// #pragma GCC target("avx")  // CPU 処理並列化
// #pragma GCC optimize("O3")  // CPU 処理並列化
// #pragma GCC optimize("unroll-loops")  // 条件処理の呼び出しを減らす
#include <bits/stdc++.h>
#include <complex>

// #include <iostream> // cout, endl, cin
// #include <string> // string, to_string, stoi
// #include <vector> // vector
// #include <algorithm> // min, max, swap, sort, reverse, lower_bound, upper_bound
// #include <utility> // pair, make_pair
// #include <tuple> // tuple, make_tuple
// #include <cstdint> // int64_t, int*_t
// #include <cstdio> // printf
// #include <map> // map
// #include <queue> // queue, priority_queue
// #include <set> // set
// #include <stack> // stack
// #include <deque> // deque
// #include <unordered_map> // unordered_map
// #include <unordered_set> // unordered_set
// #include <bitset> // bitset
// #include <climits>
// #include <cmath>
// #include <iomanip>
// #include <functional>
// #include <numeric>
// #include <random>
 
using namespace std;

#define int long long
#define pb push_back
#define eb emplace_back
// #define F first
// #define S second
#define FOR(i,a,b) for(int (i)=(a);(i)<(int)(b);(i)++)
#define rep(i,n) FOR(i,0,n)
#define RFOR(i,a,b) for(int (i)=(a);(i)>=(int)(b);(i)--)
#define rrep(i,n) RFOR(i,n,0)
#define all(a) (a).begin(),(a).end()
#define rall(a) (a).rbegin(),(a).rend()
#define ve vector
#define vi vector<int>
#define vp vector<pair<int,int>>
#define vvi vector<vector<int>>
#define UNIQUE(a) sort(all(a)), a.erase(unique(all(a)), a.end())
#define Double double
// #define endl '\n'
 
template<typename T> using pq = priority_queue<T,vector<T>,greater<T>>; 
using ll = long long;
using UnWeightedGraph = vector<vector<int>>;
ll INF = LLONG_MAX / 4 - 100;
int IINF = INT_MAX / 4;
ll mod = 1e9 + 7;
int dx[] = {1,0,-1,0}, dy[] = {0,1,0,-1};
vector<ll> prime;
 
long double pi = 3.141592653589793238;
    
class fact {
public:
    long long fmod = 1e9+7;
    vector<long long> fac, finv, inv;
    fact (int n, long long Mod = 1e9+7) {
        fmod = Mod;
        fac = vector<long long>(n + 1, 0);
        finv = vector<long long>(n + 1, 0);
        inv = vector<long long>(n + 1, 0);
        fac[0] = fac[1] = 1;
        finv[0] = finv[1] = 1;
        inv[1] = 1; 
        for (int i = 2; i < n + 1; i++) {
            fac[i] = fac[i-1] * i % fmod;
            inv[i] = mod - inv[mod%i] * (mod/i) % mod;
            finv[i] = finv[i-1] * inv[i] % mod;
        }
    }
    ll nCr(ll n, ll r) {if(n < r) return 0; return fac[n] * finv[r] % fmod * finv[n-r] % fmod;}
    ll POW(ll a, ll b) {ll c = 1; while (b > 0) {if (b & 1) {c = a * c%fmod;}a = a * a%fmod; b >>= 1;}return c;}
    inline int operator [] (int i) {return fac[i];}
    ll DeBuG(ll n, ll r);
};
    
void DEBUG(vector<int> a) {for(int i=0;i<a.size();i++)cout<<a[i]<<" ";cout<<endl;}
void EMP(int x) {cout<<"!!!"<<x<<"!!!"<<endl;}
ll GCD(ll a, ll b) {ll c; while (b != 0) {c = a % b; a = b; b = c;}return a;}
ll LCM(ll a, ll b) {return (a / GCD(a, b)) * (b / GCD(a, b)) * GCD(a, b);}
ll POW(ll a, ll b) {ll c = 1; while (b > 0) {if (b & 1) {c = a * c%mod;}a = a * a%mod; b >>= 1;}return c;}
void PRI(ll n) {bool a[n + 1]; for (int i = 0; i < n + 1; i++) {a[i] = 1;}for (int i = 2; i < n + 1; i++) {if (a[i]) {prime.pb(i); ll b = i; while (b <= n) {a[b] = 0; b += i;}}}}
template <typename T> T chmin(T& a, T b) {if(a>b)a=b;return a;}
template <typename T> T chmax(T& a, T b) {if(a<b)a=b;return b;}
bool isSqrt(ll a) {return pow(sqrt(a),2) == a ? 1 : 0;}
void YesNo(bool a) {if (a) cout << "Yes"; else cout << "No"; cout << endl;}
void yesno(bool a) {if (a) cout << "yes"; else cout << "no"; cout << endl;}
void YESNO(bool a) {if (a) cout << "YES"; else cout << "NO"; cout << endl;}
double dis(int x1, int x2, int y1, int y2) {
    return sqrt((double)abs(x1-x2)*(double)abs(x1-x2)+(double)abs(y1-y2)*(double)abs(y1-y2));
}
int ceili(int x, int y) {
    if (x % y == 0) return x / y;
    else return x / y + 1;
}
unsigned int randxor() {
    static unsigned int x=123456789,y=362436069,z=521288629,w=88675123;
    unsigned int t;
    t=(x^(x<<11));x=y;y=z;z=w; return( w=(w^(w>>19))^(t^(t>>8)) );
}
 
uint64_t xor64(void) {
    static uint64_t x = 88172645463325252ULL;
    x = x ^ (x << 7);
    return x = x ^ (x >> 9);
}

vector<complex<Double>> fft(vector<complex<Double>> a, bool inverse = false) {
    int n = a.size();
    int h = 0; // h = log_2(n)
    for (int i = 0; 1 << i < n; i++) h++;
    // バタフライ演算用の配置入れ替え
    for (int i = 0; i < n; i++) {
        int j = 0;
        for (int k = 0; k < h; k++) j |= (i >> k & 1) << (h - 1 - k);
        if (i < j) swap(a[i], a[j]);
    }
    // バタフライ演算
    for (int b = 1; b < n; b *= 2) {
        // 第 log_2(b) + 1 段
        // ブロックサイズ = b * 2
        for (int j = 0; j < b; j++) {
            // ブロック内 j 個目
            // 重み w = (1 の原始 2b 乗根の j 乗)
            complex<Double> w =
                polar((Double)1.0, (2 * (Double)M_PI) / (2 * b) * j * (inverse ? 1 : -1));
            for (int k = 0; k < n; k += b * 2) {
                // k を先頭とするブロック
                complex<Double> s = a[j + k];         // 前
                complex<Double> t = a[j + k + b] * w; // 後
                a[j + k] = s + t;                     // 前の更新
                a[j + k + b] = s - t;                 // 後の更新
            }
        }
    }
    // 逆変換時にサイズで割る調整
    if (inverse)
        for (int i = 0; i < n; i++) a[i] /= n;
    return a;
}
// Cooley–Tukey FFT algorithm O(N log N)
vector<complex<Double>> fft(vector<Double> a, bool inverse = false) {
    vector<complex<Double>> a_complex(a.size());
    for (int i = 0; i < a.size(); i++) a_complex[i] = complex<Double>(a[i], 0);
    return fft(a_complex, inverse);
}
 
// FFT による畳み込み O(N log N)
vector<Double> convolve(vector<Double> a, vector<Double> b) {
    int s = a.size() + b.size() - 1; // 畳み込み結果のサイズ
    int t = 1; // FFT に使う配列のサイズ（2 の累乗）
    while (t < s) t *= 2;
    a.resize(t); // FFT するためにリサイズ
    b.resize(t); // FFT するためにリサイズ
    vector<complex<Double>> A = fft(a);
    vector<complex<Double>> B = fft(b);
    for (int i = 0; i < t; i++) {
        A[i] *= B[i]; // 畳み込み結果の FFT 結果を得る
    }
    A = fft(A, true); // IFFT で畳み込み結果を得る
    a.resize(s);      // 畳み込み結果を入れるためにリサイズ
    for (int i = 0; i < s; i++) a[i] = A[i].real(); // 実部が答え
    return a;
}

vector<int> A;
vector<int> saiki(int l, int r) {
    if (r == l + 1) return vector<int>{A[l], 1};
    vector<int> tL = saiki(l, (l+r)/2), tR = saiki((l+r)/2, r);
    vector<Double> L, R;
    for (auto &&elm : tL) L.pb(elm);
    for (auto &&elm : tR) R.pb(elm);
    auto res = convolve(L, R);
    vector<int> ret;
    for (auto &&elm : res) {
        ret.pb(((int)(elm+(Double)0.5))%mod);
        // if (ret.back() < 0) {
        //     FOR (i, l, r) cout << A[i] << " "; cout << endl;
        //     rep (i, r-l+1) cout << res[i] << " "; cout << endl;
        //     exit(0);
        // }
    }
    return ret;
}

void solve() {
    mod = 998244353;
    int n; cin >> n; int m; cin >> m;
    A.resize(n); rep (i, n) cin >> A[i], A[i]--;
    auto ans = saiki(0, n);
    rep (i, m) {
      int a; cin >> a;
      cout << ans[a] << endl;
    }
}
 
signed main() {
    cin.tie(0);
    ios::sync_with_stdio(false);
    solve();
}