	{"id":293089,"date":"2024-12-03T17:19:23","date_gmt":"2024-12-03T17:19:23","guid":{"rendered":"https:\/\/www.artefact.com\/?post_type=blog&#038;p=293089"},"modified":"2024-12-03T17:20:10","modified_gmt":"2024-12-03T17:20:10","slug":"is-preference-alignment-always-the-best-option-to-enhance-llm-based-translation-an-empirical-analysis","status":"publish","type":"blog","link":"https:\/\/www.artefact.com\/de\/blog\/is-preference-alignment-always-the-best-option-to-enhance-llm-based-translation-an-empirical-analysis\/","title":{"rendered":"Ist Preference Alignment immer die beste Option zur Verbesserung von LLM-basierten \u00dcbersetzungen? Eine empirische Analyse"},"content":{"rendered":"<p><div class=\"fusion-fullwidth fullwidth-box fusion-builder-row-1 fusion-flex-container nonhundred-percent-fullwidth non-hundred-percent-height-scrolling\" style=\"--link_color: var(--awb-color6);--awb-border-radius-top-left:0px;--awb-border-radius-top-right:0px;--awb-border-radius-bottom-right:0px;--awb-border-radius-bottom-left:0px;--awb-background-color:var(--awb-color1);--awb-flex-wrap:wrap;\" ><div class=\"fusion-builder-row fusion-row fusion-flex-align-items-flex-start fusion-flex-content-wrap\" style=\"max-width:calc( 1440px + 20px );margin-left: calc(-20px \/ 2 );margin-right: calc(-20px \/ 2 );\"><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-0 fusion_builder_column_1_1 1_1 fusion-flex-column\" style=\"--awb-bg-size:cover;--awb-width-large:100%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:100%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><div class=\"fusion-column-wrapper fusion-column-has-shadow fusion-flex-justify-content-flex-start fusion-content-layout-column\"><div class=\"fusion-text fusion-text-1 description\" style=\"--awb-text-color:var(--awb-color5);--awb-text-font-family:&quot;PT Serif&quot;;--awb-text-font-style:normal;--awb-text-font-weight:400;\"><p>Hippolyte Gisserot-Boukhlef, Ricardo Rei, Emmanuel Malherbe, C\u00e9line Hudelot, Pierre Colombo, Nuno M. Guerreiro<\/p>\n<p>Artefact Research Center, Unbabel, Equall, MICS CentraleSup\u00e9lec, Universit\u00e9 Paris-Saclay, Instituto de Telecomunica\u00e7\u00f5es, Instituto Superior T\u00e9cnico &amp; Universidade de Lisboa (Lisbon ELLIS Unit)<\/p>\n<\/div><\/div><\/div><\/div><\/div><article class=\"fusion-fullwidth fullwidth-box fusion-builder-row-2 fusion-flex-container nonhundred-percent-fullwidth non-hundred-percent-height-scrolling\" style=\"--link_color: var(--awb-color6);--awb-border-radius-top-left:0px;--awb-border-radius-top-right:0px;--awb-border-radius-bottom-right:0px;--awb-border-radius-bottom-left:0px;--awb-background-color:var(--awb-color1);--awb-flex-wrap:wrap;\" ><div class=\"fusion-builder-row fusion-row fusion-flex-align-items-flex-start fusion-flex-justify-content-center fusion-flex-content-wrap\" style=\"max-width:calc( 1440px + 20px );margin-left: calc(-20px \/ 2 );margin-right: calc(-20px \/ 2 );\"><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-1 fusion_builder_column_1_1 1_1 fusion-flex-column\" style=\"--awb-bg-size:cover;--awb-width-large:100%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:100%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><div class=\"fusion-column-wrapper fusion-column-has-shadow fusion-flex-justify-content-flex-start fusion-content-layout-column\"><div style=\"text-align:center;\"><a class=\"fusion-button button-flat fusion-button-default-size button-default fusion-button-default button-1 fusion-button-default-span fusion-button-default-type\" target=\"_self\" href=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/12\/Is-Preference-Alignment-Always-the-Best-Option-to-Enhance-LLM-Based.pdf\"><span class=\"fusion-button-text awb-button__text awb-button__text--default\">Den Artikel lesen <\/span><\/a><\/div><div class=\"fusion-text fusion-text-2\" style=\"--awb-font-size:20px;--awb-line-height:1.6;--awb-letter-spacing:var(--awb-typography4-letter-spacing);--awb-text-transform:var(--awb-typography4-text-transform);--awb-text-color:var(--awb-color5);--awb-text-font-family:var(--awb-typography4-font-family);--awb-text-font-weight:var(--awb-typography4-font-weight);--awb-text-font-style:var(--awb-typography4-font-style);\"><p>Wir freuen uns, Ihnen den neuesten Forschungsartikel unseres Doktoranden Hippolyte Gisserot-Boukhlef vorstellen zu k\u00f6nnen, der f\u00fcr die Neunte Konferenz f\u00fcr Maschinelle \u00dcbersetzung (WMT24) im November 2024 ausgew\u00e4hlt wurde.<\/p>\n<\/div><div class=\"fusion-title title fusion-title-1 fusion-sep-none fusion-title-text fusion-title-size-two\" style=\"--awb-text-color:var(--awb-color6);--awb-margin-bottom-small:8px;--awb-font-size:30px;\"><h2 class=\"fusion-title-heading title-heading-left fusion-responsive-typography-calculated\" style=\"font-family:&quot;PT Serif&quot;;font-style:normal;font-weight:700;margin:0;letter-spacing:1.6px;font-size:1em;--fontSize:30;line-height:1.47;\">Abstrakt<\/h2><\/div><div class=\"fusion-text fusion-text-3\" style=\"--awb-font-size:20px;--awb-line-height:1.6;--awb-letter-spacing:var(--awb-typography4-letter-spacing);--awb-text-transform:var(--awb-typography4-text-transform);--awb-text-color:var(--awb-color5);--awb-text-font-family:var(--awb-typography4-font-family);--awb-text-font-weight:var(--awb-typography4-font-weight);--awb-text-font-style:var(--awb-typography4-font-style);\"><p>Das Papier untersucht die Effektivit\u00e4t von Pr\u00e4ferenzoptimierungstechniken, insbesondere im Vergleich zum Supervised Fine-Tuning. W\u00e4hrend die Optimierung auf der Grundlage von Pr\u00e4ferenzen data eine g\u00e4ngige Praxis in der maschinellen \u00dcbersetzung ist - wobei oft hochwertige Ergebnisse von externen Modellen wie GPT-4 genutzt werden -, sind die weitergehenden Auswirkungen dieses Ansatzes noch nicht vollst\u00e4ndig bekannt. Interessanterweise deuten unsere Ergebnisse darauf hin, dass durch die Verwendung des Modells selbst als Selbstlehrer eine vergleichbare \u00dcbersetzungsqualit\u00e4t erreicht werden kann, w\u00e4hrend die Komplexit\u00e4t und die Einschr\u00e4nkungen, die mit dem Verlassen auf externe Systeme verbunden sind, beseitigt werden.<\/p>\n<\/div><div style=\"text-align:center;\"><a class=\"fusion-button button-flat fusion-button-default-size button-default fusion-button-default button-2 fusion-button-default-span fusion-button-default-type\" target=\"_self\" href=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2024\/12\/presentation_vdef.pptx.pdf\"><span class=\"fusion-button-text awb-button__text awb-button__text--default\">Zugriff auf das Slide Deck <\/span><\/a><\/div><\/div><\/div><\/div><\/article><\/p>","protected":false},"excerpt":{"rendered":"<p>Neuronale Metriken f\u00fcr die maschinelle \u00dcbersetzung (MT) zur Evaluierung sind aufgrund ihrer \u00fcberlegenen Korrelation mit menschlichen Urteilen im Vergleich zu traditionellen lexikalischen Metriken immer prominenter geworden.<\/p>","protected":false},"featured_media":293097,"parent":0,"template":"","meta":{"_acf_changed":false,"ep_exclude_from_search":false},"blog-category":[21939],"blog-language":[2991,2993],"class_list":["post-293089","blog","type-blog","status-publish","has-post-thumbnail","hentry","blog-category-medium","blog-language-en","blog-language-fr"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/blog\/293089","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/blog"}],"about":[{"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/types\/blog"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/media\/293097"}],"wp:attachment":[{"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/media?parent=293089"}],"wp:term":[{"taxonomy":"blog-category","embeddable":true,"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/blog-category?post=293089"},{"taxonomy":"blog-language","embeddable":true,"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/blog-language?post=293089"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}