//header
#ifdef LOCAL
    #include "cxx-prettyprint-master/prettyprint.hpp"
    #define debug(x) cout << x << endl
#else
    #define debug(...) 42
#endif
    #pragma GCC optimize("Ofast")
    #include <bits/stdc++.h>
    //types
    using namespace std;
    using ll = long long;
    using ul = unsigned long long;
    using ld = long double;
    typedef pair < ll , ll > Pl;        
    typedef pair < int, int > Pi;
    typedef vector<ll> vl;
    typedef vector<int> vi;
    template< typename T >
    using mat = vector< vector< T > >;
    template< int mod >
    struct modint {
        int x;

        modint() : x(0) {}

        modint(int64_t y) : x(y >= 0 ? y % mod : (mod - (-y) % mod) % mod) {}

        modint &operator+=(const modint &p) {
            if((x += p.x) >= mod) x -= mod;
            return *this;
        }

        modint &operator-=(const modint &p) {
            if((x += mod - p.x) >= mod) x -= mod;
            return *this;
        }

        modint &operator*=(const modint &p) {
            x = (int) (1LL * x * p.x % mod);
            return *this;
        }

        modint &operator/=(const modint &p) {
            *this *= p.inverse();
            return *this;
        }

        modint operator-() const { return modint(-x); }

        modint operator+(const modint &p) const { return modint(*this) += p; }

        modint operator-(const modint &p) const { return modint(*this) -= p; }

        modint operator*(const modint &p) const { return modint(*this) *= p; }

        modint operator/(const modint &p) const { return modint(*this) /= p; }

        bool operator==(const modint &p) const { return x == p.x; }

        bool operator!=(const modint &p) const { return x != p.x; }

        modint inverse() const {
            int a = x, b = mod, u = 1, v = 0, t;
            while(b > 0) {
            t = a / b;
            swap(a -= t * b, b);
            swap(u -= t * v, v);
            }
            return modint(u);
        }

        modint pow(int64_t n) const {
            modint ret(1), mul(x);
            while(n > 0) {
            if(n & 1) ret *= mul;
            mul *= mul;
            n >>= 1;
            }
            return ret;
        }

        friend ostream &operator<<(ostream &os, const modint &p) {
            return os << p.x;
        }

        friend istream &operator>>(istream &is, modint &a) {
            int64_t t;
            is >> t;
            a = modint< mod >(t);
            return (is);
        }

        static int get_mod() { return mod; }
    };
    //abreviations
    #define all(x) (x).begin(), (x).end()
    #define rall(x) (x).rbegin(), (x).rend()
    #define rep_(i, a_, b_, a, b, ...) for (int i = (a), max_i = (b); i < max_i; i++)
    #define rep(i, ...) rep_(i, __VA_ARGS__, __VA_ARGS__, 0, __VA_ARGS__)
    #define rrep_(i, a_, b_, a, b, ...) for (int i = (b-1), min_i = (a); i >= min_i; i--)
    #define rrep(i, ...) rrep_(i, __VA_ARGS__, __VA_ARGS__, 0, __VA_ARGS__)
    #define SZ(x) ((int)(x).size())
    #define pb(x) push_back(x)
    #define eb(x) emplace_back(x)
    #define mp make_pair
    #define print(x) cout << x << endl
    #define vsum(x) accumulate(x, 0LL)
    #define vmax(a) *max_element(all(a))
    #define vmin(a) *min_element(all(a))
    #define lb(c, x) distance((c).begin(), lower_bound(all(c), (x)))
    #define ub(c, x) distance((c).begin(), upper_bound(all(c), (x)))
    //functions
    ll gcd(ll a, ll b) { return b ? gcd(b, a%b) : a; }
    ll lcm(ll a, ll b) { return a/gcd(a, b)*b;}
    template<class T>bool chmax(T &a, const T &b) { if (a<b) { a=b; return 1; } return 0; }
    template<class T>bool chmin(T &a, const T &b) { if (b<a) { a=b; return 1; } return 0; }
    template< typename T >
    T mypow(T x, ll n) {
        T ret = 1;
        while(n > 0) {
            if(n & 1) (ret *= x);
            (x *= x);
            n >>= 1;
        }
        return ret;
    }
    ll modpow(ll x, ll n, const ll mod) {
        ll ret = 1;
        while(n > 0) {
            if(n & 1) (ret *= x);
            (x *= x);
            n >>= 1;
            x%=mod;
            ret%=mod;
        }
        return ret;
    }
    uint64_t my_rand(void) {
        static uint64_t x = 88172645463325252ULL;
        x = x ^ (x << 13); x = x ^ (x >> 7);
        return x = x ^ (x << 17);
    }
    //graph template
    template< typename T >
    struct edge {
        int src, to;
        T cost;

        edge(int to, T cost) : src(-1), to(to), cost(cost) {}

        edge(int src, int to, T cost) : src(src), to(to), cost(cost) {}

        edge &operator=(const int &x) {
            to = x;
            return *this;
        }
        operator int() const { return to; }
    };
    template< typename T >
    using Edges = vector< edge< T > >;
    template< typename T >
    using WeightedGraph = vector< Edges< T > >;
    using UnWeightedGraph = vector< vector< int > >;

//constant
//#define inf 1000000005LL
#define inf 4000000000000000005LL
#define mod 1000000007LL
#define endl '\n'
typedef modint<mod> mint;
const long double eps = 0.0001;
const long double PI  = 3.141592653589793;
//library
//実装上当然正の値しか入らないことに注意!!!!!
struct SuccinctIndexableDictionary {
  size_t length;
  size_t blocks;
  vector< unsigned > bit, sum;

  SuccinctIndexableDictionary() = default;

  SuccinctIndexableDictionary(size_t length) : length(length), blocks((length + 31) >> 5) {
    bit.assign(blocks, 0U);
    sum.assign(blocks, 0U);
  }

  void set(int k) {
    bit[k >> 5] |= 1U << (k & 31);
  }

  void build() {
    sum[0] = 0U;
    for(int i = 1; i < blocks; i++) {
      sum[i] = sum[i - 1] + __builtin_popcount(bit[i - 1]);
    }
  }

  bool operator[](int k) {
    return (bool((bit[k >> 5] >> (k & 31)) & 1));
  }

  int rank(int k) {
    return (sum[k >> 5] + __builtin_popcount(bit[k >> 5] & ((1U << (k & 31)) - 1)));
  }

  int rank(bool val, int k) {
    return (val ? rank(k) : k - rank(k));
  }
};
template< typename T, int MAXLOG >
struct WaveletMatrix {
  size_t length;
  SuccinctIndexableDictionary matrix[MAXLOG];
  int mid[MAXLOG];

  WaveletMatrix() = default;

  WaveletMatrix(vector< T > v) : length(v.size()) {
    vector< T > l(length), r(length);
    for(int level = MAXLOG - 1; level >= 0; level--) {
      matrix[level] = SuccinctIndexableDictionary(length + 1);
      int left = 0, right = 0;
      for(int i = 0; i < length; i++) {
        if(((v[i] >> level) & 1)) {
          matrix[level].set(i);
          r[right++] = v[i];
        } else {
          l[left++] = v[i];
        }
      }
      mid[level] = left;
      matrix[level].build();
      v.swap(l);
      for(int i = 0; i < right; i++) {
        v[left + i] = r[i];
      }
    }
  }

  pair< int, int > succ(bool f, int l, int r, int level) {
    return {matrix[level].rank(f, l) + mid[level] * f, matrix[level].rank(f, r) + mid[level] * f};
  }

  // v[k]
  T access(int k) {
    T ret = 0;
    for(int level = MAXLOG - 1; level >= 0; level--) {
      bool f = matrix[level][k];
      if(f) ret |= T(1) << level;
      k = matrix[level].rank(f, k) + mid[level] * f;
    }
    return ret;
  }

  T operator[](const int &k) {
    return access(k);
  }

  // count i s.t. (0 <= i < r) && v[i] == x
  int rank(const T &x, int r) {
    int l = 0;
    for(int level = MAXLOG - 1; level >= 0; level--) {
      tie(l, r) = succ((x >> level) & 1, l, r, level);
    }
    return r - l;
  }

  // k-th(0-indexed) smallest number in v[l,r)
  T kth_smallest(int l, int r, int k) {
    assert(0 <= k && k < r - l);
    T ret = 0;
    for(int level = MAXLOG - 1; level >= 0; level--) {
      int cnt = matrix[level].rank(false, r) - matrix[level].rank(false, l);
      bool f = cnt <= k;
      if(f) {
        ret |= T(1) << level;
        k -= cnt;
      }
      tie(l, r) = succ(f, l, r, level);
    }
    return ret;
  }

  // k-th(0-indexed) largest number in v[l,r)
  T kth_largest(int l, int r, int k) {
    return kth_smallest(l, r, r - l - k - 1);
  }

  // count i s.t. (l <= i < r) && (v[i] < upper)
  int range_freq(int l, int r, T upper) {
    int ret = 0;
    for(int level = MAXLOG - 1; level >= 0; level--) {
      bool f = ((upper >> level) & 1);
      if(f) ret += matrix[level].rank(false, r) - matrix[level].rank(false, l);
      tie(l, r) = succ(f, l, r, level);
    }
    return ret;
  }

  // count i s.t. (l <= i < r) && (lower <= v[i] < upper)
  int range_freq(int l, int r, T lower, T upper) {
    return range_freq(l, r, upper) - range_freq(l, r, lower);
  }

  // max v[i] s.t. (l <= i < r) && (v[i] < upper)
  T prev_value(int l, int r, T upper) {
    int cnt = range_freq(l, r, upper);
    return cnt == 0 ? T(-1) : kth_smallest(l, r, cnt - 1);
  }

  // min v[i] s.t. (l <= i < r) && (lower <= v[i])
  T next_value(int l, int r, T lower) {
    int cnt = range_freq(l, r, lower);
    return cnt == r - l ? T(-1) : kth_smallest(l, r, cnt);
  }
};

int main(){
    cin.tie(0);
    ios::sync_with_stdio(0);
    cout << setprecision(20);
    int n; cin>>n;
    vl a(n), l(n), r(n);
    rep(i, n)cin>>a[i], a[i]+=mod;
    rep(i, n)cin>>l[i]>>r[i];
    vi id(n);
    iota(all(id), 0);
    sort(all(id), [&](ll a, ll b){return a<b;});
    vl v(n);
    rep(i, n){
        int j = id[i];
        v[i] = a[j]-l[j];
    }
    WaveletMatrix<ll, 32> wm(v);
    ll ans = 0;
    rep(i, n){
        int j = id[i];
        int k = ub(a, a[j]+r[j]);
        ans+=wm.range_freq(i+1, k, a[j]+1);
    }
    print(ans);
}