#include<bits/stdc++.h>
using namespace std;
typedef vector<int> poly;
const int N = 1e6;
const int p = 998244353;
inline int Qpow(int x, int y = p - 2) {
    int ans = 1;
    for (; y; y >>= 1, x = 1ll * x * x % p) if (y & 1) ans = 1ll * ans * x % p;
    return ans;
}
namespace Core {
    int a[N], b[N], w[N], rev[N];
    inline void NTT(int *a, int len) {
        for (int i = 1; i < len; ++i) if (i > rev[i]) swap(a[i], a[rev[i]]);
        for (int d = 1; d < len; d <<= 1) 
            for (int m = d << 1, i = 0; i < len; i += m) 
                for (int j = 0; j < d; ++j) {
                    int x = a[i + j], y = 1ll * a[i + j + d] * w[len / m * j] % p;
                    a[i + j] = (x + y >= p ? x + y - p : x + y);
                    a[i + j + d] = (x - y < 0 ? x - y + p : x - y);
                }
    }
}
inline poly operator*(poly va, poly vb) {
    // poly c(a.size() + b.size() - 1);
    // for (int i = 0; i < a.size(); ++i)
    //     for (int j = 0; j < b.size(); ++j)
    //         c[i + j] = (c[i + j] + 1ll * a[i] * b[j]) % p;
    using namespace Core;
    int len = 1; while (len < (int)va.size() + (int)vb.size() - 1) len <<= 1;
    for (int i = 0; i < len; ++i) a[i] = (i < (int)va.size() ? va[i] : 0);
    for (int i = 0; i < len; ++i) b[i] = (i < (int)vb.size() ? vb[i] : 0);
    for (int i = 1; i < len; ++i) rev[i] = rev[i >> 1] >> 1 | (i & 1 ? len >> 1 : 0);
    w[0] = 1; w[1] = Qpow(3, (p - 1) / len);
    for (int i = 2; i < len; ++i) w[i] = 1ll * w[i - 1] * w[1] % p;
    NTT(a, len); NTT(b, len);
    for (int i = 0; i < len; ++i) a[i] = 1ll * a[i] * b[i] % p;
    reverse(a + 1, a + len); NTT(a, len);
    poly c(va.size() + vb.size() - 1);
    for (int i = 0, invlen = Qpow(len); i < (int)c.size(); ++i) c[i] = 1ll * a[i] * invlen % p;
    return c;
}
inline poly Fixlen(poly a, int len) {
    a.resize(len);
    return a;
}
inline poly Qpow(int lim, poly x, int y = p - 2) {
    poly ans = x; --y;
    for (; y; y >>= 1, x = Fixlen(x * x, lim)) if (y & 1) ans = Fixlen(x * ans, lim);
    return ans;
}
int main() {
    ios::sync_with_stdio(0);
    cin.tie(0);
    int n, m;
    cin >> n >> m;
    vector<int> a(m + 1), finv(m + 1), fac(m + 1);
    a[0] = 1;
    finv[0] = 1;
    fac[0] = 1;
    for (int i = 1; i <= m; ++i) {
        finv[i] = 1ll * finv[i - 1] * Qpow(i) % p;
        fac[i] = 1ll * fac[i - 1] * i % p;
        a[i] = 1ll * Qpow(i + 1, i - 1) * finv[i] % p;
    }
    vector<int> ans = Qpow(m + 1, a, n - m + 1);
    int rst = 1ll * ans[m] * fac[m] % p;
    cout << rst << endl;
    return 0;
}