Overview
Liquid AI ndeye MIT spinout chivakwa Liquid Foundation Models (LFMs) iyo inosiya yakajairwa Transformer yeakasimba-masystem-akafuridzirwa ekuvaka. Chinangwa idiki, chinokurumidza, chiyeuchidzo-chinoshanda mhando dzinomhanya pamafoni uye edge zvishandiso pasina kupa yakawandisa mhando.
Liquid AI uye Liquid Foundation Models inonzwisiswa zvakanyanya mumamiriro ehurongwa, kuwana modhi, sarudzo dzepuratifomu, uye ecosystem kudyidzana.
Deep Dive
Liquid AI yakatangwa mu2023 naRamin Hasani, Mathias Lechner, Alexander Amini, uye Daniela Rus, boka reMIT CSAIL shure kwe 'liquid neural networks.' Idzo dzakabva pakudzidza nematode worm C. elegans, ine uropi hudiki 302-neuron hwakafemerwa Liquid Time-constant (LTC) network uko maitiro euron ega ega anochinja nguva nenguva kuburikidza neakasiyana equation. Zvigadzirwa zvekutengesa zveLiquid, iyo Liquid Foundation Models (LFM-1B, 3B, 40B), inogadzirisa pfungwa iyi kupfuura Transformers. Chinhu chakamira chiri pedyo-chinogara chiripo chekuyeuka tsoka pazvinokura, kusiyana neTransformers iyo inotarisisa cache mabharuni ane kutevedzana kureba. Muna 2024 kambani yakasimudza hombe Series A (yakashumwa inosvika mazana maviri nemakumi mashanu emamiriyoni emadhora) uye yakazoburitsa LFM2, yakarongedzerwa kutumirwa-pa-mudziyo pamalaptop, nhare, uye mota.
Technical Insight
Shanduko dzinochengeta kiyi-yakakosha cache inokura zvakatevedzana nehurefu hwekupinza, saka mamiriro akareba anodya ndangariro. MaLFM panzvimbo pacho anoshandisa 'mvura' computational mauniti akavakwa kubva kune yakarongwa mamiriro-nzvimbo uye dynamical-system maopareta anodzvanya ruzivo rwekare kuita yakagadziriswa-saizi inodzokororwa mamiriro. Computation inotsanangurwa neinoenderera-nguva equations iyo paramita (senge nguva inochinja) inochinjika kune yekuisa, ichirega modhi inobata kutevedzana kwakareba neinenge flat memory uye inofanotaurwa latency, iyo yakanakira resource-limited edge Hardware.
Mastering Liquid AI uye Liquid Foundation Models
Liquid AI ndeye MIT spinout chivakwa Liquid Foundation Models (LFMs) iyo inosiya yakajairwa Transformer yeakasimba-masystem-akafuridzirwa ekuvaka. Chinangwa idiki, chinokurumidza, chiyeuchidzo-chinoshanda mhando dzinomhanya pamafoni uye edge zvishandiso pasina kupa yakawandisa mhando. Liquid AI uye Liquid Foundation Models inonzwisiswa zvakanyanya mumamiriro ehurongwa, kuwana modhi, sarudzo dzepuratifomu, uye ecosystem kudyidzana. Kuti uvake kunzwisisa kwakadzama, bata Liquid AI uye Liquid Foundation Models semuenzaniso wekushandisa, kwete chinhu chimwe chete: tsanangura zvinodikanwa, kujekesa fungidziro, uye patsanura izvo system inogona kuita nekuvimbika kubva kune ichiri kuda kutonga kwenyanzvi.
Mukuita, zvikwata zvakasimba zvinoshandisa Liquid AI uye Liquid Foundation Models inoongorora zano revatengesi, kuvimbika kwemepu yemugwagwa, uye njodzi yekuvhara isati yaita. Ivo vanonyora zvakajeka maitiro ebudiriro, bvunzo vachipokana ne data rechokwadi uye mafambiro ebasa, uye iterate zvichibva pane zvakacherechedzwa maitiro ekutadza kwete kuhwina-nguva imwe chete yebhenji. Apa ndipo apo kunzwisisa kwe theoretical kunoshanduka kuve kugona kwakasimba pane chigadzirwa, mutemo, uye mashandiro.
Mamepu emigwagwa emutengesi anopesvedzera izvo izvo timu yako inogona kugadzira inotevera. Panguva imwecheteyo, zviziviso zveLaunch zvinogona kupfuura kugadzikana mune chaiyo yekugadzira workflows. Nzira yakatsiga ndeyekubatanidza kukurumidza kuyedza nekutonga: mhanyisa vatyairi vendege, tora humbowo, buritsa matanda esarudzo, uye urambe uchivandudza chengetedzo semaitiro emuenzaniso, zvinotarisirwa nemushandisi, uye zvinodikanwa zvekutonga.
Strategic Impact
Mamepu emigwagwa emutengesi anopesvedzera izvo izvo timu yako inogona kugadzira inotevera.
Mamepu emigwagwa emutengesi anopesvedzera izvo izvo timu yako inogona kugadzira inotevera. Mukutumirwa kwemhando yepamusoro, izvi zvinoshandurirwa kuita mitemo inoyerwa yekushanda, miganhu yevaridzi, uye tsika dzekudzokorora dzinodzokororwa kuitira kuti zvikwata zvikwire kuvimba pane kukwidza kusajeka.
Mamiriro ezvekutengeserana uye sarudzo dzekuendesa dzinokanganisa mutengo wenguva refu uye njodzi.
Mamiriro ezvekutengeserana uye sarudzo dzekuendesa dzinokanganisa mutengo wenguva refu uye njodzi. Mukutumirwa kwemhando yepamusoro, izvi zvinoshandurirwa kuita mitemo inoyerwa yekushanda, miganhu yevaridzi, uye tsika dzekudzokorora dzinodzokororwa kuitira kuti zvikwata zvikwire kuvimba pane kukwidza kusajeka.
Kambani inokurudzira inogadzirisa kusarudzika kwechigadzirwa, mamiriro ekuchengetedza, uye kuvhurika.
Kambani inokurudzira inogadzirisa kusarudzika kwechigadzirwa, mamiriro ekuchengetedza, uye kuvhurika. Mukutumirwa kwemhando yepamusoro, izvi zvinoshandurirwa kuita mitemo inoyerwa yekushanda, miganhu yevaridzi, uye tsika dzekudzokorora dzinodzokororwa kuitira kuti zvikwata zvikwire kuvimba pane kukwidza kusajeka.
Real-World Implementation
Kumhanyisa anokwanisa kutaura mubatsiri zvachose pasina Indaneti pa-smartphone kuitira kushandiswa kwekuvanzika
Kumisikidza kunzwisiswa kwemutauro wakaderera mumotokari yezvidzoreso zvezwi pasina makore-kufamba-nzendo
Kugadzira magwaro akareba kwazvo kana matanda palaptop uko Transformer's memory cache ingave yakakura kwazvo
Powering edge marobhoti uye IoT zvishandiso uko yekutanga C. elegans-yakafemerwa mvura network inokunda pakuenderera kwekutonga.
Maitiro Ekuita
Liquid AI uye Liquid Foundation Models mukuita
Kumhanyisa anokwanisa kutaura mubatsiri zvachose pasina Indaneti pa-smartphone kuitira kushandiswa kwekuvanzika.
Kumhanyisa mubatsiri wekutaura anokwanisa zvachose pasina pamhepo pa-smartphone kuitira kuvanzika-sensitive kushandisa Matimu anowanzo kuwana mhedzisiro iri nani kana achinge atsanangura emhando yepamusoro kumberi, chengetedza nzira yekukwira kwevanhu yemakesi ekumucheto, uye kuteedzera zvese zvakawanikwa zvechigadzirwa uye mutengo wekukanganisa nekufamba kwenguva.
Liquid AI uye Liquid Foundation Models mukuita
Kumisikidza kunzwisiswa kwemutauro wakaderera mumotokari yezvidzoreso zvezwi pasina makore-kufamba-nzendo.
Kupinza yakaderera-latency mutauro kunzwisisa mumotokari dzekudzora inzwi pasina gore-kutenderera-nzendo Zvikwata zvinowanzowana mibairo iri nani kana vachitsanangudza mhando kumberi, chengetedza nzira yekukwira kwevanhu yemakesi emupendero, uye kuteedzera zvese zvakawanikwa zvechigadzirwa nemitengo yekukanganisa nekufamba kwenguva.
Liquid AI uye Liquid Foundation Models mukuita
Kugadzira magwaro akareba kwazvo kana matanda palaptop uko Transformer's memory cache ingave yakakura kwazvo.
Kugadzira magwaro akareba kwazvo kana matanda palaptop uko Transformer's memory cache ingave yakakura kwazvo Matimu anowanzo kuwana mhedzisiro iri nani kana achinge atsanangura emhando yepamusoro kumberi, chengetedza nzira yekukwira kwevanhu yemakesi ekumucheto, uye kuteedzera zvese zvakawanikwa zvechigadzirwa uye kukanganisa mutengo nekufamba kwenguva.
Liquid AI uye Liquid Foundation Models mukuita
Powering edge marobhoti uye IoT zvishandiso uko yekutanga C. elegans-yakafemerwa mvura network inokunda pakuenderera kwekutonga.
Powering edge marobhoti neIoT zvishandiso uko yekutanga C. elegans-yakafuridzirwa mvura network inokunda pakuenderera kwekutonga Matimu anowanzo kuwana mhedzisiro iri nani kana achinge atsanangura emhando yepamusoro kumberi, chengetedza nzira yekukwira kwevanhu yemakesi emupendero, uye kuteedzera zvese zvakawanikwa zvechigadzirwa uye mutengo wekukanganisa nekufamba kwenguva.
Njodzi & Guardrails
Zviziviso zvekutanga zvinogona kupfuura kugadzikana mune chaiyo yekugadzira workflows.
Mitengo yeAPI kana shanduko yepolicy inogona kukanganisa fungidziro husiku.
Kutsamira kune mumwe-mutengesi kunowedzera kukiya-mukati uye mari yekufambisa.
Implementation Roadmap
Ongorora vanopa uchishandisa ako ega mabasa uye dataset.
Ongorora vanopa uchishandisa ako ega mabasa uye dataset. Bata nhanho yega yega segedhi rehumbowo: kana maitiro asina kusangana, imbomira kuburitsa, vhara gaka, uye wobva wawedzera kushandiswa.
Wongorora zvakavanzika, chengetedzo, uye mazwi emutemo usati wabatanidzwa.
Wongorora zvakavanzika, chengetedzo, uye mazwi emutemo usati wabatanidzwa. Bata nhanho yega yega segedhi rehumbowo: kana maitiro asina kusangana, imbomira kuburitsa, vhara gaka, uye wobva wawedzera kushandiswa.
Chengetedza chirongwa chekudzokera kumashure kune mamodheru kana vatengesi.
Chengetedza chirongwa chekudzokera kumashure kune mamodheru kana vatengesi. Bata nhanho yega yega segedhi rehumbowo: kana maitiro asina kusangana, imbomira kuburitsa, vhara gaka, uye wobva wawedzera kushandiswa.
Tarisa zvinyorwa zvekuburitsa kuitira kuti shanduko yemigwagwa isashamise zvikwata.
Tarisa zvinyorwa zvekuburitsa kuitira kuti shanduko yemigwagwa isashamise zvikwata. Bata nhanho yega yega segedhi rehumbowo: kana maitiro asina kusangana, imbomira kuburitsa, vhara gaka, uye wobva wawedzera kushandiswa.