luoguP5409 第一类斯特林数·列

传送门

$\left[\begin{array}{c}n\\ m\end{array}\right]$表示将$n$个不同的数分成$m$个无差别环的方案数。
令$F(x)$为一个环的指数型生成函数。对于一个大小为$i$的环，我们固定一个$1$号点，剩下的数有$(i-1)!$种排列方法。当$i$等于$0$时，无法构成环，于是方案数为$0$。那么有：
$$F(x)=\sum _{i=1}^{\infty }\frac{(i-1)!}{i!}{x}^i=\sum _{i=1}^{\infty }\frac{1}{i}{x}^i$$
那么$\left[\begin{array}{c}i\\ k\end{array}\right]=\frac{[x^i]F^k(x)}{k!}\ast i!$
($x^i$对应的系数)
除以$k!$是因为要除去环的排列。相当于是生成集合。
注意，快速幂要套加强版的板子，因为常数项为$0$。

// luogu-judger-enable-o2
#include<bits/stdc++.h>
using namespace std;
typedef long long ll;
typedef vector<int> poly;
const int g=3,Log=19,maxn=1<<Log,mod=167772161;
inline int add(int x,int y){return x+y>=mod?x+y-mod:x+y;}
inline int mul(int x,int y){return (ll)x*y%mod;}
inline int dec(int x,int y){return x-y<0?x-y+mod:x-y;}
inline int quickpow(int a,int b,int ret=1){for(;b;b>>=1,a=mul(a,a)) if(b&1) ret=mul(ret,a);return ret;}
int rev[maxn],inv[maxn],ifac[maxn],fac[maxn],*w[Log+1];
inline void print(poly A){
	for(int i=0;i<A.size();++i)
		printf("%d ",A[i]);
	puts("");
}
inline void init(int limit){
	for(int i=0;i<limit;++i)
		rev[i]=(rev[i>>1]>>1)|((i&1)*(limit>>1));
}
inline void prework(){
	inv[1]=1;for(int i=2;i<maxn;++i) inv[i]=mul(mod-mod/i,inv[mod%i]);
	fac[0]=1;for(int i=1;i<maxn;++i) fac[i]=mul(fac[i-1],i);
	ifac[0]=1;for(int i=1;i<maxn;++i) ifac[i]=mul(ifac[i-1],inv[i]);
	for(int i=1;i<=Log;++i) w[i]=new int[1<<(i-1)];
	int wn=quickpow(g,(mod-1)/(1<<Log));w[Log][0]=1;
	for(int i=1;i<(1<<(Log-1));++i) w[Log][i]=mul(w[Log][i-1],wn);
	for(int i=Log-1;i;--i)
		for(int j=0;j<(1<<(i-1));++j)
			w[i][j]=w[i+1][j<<1];
}
inline void NTT(poly &f,int limit,int type){
	for(int i=0;i<limit;++i) if(rev[i]>i) swap(f[rev[i]],f[i]);
	for(int mid=1,l=1;mid<limit;mid<<=1,++l)
		for(int i=0;i<limit;i+=(mid<<1))
			for(int j=0,p0,p1;j<mid;++j)
				p0=f[i+j],p1=mul(w[l][j],f[i+j+mid]),
				f[i+j]=add(p0,p1),f[i+j+mid]=dec(p0,p1);
	if(type==-1&&(reverse(f.begin()+1,f.begin()+limit),1)){
		int inv=quickpow(limit,mod-2);
		for(int i=0;i<limit;++i) f[i]=mul(f[i],inv);
	}
}
inline poly operator*(poly A,poly B){
	int len=A.size()+B.size()-2,limit=1;
	while(limit<=len) limit<<=1;init(limit);
	A.resize(limit),NTT(A,limit,1);
	B.resize(limit),NTT(B,limit,1);
	for(int i=0;i<limit;++i) A[i]=mul(A[i],B[i]);
	return NTT(A,limit,-1),A.resize(len+1),A;
}
inline poly operator*(poly A,int B){
	for(int i=0;i<A.size();++i) A[i]=mul(A[i],B);
	return A;
}
inline poly Inv(poly A){
	int len=A.size();poly C,B(1,quickpow(A[0],mod-2));
	for(int limit=4;limit<(len<<2);limit<<=1){
		init(limit),C=A;
		C.resize(limit>>1),B.resize(limit>>2);
		C.resize(limit),NTT(C,limit,1);
		B.resize(limit),NTT(B,limit,1);
		for(int i=0;i<limit;++i) B[i]=mul(B[i],dec(2,mul(B[i],C[i])));
		NTT(B,limit,-1);
	}return B.resize(len),B;
}
inline poly Deriv(poly A){
	for(int i=0;i<A.size()-1;++i) A[i]=mul(A[i+1],i+1);
	return A.pop_back(),A;
}
inline poly Integ(poly A){
	A.push_back(0);
	for(int i=A.size()-1;i;--i) A[i]=mul(A[i-1],inv[i]);
	return A[0]=0,A;
}
inline poly Ln(poly A){
	int len=A.size();
	A=Integ(Deriv(A)*Inv(A));
	return A.resize(len),A;
}
inline poly Exp(poly A){
	int len=A.size();
	poly B(1,1),C;A.resize(len<<1);
	for(int limit=2;limit<(len<<1);limit<<=1){
		B.resize(limit),C=Ln(B);
		for(int i=0;i<limit;++i) C[i]=dec(A[i],C[i]);
		C[0]=add(C[0],1),B=B*C;
	}return B.resize(len),B;
}
inline poly quickpow(poly A,int b){
	int len=A.size(),low=0;poly ret(len,0);
	for(int i=0;i<len;++i)
		if(A[i]){low=i;break;}
	int inv=quickpow(A[low],mod-2),pinv=quickpow(A[low],b);
	if(low){
		for(int i=0;i<len-low;++i) A[i]=A[i+low];
		for(int i=len-low;i<len;++i) A[i]=0;
	}for(int i=0;i<len;++i) A[i]=mul(A[i],inv);
	if(1ll*low*b>=1ll*len) return ret;
	A=Ln(A)*b,A=Exp(A),low=low*b;
	for(int i=0;i<low;++i) ret[i]=0;
	for(int i=low;i<len;++i) ret[i]=mul(A[i-low],pinv);
	return ret;
}
int n,k;poly A;
int main(){
	//freopen("luogu5409.in","r",stdin);
	prework(),scanf("%d%d",&n,&k),A.push_back(0);
	for(int i=1;i<=n;++i) A.push_back(inv[i]);
	A=quickpow(A,k);
	for(int i=0;i<=n;++i)
		printf("%d ",mul(mul(A[i],ifac[k]),fac[i]));
}