From 3bc5705d30929500a3d0218457c837ffd096d187 Mon Sep 17 00:00:00 2001 From: TravelLeraLone <745111625@qq.com> Date: Thu, 7 Mar 2024 21:38:34 +0800 Subject: [PATCH] Update transparent main for webpage --- docs/index.html | 2 +- docs/static/images/main_transparent.png | Bin 0 -> 1399793 bytes 2 files changed, 1 insertion(+), 1 deletion(-) create mode 100644 docs/static/images/main_transparent.png diff --git a/docs/index.html b/docs/index.html index db9abdf..f87952c 100644 --- a/docs/index.html +++ b/docs/index.html @@ -166,7 +166,7 @@
-
+
Rjy-9l)Jf==X`@f;KmR7QD{A49&a^-#+!M`9Rp4)k)J H^nduq7Tp&N@sCiYN6<)`>r|pEj1Y-y!$B`1#6m)|Mk*%T-#~KgvJvCC*
zw?-#YjgNX)M&tUeIXm2(UDqGol1rscoF)3Jy=iWjb?7MNBA|9Q*mdyM^5$Wlq{qY~
zHBw`>eIxBRy*+I2ge$K;tNUT#w0$Q&xk9eozI=uH{EMurrAo|FV^Ggs^pRrxu+H<0
z{(2dHv;D$=uv7)}HTEnPGwxgK twOFCyN
z|FfFfxgPx&?t_KTT;iNcmpg@L5X}BT?TU&6rX?(^U4*b*!+z{+DDduHC?ITGGtJ2Q
z?*3^v$ePo|sjOSG`VafEwH?6Bf`aZOUVl 7C=RwlpKzn6l_c#H6Y0KZX2St?0_{K87h1gifI_7b|
zD`fPGtPmwGwW-n~+#b1EwsrRu*lZSaSs*6Leq6n7@=EZVp5;K+5ZyUcZnB-yZ_o&7
zx=MzNz5=31C85whS}RTKv3KY7u(M^S?y8!>u0;?w>qI?o*9&seH1`hgiyG0rwipEZ
zM;_y@A9c5Z{o8@eCeKAS?8V-;tlK`hWNv@5#(gx)dj2|A*KNQrN2|-CXmtrZ#i#cQ
zU#^c#MolqkW8%I<$jf%L!^q6HFwokB#^
Large language models (LLMs) have established great success in the general domain of natural language processing. Their emerging task generalization and free-form dialogue capabilities can greatly help to design Chemical General Intelligence (CGI) to assist real-world research in chemistry. However, the existence of specialized language and knowledge in the field of chemistry, such as the highly informative SMILES notation, hinders the performance of general-domain LLMs in chemistry. To this end, we develop ChemDFM, the first LLM towards CGI. ChemDFM-13B is trained on 34B tokens from chemical literature, textbooks, and instructions as well as various data from the general domain. Therefore, it can store, understand, and reason over chemical knowledge and languages while still possessing advanced free-form language comprehension capabilities. Extensive quantitative evaluation shows that ChemDFM can significantly outperform the representative open-sourced LLMs. Moreover, ChemDFM can also surpass GPT-4 on a great portion of chemical tasks, despite the significant size difference. Further qualitative evaluations demonstrate the efficiency and effectiveness of ChemDFM in real-world research scenarios.
diff --git a/docs/static/images/main_transparent.png b/docs/static/images/main_transparent.png
new file mode 100644
index 0000000000000000000000000000000000000000..b1d4391193547b26e242a781ebffdc2b976e477f
GIT binary patch
literal 1399793
zcmbTdXHZk|w>1o+0@6XH69fbcMMOFT6a*9nJ4lB}QJN4Efe;Wuij)W{y^0Dd5PB!{
zqLfgRkWi#0)C345kPv*m-=6=S`#g8%T{CCSS?9y~^gDa6S$pmHdsgOrrzB3Xv9a;p
zvM_zX#wI{vV>{`?b@Jbvz%T`ke>ZmU1M?ef<-<~||HyIA>vylSu~i~?4qZ?D#SyNrqpk*bSe+w95tr}X_a{G~?D
zg%~Y7Ok{;jIupKaLX(L}H72hlmJ)1lB0rzIZtdbC?_;$@a;S{Kxk<}|y!8N}0>c^l
z<$Z*dIe&as#CkQ{VW3(+`6y2dg+5`TZ_*u{JC4D=*0OO;SdK2IU)JubOD{W;
#mQsLfG(<6x%zbVzH`ASc5T7twrdY;TlHZ)dh)DqDipQ{(P2+A;tjRA#
zg_3Y`Ti`(c)t|lQ`4+n}0s@sfN4iARj7ENTl^M&u7Ov+!YPKV?Lw!K`L(WNbAVnRG
z!B+QD6TxUkHvpGgTt-!yGcbm%^#(>oNTMML@{J^PM-}X2yexh}L`fRO;4N5k+H3MH
zO|NSQr(;OcQzoNMXz5PLgsJn9EBnw26ZJ#{cq6JH>eZ^b?8UJe!bZmOegXl&e7|1FVz
zi64EHE`GUsl_#7ngMUGj4`?o~_I8Nk#%2-L(-6w~
-#ddJ(oPN+r=wGBh$XvFdz@n0~pMCa)a6K+-fmY*H%1Gm<
zGi~