<!DOCTYPE html><html lang="en" xmlns="http://www.w3.org/1999/xhtml" xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" style="font-size:16px;"><head></head><head><meta charset="utf-8"/><!--[if !mso]><!--><meta http-equiv="X-UA-Compatible" content="IE=edge"/><!--<![endif]--><meta name="viewport" content="width=device-width,initial-scale=1"/><meta name="x-apple-disable-message-reformatting"/><meta name="format-detection" content="telephone=no,address=no,email=no,date=no,url=no"/><meta name="color-scheme" content="light"/><meta name="supported-color-schemes" content="light"/><title>LLM That Can Modify Itself?</title><!--[if mso]><xml><o:OfficeDocumentSettings><o:AllowPNG/><o:PixelsPerInch>96</o:PixelsPerInch></o:OfficeDocumentSettings></xml><![endif]--><style>
:root { color-scheme: light; supported-color-schemes: light; }
body { margin: 0; padding: 0; min-width: 100%!important; -ms-text-size-adjust: 100% !important; -webkit-transform: scale(1) !important; -webkit-text-size-adjust: 100% !important; -webkit-font-smoothing: antialiased !important; }
.body { word-wrap: normal; word-spacing:normal; }
table.mso { width: 100%; border-collapse: collapse; padding: 0; table-layout: fixed; }
img { border: 0; outline: none; }
table { mso-table-lspace: 0px; mso-table-rspace: 0px; }
td, a, span { mso-line-height-rule: exactly; }
#root [x-apple-data-detectors=true],
a[x-apple-data-detectors=true],
#MessageViewBody a { color: inherit !important; text-decoration: inherit !important; font-size: inherit !important; font-family: inherit !important; font-weight: inherit !important; line-height: inherit !important; }
span.MsoHyperlink { color: inherit !important; mso-style-priority: 99 !important; }
span.MsoHyperlinkFollowed { color: inherit !important; mso-style-priority: 99 !important; }
.a { background-color:#dedede; }
.b { background-color:#2a2a2a; }
.c { background-color:#ffffff; }
.d { background-color:#fff0c8; }
.d2 { background-color:#FFFFFF; }
.d3 { background-color:#FFFFFF; }
h1 a { text-decoration:none;color:#2C81E5;font-style:italic; }
h2 a { text-decoration:none;color:#2C81E5;font-style:italic; }
h3 a { text-decoration:none;color:#2C81E5;font-style:italic; }
h4 a { text-decoration:none;color:#2C81E5;font-style:italic; }
h5 a { text-decoration:none;color:#2C81E5;font-style:italic; }
h6 a { text-decoration:none;color:#2C81E5;font-style:italic; }
h1, h1 a, h2, h2 a, h3, h3 a, h4, h4 a, h5, h5 a, h6, h6 a, ul, li, ol, p, p a { margin: 0;padding: 0; }
h1 { font-family:'Trebuchet MS','Lucida Grande',Tahoma,sans-serif;font-weight:700;font-size:28px;color:#2A2A2A;line-height:42px;padding-bottom:4px;padding-top:16px;mso-margin-top-alt:16px;mso-margin-bottom-alt:4px }
h2 { font-family:'Trebuchet MS','Lucida Grande',Tahoma,sans-serif;font-weight:700;font-size:24px;color:#2A2A2A;line-height:36px;padding-bottom:4px;padding-top:16px;mso-margin-top-alt:16px;mso-margin-bottom-alt:4px }
h3 { font-family:'Trebuchet MS','Lucida Grande',Tahoma,sans-serif;font-weight:400;font-size:20px;color:#2A2A2A;line-height:30px;padding-bottom:4px;padding-top:16px;mso-margin-top-alt:16px;mso-margin-bottom-alt:4px }
h4 { font-family:'Trebuchet MS','Lucida Grande',Tahoma,sans-serif;font-weight:400;font-size:18px;color:#2A2A2A;line-height:27px;padding-bottom:4px;padding-top:16px;mso-margin-top-alt:16px;mso-margin-bottom-alt:4px }
h5 { font-family:'Trebuchet MS','Lucida Grande',Tahoma,sans-serif;font-weight:400;font-size:16px;color:#2A2A2A;line-height:24px;padding-bottom:4px;padding-top:16px;mso-margin-top-alt:16px;mso-margin-bottom-alt:4px }
h6 { font-family:'Trebuchet MS','Lucida Grande',Tahoma,sans-serif;font-weight:400;font-size:14px;color:#2A2A2A;line-height:21px;padding-bottom:4px;padding-top:16px;mso-margin-top-alt:16px;mso-margin-bottom-alt:4px }
p { font-family:'Georgia','Times New Roman',serif;font-weight:400;color:#2D2D2D;font-size:16px;line-height:24px;padding-bottom:8px;padding-top:8px;mso-margin-top-alt:8px;mso-margin-bottom-alt:8px; }
p a, .e a, ul a, li a, .h a, .h2 a, .h3 a { word-break:break-word;color:#2C81E5 !important;text-decoration:none;font-style:italic; }
p a span, .e a span, ul a span, li a span { color: inherit }
p .bold { font-weight:bold;color:#2D2D2D; }
p span[style*="font-size"] { line-height: 1.6; }
.f p { font-size:12px;line-height:15px;color:#2D2D2D;padding:0; }
.f p a { color:#2D2D2D !important; }
.g p { font-family:'Helvetica',Arial,sans-serif;font-size:14px;line-height:20px;font-weight:normal;margin:0; }
.g p a { text-decoration: underline; }
.i p { font-family:'Helvetica',Arial,sans-serif;line-height:23px;font-size:15px;color:#2D2D2D; }
.i p a { color:#2D2D2D !important; }
.i2 p { font-family:'Helvetica',Arial,sans-serif;line-height:23px;font-size:15px;color:#2D2D2D; }
.i2 p a { color:#2D2D2D !important; }
.i3 p { font-family:'Helvetica',Arial,sans-serif;line-height:43px;font-size:24px;color:#2D2D2D; }
.i3 p a { color:#2D2D2D !important; }
.h p a { color:#595959 !important; }
.h2 p a { color:#595959 !important; }
.h3 p a { color:#595959 !important; }
.f p a, .i p a, .i2 p a, .i3 p a, .h p a, .h2 p a, .h3 p a { text-decoration:underline; }
.j { border-top:3px solid #ffeb2d; }
.k p { padding-left:15px;padding-bottom:0px;padding-top:6px;mso-margin-top-alt:6px;mso-margin-bottom-alt:0px;mso-margin-left-alt:15px; }
.o { background-color:#FFFFFF;border:1px solid #F1F1F1;border-radius:5px; }
.o p { font-family:'Helvetica',Arial,sans-serif;padding:0px;margin:0px; }
.l p,
.l p a { font-size:14px;line-height:20px;font-weight: bold;color:#2D2D2D;padding-bottom:6px;mso-margin-bottom-alt:6px;text-decoration:none; }
.m p,
.m p a { font-size:13px;line-height:18px;font-weight:400;color:#2D2D2D;padding-bottom:6px;mso-margin-bottom-alt:6px;text-decoration:none; }
.n p,
.n p a { font-size:12px;line-height:17px;font-weight:400;color:#2D2D2D;padding-bottom:6px;mso-margin-bottom-alt:6px;text-decoration:none; }
.p { background-color:#FFFFFF;max-width:520px;border:1px solid #E1E8ED;border:1px solid rgba(80, 80, 80, 0.3);border-radius:5px; }
.q { font-size:16px;font-family:Helvetica,Roboto,Calibri,sans-serif !important;border:1px solid #e1e8ed;border:1px solid rgba(80, 80, 80, 0.3);border-radius:10px;background-color:#FFFFFF; }
.q p { font-size:16px;font-family:system-ui,Helvetica,Roboto,Calibri,sans-serif !important;color:#222222;padding:4px 0; }
.r { border:1px solid #E1E8ED !important;border-radius:5px; }
.s p { font-size: 14px; line-height: 17px; font-weight: 400; color: #697882; text-decoration: none; }
.t p { font-family:'Helvetica',Arial,sans-serif;font-size:12px;line-height:18px;font-weight:400;color:#000000;font-style:italic;padding:4px 0px 0px; }
.v { border-radius:10px;border:solid 0px #DFD150;background-color:#2C81E5;font-family:'Open Sans','Segoe UI','Apple SD Gothic Neo','Lucida Grande','Lucida Sans Unicode',sans-serif;color:#FFFFFF; }
.v a { text-decoration:none;display:block;color:#FFFFFF; }
.w p { font-size:12px;line-height:15px;font-weight:400;color:#FFFFFF; }
.w p a { text-decoration: underline !important;color:#FFFFFF !important; }
ul { font-family:'Helvetica',Arial,sans-serif;margin:0px 0px 0px 25px !important;padding:0px !important;color:#2D2D2D;line-height:24px;list-style:disc;font-size:16px; }
ul > li { font-family:'Helvetica',Arial,sans-serif;margin:10px 0px 0px 0px !important;padding: 0px 0px 0px 0px !important; color: #2D2D2D; list-style:disc; }
ol { font-family:'Helvetica',Arial,sans-serif;margin: 0px 0px 0px 25px !important;padding:0px !important;color:#2D2D2D;line-height:24px;list-style:decimal;font-size:16px; }
ol > li { font-family:'Helvetica',Arial,sans-serif;margin:10px 0px 0px 0px !important;padding: 0px 0px 0px 0px !important; color: #2D2D2D; list-style:decimal; }
.e h3,
.e p,
.e span { padding-bottom:0px;padding-top:0px;mso-margin-top-alt:0px;mso-margin-bottom-alt:0px; }
.e span,
.e li { font-family:'Helvetica',Arial,sans-serif;font-size:16px;color:#2D2D2D;line-height:24px; }
.rec { font-family: ui-sans-serif, system-ui, -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, "Helvetica Neue", Arial, "Noto Sans", sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol", "Noto Color Emoji" !important; }
.rec__button:hover { background-color: #f9fafb !important; }
.copyright a {color: inherit !important; text-decoration: none !important; font-size: inherit !important; font-family: inherit !important; font-weight: inherit !important; line-height: inherit !important;}
.txt_social p { padding: 0; word-break: break-all; }
.table, .table-c, .table-h { border: 1px solid #C0C0C0; }
.table-c { padding:5px; background-color:#FFFFFF; }
.table-c p { color: #2D2D2D; font-family:'Helvetica',Arial,sans-serif !important;overflow-wrap: break-word; }
.table-h { padding:5px; background-color:#F1F1F1; }
.table-h p { color: #2A2A2A; font-family:'Trebuchet MS','Lucida Grande',Tahoma,sans-serif !important;overflow-wrap: break-word; }
@media only screen and (max-width:667px) {
.aa { width: 100% !important; }
.bb img { width: 100% !important; height: auto !important; max-width: none !important; }
.cc { padding: 0px 8px !important; }
.ee { padding-top:10px !important;padding-bottom:10px !important; }
.ff ul, .ff ol { margin: 0px 0px 0px 10px !important;padding: 0px !important; }
.ff li { margin:10px 0px 0px 10px !important; }
.r {height:140px !important;}
.s p { font-size:13px !important;line-height:15px !important; }
.mob-hide {display:none !important;}
.mob-stack {display:block !important;width:100% !important;}
.mob-w-full {width:100% !important;}
.mob-block {display:block !important;}
.embed-img {padding:0px 0px 12px 0px !important;}
.socialShare {padding-top:15px !important;}
.rec { padding-left:15px!important;padding-right:15px!important; }
.bodyWrapper { padding:7px 4px 7px 4px !important; }
.social-mobile {float:left !important;margin-top:10px !important;}
}
@media screen and (max-width: 480px) {
u + .a .gg { width: 100% !important; width: 100vw !important; }
.tok-heart { padding-top:75% !important; }
.tok-play { padding-top: 250px !important; }
}
@media screen and (max-width: 320px) {
.tok-heart { padding-top:65% !important; }
}
.u { border: 1px solid #CACACA !important; border-radius: 2px !important; background-color: #ffffff !important; padding: 0px 13px 0px 13px !important; font-family:ui-sans-serif,system-ui,-apple-system,BlinkMacSystemFont,"Segoe UI",Roboto,"Helvetica Neue",Arial,"Noto Sans",sans-serif !important;font-size: 12px !important; color: #767676 !important; }
.u a { text-decoration: none; display: block !important; color: #767676 !important; margin: 0px !important; }
.u span, .u img { color: #767676 !important;margin:0px !important; max-height:32px !important;background-color:#ffffff !important; }
</style><!--[if mso]><style type="text/css">
h1, h2, h3, h4, h5, h6 {font-family: Arial, sans-serif !important;}
body, table, td, p, a, span {font-family: Arial, sans-serif !important;}
sup { font-size: 100% !important;vertical-align: .5em !important;mso-text-raise: -1.5% !important;line-height: 0 !important; }
ul { margin-left:0px !important; margin-right:10px !important; margin-top:20px !important; margin-bottom:20px !important; }
ul li { margin-left: 0px !important; mso-special-format: decimal; }
ol { margin-left:0px !important; margin-right:10px !important; margin-top:20px !important; margin-bottom:20px !important; }
ol li { margin-left: 0px !important; mso-special-format: decimal; }
li.listItem { margin-left:15px !important; margin-top:0px !important; }
.paddingDesktop { padding: 10px 0 !important; }
.edm_outlooklist { margin-left: -20px !important; }
.embedImage { display:none !important; }
</style><![endif]--><style>
@font-face {
font-family: 'Open Sans';
font-style: normal;
font-weight: 700;
font-display: swap;
src: url('https://fonts.gstatic.com/s/opensans/v40/memSYaGs126MiZpBA-UvWbX2vVnXBbObj2OVZyOOSr4dVJWUgsg-1x4gaVIUwaEQbjA.woff2') format('woff2');
}
@font-face {
font-family: 'Open Sans';
font-style: italic;
font-weight: 700;
font-display: swap;
src: url('https://fonts.googleapis.com/css2?family=Open+Sans:ital,wght@1,700&display=swap') format('woff2');
}
</style></head><body class="a" style="margin:0px auto;padding:0px;word-wrap:normal;word-spacing:normal;background-color:#dedede;"><div role="article" aria-roledescription="email" aria-label="email_name" lang="en" style="font-size:1rem"><div style="display:none;max-height:0px;overflow:hidden;"> Plus more about "The Diffusion Duality" and "Reinforcement Pre-Training"  ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ </div><table role="none" width="100%" border="0" cellspacing="0" align="center" cellpadding="0" class="gg"><tr><td align="center" valign="top"><table role="none" width="670" border="0" cellspacing="0" cellpadding="0" class="aa" style="width:670px;table-layout:fixed;"><tr><td class="bodyWrapper" align="center" valign="top" style="padding:7px 7px 7px 7px;"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td align="center" valign="top" style="border-width:0px 0px 0px 0px;border-style: solid; border-color: #2a2a2a;border-radius:10px 10px 0px 0px;background-color:#ffffff;" class="c"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center"><tr id="header"><td style="padding:28px 28px 0px 28px;"><div style="padding-top:0px;padding-right:0px;padding-bottom:20px;padding-left:0px;"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td class="f" align="right" valign="top"><p> June 17, 2025 | <a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.c6q0w4g5sodbtO4I1B_pxSdB5RCIH6yy1Fm1CYma3EwVR78K3s_E1oHSiNURYmTtW8dcWsxknbqk_pOlNoq_gZhHD_hT4FsZImxicyuo5M8I1a5WqmhVzQ8dlA3D6HW6vuWS_AJ2YDptK0UmUZG7-OTJBB9COUmopET7zP9x-vqmXAx1U448hzLeBQjEnBKEyaIbBbFR_DQF5UKd4Nvc-P63tjY5bxHjFFjH2YVp7yY/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h0/h001.dSSrnxLeantuO3YV-nVeJIWc1Njfx_Xxt2mWOzfule0">Read Online</a></p></td></tr><tr><td class="dd" align="center" valign="top" style="padding:15px 0;"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td align="center" valign="top"><h1 style="text-align:left;font-family:'Open Sans','Segoe UI','Apple SD Gothic Neo','Lucida Grande','Lucida Sans Unicode',sans-serif;font-weight:Bold;font-size:32px;color:#2A2A2A;padding:2px 0;line-height:38px;"> LLM That Can Modify Itself? </h1><p style="text-align:left;font-family:'Helvetica',Arial,sans-serif;font-weight:normal;font-size:20px;color:#3E3E3E;padding:5px 0;line-height:24px;"> Plus more about "The Diffusion Duality" and "Reinforcement Pre-Training" </p></td></tr></table></td></tr><tr><td style="height:0px;width:0px;"><div style="height:1px;" data-open-tracking="true"> <img src="https://elink4f7.mail.bycloud.ai/ss/o/u001.3wmUuY8gEWd4_869a_eXcg/4hf/9EAxPIReQ9K2uiAP9qmNaQ/ho.gif" alt="" width="1" height="1" border="0" style="height:1px !important;width:1px !important;border-width:0 !important;margin-top:0 !important;margin-bottom:0 !important;margin-right:0 !important;margin-left:0 !important;padding-top:0 !important;padding-bottom:0 !important;padding-right:0 !important;padding-left:0 !important;"/> </div></td></tr></table></div></td></tr><tr id="content-blocks"><td class="email-card-body" align="center" valign="top" style="padding-bottom:28px;"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td id="nov-18-th-nov-24-th-33-latest-ai-re" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:normal;padding:0px 28px;text-align:left;"><h6 style="color:#2A2A2A;font-weight:normal;mso-line-height-alt:87.5%;"><i>JUNE 10th ~ JUNE 16th</i><br><i>#60 Latest AI Research Explained Simply</i></h6></td></tr><tr><td><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" style=""><tr><td bgcolor="#222222" style="background-color:#222222;padding:0.0px 0.0px 0.0px 0.0px;"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0"><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"></p></td></tr></table></td></tr></table></td></tr><tr><td id="industry-news-in-1-line" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:Bold;padding:0px 28px;text-align:left;"><h2 style="color:#2A2A2A;font-weight:Bold;mso-line-height-alt:150.0%;">🗞️ Industry News in 1 Line</h2></td></tr><tr><td style="padding-bottom:12px;padding-left:50px;padding-right:40px;padding-top:12px;" class="ee"><div style="margin-left:0px;" class="edm_outlooklist"><ol start="1" style="list-style-type:decimal;margin:0px 0px;padding:0px 0px 0px 0px;"><li class="listItem ultext"><p style="mso-line-height-alt:150.0%;padding:0px;text-align:left;word-break:break-word;"><span style="background-color:#e0e0e0;"><span style="color:rgb(255, 58, 58);font-size:0.6rem;"> HOT </span></span> Google has officially launched the <a class="link" href="https://elink4f7.mail.bycloud.ai/ss/c/u001.oB7zuO_W-X4Toa45C28ng3N7PLrWqoSDhsskwKrBQeiF5FxniVLW1TvSb0TUwtAMoqz0Kz8xFmAQ1v0jeZGze2szNOLz-eq20Iuvgk3FQM-57Tyisf_zy3vWqhJ8y3Tomk-NUQZ-lO9oygMhW8dLCh3EWS_dZy_q8wGtyVSRmDtnJ9XWJUekqF5XPv31ZK_a/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h1/h001.mxlxCicpdcGzB97X0Cwiy-HsnaHzf_74GqOSRiUwjFA" target="_blank" rel="noopener noreferrer nofollow"><span>Gemini 2.5 family</span></a>, with an addition of Gemini 2.5 Flash Lite preview. </p><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:420px;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.oB7zuO_W-X4Toa45C28ng3N7PLrWqoSDhsskwKrBQeiF5FxniVLW1TvSb0TUwtAMoqz0Kz8xFmAQ1v0jeZGze2szNOLz-eq20Iuvgk3FQM-57Tyisf_zy3vWqhJ8y3ToJrGmy7bL85Z_jbO2NNany-3LuViLQEb4Dn5Qm8kMvyEaKFsHVpGVuRE-lLetvWHL/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h2/h001.Xgwu5kpNnUu0Tt7GtkVV9z2WQAcWtwkrrH1_EsvNN4o" target="_blank" rel="noopener noreferrer nofollow" style="text-decoration:none;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/0c6ea2be-ba77-48ff-8c88-13ef3098847c/gemini_2-5_benchmarks_margin_light2x_1.gif?t=1750177507" alt="" height="auto" width="420" style="display:block;width:100%;" border="0"/></a></td></tr><tr><td align="center" valign="top" class="t" style="width:420px;"><p>Gemini 2.5 family benchmark</p></td></tr></table></li><li class="listItem ultext"><p style="mso-line-height-alt:150.0%;padding:0px;text-align:left;word-break:break-word;"><span style="background-color:#e0e0e0;"><span style="color:rgb(255, 58, 58);font-size:0.6rem;">♥ 5.5k</span></span> OpenAI o3-pro now available for all Pro users in ChatGPT and in the API. </p><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:510px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/56dad4f6-c0f4-4fbb-b052-d12bb0d8b847/GtG4epLbMAE0pGm.png?t=1750177604" alt="" height="auto" width="510" style="display:block;width:100%;" border="0"/></td></tr><tr><td align="center" valign="top" class="t" style="width:510px;"><p>o3 Benchmark</p></td></tr></table></li><li class="listItem ultext"><p style="mso-line-height-alt:150.0%;padding:0px;text-align:left;word-break:break-word;"><span style="background-color:#e0e0e0;"><span style="color:rgb(255, 58, 58);font-size:0.6rem;">♥ 1.1k</span></span> Hailuo AI released <a class="link" href="https://elink4f7.mail.bycloud.ai/ss/c/u001.1muhFWIqieRYpaJ-FbWSCUiRXUf67lKeQoQaLwT7aa8ecHZQm4cS32_KiamKtLNVbkSRg-7iE4C4CKz06J3YACagIk_MexKIietzJwbp8TS8UuGtVtzAon2l6FgBxKwoBKGViz1LYyIZmwrpeT6jLQ/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h3/h001.uyCeqT6P9MdzB7TN4XxXjPOJ4jNVGO2Xr5AARREuWz0" target="_blank" rel="noopener noreferrer nofollow"><span>MiniMax-M1</span></a>, which is a 456B hybrid MoE model with lightning attention, supporting 1M-token contexts. Currently it is a bit <a class="link" href="https://elink4f7.mail.bycloud.ai/ss/c/u001.1muhFWIqieRYpaJ-FbWSCRL4eJbpht1CQdP96tsHazdUWrf3QssbjYyAO6gOKY4ZbVVyQBD9iT8shntRBDTxXcv72jsMIgdJpppDCykPz0_GyAvDNjb5v-d7TNsIqvPsp_HCXsg9RdrmpMjueslEZg/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h4/h001.ceQWTr5Q5BLhmLP-VDxlbThNcIwrKizfTkKYvnBT_HU" target="_blank" rel="noopener noreferrer nofollow"><span>controversial</span></a> due to how they didn’t properly compare the test-time scaling vs FLOPs performance. </p><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:450px;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.VomAAYwkCjux8i_FMc4kJXVKJUDFWFdYMzacFEcbOUbrOgwzGlE0h0O7mwYjJIJFuZ1b6z8lOTgScJdetHvTa1JpUn1LJ8pkdrzgWNQvCiO4jB8Hz3srfg0UK4c_cnXCtb3Yx-88wbE63E8ypS3QwBftw9wk1Izw6GtGuVfgbFWQcmLgOrAXH8DfhOG0ic5t/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h5/h001.i0YoQR3sY4VXVlkDQ2LQXZIeCdSLTEGzrOfZTAlUamc" target="_blank" rel="noopener noreferrer nofollow" style="text-decoration:none;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/95c24434-760a-444b-90b1-1a350437e2f6/Screenshot_2025-06-17_122359.png?t=1750177708" alt="" height="auto" width="450" style="display:block;width:100%;" border="0"/></a></td></tr><tr><td align="center" valign="top" class="t" style="width:450px;"><p>MiniMax-M1</p></td></tr></table></li><li class="listItem ultext"><p style="mso-line-height-alt:150.0%;padding:0px;text-align:left;word-break:break-word;"><span style="background-color:#e0e0e0;"><span style="color:rgb(255, 58, 58);font-size:0.6rem;">♥ 3.7k</span></span> <a class="link" href="https://elink4f7.mail.bycloud.ai/ss/c/u001.DUiN96-Eq7pUHzwEhy5j2yo571wOI2ayM48Skcu9lJLbYBo25FTwAYdRr4o2NPN19ehmeItiqG2IPtJrSYpI05tqaGYFMjgDZ2_WcYLtzQZky_X3zISnLL9cp8EuCk1edjc_FIeMoMFpM2u13ywtp2liiuzBCX3i0RhicrrFTgvECI330Cf1RIWkpG2ENjr47IRSCTxqpmOhvht8RqMR0g/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h6/h001.0ewYf3gKAbI-CQIyNTkJs6zGqN0BRainPpSjOpMCq7Y" target="_blank" rel="noopener noreferrer nofollow"><span>New Anthropic Research</span></a> dives in how they build a multi-agent system where LLMs can autonomously coordinate tool usage and parallelize web searches. Offering insights into scalable system architecture, agent reliability, and prompt design for production-ready deployments. </p><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:510px;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.DUiN96-Eq7pUHzwEhy5j2yo571wOI2ayM48Skcu9lJLbYBo25FTwAYdRr4o2NPN19ehmeItiqG2IPtJrSYpI05tqaGYFMjgDZ2_WcYLtzQZky_X3zISnLL9cp8EuCk1eyn6Rb11Tk1XlferlZeZR4mDPrBXiboQaBv5gMHdWv2UZpxEP1MOK9yj2z24p7gcuFgqXeanTbQqkB31jbM4ptg/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h7/h001.ZsUC8WG_x246HFGiO-TyBjsyd4AbIvvjnB7RFIlm5Fo" target="_blank" rel="noopener noreferrer nofollow" style="text-decoration:none;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/8e3fb975-88a0-42f3-98c0-fbb2a3eda369/1198befc0b33726c45692ac40f764022f4de1bf2-4584x2579.jpg?t=1750177937" alt="" height="auto" width="510" style="display:block;width:100%;" border="0"/></a></td></tr><tr><td align="center" valign="top" class="t" style="width:510px;"><p> multi-agent system diagram</p></td></tr></table><p style="mso-line-height-alt:150.0%;padding:0px;text-align:left;word-break:break-word;"></p></li></ol></div></td></tr><tr><td><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" style=""><tr><td bgcolor="#222222" style="background-color:#222222;padding:0.0px 0.0px 0.0px 0.0px;"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0"><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"></p></td></tr></table></td></tr></table></td></tr><tr><td><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" style=""><tr><td bgcolor="transparent" style="background-color:transparent;border-color:#2C81E5;border-style:solid;border-width:5px;padding:0.0px 0.0px 0.0px 0.0px;"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0"><tr><td class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:Bold;padding:0px 28px;text-align:left;"><h2 style="color:#2A2A2A;font-weight:Bold;mso-line-height-alt:150.0%;"><span style="">The AI Timeline: Premium Insights</span></h2></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"><span style="color:rgb(34, 34, 34);font-family:Georgia, "Times New Roman", serif;font-size:16px;">Recently, we introduced a premium membership for The AI Timeline! </span></p></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"><span style="color:rgb(34, 34, 34);font-family:Georgia, "Times New Roman", serif;font-size:16px;">With the membership, you would receive exclusive insights/explainers into technical AI topics and monthly research trend reports that contains my analysis of up to 40+ papers.</span></p></td></tr><tr><td align="center" valign="top" class="dd"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.c6q0w4g5sodbtO4I1B_pxSdB5RCIH6yy1Fm1CYma3EzRMZU7syrc4B7ZADDNO77fF1CP9dNVR_P9aEWpRp8MUJs4-GxSOtnm-_GrfclATvzA6KgzXx0OKmlBOTkzUXcWuUvIqb69Luo8oST5epIEF1L5eLtOK2SFiI5C3CpPyH1J1vbMfrQi_LEevLYMQI3vu0yrEjOZOs9VlZ6Aw-R4SQmNjl9X0TUEx42hD3durRY/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h8/h001.HXrrrHlteVmugdhm5f26Ea_neKowwt5YoxTxHrVEUt0" style="text-decoration:none;" target="_blank"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center" style="margin-bottom:12px;margin-top:12px;padding-left:12px;padding-right:12px;"><tr><td align="center" valign="top" class="o" style="padding:12px 12px 12px 12px;;background-color:#FFFFFF;border-color:#F1F1F1;border-radius:5px 5px 5px 5px;border-width:1px 1px 1px 1px;"><table role="none" border="0" cellspacing="0" cellpadding="0" align="right" width="100%"><tr><!--[if mso]><td width="0"><table cellpadding="0" cellspacing="0" border="0" role="presentation" style="display:none;"><tr><![endif]--><td class="embed-img mob-stack" align="center" valign="top" style="width:35%;min-height:100px;vertical-align:middle;display:none;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.c6q0w4g5sodbtO4I1B_pxSdB5RCIH6yy1Fm1CYma3EzRMZU7syrc4B7ZADDNO77fF1CP9dNVR_P9aEWpRp8MUEodPNBwgKzPBQvZX63W3wlH_CQ84U-MKW9KM6tyxKLmBqgYIZojPTieGcmRSpceKN0OtF2rMRFtCnYPgtLRe1b3EJ8pp6XyrLC3xK81m7nPh5QkDc5GYECp4PkepT0jK7lyRU7ODccKg1qAIlYmWhU/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h9/h001.v-s_s06BWjiEpcqfsfdfsH93VPtYzh8vtp5K882wjn4" style="text-decoration:none;" target="_blank"><img src="https://beehiiv-images-production.s3.amazonaws.com/uploads/asset/file/ec385139-4e87-42bb-933d-9a33db719a7f/May_2025_research_trend_report.jpg?t=1748926968" width="100%" style="display:block;"/></a></td><!--[if mso]></tr></table></td><![endif]--></tr><tr><td align="center" valign="top" class="cc"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td align="left" valign="top" class="l"><p>Research Trend Report</p></td></tr><tr><td align="left" valign="top" class="m"><p>May 2025</p></td></tr><tr><td align="left" valign="bottom" class="n" style="vertical-align:bottom;padding-top:12px;"><p style="word-break:break-word;">https://mail.bycloud.ai/p/may-2025-research-trend-report</p></td></tr></table></td><!--[if mso]><td width="0"><table cellpadding="0" cellspacing="0" border="0" role="presentation" style="display:none;"><tr><![endif]--><td class="mob-hide" align="center" valign="top" style="width:35%;min-height:100px;padding:0px 0px 0px 12px;vertical-align:middle;"><img src="https://beehiiv-images-production.s3.amazonaws.com/uploads/asset/file/ec385139-4e87-42bb-933d-9a33db719a7f/May_2025_research_trend_report.jpg?t=1748926968" width="100%" style="display:block;"/></td><!--[if mso]></tr></table></td><![endif]--></tr></table></td></tr></table></a></td></tr><tr><td align="center" valign="top" class="dd"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.c6q0w4g5sodbtO4I1B_pxSdB5RCIH6yy1Fm1CYma3ExPOg33txZ1eVAfVJaQKmxhdy-hRaFi6eQTvWoxLPaZXPYxlTcgeXPgGuuQ8rCIx1uJ0yY1m6TsrQmsAdzcVUjRZzXqAMHLQs4tUtcrPmPnMJ74ODhnnNTiuUF10kqy_d9VXEMpNKQ8FQyFN8_ynRGBit8W7wjeg7RH9iFN_E8SLAfRIsjFVVW-jS3MSyshIAC889sBhyd2g1pUv5-SJcHT/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h10/h001.wYpiY7-MYoKStL2FB-AqQQ3-HYzNoR_1k-m6iDx4XQA" style="text-decoration:none;" target="_blank"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center" style="margin-bottom:12px;margin-top:12px;padding-left:12px;padding-right:12px;"><tr><td align="center" valign="top" class="o" style="padding:12px 12px 12px 12px;;background-color:#FFFFFF;border-color:#F1F1F1;border-radius:5px 5px 5px 5px;border-width:1px 1px 1px 1px;"><table role="none" border="0" cellspacing="0" cellpadding="0" align="right" width="100%"><tr><!--[if mso]><td width="0"><table cellpadding="0" cellspacing="0" border="0" role="presentation" style="display:none;"><tr><![endif]--><td class="embed-img mob-stack" align="center" valign="top" style="width:35%;min-height:100px;vertical-align:middle;display:none;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.c6q0w4g5sodbtO4I1B_pxSdB5RCIH6yy1Fm1CYma3ExPOg33txZ1eVAfVJaQKmxhdy-hRaFi6eQTvWoxLPaZXJ-PFkw3tGErU3BE-pWiX5PE8Avhvcq-A13TPrPO8apTnH8CbGbWf6nqcFNEmdEwj8q-YqfOsZGWXNXYUpR5ybi5Z5BRkttkYUi42HZw2XRLoOOFYL1CyqUC-Fp-AzLY0fsJvN2ohWemao1j4_zULaTVT0hAlUA84lL4aUKyzeC9/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h11/h001.ghWcQj27w6fsLREYmkVyE_a020F9W5Sw1UzUzKxwIzU" style="text-decoration:none;" target="_blank"><img src="https://beehiiv-images-production.s3.amazonaws.com/uploads/asset/file/9a3842a5-880f-4f74-b950-e7a687ce9f40/premium_insights_fp8.jpg?t=1749757642" width="100%" style="display:block;"/></a></td><!--[if mso]></tr></table></td><![endif]--></tr><tr><td align="center" valign="top" class="cc"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td align="left" valign="top" class="l"><p>Understanding Floating Points in LLMs</p></td></tr><tr><td align="left" valign="top" class="m"><p>Introduction into Floating Points through DeepSeek-V3 </p></td></tr><tr><td align="left" valign="bottom" class="n" style="vertical-align:bottom;padding-top:12px;"><p style="word-break:break-word;">mail.bycloud.ai/p/understanding-floating-points-in-llms</p></td></tr></table></td><!--[if mso]><td width="0"><table cellpadding="0" cellspacing="0" border="0" role="presentation" style="display:none;"><tr><![endif]--><td class="mob-hide" align="center" valign="top" style="width:35%;min-height:100px;padding:0px 0px 0px 12px;vertical-align:middle;"><img src="https://beehiiv-images-production.s3.amazonaws.com/uploads/asset/file/9a3842a5-880f-4f74-b950-e7a687ce9f40/premium_insights_fp8.jpg?t=1749757642" width="100%" style="display:block;"/></td><!--[if mso]></tr></table></td><![endif]--></tr></table></td></tr></table></a></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"><span style="color:rgb(34, 34, 34);font-family:Georgia, "Times New Roman", serif;font-size:16px;">Plus, we are also scheduling more content in the future, so don’t miss out!</span></p></td></tr><tr><td align="center" valign="top" style="padding-bottom:14px;padding-left:28px;padding-right:28px;padding-top:14px;text-align:center;width:100%;word-break:break-word;" class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" align="center" style="margin:14px auto 14px auto;"><tr><td align="center" valign="middle" height="44.75" style="height:44.75px;background-color:#2C81E5;border-color:#DFD150;border-radius:10px 10px 10px 10px;border-style:solid;border-width:0px 0px 0px 0px;color:#FFFFFF;font-family:'Open Sans','Segoe UI','Apple SD Gothic Neo','Lucida Grande','Lucida Sans Unicode',sans-serif;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.c6q0w4g5sodbtO4I1B_pxcTFNXiisgoxAScOL72N4DHVytoeFQ-mBBAkS5Q6m2B0JsVl7Ceocxpip3rVxsdDNU-abgLsphn3E_51ocYe78bpPmEhUU8HruSVyMtxjK0WyvPFT05xO4_m-Ra5-m3Np9R9Bgr6lgl61AHePhdhsS-X0LJJKWGwtE1YLe4k92AXsmrE9MKeb1AAg9am9gwgRA/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h12/h001.uoGDT4Mk_GyGZZyirbFMOht6Angy0PToh96SJT6myEg" target="_blank" rel="noopener noreferrer nofollow" style="color:#FFFFFF;display:block;font-size:16px;font-size:16px;font-weight:normal;padding:0px 14px;padding:14px 14px 14px 14px;text-decoration:none;"> Upgrade To Premium Insights </a></td></tr></table></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"><span style=""><sub><a class="link" href="https://elink4f7.mail.bycloud.ai/ss/c/u001.tLfGW26lAwaS9gFg17HSoGymQ3NNPtd5dE5MV_8UgjIDFPVXngz8pvQBldSW42yhUe_Qiq6DgEPMEBuPL9yfRpXelTiuu2kS8pLFvsoem_XoZoy_n13sTKUhZIbl0VH6/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h13/h001.Afc8LYIlOoxqI5PfOEiVDdGFWGk6OLogbqJJCLPqqrY" target="_blank" rel="noopener noreferrer nofollow"><span>Advertise with The AI Timeline!</span></a></sub></span></p></td></tr></table></td></tr></table></td></tr><tr><td><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" style=""><tr><td bgcolor="#222222" style="background-color:#222222;padding:0.0px 0.0px 0.0px 0.0px;"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0"><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"></p></td></tr></table></td></tr></table></td></tr><tr><td id="the-diffusion-duality" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:Bold;padding:0px 28px;text-align:left;"><h2 style="color:#2A2A2A;font-weight:Bold;mso-line-height-alt:150.0%;">The Diffusion Duality </h2></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"><span style=""><i>Sahoo et al. [Cornell Tech, EPFL Lausanne]</i></span></p></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"><span style="background-color:#e0e0e0;"><span style="color:rgb(255, 58, 58);font-size:0.6rem;"> ♥ 365 </span></span><span style="color:rgb(44, 129, 229);font-size:0.6rem;"> </span><span style="background-color:#e0e0e0;"><span style="color:rgb(44, 129, 229);font-size:0.6rem;"> LLM Diffusion </span></span><span style="color:rgb(44, 129, 229);font-size:0.6rem;"> </span><span style="background-color:#e0e0e0;"><span style="color:rgb(44, 129, 229);font-size:0.6rem;"> bycloud’s pick </span></span><span style="color:rgb(44, 129, 229);font-size:0.6rem;"> </span></p></td></tr><tr><td id="unlocking-faster-text-generation-wi" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:normal;padding:0px 28px;text-align:left;"><h3 style="color:#2A2A2A;font-weight:normal;mso-line-height-alt:125.0%;"><span style="color:rgb(67, 67, 67);">Unlocking Faster Text Generation with Diffusion Duality</span></h3></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> We have seen recent papers where researchers have tried to create text generation models using diffusion processes. These models promise efficient self-correction, but discrete variants like Uniform-state Diffusion Models (USDMs) are often worse than autoregressive and masked diffusion approaches in speed and quality. </p></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> The biggest challenge with the USDMs approach is lack the advanced training and sampling techniques that power their continuous counterparts. This paper introduces Duo, which is a method that bridges Gaussian and discrete diffusion by revealing a hidden connection: discrete states naturally emerge from underlying Gaussian processes. This duality unlocks game-changing optimizations, accelerating both training and sampling while closing the performance gap. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/938b1fb4-3aef-4371-a72b-b1659f88adab/duo_schematic_graphical.png?t=1750174561" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td id="how-duo-rewires-diffusion-mechanics" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:normal;padding:0px 28px;text-align:left;"><h3 style="color:#2A2A2A;font-weight:normal;mso-line-height-alt:125.0%;"><span style="color:rgb(67, 67, 67);">How Duo Rewires Diffusion Mechanics</span></h3></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> The Duo mechanism maps the Gaussian diffusion latents to discrete states via a simple <span style="font-weight:700;"><b>argmax operation</b></span>. When this transformation is applied to noisy Gaussian vectors, argmax transforms them into categorical distributions matching USDMs, with diffusion parameters remapped through a <span style=""><i>diffusion transformation operator</i></span>. This approach has several practical implications, first, <span style="font-weight:700;"><b>curriculum learning</b></span> leverages the Gaussian backbone to reduce training variance. By starting with a tempered softmax approximation of argmax (easing reconstruction) and gradually hardening it to true argmax, models learn faster. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/f6b9ff09-ad05-4e1d-b1b5-18388676167c/image.png?t=1750174627" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> Second, the duality enables <span style="font-weight:700;"><b>Discrete Consistency Distillation (DCD)</b></span>. Since USDMs lack deterministic Probability Flow ODEs, Duo constructs proxy trajectories in Gaussian space: clean data and noise are combined into continuous paths, then discretized via argmax. A student model distills knowledge from a teacher by matching output distributions across these discrete points, by skipping stochastic sampling. This results in reduction of sampling steps and speeds up the entire workflow. </p></td></tr><tr><td align="center" valign="top" class="dd"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.qZVn6KJQuMivjuNasJr7ICx0t4Bjk7tE9zMQyBNaMqJr_eL2rl0QZALsCDkB8KyWhnvk8dX00L9D5vFnB8AivLP7seGY2JEIxDp-sHj6nCk/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h14/h001.yL0Hd7DauQ40HfiaqUJcKbXrhrsXTMJvAvOaaYH1Wlg" style="text-decoration:none;" target="_blank"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center" style="margin-bottom:12px;margin-top:12px;padding-left:12px;padding-right:12px;"><tr><td align="center" valign="top" class="o" style="padding:12px 12px 12px 12px;;background-color:#FFFFFF;border-color:#F1F1F1;border-radius:5px 5px 5px 5px;border-width:1px 1px 1px 1px;"><table role="none" border="0" cellspacing="0" cellpadding="0" align="right" width="100%"><tr><!--[if mso]><td width="0"><table cellpadding="0" cellspacing="0" border="0" role="presentation" style="display:none;"><tr><![endif]--><td class="embed-img mob-stack" align="center" valign="top" style="width:35%;min-height:100px;vertical-align:middle;display:none;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.qZVn6KJQuMivjuNasJr7ICx0t4Bjk7tE9zMQyBNaMqJTl5yQa2cq3o_0t_mvaRySGRuHjs7OxvbdjmuunlMAhGTln1Moxv7ysra9byBqnLM/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h15/h001.ERvfMdPSoh1EA3yLDm2LJBKg5gFvVwnFIyo0w5L4ReY" style="text-decoration:none;" target="_blank"><img src="https://static/images/duo_schematic.png" width="100%" style="display:block;"/></a></td><!--[if mso]></tr></table></td><![endif]--></tr><tr><td align="center" valign="top" class="cc"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td align="left" valign="top" class="l"><p>DUO project page</p></td></tr><tr><td align="left" valign="top" class="m"><p>The Diffusion Duality</p></td></tr><tr><td align="left" valign="bottom" class="n" style="vertical-align:bottom;padding-top:12px;"><p style="word-break:break-word;">s-sahoo.com/duo</p></td></tr></table></td><!--[if mso]><td width="0"><table cellpadding="0" cellspacing="0" border="0" role="presentation" style="display:none;"><tr><![endif]--><td class="mob-hide" align="center" valign="top" style="width:35%;min-height:100px;padding:0px 0px 0px 12px;vertical-align:middle;"><img src="https://static/images/duo_schematic.png" width="100%" style="display:block;"/></td><!--[if mso]></tr></table></td><![endif]--></tr></table></td></tr></table></a></td></tr><tr><td id="benchmark-gains-and-future-horizons" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:normal;padding:0px 28px;text-align:left;"><h3 style="color:#2A2A2A;font-weight:normal;mso-line-height-alt:125.0%;"><span style="color:rgb(67, 67, 67);">Benchmark Gains and Future Horizons</span></h3></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> The benchmark results show that Duo’s curriculum-trained models outperform autoregressive baselines in zero-shot perplexity on 3 of 7 language tasks. This shows that USDMs can compete with established methods. We also saw that the sampling efficiency of the models improved significantly. The DCD approach <span style="font-weight:700;"><b>reduces inference costs by 100×</b></span> while maintaining quality, which notably outpaced the masked diffusion in few-step regimes. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/ead21cd1-3cb7-46cb-92a2-2ca1f505f361/image.png?t=1750174657" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> However, it is not all sunshine and roses, there are still a few limitations. For example, when handling large vocabularies where the diffusion transformation narrows, but the path forward is clear. By borrowing from Gaussian diffusion’s rich toolkit, Duo shows that USDMs can be considered a viable alternative for real-time applications. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/98b330fb-9d19-4d95-8b2b-37961b529370/image.png?t=1750174687" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td align="center" valign="top" style="padding-bottom:14px;padding-left:28px;padding-right:28px;padding-top:14px;text-align:center;width:100%;word-break:break-word;" class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" align="center" style="margin:14px auto 14px auto;"><tr><td align="center" valign="middle" height="44.75" style="height:44.75px;background-color:#2C81E5;border-color:#DFD150;border-radius:10px 10px 10px 10px;border-style:solid;border-width:0px 0px 0px 0px;color:#FFFFFF;font-family:'Open Sans','Segoe UI','Apple SD Gothic Neo','Lucida Grande','Lucida Sans Unicode',sans-serif;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.fUNb4GdFo9D3F8WuLArtoV5sElgytBlvJRzI9WtI92bHcebzdIzGFis2mAkJfB47MBexhLnmAPvMEyAxrUjRmzA62BPohrMFbbsS0EXbEzpNkb5vXdXoVkqNQLbxvDw-/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h16/h001.KQe3_hZi36v9Dz2t8iUd-7bUmjBxbPqxK7s_rGXfv-c" target="_blank" rel="noopener noreferrer nofollow" style="color:#FFFFFF;display:block;font-size:16px;font-size:16px;font-weight:normal;padding:0px 14px;padding:14px 14px 14px 14px;text-decoration:none;"> Read Full Paper </a></td></tr></table></td></tr><tr><td><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" style=""><tr><td bgcolor="#222222" style="background-color:#222222;padding:0.0px 0.0px 0.0px 0.0px;"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0"><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"></p></td></tr></table></td></tr></table></td></tr><tr><td id="self-adapting-language-models" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:Bold;padding:0px 28px;text-align:left;"><h2 style="color:#2A2A2A;font-weight:Bold;mso-line-height-alt:150.0%;">Self-Adapting Language Models </h2></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"><span style=""><i>Zweiger et al. [Massachusetts Institute of Technology]</i></span></p></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"><span style="background-color:#e0e0e0;"><span style="color:rgb(255, 58, 58);font-size:0.6rem;"> ♥ 3.1k </span></span><span style="color:rgb(44, 129, 229);font-size:0.6rem;"> </span><span style="background-color:#e0e0e0;"><span style="color:rgb(44, 129, 229);font-size:0.6rem;"> LLM RL </span></span><span style="color:rgb(44, 129, 229);font-size:0.6rem;"> </span></p></td></tr><tr><td id="introduction-to-self-adapting-ll-ms" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:normal;padding:0px 28px;text-align:left;"><h3 style="color:#2A2A2A;font-weight:normal;mso-line-height-alt:125.0%;"><span style="color:rgb(67, 67, 67);">Introduction to Self-Adapting LLMs</span></h3></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> Large language models often feel frozen in time as they are unable to integrate new knowledge or adapt to tasks beyond their initial training. This limitation forces reliance on in-context learning or resource-heavy finetuning, which struggles with sparse data or suboptimal formats. </p></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> This paper introduces SEAL, a framework that enables models to generate their own training directives, self-edit their weights, and evolve autonomously. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/9cedefe5-ea7c-4c2f-bba3-1e0113224a56/Screenshot_2025-06-09_at_1.28.39_PM.png?t=1750174725" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr><tr><td align="center" valign="top" class="t" style="width:600px; padding: 4px 0px 4px 0px;"><p>Overview of SEAL. In each RL outer loop iteration, the model generates candidate self-edits (SE) — directives on how to update the weights, applies corresponding updates, evaluates performance on a downstream task, and uses the resulting rewards to improve the self-edit generation policy.</p></td></tr></table></td></tr><tr><td id="inner-workings-of-seal" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:normal;padding:0px 28px;text-align:left;"><h3 style="color:#2A2A2A;font-weight:normal;mso-line-height-alt:125.0%;"><span style="color:rgb(67, 67, 67);">Inner Workings of SEAL</span></h3></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> The SEAL architecture uses a multi-loop design to iteratively improve. First, the model processes a task context, like a factual passage or few-shot examples, and generates a "self-edit", which is a natural-language instruction specifying synthetic data (e.g., implications of a text) or optimization parameters (e.g., learning rates). For knowledge integration, it rewrites passages into distilled facts and for few-shot reasoning, it selects data augmentations like rotations or resizing. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/818ef9ba-8abb-4613-8d94-c8f1e3136e4e/Screenshot_2025-06-09_at_2.04.38_PM.png?t=1750174736" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr><tr><td align="center" valign="top" class="t" style="width:600px; padding: 4px 0px 4px 0px;"><p>SEAL Reinforcement Learning Loop. The specific format of the self-edits (SE) are defined per task domain. </p></td></tr></table></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/b5c6cb2a-8e09-4c8b-8484-8993f24a4973/Screenshot_2025-06-09_at_2.16.16_PM.png?t=1750174765" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> These self-edits trigger a supervised fine-tuning process which updates the model’s weights via lightweight LoRA adapters. This reinforcement learning loop trains the self-edit policy which contains the model samples multiple edits, applies them, and receives rewards based on downstream performance (e.g., QA accuracy). Only edits boosting performance are reinforced via rejection sampling. This <span style="font-weight:700;"><b>dual-loop design</b></span>, generation followed by validation, transforms static models into adaptive learners. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/5a22bc64-77c8-4576-bccf-a351fc881723/Screenshot_2025-06-09_at_2.15.54_PM.png?t=1750174750" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td id="evaluation-and-benchmark-results" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:normal;padding:0px 28px;text-align:left;"><h3 style="color:#2A2A2A;font-weight:normal;mso-line-height-alt:125.0%;"><span style="color:rgb(67, 67, 67);">Evaluation and Benchmark Results</span></h3></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> In knowledge incorporation tests using SQuAD passages, SEAL lifted no-context QA accuracy from 33.5% (base model) to <span style="font-weight:700;"><b>47.0%</b></span> after two RL iterations which <span style="font-weight:700;"><b>outperformed GPT-4.1-generated synthetic data</b></span>. For few-shot reasoning on ARC tasks, it achieved 72.5% success by autonomously configuring augmentations and hyperparameters, vastly exceeding non-RL baselines (20%). However, sequential updates revealed catastrophic forgetting, and computational overhead remains high due to per-edit finetuning. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/2689a90e-21cb-4436-948b-e23ef866b899/Screenshot_2025-06-12_at_9.51.20_PM.png?t=1750174830" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> These results spotlight SEAL’s potential for data-scarce settings, where models must self-distill knowledge. Future work could tackle forgetting via retention-focused rewards or expand to continual pretraining. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:480px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/0c62610b-ac8b-4186-88c9-e4ebf7f032eb/Screenshot_2025-06-12_at_10.47.10_AM.png?t=1750174842" alt="" height="auto" width="480" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td align="center" valign="top" style="padding-bottom:14px;padding-left:28px;padding-right:28px;padding-top:14px;text-align:center;width:100%;word-break:break-word;" class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" align="center" style="margin:14px auto 14px auto;"><tr><td align="center" valign="middle" height="44.75" style="height:44.75px;background-color:#2C81E5;border-color:#DFD150;border-radius:10px 10px 10px 10px;border-style:solid;border-width:0px 0px 0px 0px;color:#FFFFFF;font-family:'Open Sans','Segoe UI','Apple SD Gothic Neo','Lucida Grande','Lucida Sans Unicode',sans-serif;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.fUNb4GdFo9D3F8WuLArtoV5sElgytBlvJRzI9WtI92Y3ZdeZfJ6xDXgnNR5upy0xX6akuEmG_D8gEakD8xACzd8LqJ-Rwa3lXuVjqrSyHip9tBtW_Z4xscXV7naiDqFN/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h17/h001.Cw3ZhWsomBce48Pg7tUQdwniVP7fggfzCMr2-KDe0I0" target="_blank" rel="noopener noreferrer nofollow" style="color:#FFFFFF;display:block;font-size:16px;font-size:16px;font-weight:normal;padding:0px 14px;padding:14px 14px 14px 14px;text-decoration:none;"> Read Full Paper </a></td></tr></table></td></tr><tr><td><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" style=""><tr><td bgcolor="#222222" style="background-color:#222222;padding:0.0px 0.0px 0.0px 0.0px;"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0"><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"></p></td></tr></table></td></tr></table></td></tr><tr><td id="reinforcement-pre-training" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:Bold;padding:0px 28px;text-align:left;"><h2 style="color:#2A2A2A;font-weight:Bold;mso-line-height-alt:150.0%;">Reinforcement Pre-Training</h2></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"><span style=""><i>Dong et al. [Microsoft Research, Peking University, Tsinghua University]</i></span></p></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"><span style="background-color:#e0e0e0;"><span style="color:rgb(255, 58, 58);font-size:0.6rem;"> ♥ 424 </span></span><span style="color:rgb(44, 129, 229);font-size:0.6rem;"> </span><span style="background-color:#e0e0e0;"><span style="color:rgb(44, 129, 229);font-size:0.6rem;"> LLM Training </span></span></p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/2da19c6e-0909-4b4c-a305-baf0b16ef42c/rpt.png?t=1750174898" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td id="reinforcement-pre-training-for-lang" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:normal;padding:0px 28px;text-align:left;"><h3 style="color:#2A2A2A;font-weight:normal;mso-line-height-alt:125.0%;"><span style="color:rgb(67, 67, 67);">Reinforcement Pre-Training for Language Models</span></h3></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> Large language models have improved and gotten better on self-supervised pre-training by training on large amounts of data scraped from the internet. However, integrating reinforcement learning (RL) into this process has faced several hurdles such as costly human feedback data risks reward hacking, while verifiable-reward approaches struggle with limited annotated datasets. This paper introduces Reinforcement Pre-Training (RPT) which bridges this gap by transforming next-token prediction into a reasoning task trained with scalable RL. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/68c6e1f4-f75b-4c0d-92cb-e4137533f626/image.png?t=1750174986" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td id="how-reinforcement-pre-training-rein" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:normal;padding:0px 28px;text-align:left;"><h3 style="color:#2A2A2A;font-weight:normal;mso-line-height-alt:125.0%;"><span style="color:rgb(67, 67, 67);">How Reinforcement Pre-Training Reinvents Language Modeling</span></h3></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> The Reinforcement Pre-Training modifies the next-token prediction as a reasoning challenge. For any text snippet, the model generates multiple "thinking trajectories", chains of thought exploring why a token should follow, before predicting the next token. Each trajectory earns a verifiable reward: 1 if the prediction matches the ground-truth token from the corpus, 0 otherwise. This rule-based reward sidesteps reward hacking and leverages unannotated text as RL training data. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/d7bb379d-af36-4631-ad68-ec3f01708b87/image.png?t=1750175006" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> To optimize learning, RPT prioritizes challenging tokens. Before training, a proxy model identifies high-entropy tokens (where predictions are uncertain), while focusing computational effort where reasoning matters most. During rollout, the model samples multiple reasoning paths per context, using a prefix-matching reward that validates predictions against token boundaries in the corpus. This encourages the model to explore hypotheses, self-correct, and deduce patterns rather than memorize. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/7edb6f42-4677-4275-8702-e8009fab7e1a/image.png?t=1750175032" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> The approach uses established RL algorithms like GRPO for training, with dynamic sampling to boost efficiency. By integrating reasoning directly into pre-training, RPT aligns the model’s internal "thought process" with token prediction and effectively scales inference-time computation during training itself. </p></td></tr><tr><td id="performance-gains-and-scaling-poten" class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:normal;padding:0px 28px;text-align:left;"><h3 style="color:#2A2A2A;font-weight:normal;mso-line-height-alt:125.0%;"><span style="color:rgb(67, 67, 67);">Performance Gains and Scaling Potential of Reinforcement Pre-Training</span></h3></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> RPT significantly boosts next-token accuracy, especially on complex tasks. When tested on math-heavy datasets, a 14B-parameter RPT model matched the performance of a 32B-parameter baseline and achieved up to 45% accuracy on hard tokens, which is nearly 3× higher than standard methods. Additionally, RPT exhibits predictable scaling: accuracy improves consistently with compute across easy, medium, and hard tasks, following a power-law curve. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/33bdfef4-ecac-4e81-9b6b-984b451a823d/image.png?t=1750175066" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> For building new applications, the RPT approach can serve as a robust foundation. Fine-tuning it with RL on specialized tasks like competition-level math gives faster convergence and higher performance ceilings. Zero-shot evaluations on MMLU-Pro and SuperGPQA showed gains of 7–22 points over baselines, which highlights its generalization power. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; " class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 auto;"><tr><td align="center" valign="top" style="width:600px;"><img src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/ad087afa-f7db-4c27-84f1-085be88a46af/image.png?t=1750175085" alt="" height="auto" width="600" style="display:block;width:100%;" border="0"/></td></tr></table></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> These results indicate that RPT can be considered as a scalable alternative to conventional pre-training, minimizing the gap between self-supervised learning and reinforcement fine-tuning. </p></td></tr><tr><td align="center" valign="top" style="padding-bottom:14px;padding-left:28px;padding-right:28px;padding-top:14px;text-align:center;width:100%;word-break:break-word;" class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" align="center" style="margin:14px auto 14px auto;"><tr><td align="center" valign="middle" height="44.75" style="height:44.75px;background-color:#2C81E5;border-color:#DFD150;border-radius:10px 10px 10px 10px;border-style:solid;border-width:0px 0px 0px 0px;color:#FFFFFF;font-family:'Open Sans','Segoe UI','Apple SD Gothic Neo','Lucida Grande','Lucida Sans Unicode',sans-serif;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.fUNb4GdFo9D3F8WuLArtoV5sElgytBlvJRzI9WtI92b-MtNB_n18BpwVjlhRn9eNeGokVhoD8oCSM-apP_Hqent-zeh3DtDhjBMF9NOfmeSdTXFPw0H0AeaKq14X2ww2/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h18/h001.b3hagvb6147v977qToBodxpHyqeMdHV1VAcqZRxfHbQ" target="_blank" rel="noopener noreferrer nofollow" style="color:#FFFFFF;display:block;font-size:16px;font-size:16px;font-weight:normal;padding:0px 14px;padding:14px 14px 14px 14px;text-decoration:none;"> Read Full Paper </a></td></tr></table></td></tr><tr><td align="center" valign="top"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td align="center" valign="top" style="font-size:0px;line-height:0px;padding:30px 0px 30px;" class="dd"><table class="j" role="none" width="50%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td> </td></tr></table></td></tr><tr><td class="dd" align="left" valign="top" style="color:#2A2A2A;font-weight:Bold;padding:0px 28px;text-align:left;"><h2 style="color:#2A2A2A;font-weight:Bold;mso-line-height-alt:150.0%;">Share The AI Timeline</h2></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> You currently have <strong>0</strong> referrals. </p></td></tr><tr><td align="left" valign="top" style="padding-bottom:20px;padding-left:28px;padding-right:28px;padding-top:20px; display:none;width:0px;max-height:0px;overflow:hidden;mso-hide:all;height:0;font-size:0;max-height:0;line-height:0;margin:0 auto;" class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" style="margin:0 auto 0 0;"><tr><td align="center" valign="top" style="width:300px;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.c6q0w4g5sodbtO4I1B_pxWc4htTObwdorovK0nFHVH-4pUdVE0ELYH5DsNemk732SjNwhPNJ25r0O8B5vYifsGNUqyW5TiZkyMsF1yreu0byy2KW36J1wDdpoLuXg2TU1F1OW8OHoHaU4-ZmrZpPU4RN-crQCEimD190CSn9fPvQfHEYx7CTSXIidbla7LLK/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h19/h001.HKnp0uHT5kbWlKqSIBqtaylhsOBpZ9LoQTPTWYPim9k" rel="noopener noreferrer nofollow" style="text-decoration:none;" target="_blank"><img src="" alt="" height="auto" width="300" style="display:block;width:100%;" border="0"/></a></td></tr></table></td></tr><tr><td align="left" valign="top" style="padding-bottom:14px;padding-left:28px;padding-right:28px;padding-top:14px;text-align:left;width:100%;word-break:break-word;" class="dd"><table role="none" border="0" cellspacing="0" cellpadding="0" align="left" style="margin:14px auto 14px auto;"><tr><td align="center" valign="middle" height="44.75" style="height:44.75px;background-color:#2C81E5;border-color:#DFD150;border-radius:10px 10px 10px 10px;border-style:solid;border-width:0px 0px 0px 0px;color:#FFFFFF;font-family:'Open Sans','Segoe UI','Apple SD Gothic Neo','Lucida Grande','Lucida Sans Unicode',sans-serif;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.c6q0w4g5sodbtO4I1B_pxWc4htTObwdorovK0nFHVH-4pUdVE0ELYH5DsNemk732SjNwhPNJ25r0O8B5vYifsGNUqyW5TiZkyMsF1yreu0byy2KW36J1wDdpoLuXg2TU1F1OW8OHoHaU4-ZmrZpPU4RN-crQCEimD190CSn9fPvQfHEYx7CTSXIidbla7LLK/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h20/h001.kZ9WY0A30AR04NHtEgR35OJOvc8TuDh2Bp-5G5P8XSY" target="_blank" rel="noopener noreferrer nofollow" style="color:#FFFFFF;display:block;font-size:16px;font-size:16px;font-weight:normal;padding:0px 14px;padding:14px 14px 14px 14px;text-decoration:none;"> Click to Share </a></td></tr></table></td></tr><tr><td class="dd" align="left" style="padding:0px 28px;text-align:left;word-break:break-word;"><p style="mso-line-height-alt:150.0%;"> Or copy and paste this link to others: <a class="link" href="https://mail.bycloud.ai/subscribe?ref=6SqUHb8KiF&_bhlid=bf7a73b936aab597b0df9777ef50b28c5a049d32" target="_blank" rel="noopener noreferrer nofollow" clicktracking="off"><span>https://mail.bycloud.ai/subscribe?ref=6SqUHb8KiF</span></a></p></td></tr><tr><td align="center" valign="top" style="font-size:0px;line-height:0px;padding:30px 0px 30px;" class="dd"><table class="j" role="none" width="50%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td> </td></tr></table></td></tr></table></td></tr><tr><td class="dd" style="padding: 20px;"><table width="100%" cellpadding="0" cellspacing="0" role="none" style="max-width:520px;margin:0 auto;"><tr><td class="q" style="padding:16px 16px 6px 16px;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.tLfGW26lAwaS9gFg17HSoDDFT6eh5Nsg0xYVQj-h6I3o9m2k79_qw4izMYhmcI36cfBKuy-1LSQ1GoZg9de-lPRGsDvhTHltGGFBr2zzg9vEbnxMeILinHM8ka02lSeEBUh0n1_tyakLrpojx9Q7pXhcYpYK-XNl9vJ7YL3pLSg/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h21/h001.SlU1-9qFPciA9J_SJnnv8fJgIDJwATc21QghoEdZvXw" style="text-decoration:none !important;"><table width="100%" cellpadding="0" cellspacing="0" border="0" role="none"><tr><td width="100%" style="padding: 0 0 14px 0;text-decoration:none;width:100%;"><table width="100%" cellpadding="0" cellspacing="0" border="0" role="none"><tr><td width="36" style="width:36px;"><img src="https://pbs.twimg.com/profile_images/1698572487909400576/BvncwnrP_normal.jpg" alt="tw profile: The AI Timeline" style="display:block;width:36px;height:36px;border-radius:50%;border:0;"/></td><td width="400" style="padding:0 0 0 8px;text-decoration:none;"><span style="display:block;font-size:14px;color:#1c2022;font-weight:700;"> The AI Timeline </span><span style="display:block;color:#697882;font-size:14px;"> @TheAITimeline </span></td><td width="24" align="right" style="vertical-align:text-top;"><img width="24" height="24" loading="lazy" alt="tw" style="border:0;" src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/static_assets/x_logo.png"/></td></tr></table></td></tr><tr></tr><tr><td style="word-break:break-word;"><p>🚨This week's top AI/ML research papers:</p><p>- Self-Adapting Language Models <br>- V-JEPA 2 <br>- The Illusion of the Illusion of Thinking <br>- Magistral <br>- Reinforcement Pre-Training <br>- VideoDeepResearch <br>- Unsupervised Elicitation of LMs <br>- CoRT <br>- The Diffusion Duality <br>- Ming-Omni <br>- One</p></td></tr><tr><td style="padding:12px 0 0 0;"></td></tr><tr><td align="center" style="padding:8px 0 0 0;width:480px;"><img src="https://pbs.twimg.com/media/GthTf2aXIAAev0b.jpg" width="480" height="auto" style="display:block;border:1px solid #E1E8ED;border-radius:5px;width:100%;max-width:480px;height:auto;"/></td></tr><tr><td height="8" style="line-height:1px;font-size:1px;height:8px;"> </td></tr><tr><td align="left" valign="top" class="s"><p>11:15 PM • Jun 15, 2025</p></td></tr><tr><td height="10" style="line-height: 1px; font-size: 1px; height: 10px;"> </td></tr><tr><td height="1" bgcolor="#e1e8ed" style="line-height:0px;font-size:0px;height:1px;"></td></tr><tr><td height="10" style="line-height:1px;font-size:1px;height:10px;"> </td></tr><tr><td align="left" valign="top" class="s"><p><b style="color:#1C2022">668</b> Likes <b style="color:#1C2022">78</b> Retweets </p></td></tr><tr><td align="left" valign="top" class="s"><div align="center" style="text-align:center;margin-top:4px;margin-bottom:4px;padding:8px;border:1px solid #ccd6dd;border-radius:9999px;color:#1B95E0"><b>7 Replies</b></div></td></tr></table></a></td></tr></table></td></tr><tr><td class="dd" align="center" valign="top" style="padding:20px;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.amatuKKICSickUKplYJXmPHDV6nnitgWrrZbWcRODjlmO6G1jbNTa92dAk4b9zf1N5LJq8ADDLT5qzrimucAxdObhakig2nX8Ngo9ujvw58OhHPzWbglBX-5oO4tPTal/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h22/h001.KRURrnBrrWimHsEXYAaDnQLuHQ9TZ2b_sZOVTQlCTkA" style="text-decoration:none;"><table align="center" width="100%" cellpadding="0" cellspacing="0" border="0" role="none" style="max-width:520px;margin:0 auto;"><tr><td class="p" width="100%" style="padding:2px;border:none;"><table width="100%" cellpadding="0" cellspacing="0" border="0" role="none"><tr><td align="center" valign="top" style="width:100%;"><div style="max-height:0;position:relative;opacity:0.999;width:100%;mso-hide:all;"><div style="display:inline-block;width:100%;padding-top:25%;"><img width="20%" height="auto" loading="lazy" alt="" style="border:0;" src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/static_assets/youtube_play_icon.png"/></div></div><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.amatuKKICSickUKplYJXmPHDV6nnitgWrrZbWcRODjlhNd8BZr9txwOZhl2XxCPy3v6mNj5DswvATDeVnJJoZLheKd6O9GH3cNfCo2UjQlHgERpWmiwFWMbqfm6A1ElL/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h23/h001.nCO1eqRIYvpAjNWVqTPMpa5PsIbqHLr6DefPRO0lnRs" style="text-decoration:none;"><img src="https://i.ytimg.com/vi/obSNYqgL53k/maxresdefault.jpg" width="480" height="auto" loading="lazy" alt="YouTube video by bycloud" style="display:block;height:auto;border:0;outline:none;text-decoration:none;background-color:#000000;width:100%;"/></a></td></tr><tr><td><p style="font-size:12px;font-weight:500;font-style:italic;font-family:Helvetica, Calibri, sans-serif;color: #686a6d; padding-top:0 !important;padding-bottom:6px !important; padding-left:4px !important;"> Native MoE Multimodal LLM Will Be The Next AI Frontier </p></td></tr></table></td></tr></table></a></td></tr></table></td></tr></table></td></tr><tr><td align="center" valign="top"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td><tr><td class="b" align="center" valign="top" bgcolor="#2a2a2a" style="padding:0px 0px 0px 0px;border-style:solid;border-width: 0px 0px 0px 0px;border-color: #2a2a2a;border-bottom-left-radius:10px;border-bottom-right-radius:10px;"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td align="center" valign="top" bgcolor="#73ddff" style="padding:12px"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td><span style="padding-left:1px;"></span></td><td align="center" valign="middle" width="75" style="width:75px;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.1muhFWIqieRYpaJ-FbWSCQqcWoV4NNHHr5SkP9THApWuHAAlWLQxI3Q_IqFmt_DcyAxeC8jDApCnHmMSBGpBb5sgtimvBYgxRX-Rp7s0F3LjCHoSwdhr83OBqRFhJ1y_/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h24/h001.3GwphtqsaLvpBkl0yk7FnmUJ_eSYn8oy6ELGexffNRY" style="text-decoration:none;"><img width="22" height="22" alt="tw" border="0" style="display:block;max-width:22px;color:Dark" src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/static_assets/x_dark.png"/></a></td><td align="center" valign="middle" width="75" style="width:75px;"><a href="https://elink4f7.mail.bycloud.ai/ss/c/u001.amatuKKICSickUKplYJXmBoQnQ9VXnB2zTxBG4HeHBgjMqVxpoXRdj01cjwyoVlHgiebEOgBvwHtevoVpsSvpn3Q1di2ml6sb3cBM-X6IStQbj_zQSVGWJ8AAmPw2en2/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h25/h001.h4r-VEGOLdSZQkVYfNMWkS6GmLy1LGjd499Yl424GK0" style="text-decoration:none;"><img width="22" height="16" alt="yt" border="0" style="display:block;max-width:22px;color:Dark" src="https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/static_assets/youtube_dark.png"/></a></td><td><span style="padding-left:1px;"></span></td></tr></table></td></tr><tr><td height="10" style="line-height:1px;font-size:1px;height:10px;"> </td></tr><tr><td class="w" align="center" valign="top" style="padding:15px 15px 15px 15px;"><table role="none" width="100%" border="0" cellspacing="0" cellpadding="0" align="center"><tr><td align="center" valign="top"><p style="font-family:'Verdana',Geneva,sans-serif;color:#FFFFFF!important;"> Update your email preferences or unsubscribe <a class="link" href="https://elink4f7.mail.bycloud.ai/ss/c/u001.c6q0w4g5sodbtO4I1B_pxWc4htTObwdorovK0nFHVH-4pUdVE0ELYH5DsNemk732SjNwhPNJ25r0O8B5vYifsBhEpz-DJgyVFmavJPa0OyKRRnvw4o7XGyvIv7PRofnm1QoKBWdNBEfwiDJ9aIJ2Lin0V54UgS5GRvDN-EsahJ6iXgBmsq48zpKXhXmi7XAmq2qa5vbreGtZ19NLzM-9x_9Tar5LP4naiL3OQ24s02ZKtl_yz9uVUqa8uoJ4eQSc8l_g6ShgZrE6TVqI9rtHCUzRPaWSKZ0MRZp5byTt1n56qvcuH7qofXiBI-Ocn-n9_VUDNHgM_kgBvGod0SoWkJmVAbwOcKCtoxahHmulEXZk-2nQjlt4VVMK5QAQc5ks2mr_A7SMWMFTdNqgVMqyBpr89jjgDYbDImz8b50NK7g/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h26/h001.zTd7iZysIRAUVzc4Sus1G_fdkwipI7tlRg5pRFTAZZs" style="text-decoration:underline;text-decoration-color:#FFFFFF!important;color:#FFFFFF!important;"> here</a></p><p class="copyright" style="font-family:'Verdana',Geneva,sans-serif;color:#FFFFFF!important;"> © 2025 bycloudai </p><p style="font-family:'Verdana',Geneva,sans-serif;color:#FFFFFF!important;"> 228 Park Ave S, #29976, New York, New York 10003, United States </p></td></tr><tr style="display: table-row !important;"><td align="center" valign="top" style="padding-top:20px;" style="display:table-cell !important;"><table role="none" border="0" cellspacing="0" cellpadding="0" align="center" style="display:table !important;"><tr style="display:table-row !important;"><td class="u" align="center" valign="middle" height="32" style="height:32px;display:table-cell !important; max-height: 32px !important;margin:0px !important; background-color: #ffffff !important;"><a style="line-height:32px !important;text-decoration:none;display:block !important;" href="https://elink4f7.mail.bycloud.ai/ss/c/u001.DUiN96-Eq7pUHzwEhy5j28olDWFpV5DDKfdk_OdOKOhKVau2gPjNlM5rS5JH6X9I9W6SGfe8WvUFK7GDp0Du19ldsXSa5u35e7gKAX9GQ9YYED0N0QSveuzy5rKq7w9gtR87ymGCXkFEC_sn9STdZSCvfbKVOuW0_MOs0TxJUXEO00xuX2B1iuLeSGhnlekoQqJ6wjEJcF3ypn8fb3dv1KjZe6p33tKOgjOLGnm4JI9Y3_zojHT8izkapkc49ocr/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h27/h001.KyCgiAxibKaK2StcaIe0Xx1p-BQ13OujUetvHl_Ql44"><img src="https://media.beehiiv.com/output-onlinepngtools.png" width="16" alt="beehiiv logo" style="display:inline-block !important;max-width:16px !important; vertical-align:-3px !important;width: 16px !important;" border="0"/><span style="padding-left:11px !important;display: inline-block !important;">Powered by beehiiv</span></a></td></tr></table></td></tr><tr><td align="left" valign="top" height="2" style="height:2px;"><a href='https://elink4f7.mail.bycloud.ai/ss/c/u001.CxDkkVpJsBdVoe83c_tBWsHIaP4XNp0WgUYqLvHcKk_3uqk_KIkz4ddLinhFbud6JuxLFdSUhYnR7b1NSsmbtzXNGNblnEEMKUtkCAjkn8Y/4hf/9EAxPIReQ9K2uiAP9qmNaQ/h28/h001.ushpRqnhC8vTKwGRi0B0lqfVOdjLZXCA_jJ_XaomXEk' style="color: #2a2a2a !important; cursor: default; font-size: 1px; text-decoration: none;"> Terms of Service </a></td></tr></table></td></tr></table></td></tr></td></tr></table></td></tr></table></td></tr></table></td></tr></table></div></body></html>