/>\n<meta property=\"article:modified_time\" content=\"2025-10-02T22:59:34+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/megagon.ai\/wp-content\/uploads\/2025\/02\/Publications.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1600\" \/>\n\t<meta property=\"og:image:height\" content=\"900\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:title\" content=\"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual Data\" \/>\n<meta name=\"twitter:description\" content=\"LCLM performance is more sensitive to how much information is packed into the context than to the length of that context. Moreover, tasks requiring aggregation of multiple facts across the input lead to noticeable performance drops, especially as complexity increases.These insights reveal a critical bottleneck in current LCLMs and point to where future work must focus: not just expanding context windows, but improving how models reason within them.\" \/>\n<meta name=\"twitter:image\" content=\"https:\/\/megagon.ai\/wp-content\/uploads\/2025\/02\/Publications.png\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/publications\\\/holistic-reasoning-with-long-context-lms\\\/\",\"url\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/publications\\\/holistic-reasoning-with-long-context-lms\\\/\",\"name\":\"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual Data - Megagon\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/publications\\\/holistic-reasoning-with-long-context-lms\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/publications\\\/holistic-reasoning-with-long-context-lms\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/megagon.ai\\\/wp-content\\\/uploads\\\/2025\\\/02\\\/holo_bench.png\",\"datePublished\":\"2025-01-31T23:56:40+00:00\",\"dateModified\":\"2025-10-02T22:59:34+00:00\",\"description\":\"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual... Megagon Labs paper. Presented at ICLR 2025.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/publications\\\/holistic-reasoning-with-long-context-lms\\\/#breadcrumb\"},\"inLanguage\":\"ja-JP\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/megagon.ai\\\/jp\\\/publications\\\/holistic-reasoning-with-long-context-lms\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"ja-JP\",\"@id\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/publications\\\/holistic-reasoning-with-long-context-lms\\\/#primaryimage\",\"url\":\"https:\\\/\\\/megagon.ai\\\/wp-content\\\/uploads\\\/2025\\\/02\\\/holo_bench.png\",\"contentUrl\":\"https:\\\/\\\/megagon.ai\\\/wp-content\\\/uploads\\\/2025\\\/02\\\/holo_bench.png\",\"width\":1600,\"height\":900,\"caption\":\"HoloBench\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/publications\\\/holistic-reasoning-with-long-context-lms\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Publications\",\"item\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/publications\\\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual Data\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/#website\",\"url\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/\",\"name\":\"Megagon Labs\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"ja-JP\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/megagon.ai\\\/jp\\\/#organization\",\"name\":\"Megagon Labs\",\"url\":\"https:\\\/\\\/megagon.ai\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"url\":\"https:\\\/\\\/megagon.ai\\\/wp-content\\\/uploads\\\/2025\\\/02\\\/Logo-Megagon-Labs.webp\",\"caption\":\"Megagon Labs\"},\"image\":{\"url\":\"https:\\\/\\\/megagon.ai\\\/wp-content\\\/uploads\\\/2025\\\/02\\\/Logo-Megagon-Labs.webp\"},\"description\":\"Megagon Labs is an AI research organization conducting research in compound AI systems, large language models, data-AI symbiosis, and human-centered AI. Megagon Labs shares its findings with the broader community through open-source tools, datasets, publications, workshops, and an invited speaker series.\",\"sameAs\":[\"https:\\\/\\\/github.com\\\/megagonlabs\",\"https:\\\/\\\/twitter.com\\\/megagonlabs\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/megagon-labs\\\/\",\"https:\\\/\\\/www.facebook.com\\\/megagonlabs\\\/\"],\"address\":{\"@type\":\"PostalAddress\",\"streetAddress\":\"444 Castro Street\",\"addressLocality\":\"Mountain View\",\"addressRegion\":\"CA\",\"postalCode\":\"94041\",\"addressCountry\":\"US\"},\"contactPoint\":{\"@type\":\"ContactPoint\",\"email\":\"contactus@megagon.ai\",\"contactType\":\"general inquiries\"},\"parentOrganization\":{\"@type\":\"Organization\",\"name\":\"Recruit Holdings\",\"url\":\"https:\\\/\\\/recruit-holdings.com\\\/en\\\/\"}}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual Data - Megagon","description":"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual... Megagon Labs paper. Presented at ICLR 2025.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/megagon.ai\/jp\/publications\/holistic-reasoning-with-long-context-lms\/","og_locale":"ja_JP","og_type":"article","og_title":"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual Data","og_description":"LCLM performance is more sensitive to how much information is packed into the context than to the length of that context. Moreover, tasks requiring aggregation of multiple facts across the input lead to noticeable performance drops, especially as complexity increases.These insights reveal a critical bottleneck in current LCLMs and point to where future work must focus: not just expanding context windows, but improving how models reason within them.","og_url":"https:\/\/megagon.ai\/jp\/publications\/holistic-reasoning-with-long-context-lms\/","og_site_name":"Megagon","article_publisher":"https:\/\/www.facebook.com\/megagonlabs\/","article_modified_time":"2025-10-02T22:59:34+00:00","og_image":[{"width":1600,"height":900,"url":"https:\/\/megagon.ai\/wp-content\/uploads\/2025\/02\/Publications.png","type":"image\/png"}],"twitter_card":"summary_large_image","twitter_title":"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual Data","twitter_description":"LCLM performance is more sensitive to how much information is packed into the context than to the length of that context. Moreover, tasks requiring aggregation of multiple facts across the input lead to noticeable performance drops, especially as complexity increases.These insights reveal a critical bottleneck in current LCLMs and point to where future work must focus: not just expanding context windows, but improving how models reason within them.","twitter_image":"https:\/\/megagon.ai\/wp-content\/uploads\/2025\/02\/Publications.png","schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/megagon.ai\/jp\/publications\/holistic-reasoning-with-long-context-lms\/","url":"https:\/\/megagon.ai\/jp\/publications\/holistic-reasoning-with-long-context-lms\/","name":"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual Data - Megagon","isPartOf":{"@id":"https:\/\/megagon.ai\/jp\/#website"},"primaryImageOfPage":{"@id":"https:\/\/megagon.ai\/jp\/publications\/holistic-reasoning-with-long-context-lms\/#primaryimage"},"image":{"@id":"https:\/\/megagon.ai\/jp\/publications\/holistic-reasoning-with-long-context-lms\/#primaryimage"},"thumbnailUrl":"https:\/\/megagon.ai\/wp-content\/uploads\/2025\/02\/holo_bench.png","datePublished":"2025-01-31T23:56:40+00:00","dateModified":"2025-10-02T22:59:34+00:00","description":"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual... Megagon Labs paper. Presented at ICLR 2025.","breadcrumb":{"@id":"https:\/\/megagon.ai\/jp\/publications\/holistic-reasoning-with-long-context-lms\/#breadcrumb"},"inLanguage":"ja-JP","potentialAction":[{"@type":"ReadAction","target":["https:\/\/megagon.ai\/jp\/publications\/holistic-reasoning-with-long-context-lms\/"]}]},{"@type":"ImageObject","inLanguage":"ja-JP","@id":"https:\/\/megagon.ai\/jp\/publications\/holistic-reasoning-with-long-context-lms\/#primaryimage","url":"https:\/\/megagon.ai\/wp-content\/uploads\/2025\/02\/holo_bench.png","contentUrl":"https:\/\/megagon.ai\/wp-content\/uploads\/2025\/02\/holo_bench.png","width":1600,"height":900,"caption":"HoloBench"},{"@type":"BreadcrumbList","@id":"https:\/\/megagon.ai\/jp\/publications\/holistic-reasoning-with-long-context-lms\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/megagon.ai\/jp\/"},{"@type":"ListItem","position":2,"name":"Publications","item":"https:\/\/megagon.ai\/jp\/publications\/"},{"@type":"ListItem","position":3,"name":"Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual Data"}]},{"@type":"WebSite","@id":"https:\/\/megagon.ai\/jp\/#website","url":"https:\/\/megagon.ai\/jp\/","name":"Megagon Labs","description":"","publisher":{"@id":"https:\/\/megagon.ai\/jp\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/megagon.ai\/jp\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"ja-JP"},{"@type":"Organization","@id":"https:\/\/megagon.ai\/jp\/#organization","name":"Megagon Labs","url":"https:\/\/megagon.ai\/","logo":{"@type":"ImageObject","url":"https:\/\/megagon.ai\/wp-content\/uploads\/2025\/02\/Logo-Megagon-Labs.webp","caption":"Megagon Labs"},"image":{"url":"https:\/\/megagon.ai\/wp-content\/uploads\/2025\/02\/Logo-Megagon-Labs.webp"},"description":"Megagon Labs is an AI research organization conducting research in compound AI systems, large language models, data-AI symbiosis, and human-centered AI. Megagon Labs shares its findings with the broader community through open-source tools, datasets, publications, workshops, and an invited speaker series.","sameAs":["https:\/\/github.com\/megagonlabs","https:\/\/twitter.com\/megagonlabs","https:\/\/www.linkedin.com\/company\/megagon-labs\/","https:\/\/www.facebook.com\/megagonlabs\/"],"address":{"@type":"PostalAddress","streetAddress":"444 Castro Street","addressLocality":"Mountain View","addressRegion":"CA","postalCode":"94041","addressCountry":"US"},"contactPoint":{"@type":"ContactPoint","email":"contactus@megagon.ai","contactType":"general inquiries"},"parentOrganization":{"@type":"Organization","name":"Recruit Holdings","url":"https:\/\/recruit-holdings.com\/en\/"}}]}},"_links":{"self":[{"href":"https:\/\/megagon.ai\/jp\/wp-json\/wp\/v2\/publications\/7359","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/megagon.ai\/jp\/wp-json\/wp\/v2\/publications"}],"about":[{"href":"https:\/\/megagon.ai\/jp\/wp-json\/wp\/v2\/types\/publications"}],"version-history":[{"count":1,"href":"https:\/\/megagon.ai\/jp\/wp-json\/wp\/v2\/publications\/7359\/revisions"}],"predecessor-version":[{"id":7362,"href":"https:\/\/megagon.ai\/jp\/wp-json\/wp\/v2\/publications\/7359\/revisions\/7362"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/megagon.ai\/jp\/wp-json\/wp\/v2\/media\/7360"}],"wp:attachment":[{"href":"https:\/\/megagon.ai\/jp\/wp-json\/wp\/v2\/media?parent=7359"}],"wp:term":[{"taxonomy":"publications-tags","embeddable":true,"href":"https:\/\/megagon.ai\/jp\/wp-json\/wp\/v2\/publications-tags?post=7359"},{"taxonomy":"conference-year","embeddable":true,"href":"https:\/\/megagon.ai\/jp\/wp-json\/wp\/v2\/conference-year?post=7359"},{"taxonomy":"conference","embeddable":true,"href":"https:\/\/megagon.ai\/jp\/wp-json\/wp\/v2\/conference?post=7359"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}