#ifndef LOCAL
#define FAST_IO
#endif

// ============
#include <algorithm>
#include <array>
#include <bitset>
#include <cassert>
#include <cmath>
#include <iomanip>
#include <iostream>
#include <list>
#include <map>
#include <numeric>
#include <queue>
#include <random>
#include <set>
#include <stack>
#include <string>
#include <tuple>
#include <unordered_map>
#include <unordered_set>
#include <utility>
#include <vector>

#define OVERRIDE(a, b, c, d, ...) d
#define REP2(i, n) for (i32 i = 0; i < (i32)(n); ++i)
#define REP3(i, m, n) for (i32 i = (i32)(m); i < (i32)(n); ++i)
#define REP(...) OVERRIDE(__VA_ARGS__, REP3, REP2)(__VA_ARGS__)
#define PER(i, n) for (i32 i = (i32)(n) - 1; i >= 0; --i)
#define ALL(x) begin(x), end(x)

using namespace std;

using u32 = unsigned int;
using u64 = unsigned long long;
using i32 = signed int;
using i64 = signed long long;
using f64 = double;
using f80 = long double;

template <typename T>
using Vec = vector<T>;

template <typename T>
bool chmin(T &x, const T &y) {
    if (x > y) {
        x = y;
        return true;
    }
    return false;
}
template <typename T>
bool chmax(T &x, const T &y) {
    if (x < y) {
        x = y;
        return true;
    }
    return false;
}

#ifdef INT128

using u128 = __uint128_t;
using i128 = __int128_t;

istream &operator>>(istream &is, i128 &x) {
    i64 v;
    is >> v;
    x = v;
    return is;
}
ostream &operator<<(ostream &os, i128 x) {
    os << (i64)x;
    return os;
}
istream &operator>>(istream &is, u128 &x) {
    u64 v;
    is >> v;
    x = v;
    return is;
}
ostream &operator<<(ostream &os, u128 x) {
    os << (u64)x;
    return os;
}

#endif

[[maybe_unused]] constexpr i32 INF = 1000000100;
[[maybe_unused]] constexpr i64 INF64 = 3000000000000000100;
struct SetUpIO {
    SetUpIO() {
#ifdef FAST_IO
        ios::sync_with_stdio(false);
        cin.tie(nullptr);
#endif
        cout << fixed << setprecision(15);
    }
} set_up_io;
// ============

#ifdef DEBUGF
#else
#define DBG(x) (void) 0
#endif

// ============

// ============

#include <algorithm>
#include <iostream>
#include <atcoder/convolution>

namespace poly {

using Mint = atcoder::modint998244353;
using Poly = std::vector<Mint>;

Poly add(Poly f, Poly g) {
    if (f.size() < g.size()) {
        std::swap(f, g);
    }
    for (int i = 0; i < (int)g.size(); ++i) {
        f[i] += g[i];
    }
    return f;
}

Poly sub(Poly f, Poly g) {
    if (f.size() < g.size()) {
        std::swap(f, g);
    }
    for (int i = 0; i < (int)g.size(); ++i) {
        f[i] -= g[i];
    }
    return f;
}

Poly mul(const Poly &f, const Poly &g) {
    return atcoder::convolution(f, g);
}

void dft(Poly &f) {
    atcoder::internal::butterfly(f);
}

void idft(Poly &f) {
    atcoder::internal::butterfly_inv(f);
    Mint inv = Mint::raw(f.size()).inv();
    for (Mint &cf : f) {
        cf *= inv;
    }
}

} // namespace poly
// ============

namespace poly {

class Factorial {
    std::vector<Mint> fac;
    std::vector<Mint> ifac;
    
public:
    Factorial() : fac(1, Mint::raw(1)), ifac(1, Mint::raw(1)) {}
    
    void reserve(int n) {
        int m = (int)fac.size() - 1;
        if (n <= m) {
            return;
        }
        fac.resize(n + 1);
        for (int i = m + 1; i <= n; ++i) {
            fac[i] = fac[i - 1] * Mint::raw(i);
        }
        ifac.resize(n + 1);
        ifac[n] = fac[n].inv();
        for (int i = n - 1; i >= m; --i) {
            ifac[i] = ifac[i + 1] * Mint::raw(i + 1);
        }
    }
    
    Mint fact(int n) const {
        return fac[n];
    }
    
    Mint inv_fact(int n) const {
        return ifac[n];
    }
    
    Mint inv_n(int n) const {
        return ifac[n] * fac[n - 1];
    }
    
    Mint binom(int n, int r) const {
        return fac[n] * ifac[r] * ifac[n - r];
    }
} factorial;

} // namespace poly
// ============

using namespace poly;

Poly pow_of_fps(const Poly &f, i32 n) {
    // fps exp まだ移植できてないので log 2 つで...
    if (n == 0) {
        Poly g(f.size());
        g[0] = Mint::raw(1);
        return g;
    }
    Poly g = pow_of_fps(f, n / 2);
    g = mul(g, g);
    g.resize(f.size());
    if (n % 2 == 1) {
        g = mul(f, g);
        g.resize(f.size());
    }
    return g;
}

int main() {
    i32 n, m;
    cin >> n >> m;
    
    if (m == 1) {
        cout << 0 << '\n';
        exit(0);
    }
    
    factorial.reserve(n);
    
    Poly f(n + 1);
    REP(i, n + 1) {
        f[i] = factorial.inv_fact(i) * factorial.inv_fact(i);
    }
    Poly g = pow_of_fps(f, m - 2);
    // 1 と 2 が swap される回数を数える。
    // この値を M(M-1)/2 倍すれば答えが求まる。
    Mint ans;
    REP(i, 2, n + 1) {
        Mint c = g[n - i] * factorial.fact(n - i) * factorial.fact(n - i) * factorial.binom(n, i) * factorial.binom(n, i);
        Mint sub = Mint::raw(2).pow(2 * i - 3) * Mint::raw(i - 1);
        ans += c * sub;
    }
    ans *= Mint::raw(m) * Mint::raw(m - 1) / Mint::raw(2);
    cout << ans.val() << '\n';
}