// Judges with GCC >= 12 only needs Ofast
// #pragma GCC optimize("O3,no-stack-protector,fast-math,unroll-loops,tree-vectorize")
// MLE optimization
// #pragma GCC optimize("conserve-stack")
// Old judges
// #pragma GCC target("sse4.2,popcnt,lzcnt,abm,mmx,fma,bmi,bmi2")
// New judges. Test with assert(__builtin_cpu_supports("avx2"));
// #pragma GCC target("avx2,popcnt,lzcnt,abm,bmi,bmi2,fma,tune=native")
// Atcoder
// #pragma GCC target("avx2,popcnt,lzcnt,abm,bmi,bmi2,fma")
/*
#include <ext/pb_ds/assoc_container.hpp>
#include <ext/pb_ds/tree_policy.hpp>
using namespace __gnu_pbds;
typedef tree<int,null_type,less<int>,rb_tree_tag,tree_order_statistics_node_update> ordered_set;
- insert(x)
- find_by_order(k): return iterator to the k-th smallest element
- order_of_key(x): the number of elements that are strictly smaller
*/
#include<bits/stdc++.h>
using namespace std;
mt19937_64 rng(chrono::steady_clock::now().time_since_epoch().count());
uniform_real_distribution<> pp(0.0,1.0);
#define int long long
#define ld long double
#define pii pair<int,int>
#define piii pair<int,pii>
#define fi first
#define se second
const int inf=1e18;
const int mod=998244353;
const int mod2=1e9+7;
const int maxn=200005;
const int bl=650;
const int maxs=650;
const int maxm=200005;
const int maxq=500005;
const int maxl=20;
const int maxa=1000000;
int power(int a,int n){
    int res=1;
    while(n){
        if(n&1) res=res*a%mod;
        a=a*a%mod;n>>=1;
    }
    return res;
}
int g(int x,int st){
    int l=1,r=st,res=1;
    while(r>=l){
        int mid=(r+l)>>1;
        if(mid*(mid+1)/2+st-mid<=x){res=mid;l=mid+1;}
        else r=mid-1;
    }
    return res;
}
void solve(){
    int n,m;cin >> n >> m;
    vector<vector<int>> x(n,vector<int>(m,0));
    if(n%4!=3 && m%4!=3 && n%4!=0 && m%4!=0){
        cout << "No\n";
        return;
    }
    if(n%4==3 || n%4==0){
        int cur=n*(n+1)/4,st=n;
        vector<int> p;
        while(cur){
            int f=g(cur,st);
            p.push_back(f);
            cur-=f*(f+1)/2;st-=f;
        }
        int nw=0;
        for(int i=0;i<(int)p.size();i++){
            for(int j=nw;j<nw+p[i];j++) x[j].assign(m,i%2);
            nw+=p[i];
        }
    }
    else{
        int cur=m*(m+1)/4,st=m;
        vector<int> p;
        while(cur){
            int f=g(cur,st);
            p.push_back(f);
            cur-=f*(f+1)/2;st-=f;
        }
        int nw=0;
        for(int i=0;i<(int)p.size();i++){
            for(int j=nw;j<nw+p[i];j++){
                for(int k=0;k<n;k++) x[k][j]=i%2;
            }
            nw+=p[i];
        }
    }
    cout << "Yes\n";
    for(int i=0;i<n;i++){
        for(int j=0;j<m;j++) cout << x[i][j] << ' ';
        cout << '\n';
    }
}
signed main(){
    ios_base::sync_with_stdio(false);
    cin.tie(NULL);cout.tie(NULL);
    int test=1;cin >> test;
    while(test--) solve();
}