diff --git a/convolution/min_plus_convolution_concave_arbitrary/checker.cpp b/convolution/min_plus_convolution_concave_arbitrary/checker.cpp
new file mode 100644
index 00000000..6a66d533
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/checker.cpp
@@ -0,0 +1,62 @@
+// https://github.com/MikeMirzayanov/testlib/blob/master/checkers/wcmp.cpp
+
+// The MIT License (MIT)
+
+// Copyright (c) 2015 Mike Mirzayanov
+
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files (the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions:
+
+// The above copyright notice and this permission notice shall be included in all
+// copies or substantial portions of the Software.
+
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+// SOFTWARE.
+
+#include "testlib.h"
+
+using namespace std;
+
+int main(int argc, char * argv[])
+{
+    setName("compare sequences of tokens");
+    registerTestlibCmd(argc, argv);
+
+    int n = 0;
+    string j, p;
+
+    while (!ans.seekEof() && !ouf.seekEof()) 
+    {
+        n++;
+
+        ans.readWordTo(j);
+        ouf.readWordTo(p);
+        
+        if (j != p)
+            quitf(_wa, "%d%s words differ - expected: '%s', found: '%s'", n, englishEnding(n).c_str(), compress(j).c_str(), compress(p).c_str());
+    }
+
+    if (ans.seekEof() && ouf.seekEof())
+    {
+        if (n == 1)
+            quitf(_ok, "\"%s\"", compress(j).c_str());
+        else
+            quitf(_ok, "%d tokens", n);
+    }
+    else
+    {
+        if (ans.seekEof())
+            quitf(_wa, "Participant output contains extra tokens");
+        else
+            quitf(_wa, "Unexpected EOF in the participants output");
+    }
+}
diff --git a/convolution/min_plus_convolution_concave_arbitrary/gen/common.hpp b/convolution/min_plus_convolution_concave_arbitrary/gen/common.hpp
new file mode 100644
index 00000000..fd0393f9
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/gen/common.hpp
@@ -0,0 +1,50 @@
+
+using namespace std;
+using ll = long long;
+
+vector<ll> rand_gen(Random& gen, int N, int min_slope, int max_slope) {
+  vector<ll> A(N);
+  while (1) {
+    vector<int> slope(N - 1);
+    for (int i = 0; i < N - 1; ++i) {
+      slope[i] = gen.uniform<int>(min_slope, max_slope);
+    }
+    sort(slope.begin(), slope.end());
+    for (int i = 0; i < N - 1; ++i) A[i + 1] = A[i] + slope[i];
+    ll mi = *(min_element(A.begin(), A.end()));
+    ll ma = *(max_element(A.begin(), A.end()));
+    if (ma - mi > A_MAX) continue;
+    for (int i = 0; i < N; ++i) A[i] -= mi;
+    ma -= mi;
+    int add = gen.uniform<int>(0, A_MAX - ma);
+    for (int i = 0; i < N; ++i) A[i] += add;
+    break;
+  }
+  return A;
+}
+
+vector<ll> rand_B(Random& gen, int N) {
+  vector<ll> B(N);
+  for (int i = 0; i < N; ++i) B[i] = gen.uniform<int>(A_MIN, A_MAX);
+  return B;
+}
+
+void upside_down(vector<ll>& A){
+  int N = A.size();
+  for (int i = 0; i < N; ++i) A[i] = A_MAX + A_MIN - A[i];
+}
+
+void out(vector<ll>& A, vector<ll>& B) {
+  int n = A.size(), m = B.size();
+  printf("%d %d\n", n, m);
+  for (int i = 0; i < n; ++i) {
+    if (i) printf(" ");
+    printf("%lld", A[i]);
+  }
+  printf("\n");
+  for (int i = 0; i < m; ++i) {
+    if (i) printf(" ");
+    printf("%lld", B[i]);
+  }
+  printf("\n");
+}
\ No newline at end of file
diff --git a/convolution/min_plus_convolution_concave_arbitrary/gen/example_00.in b/convolution/min_plus_convolution_concave_arbitrary/gen/example_00.in
new file mode 100644
index 00000000..c56bd90f
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/gen/example_00.in
@@ -0,0 +1,3 @@
+4 5
+0 2 3 1
+5 1 3 3 2
diff --git a/convolution/min_plus_convolution_concave_arbitrary/gen/hack_00.in b/convolution/min_plus_convolution_concave_arbitrary/gen/hack_00.in
new file mode 100644
index 00000000..52d0e58f
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/gen/hack_00.in
@@ -0,0 +1,3 @@
+17 8
+127 127 126 124 121 117 112 106 99 91 82 72 61 49 36 22 7
+100000 100000 100000 100000 100000 100000 100000 2
diff --git a/convolution/min_plus_convolution_concave_arbitrary/gen/large_small.cpp b/convolution/min_plus_convolution_concave_arbitrary/gen/large_small.cpp
new file mode 100644
index 00000000..4f7fb2b1
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/gen/large_small.cpp
@@ -0,0 +1,26 @@
+#include <cstdio>
+#include "../params.h"
+#include "random.h"
+#include <vector>
+
+#include "common.hpp"
+
+int main(int, char* argv[]) {
+  long long seed = atoll(argv[1]);
+  auto gen = Random(seed);
+
+  int ns[] = {N_MAX, N_MAX, 1, 2};
+  int ms[] = {1, 2, N_MAX, N_MAX};
+
+  int n = ns[seed % 4];
+  int m = ms[seed % 4];
+
+  int LIM_1 = A_MAX / n;
+
+  vector<ll> A = rand_gen(gen, n, -LIM_1, +LIM_1);
+  vector<ll> B = rand_B(gen, m);
+  upside_down(A);
+  out(A, B);
+
+  return 0;
+}
diff --git a/convolution/min_plus_convolution_concave_arbitrary/gen/max_random.cpp b/convolution/min_plus_convolution_concave_arbitrary/gen/max_random.cpp
new file mode 100644
index 00000000..e3e411a7
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/gen/max_random.cpp
@@ -0,0 +1,23 @@
+#include <cstdio>
+#include "../params.h"
+#include "random.h"
+#include <vector>
+
+#include "common.hpp"
+
+int main(int, char* argv[]) {
+  long long seed = atoll(argv[1]);
+  auto gen = Random(seed);
+
+  int n = N_MAX;
+  int m = N_MAX;
+
+  int LIM_1 = A_MAX / n * 3;
+
+  vector<ll> A = rand_gen(gen, n, -LIM_1, +LIM_1);
+  vector<ll> B = rand_B(gen, m);
+  upside_down(A);
+  out(A, B);
+
+  return 0;
+}
diff --git a/convolution/min_plus_convolution_concave_arbitrary/gen/med_random.cpp b/convolution/min_plus_convolution_concave_arbitrary/gen/med_random.cpp
new file mode 100644
index 00000000..66a6ad89
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/gen/med_random.cpp
@@ -0,0 +1,23 @@
+#include <cstdio>
+#include "../params.h"
+#include "random.h"
+#include <vector>
+
+#include "common.hpp"
+
+int main(int, char* argv[]) {
+  long long seed = atoll(argv[1]);
+  auto gen = Random(seed);
+
+  int n = gen.uniform<int>(100, 1000);
+  int m = gen.uniform<int>(100, 1000);
+
+  int LIM_1 = A_MAX / n * 3;
+
+  vector<ll> A = rand_gen(gen, n, -LIM_1, +LIM_1);
+  vector<ll> B = rand_B(gen, m);
+  upside_down(A);
+  out(A, B);
+
+  return 0;
+}
diff --git a/convolution/min_plus_convolution_concave_arbitrary/gen/monotone.cpp b/convolution/min_plus_convolution_concave_arbitrary/gen/monotone.cpp
new file mode 100644
index 00000000..d6b910fd
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/gen/monotone.cpp
@@ -0,0 +1,37 @@
+#include <cstdio>
+#include "../params.h"
+#include "random.h"
+#include <vector>
+
+#include "common.hpp"
+
+int main(int, char* argv[]) {
+  long long seed = atoll(argv[1]);
+  auto gen = Random(seed);
+
+  int n = N_MAX;
+  int m = N_MAX;
+
+  int LIM = A_MAX / n * 2;
+  int lo_1 = (seed & 1 ? -LIM : 0);
+  int hi_1 = (seed & 1 ? 0 : LIM);
+
+  vector<ll> A = rand_gen(gen, n, lo_1, hi_1);
+  vector<ll> B = rand_B(gen, m);
+  sort(B.begin(), B.end());
+  if (seed & 2) reverse(B.begin(), B.end());
+
+  for (int i = 0; i < n - 1; ++i) {
+    if (!(seed & 1)) assert(A[i] <= A[i + 1]);
+    if (seed & 1) assert(A[i] >= A[i + 1]);
+  }
+  for (int i = 0; i < m - 1; ++i) {
+    if (!(seed & 2)) assert(B[i] <= B[i + 1]);
+    if (seed & 2) assert(B[i] >= B[i + 1]);
+  }
+
+  upside_down(A);
+  out(A, B);
+
+  return 0;
+}
diff --git a/convolution/min_plus_convolution_concave_arbitrary/gen/near_power_of_2.cpp b/convolution/min_plus_convolution_concave_arbitrary/gen/near_power_of_2.cpp
new file mode 100644
index 00000000..6096528e
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/gen/near_power_of_2.cpp
@@ -0,0 +1,28 @@
+#include <cstdio>
+#include "../params.h"
+#include "random.h"
+#include <vector>
+
+#include "common.hpp"
+
+int main(int, char* argv[]) {
+  long long seed = atoll(argv[1]);
+  auto gen = Random(seed);
+
+  int X = 1 << 18;
+  int a = X - 1, b = X, c = X + 1;
+  int ns[] = {a, a, a, b, b, b, c, c, c};
+  int ms[] = {a, b, c, a, b, c, a, b, c};
+
+  int n = ns[seed % 9];
+  int m = ms[seed % 9];
+
+  int LIM_1 = A_MAX / n * 3;
+
+  vector<ll> A = rand_gen(gen, n, -LIM_1, +LIM_1);
+  vector<ll> B = rand_B(gen, m);
+  upside_down(A);
+  out(A, B);
+
+  return 0;
+}
diff --git a/convolution/min_plus_convolution_concave_arbitrary/gen/only_first_small.cpp b/convolution/min_plus_convolution_concave_arbitrary/gen/only_first_small.cpp
new file mode 100644
index 00000000..0fa79b3b
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/gen/only_first_small.cpp
@@ -0,0 +1,30 @@
+#include <cstdio>
+#include "../params.h"
+#include "random.h"
+#include <vector>
+
+#include "common.hpp"
+
+int main(int, char* argv[]) {
+  long long seed = atoll(argv[1]);
+  auto gen = Random(seed);
+
+  int n = N_MAX;
+  int m = N_MAX;
+
+  int LIM_1 = A_MAX / n * 3;
+
+  vector<ll> A = rand_gen(gen, n, -LIM_1, +LIM_1);
+  vector<ll> B(m);
+  for (int i = 0; i < m; ++i) {
+    B[i] = gen.uniform<int>(A_MAX * 9 / 10, A_MAX);
+  }
+
+  int idx = (seed % 2 == 0 ? 0 : m - 1);
+  B[idx] = A_MIN;
+
+  upside_down(A);
+  out(A, B);
+
+  return 0;
+}
diff --git a/convolution/min_plus_convolution_concave_arbitrary/gen/random.cpp b/convolution/min_plus_convolution_concave_arbitrary/gen/random.cpp
new file mode 100644
index 00000000..e6c0b802
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/gen/random.cpp
@@ -0,0 +1,23 @@
+#include <cstdio>
+#include "../params.h"
+#include "random.h"
+#include <vector>
+
+#include "common.hpp"
+
+int main(int, char* argv[]) {
+  long long seed = atoll(argv[1]);
+  auto gen = Random(seed);
+
+  int n = gen.uniform<int>(1, N_MAX);
+  int m = gen.uniform<int>(1, N_MAX);
+
+  int LIM_1 = A_MAX / n * 3;
+
+  vector<ll> A = rand_gen(gen, n, -LIM_1, +LIM_1);
+  vector<ll> B = rand_B(gen, m);
+  upside_down(A);
+  out(A, B);
+
+  return 0;
+}
diff --git a/convolution/min_plus_convolution_concave_arbitrary/gen/small.cpp b/convolution/min_plus_convolution_concave_arbitrary/gen/small.cpp
new file mode 100644
index 00000000..8fb93885
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/gen/small.cpp
@@ -0,0 +1,24 @@
+#include <cstdio>
+#include "../params.h"
+#include "random.h"
+#include <vector>
+
+#include "common.hpp"
+
+int main(int, char* argv[]) {
+  long long seed = atoll(argv[1]);
+  auto gen = Random(seed);
+
+  int ns[] = {1, 1, 1, 2, 2, 2, 3, 3, 3};
+  int ms[] = {1, 2, 3, 1, 2, 3, 1, 2, 3};
+
+  int n = ns[seed % 9];
+  int m = ms[seed % 9];
+
+  vector<ll> A = rand_gen(gen, n, -A_MAX, +A_MAX);
+  vector<ll> B = rand_B(gen, m);
+  upside_down(A);
+  out(A, B);
+
+  return 0;
+}
diff --git a/convolution/min_plus_convolution_concave_arbitrary/gen/small_slopes.cpp b/convolution/min_plus_convolution_concave_arbitrary/gen/small_slopes.cpp
new file mode 100644
index 00000000..1a0adc1c
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/gen/small_slopes.cpp
@@ -0,0 +1,21 @@
+#include <cstdio>
+#include "../params.h"
+#include "random.h"
+#include <vector>
+
+#include "common.hpp"
+
+int main(int, char* argv[]) {
+  long long seed = atoll(argv[1]);
+  auto gen = Random(seed);
+
+  int n = N_MAX;
+  int m = N_MAX;
+
+  vector<ll> A = rand_gen(gen, n, -10, +10);
+  vector<ll> B = rand_B(gen, m);
+  upside_down(A);
+  out(A, B);
+
+  return 0;
+}
diff --git a/convolution/min_plus_convolution_concave_arbitrary/hash.json b/convolution/min_plus_convolution_concave_arbitrary/hash.json
new file mode 100644
index 00000000..7a42d993
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/hash.json
@@ -0,0 +1,84 @@
+{
+  "example_00.in": "3f248bae57fe16a6b3b63c0743b53e5da574c4b709aa41b14e4e81ad37251f49",
+  "example_00.out": "afcbb20cff61b123b64d2feda6e1473486f3444c3186acb8d2ff2c9c823f6a5b",
+  "hack_00.in": "2790cbcd42ffa6208ad9800a101ab2b4c22cdf148c0a3c794de95a1ea599cd4a",
+  "hack_00.out": "e58ccf2b3c94da35a211f58352b2f6aff3da42d724f9c50101d1ebe7f036fd4c",
+  "large_small_00.in": "f8df9037aedd9092a0d0f3dbe7e7c2d0ed1d496b93d8a1a8a21d35745165d3e4",
+  "large_small_00.out": "e0d5ab1ddb2947ca9d0784f985fd2970481019f79d94f0d1882b34c7e0153fce",
+  "large_small_01.in": "252533d35d2ec5a238dd8b9e1d7639eec2c959726fd58b87c89ee5896982de3e",
+  "large_small_01.out": "2370f95670527efe6113c9dea6b458453c7097269daa6dcf408f21b2a560eb0d",
+  "large_small_02.in": "9168f3bfb7862da498fc043eaa6872793bb75edd62c048898375366caa4dfe00",
+  "large_small_02.out": "4a735075f308dc62638fd7434cda39fe168fdb17da1cab780e967640040c16c2",
+  "large_small_03.in": "356dc57ad8a9a518d4ce3c7868ec13024e7af0edaf9e8c213dfb59b1a30202e3",
+  "large_small_03.out": "e29c4211d6cdfbd9224bec06877b4aec550e96126285228b0cbb4680947e66bc",
+  "max_random_00.in": "7023bd58d017c7b88a3f93ce3335bb37d797b9196d64a1332437c4d9e8f4e51c",
+  "max_random_00.out": "41021c8968f79e804afe9fdb95358530ee2610283d6c7578d419bb95c3484fee",
+  "max_random_01.in": "93b2f1be9813bc0d0a48b47aebf1a85849966bfbb0f308e565e6bbd7e7d2f7c7",
+  "max_random_01.out": "761a3663150262257cf65048538e9bb7ce42e334408444f30b365ceefc736c16",
+  "max_random_02.in": "413329b4baa246df9aab3e55e5bff2268fb2393a7b5c58ecece399d29180051f",
+  "max_random_02.out": "ee3fac52a5c4b3edd5161eb0580fa16da629273be99dfc3f4706df3075173aa7",
+  "med_random_00.in": "e93b01c36d779123425cc4fb2d72cc692b965d1347e4bd56a0df11d277b14ff5",
+  "med_random_00.out": "7361aa0f8e261d7a6394518b47150b66259f62c8cdbceb4a62e8cb95f065c5d6",
+  "med_random_01.in": "254fc9316862e78ccf06236a4710b904a5dcbc624c8e84c9cb7a5f69cbc9f39a",
+  "med_random_01.out": "14cbe95270c924f668f6b86c1e89447c5ae6885a3d824febd488dc9061abcff7",
+  "med_random_02.in": "8d742829c47bf7e0292ccaea2090b0a8b9ccf9170b27a0a00d1f3cdb894ad67a",
+  "med_random_02.out": "edafdd28496182d18403f39e79a98d4f48756052465db78ea548571cd19dffb8",
+  "monotone_00.in": "9a395b1efc7e9429180ea10bab6cd3eed88b38495ddf8dd439034d14d0ae53e5",
+  "monotone_00.out": "f2d4d7d84e153cf832651b525e39018aa9ffca1c6424d993a60096a3eaafc207",
+  "monotone_01.in": "38c8de37e07728e9c3d5baabf236a375d5ca2598bad22da73cb3972b9945a67d",
+  "monotone_01.out": "44427f85913b5413d6ae83b3971ede67db521db22a6fc6b7f6ec10b724a28747",
+  "monotone_02.in": "a4e51109a92f32a35072dd444491871b30f97deaa42405c5ad39b28b75d0af4c",
+  "monotone_02.out": "745b424b292e5b7aa7f60c3828c2a50c5047812a4300af09c558ecc8dc16d8f3",
+  "monotone_03.in": "28e46f1847c3671f746f5a9ed954aecbad56eb2e1be8d35eacbd67a84e5460a9",
+  "monotone_03.out": "a2f64d6ab1f8892f17eed0a0f0930f2ba9fbc44bb711e87bbcea7a0a227fcfab",
+  "near_power_of_2_00.in": "2f86d212ce3e9d0b2b481556ec3e7b4de97754bb45edd9e29901c8d2ed65e104",
+  "near_power_of_2_00.out": "04797591844a5ee06f596a7faf5956eec9ee7860f87061af630a6d3de16d79eb",
+  "near_power_of_2_01.in": "2fbb90f7610c5669342c866d3c1152259aa369c8d30abb7463c7ee8619a588d9",
+  "near_power_of_2_01.out": "e8e43b4a90ea9ffd75d275a17dda4161b82d8667c4878c8acbb185210d577edc",
+  "near_power_of_2_02.in": "250bb97e5dbca7681cce87d297d8bcb07b414bd70879a52bc65d33d0dab9f1eb",
+  "near_power_of_2_02.out": "321efe4bc82fe2427575124c72a84bf13e695221d672f51f51dd5e32e7a13050",
+  "near_power_of_2_03.in": "7fbb267e1fe705dab2fc5b63ac9443502306e552f13450f82aa981823b108087",
+  "near_power_of_2_03.out": "811bc2cbbb517f5d072f2062958b49db329e0a456d05f0998e48a8dfaf94d3fb",
+  "near_power_of_2_04.in": "bcd9f3e6cb14e64317a5b262174b6ef603b731e7b3758af9f0a0dfbdf4d468f1",
+  "near_power_of_2_04.out": "c816ec836ea7d2c6c3d736793c77953ff2bc8d2a03978db8269c26b6478d90f9",
+  "near_power_of_2_05.in": "616ff84d8fbf0959a4af8a8c74bcc683deae2eadc6f6f079659754c9940874f0",
+  "near_power_of_2_05.out": "efadb9add7ce3389a738e3e03b705f70ab4493a69d625294081a8e41177b3647",
+  "near_power_of_2_06.in": "d9e7f33f97d0e4b77cd27a284ce107e3377dc54bc60e326266442a9ed4718e54",
+  "near_power_of_2_06.out": "fbb1cde067c76b8001ba0a855a66d6901dfaa08a636e24d6c50d830a6dcf1d53",
+  "near_power_of_2_07.in": "b40cfda2d2428c89b5de817642ca443702cc2d2bf6233803bd934f47a5a8297d",
+  "near_power_of_2_07.out": "aa8e201138fac27ec3189dfb23201bcb239e10b96b91d8fda8ca9a2b0b086ae4",
+  "near_power_of_2_08.in": "427385f2f47d0588ff44af9c5873b65e6dbffd13db74fcb90e573ad4a32e21c5",
+  "near_power_of_2_08.out": "abd53f2257289a27e55801efacb8f34e272ae508e1c07f283c576c618e3d41f3",
+  "only_first_small_00.in": "5b2629a2a7fcb9ac2f5426f47b832065f5c4525da4aab5554ffe6b1a2d4675ce",
+  "only_first_small_00.out": "b060d9ea41e4bc137fe68b360b170cf08e6605b6643e479ff3cde3173def2770",
+  "only_first_small_01.in": "5b402e1496d9341bf0ebd419e849ca1e1d5157572d29540494cb9fc0398e277d",
+  "only_first_small_01.out": "085bcbc846ba57a2631c756e67f920a42e45df962a8478e70b5154f4a0964311",
+  "random_00.in": "c5b1d358adf5c02ce4fafe154d2506bb37853d6eccb224cb8918a4ceff07eae0",
+  "random_00.out": "c34585eaf3266c86b98ea02fea41eab9d34e66265cdbc382e589ad8ee52de786",
+  "random_01.in": "ff7140f881f9867ceae4e677be50af82eaa0806fe45db6fcec23895504a7ac95",
+  "random_01.out": "536da5f8d48ac9528b91a993f9efdcb9e1fba571c2661eec9f89f7faa986f043",
+  "random_02.in": "d67ab8ae479b52145c2b993ce9f50285f86ae580f671b4071cef04a7d41c658e",
+  "random_02.out": "bb38212721594630d7fc3250ef3e10f68242c11b9cb13e5e03051f9ed37caa3e",
+  "small_00.in": "bd5a7f2cbe20c36368f5af09d2eb7919625748740baa99518e76835a445726fe",
+  "small_00.out": "430f0f2d9e0d91e4f2a0072a9c7841e1f3ca9941e97de5adc0a07b0c60b7715a",
+  "small_01.in": "da0926405459eee4282d14e65e9668a5c0a7751c1ab7e50430a9ef316c3f0024",
+  "small_01.out": "3c97c192205eb16a2c5f8f8ddf2a7bd79775a260d8b6fcb64fa388d7d45e13b1",
+  "small_02.in": "8fdc2390e58dcdc9f668b0309043be09b316582ecf0d6358870449a8ac01cb6a",
+  "small_02.out": "7802293a5a2e4b9e016c430568de50c1767626c78112c14bf1c6be4f2790eba4",
+  "small_03.in": "dfeeb84628515e76eebe94030843be80c291075bd0327b64f97893f6a8d40624",
+  "small_03.out": "9f900fbeb6a35ba7ab44e8eaf2a84e4eecea43f2b4961206a78cb8a6b4a4852a",
+  "small_04.in": "d580838df4d46856c97af6fc78cd471a01fd79a9cbd46768ca2c8a3ef3b380bb",
+  "small_04.out": "b9d22ac93fc60a758f5e419fcbbb60d0e0408ae2d193f73574172b8d70750d4c",
+  "small_05.in": "c4a1873f7f6105d77e4c964e00c6b3dbaa3b8fe9b2ad2b5062f147cd2e83d0c4",
+  "small_05.out": "21641c230ef7cc39ea0afa969b94e822d4e43df2c9790b40ac8e6e3cd9557bdc",
+  "small_06.in": "939cb5c8dd9870fd2a315d471c52c800e12573e01ae5b8b861e055bebc0a531f",
+  "small_06.out": "13e1459a241b3f6c649876ea801fc02b13283d5be817b32ba6bdbcb3ad7db7e3",
+  "small_07.in": "2621399c9f319ae16155daefce41f57d068370ed4866b6c07781b0aef5ca1fab",
+  "small_07.out": "1f3fd66cff566e520dd156241d820d322aae2dc2a160550d62cd79293f606b09",
+  "small_08.in": "43b2a4417663df99cc254a9cd0210f3c141cad25f28b797c70869d2a8dfd1e2d",
+  "small_08.out": "a5351352805a6c73d99ea05553194ca40014b5445d6d9f3716590a18af000994",
+  "small_slopes_00.in": "ee8972534274bfa318d3443ece5d79393bb69e40b02b84cc38ed907e8d3c9893",
+  "small_slopes_00.out": "9c2e46d775f028f541f30bd635265452049d83e6e43f6cca6714302283b91a47",
+  "small_slopes_01.in": "29932d0af66b3147f607708cfd301ff2d91f7862c2d6c20b08773a6bef0314d7",
+  "small_slopes_01.out": "ccc813ef1f3604a593e20b4428555cc788ee1d01a759abbf6ca3589ddcbe8577"
+}
\ No newline at end of file
diff --git a/convolution/min_plus_convolution_concave_arbitrary/info.toml b/convolution/min_plus_convolution_concave_arbitrary/info.toml
new file mode 100644
index 00000000..ecc2392b
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/info.toml
@@ -0,0 +1,56 @@
+title = 'Min Plus Convolution (Concave and Arbitrary)'
+timelimit = 5.0
+forum = "https://github.com/yosupo06/library-checker-problems/issues/1252"
+
+[[tests]]
+    name = "example.in"
+    number = 1
+
+[[tests]]
+    name = "small.cpp"
+    number = 9
+
+[[tests]]
+    name = "med_random.cpp"
+    number = 3
+
+[[tests]]
+    name = "random.cpp"
+    number = 3
+
+[[tests]]
+    name = "max_random.cpp"
+    number = 3
+
+[[tests]]
+    name = "monotone.cpp"
+    number = 4
+
+[[tests]]
+    name = "small_slopes.cpp"
+    number = 2
+
+[[tests]]
+    name = "near_power_of_2.cpp"
+    number = 9
+
+[[tests]]
+    name = "hack.in"
+    number = 1
+
+[[tests]]
+    name = "large_small.cpp"
+    number = 4
+
+[[tests]]
+    name = "only_first_small.cpp"
+    number = 2
+
+[[solutions]]
+    name = "naive.cpp"
+    allow_tle = true
+
+[params]
+    N_MAX = 524288
+    A_MIN = 0
+    A_MAX = 1000000000
diff --git a/convolution/min_plus_convolution_concave_arbitrary/sol/correct.cpp b/convolution/min_plus_convolution_concave_arbitrary/sol/correct.cpp
new file mode 100644
index 00000000..5c3071a2
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/sol/correct.cpp
@@ -0,0 +1,128 @@
+#include <cstdio>
+#include <vector>
+#include <cassert>
+#include <limits>
+#include <functional>
+
+using namespace std;
+
+
+template <class FUNC>
+vector<int> monotone_minima(int h, int w, const FUNC& a) {
+	vector<int> j_min(h);
+
+	for (int di = 1 << 20; di > 0; di >>= 1) {
+		int di2 = 2 * di;
+		for (int i = di; i <= h; i += di2) {
+			int jL = (i - di > 0 ? j_min[i - di - 1] : 0);
+			int jR = (i + di <= h ? j_min[i + di - 1] : w - 1);
+
+			int a_min = numeric_limits<int>::max();
+			for (int j = jL; j <= jR; j++) {
+				int val = a(i - 1, j);
+				
+				if (a_min > val) {
+					a_min = val;
+					j_min[i - 1] = j;
+				}
+			}
+		}
+	}
+
+	return j_min;
+}
+
+
+vector<int> semiconcave_min_plus_convolution(const vector<int>& a, const vector<int>& b) {
+	int n = (int)a.size(), m = (int)b.size();
+	
+	// b is concave
+	for (int i = 0; i < m - 2; ++i) assert(b[i] + b[i + 2] <= 2 * b[i + 1]);
+
+	int h = n + m - 1, w = n;
+
+	vector<int> y_min(h, 0), y_max(h, w - 1);
+	for (int x = m; x <= h - 1; x++) y_min[x] = x - m + 1;
+	for (int x = 0; x <= h - m; x++) y_max[x] = x;
+
+	vector<int> x_min(w), x_max(w);
+	for (int y = 0; y < w; y++) {
+		x_min[y] = y;
+		x_max[y] = m - 1 + y;
+	}
+
+	vector<int> c(h, numeric_limits<int>::max());
+
+	function<void(int, int, int, int)> rf = [&](int x1, int x2, int y1, int y2) {
+		if (y_max[x1] >= y2 && y1 >= y_min[x2]) {
+			auto A = [&](int i, int j) {
+				return a[y2 - j] + b[(x1 + i) - (y2 - j)];
+			};
+
+			auto j_min = monotone_minima(x2 - x1 + 1, y2 - y1 + 1, A);
+
+			for (int i = x1; i <= x2; i++) {
+				int val = A(i - x1, j_min[i - x1]);
+				if (c[i] > val) c[i] = val;
+			}
+
+			return;
+		}
+
+		if ((long long)(x2 - x1) * (y2 - y1) < 1000) {
+			for (int x = x1; x <= x2; x++) {
+				int y_from = max(y_min[x], y1);
+				int y_to = min(y_max[x], y2);
+				for (int y = y_from; y <= y_to; y++) {
+					int val = a[y] + b[x - y];
+					if (c[x] > val) c[x] = val;
+				}
+			}
+			return;
+		}
+
+		if (x2 - x1 > y2 - y1) {
+			int xm = (x1 + x2) / 2;
+
+			int ny2 = min(y_max[xm], y2);
+			if (y1 <= ny2) rf(x1, xm, y1, ny2);
+
+			int ny1 = max(y_min[xm], y1);
+			if (ny1 <= y2) rf(xm + 1, x2, ny1, y2);
+		}
+		else {
+			int ym = (y1 + y2) / 2;
+
+			int nx2 = min(x_max[ym], x2);
+			if (x1 <= nx2) rf(x1, nx2, y1, ym);
+
+			int nx1 = max(x_min[ym], x1);
+			if (nx1 <= x2) rf(nx1, x2, ym + 1, y2);
+		}
+	};
+	rf(0, h - 1, 0, w - 1);
+
+	return c;
+}
+
+
+int main() {
+	int n, m;
+	scanf("%d %d", &n, &m);
+
+	vector<int> a(n), b(m);
+	for (int i = 0; i < n; i++) scanf("%d", &a[i]);
+	for (int j = 0; j < m; j++) scanf("%d", &b[j]);
+
+	vector<int> c = semiconcave_min_plus_convolution(b, a);
+
+	int K = n + m - 1;
+	for (int k = 0; k < K; k++) {
+		printf("%d", c[k]);
+
+		if (k < K - 1) printf(" ");
+		else printf("\n");
+	}
+
+	return 0;
+}
diff --git a/convolution/min_plus_convolution_concave_arbitrary/sol/naive.cpp b/convolution/min_plus_convolution_concave_arbitrary/sol/naive.cpp
new file mode 100644
index 00000000..715fb5ad
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/sol/naive.cpp
@@ -0,0 +1,26 @@
+#include <cassert>
+#include <cstdio>
+#include <limits>
+#include <vector>
+
+using namespace std;
+
+int main() {
+  int n, m;
+  scanf("%d %d", &n, &m);
+  vector<int> a(n), b(m);
+  for (int i = 0; i < n; ++i) scanf("%d", &a[i]);
+  for (int i = 0; i < m; ++i) scanf("%d", &b[i]);
+  vector<int> c(n + m - 1, numeric_limits<int>::max());
+  for (int i = 0; i < n; ++i) {
+    for (int j = 0; j < m; ++j) {
+      int x = a[i] + b[j];
+      if (c[i + j] > x) c[i + j] = x;
+    }
+  }
+  for (int k = 0; k < n + m - 1; ++k) {
+    if (k) printf(" ");
+    printf("%d", c[k]);
+  }
+  return 0;
+}
diff --git a/convolution/min_plus_convolution_concave_arbitrary/task.md b/convolution/min_plus_convolution_concave_arbitrary/task.md
new file mode 100644
index 00000000..83e8bb94
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/task.md
@@ -0,0 +1,42 @@
+## @{keyword.statement}
+
+@{lang.en}
+Given integer sequences $a_0, a_1, ..., a_{N - 1}$ and $b_0, b_1, ..., b_{M - 1}$. Here, $a$ is concave. Calculate an integer sequence $c_0, c_1, ..., c_{(N - 1) + (M - 1)}$ defined as follows:
+
+$$c_k = \min_{i+j=k} (a_i+b_j)$$
+
+@{lang.ja}
+整数列 $a_0, a_1, ..., a_{N - 1}$、$b_0, b_1, ..., b_{M - 1}$ が与えられます。ここで $a$ は凹です。次のように定義される整数列 $c_0, c_1, ..., c_{(N - 1) + (M - 1)}$ を求めてください。
+
+$$c_k = \min_{i+j=k} (a_i+b_j)$$
+
+@{lang.end}
+
+## @{keyword.constraints}
+
+@{lang.ja}
+- $1 \leq N, M \leq @{param.N_MAX}$
+- $0 \leq a_i, b_i \leq @{param.A_MAX}$
+- $a$ は凹である。つまり $0\leq i < N - 2$ に対して $a_{i+1}-a_i\geq a_{i+2}-a_{i+1}$ が成り立つ。
+@{lang.en}
+- $1 \leq N, M \leq @{param.N_MAX}$
+- $0 \leq a_i, b_i \leq @{param.A_MAX}$
+- $a$ is concave. i.e. $a_{i+1}-a_i\geq a_{i+2}-a_{i+1}$ holds for $0\leq i < N - 2$. 
+@{lang.end}
+## @{keyword.input}
+
+```
+$N$ $M$
+$a_0$ $a_1$ ... $a_{N-1}$
+$b_0$ $b_1$ ... $b_{M-1}$
+```
+
+## @{keyword.output}
+
+```
+$c_0$ $c_1$ ... $c_{(N - 1) + (M - 1)}$
+```
+
+## @{keyword.sample}
+
+@{example.example_00}
diff --git a/convolution/min_plus_convolution_concave_arbitrary/verifier.cpp b/convolution/min_plus_convolution_concave_arbitrary/verifier.cpp
new file mode 100644
index 00000000..178529e2
--- /dev/null
+++ b/convolution/min_plus_convolution_concave_arbitrary/verifier.cpp
@@ -0,0 +1,38 @@
+#include <iostream>
+
+#include "params.h"
+#include "testlib.h"
+
+using namespace std;
+
+bool is_concave(vector<int>& a) {
+  int n = a.size();
+  for (int i = 0; i < n - 2; ++i) {
+    if (a[i + 1] - a[i] < a[i + 2] - a[i + 1]) return false;
+  }
+  return true;
+}
+
+int main() {
+  registerValidation();
+  int n = inf.readInt(1, N_MAX);
+  inf.readSpace();
+  int m = inf.readInt(1, N_MAX);
+  inf.readChar('\n');
+  vector<int> a(n), b(m);
+  for (int i = 0; i < n; i++) {
+    if (i) inf.readSpace();
+    a[i] = inf.readInt(A_MIN, A_MAX);
+  }
+  inf.readChar('\n');
+  for (int i = 0; i < m; i++) {
+    if (i) inf.readSpace();
+    b[i] = inf.readInt(A_MIN, A_MAX);
+  }
+  inf.readChar('\n');
+
+  ensure(is_concave(a));
+
+  inf.readEof();
+  return 0;
+}