[{"data":1,"prerenderedAt":296},["ShallowReactive",2],{"project-46274287":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":10,"maker":11,"posted_at":12,"year":13,"points":14,"comments":15,"rank_in_year":16,"shown_in_year":17,"top":18,"tier":19,"category":20,"subcategory":21,"format":22,"keywords":23,"reception":28,"sentiment":29,"stance_mix":30,"scored_comments":35,"aspects":36,"strip":41,"post_text":43,"what":44,"said":67,"maker_in_thread":180,"shape":189,"grid":193,"attempts":257,"same_week":258,"maker_shows":31,"maker_hidden":288,"reader_version":289,"card":290},"46274287","LangSeed","Learning a Language Using Only Words You Know","title","Show HN: Learning a Language Using Only Words You Know","https:\u002F\u002Fsimedw.com\u002F2025\u002F12\u002F15\u002Flangseed\u002F","simedw.com","simedw",1765805532,2025,84,29,853,3837,23,"notable","learning_science","language learning","web_app",[21,24,25,26,27],"comprehensible input","vocabulary","llm","known words","mixed",0.143,{"positive":31,"neutral":32,"mixed":33,"critical":34},4,5,1,2,12,{"idea":37,"execution":38,"design":40},{"positive":31,"critical":33},{"positive":39,"critical":39},0,{"positive":39,"critical":33},[33,39,39,39,32,42,39,39,42,39,39,39,39,39,39,39,39,32,42,39,32,39,42,42,39,33,42,39,32],3,"A proof-of-concept language learning app that uses LLMs to generate definitions of unknown words using only previously mastered vocabulary.",{"problem":45,"audience":51,"pricing":51,"motivation":51,"time_to_build":51,"stack":52,"openness":64,"stage":65,"team_size":66},{"value":46,"source":47,"evidence":48},"Learning vocabulary when dictionary definitions use words the learner does not know yet","inferred",[49],{"id":50},"post",null,[53,58],{"value":54,"source":55,"evidence":56},"LLMs","maker",[57],{"id":50},{"value":59,"source":55,"evidence":60},"Snowball stemmer",[61],{"id":62,"by":11,"text":63},"46331832","Surprisingly easy. If the language has a lot of conjugations (e.g., polite past verb forms), running each word through Snowball first makes the process a bit easier.","open_source","prototype","unknown",{"praise":68,"objections":92,"requests":110,"questions":124,"alternatives":143},[69,85],{"value":70,"source":71,"evidence":72},"Great idea that mirrors how children learn language","commenter",[73,77,81],{"id":74,"by":75,"text":76},"46325846","englishcat","This is quite a great idea, as a native Chinese speaker, i want to say this is the way very similar how we learned Chinese when we were kids. On the other hand, the Chinese…",{"id":78,"by":79,"text":80},"46321376","closetkantian","This is genius! I love it.",{"id":82,"by":83,"text":84},"46320906","nubg","This is extremely interesting, great idea. Really both thumbs up. Looking for more ideas\u002Flifehack approaches to learning via LLMs.",{"value":86,"source":71,"evidence":87},"Useful for full immersion and vocabulary building with LLMs",[88],{"id":89,"by":90,"text":91},"46328387","bisonbear","As a fellow Mandarin learner - this is super cool! Intuitively makes a lot of sense for the \"full immersion\" component of language. I love to see exciting uses of AI for language…",[93,103],{"value":94,"source":55,"evidence":95},"Model sometimes produces awkward phrasing or grammar errors, especially in Chinese",[96,99],{"id":97,"by":11,"text":98},"46332227","Thanks for the questions. Very fair concerns. Take all of this with a fairly large pinch of salt; this is still an experiment. 1. How does it know which words I already know? It…",{"id":100,"by":101,"text":102},"46320550","andai","Cool idea! You mentioned the model struggling with Chinese a bit. Have you tried any Chinese models, e.g. DeepSeek or GLM? I imagine they probably have a lot more Chinese in the…",{"value":104,"source":71,"evidence":105},"Site is hard to use on mobile and lacks prepared texts",[106],{"id":107,"by":108,"text":109},"46326421","gamander","This site sucks on mobile. can't upload full text files? why no prepared texts to start reading right away?",[111,117],{"value":112,"source":71,"evidence":113},"Support for replacing most text with known words, leaving unknown words in English",[114],{"id":115,"by":90,"text":116},"46329353","checked out the tool and think it's a cool idea! one piece of feedback though - I actually feel like the inverse product would be more helpful for me. What I mean is replacing…",{"value":118,"source":71,"evidence":119},"Support for more languages",[120],{"id":121,"by":122,"text":123},"46323468","mog_dev","How hard would it be to add new languages ?",[125,132,139],{"value":126,"source":71,"evidence":127},"How does the system know which words the user already knows?",[128],{"id":129,"by":130,"text":131},"46320125","bryanhogan","Interesting concept! Think this would be quite cool to explore. Personally am very interested in language learning concepts \u002F apps. My first concerns though: 1. How can the…",{"value":133,"source":71,"evidence":134},"Which models handle Chinese well, such as DeepSeek, GLM or Kimi?",[135,136],{"id":100,"by":101,"text":102},{"id":137,"by":90,"text":138},"46328405","I have personally had success with using Kimi for Chinese creative writing making the same assumption that Moonshot, as a Chinese company, has more\u002Fbetter Mandarin language…",{"value":140,"source":71,"evidence":141},"Is decoding done by logit bias or token blocking?",[142],{"id":89,"by":90,"text":91},[144,158,166,173],{"value":145,"source":71,"evidence":146},"Pimsleur",[147,150,154],{"id":148,"by":101,"text":149},"46320572","There's a language learning method where you just listen to audio, until you develop a basic familiarity with the language. (Then learn reading and writing later.) You listen to…",{"id":151,"by":152,"text":153},"46322068","cblum","Pimsleur is awful for Mandarin. I wish I hadn’t wasted my time on it.",{"id":155,"by":156,"text":157},"46323683","bpev","As someone who did most of Pimsleur Spanish and Mandarin (and did a single unit in various other languages), and has since continued learning these languages (I'm currently…",{"value":159,"source":71,"evidence":160},"Anki flashcards",[161,165],{"id":162,"by":163,"text":164},"46321629","SuperNinKenDo","I recently changed all my language flashcards to be like this. Anki is probably the best option. I have the field with the Hanzi, but just configure my cards not to show it at…",{"id":155,"by":156,"text":157},{"value":167,"source":71,"evidence":168},"Graded readers and simplified texts like Easy Chinese",[169,172],{"id":170,"by":11,"text":171},"46331859","This is a simplified version: Journey to the West in Easy Chinese by Jeff Pepper and Xiao Hui Wang. Otherwise, I would definitely have waited a bit before biting off something…",{"id":155,"by":156,"text":157},{"value":174,"source":71,"evidence":175},"Dreaming Spanish and comprehensible input methods",[176,179],{"id":177,"by":152,"text":178},"46322101","To add a bit more to this: AJATT (all Japanese all the time) later evolved into MIA (mass input approach), which then became Refold. The gist of those methods is mass input +…",{"id":148,"by":101,"text":149},{"comments":32,"answered":34,"key":181},[182,183,186],{"id":170,"text":171},{"id":184,"text":185},"46331780","That's a really cool concept. Naively replacing words might work, but sometimes the context is needed. Maybe a model like gemini 2.5 flash lite would be fast enough but still…",{"id":187,"text":188},"46276370","Thanks! I think getting comfortable with characters fairly early is important, as it helps shift your mindset into the right place. That said, I don’t think this project really…",{"commenters":190,"max_depth":31,"maker_replies":32,"first_comment_mins":191,"lifespan_hours":192},20,143,265,[194,198,202,205,207,209,211,213,216,220,222,223,226,227,228,229,230,233,234,235,237,241,245,247,249,250,253,255,256],[195,39,33,196,197,39],"46320904","project\u002Fcritical\u002Fidea","dylanzhangdev",[199,39,39,200,201,39],"46321593","other\u002Fneutral\u002Fnone","gcanyon",[203,39,39,200,204,39],"46322021","bitwize",[170,39,39,206,11,33],"maker\u002Fneutral\u002Fnone",[74,33,32,208,75,39],"project\u002Fpositive\u002Fidea",[115,33,42,210,90,39],"project\u002Fmixed\u002Fidea",[184,33,39,212,11,33],"project\u002Fneutral\u002Fexecution",[214,33,39,215,75,39],"46341492","commenter\u002Fneutral\u002Fnone",[217,34,42,218,219,39],"46276114","project\u002Fneutral\u002Fidea","jtokoph",[187,34,39,221,11,33],"maker\u002Fneutral\u002Fidea",[148,34,39,200,101,39],[224,34,39,200,225,39],"46321034","armenarmen",[151,34,39,200,152,39],[155,34,39,200,156,39],[177,34,39,200,152,39],[162,34,39,200,163,39],[231,42,39,200,232,39],"46392335","johanyc",[82,31,32,208,83,39],[129,32,42,218,130,39],[97,32,39,236,11,33],"maker\u002Fneutral\u002Fexecution",[238,239,32,208,240,39],"46322266",6,"sbinnee",[242,243,39,200,244,39],"46322586",7,"NiloCK",[89,246,42,212,90,39],8,[100,248,42,212,101,39],9,[137,248,39,200,90,39],[107,251,33,252,108,39],10,"project\u002Fcritical\u002Fdesign",[121,254,42,212,122,39],11,[62,254,39,236,11,33],[78,35,32,208,79,39],[],[259,264,269,274,279,284],{"id":260,"name":261,"points":262,"category":263},"46290349","Picknplace.js",455,"languages",{"id":265,"name":266,"points":267,"category":268},"46294144","fuzzycanary",373,"security_privacy",{"id":270,"name":271,"points":272,"category":273},"46234734","bidicalc",252,"dev_tools",{"id":275,"name":276,"points":277,"category":278},"46234186","Sim",240,"ai",{"id":280,"name":281,"points":282,"category":283},"46249538","uvm32",196,"infrastructure",{"id":285,"name":286,"points":287,"category":273},"46276002","Sqlit",190,false,"v2",{"id":291,"name":5,"line":6,"ai":288,"cat":20,"year":13,"points":14,"comments":15,"rec":28,"s":292,"top":18,"kw":293,"maker":11,"domain":10,"tier":19,"attempts":39,"strip":294,"url":9,"mix":295,"praise":70,"objection":94,"sub":21},46274287,0.14,[21,24,25,26,27],"10005300300000000530503301305",[31,32,33,34],1793991136470]