diff --git a/src/frontends/paddle/src/op/round.cpp b/src/frontends/paddle/src/op/round.cpp
new file mode 100644
index 00000000000000..f981fa1e841843
--- /dev/null
+++ b/src/frontends/paddle/src/op/round.cpp
@@ -0,0 +1,22 @@
+// Copyright (C) 2018-2023 Intel Corporation
+// SPDX-License-Identifier: Apache-2.0
+//
+
+#include "default_opset.hpp"
+#include "openvino/frontend/paddle/node_context.hpp"
+
+namespace ov {
+namespace frontend {
+namespace paddle {
+namespace op {
+NamedOutputs round(const NodeContext& node) {
+    return node.default_single_output_mapping(
+        {std::make_shared<default_opset::Round>(node.get_input("X"),
+                                                ov::op::v5::Round::RoundMode::HALF_AWAY_FROM_ZERO)},
+        {"Out"});
+}
+
+}  // namespace op
+}  // namespace paddle
+}  // namespace frontend
+}  // namespace ov
diff --git a/src/frontends/paddle/src/op/set_value.cpp b/src/frontends/paddle/src/op/set_value.cpp
index 63260b60da45c4..94c851479595ed 100644
--- a/src/frontends/paddle/src/op/set_value.cpp
+++ b/src/frontends/paddle/src/op/set_value.cpp
@@ -2,6 +2,8 @@
 // SPDX-License-Identifier: Apache-2.0
 //
 
+#include <limits>
+
 #include "default_opset.hpp"
 #include "openvino/frontend/paddle/node_context.hpp"
 
@@ -20,11 +22,11 @@ std::shared_ptr<Node> handle_minus_index(const std::vector<int64_t>& node, const
     return new_node;
 }
 
-// std::shared_ptr<Node> handle_maximum_index(Output<Node>& node, const Output<Node>& update_node) {
-//     const auto maximum_node = default_opset::Constant::create(element::i64, {1}, {INT_MAX});
-//     const auto mask = std::make_shared<default_opset::Equal>(node, maximum_node);
-//     return std::make_shared<default_opset::Select>(mask, update_node, node);
-// }
+std::shared_ptr<Node> handle_maximum_index(Output<Node>& node, const Output<Node>& update_node) {
+    const auto maximum_node = default_opset::Constant::create(element::i64, {1}, {std::numeric_limits<int32_t>::max()});
+    const auto mask = std::make_shared<default_opset::Equal>(node, maximum_node);
+    return std::make_shared<default_opset::Select>(mask, update_node, node);
+}
 
 bool is_contain_minus(const std::vector<int64_t> vec) {
     for (int64_t i : vec) {
@@ -52,14 +54,14 @@ NamedOutputs set_value(const NodeContext& node) {
     // Given:
     // input_data: shape(5, 6, 7, 8, 9)
     // update_value: shape(1, 6, 3, 3)
-    // operation: input_data[:, :, 2: 7: 2, -4: -1] = update_value
+    // operation: input_data[:, :, 2: 7: 2, -4: -1, :] = update_value
     // axes = [2, 3]
     // starts = [2, -4]
     // ends = [7, -1]
     // steps = [2, 1]
     // Our process is:
     // 1. Get axes [2, 3], get shape of input [5, 6, 7, 8, 9], select dimension from shape by axes: [7, 8].
-    // 2. Get starts [2, -4] and ends [3, -1]. Process minus starts and ends. starts: [2, 4], ends: [7, 7].
+    // 2. Get starts [2, -4] and ends [7, -1]. Process minus starts and ends. starts: [2, 4], ends: [7, 7].
     // 3. Calculate starts_node, ends_node and steps_node
     //    1. Create `starts node` filled with 0. Update `starts` to `starts_node` according to axes.
     //    starts_node[axes[i]] = starts[i] for i in axes.size
@@ -92,25 +94,41 @@ NamedOutputs set_value(const NodeContext& node) {
     const auto slice_shape = default_opset::Constant::create(ov::element::i64, {1, 1}, {-1});
 
     // get positive starts ends and steps
-    if (node.has_input("StartsTensorList") && node.has_input("StepsTensorList") && node.has_input("EndsTensorList")) {
+    if (node.has_input("StartsTensorList")) {
         starts = handle_minus_index(node.get_ng_inputs("StartsTensorList"), spec_dim_node);
-        ends = handle_minus_index(node.get_ng_inputs("EndsTensorList"), spec_dim_node);
-        steps = std::make_shared<default_opset::Concat>(node.get_ng_inputs("StepsTensorList"), 0);
-    } else if (node.has_attribute("starts") && node.has_attribute("steps") && node.has_attribute("ends")) {
-        const auto start_vec = node.get_attribute<std::vector<int64_t>>("starts");
-        const auto ends_vec = node.get_attribute<std::vector<int64_t>>("ends");
-        const auto step_vec = node.get_attribute<std::vector<int64_t>>("steps");
-        if (is_contain_minus(start_vec) || is_contain_minus(ends_vec) || is_contain_minus(step_vec)) {
-            PADDLE_OP_CHECK(node, (false), "Currently not support minus start, ends and steps!");
+    } else if (node.has_attribute("starts")) {
+        auto start_vec = node.get_attribute<std::vector<int64_t>>("starts");
+        if (is_contain_minus(start_vec)) {
+            PADDLE_OP_CHECK(node, (false), "Currently not support minus start!");
         }
         starts = handle_minus_index(start_vec, spec_dim_node);
+    } else
+        PADDLE_OP_CHECK(node, (false), "Invalid arguments!");
+
+    if (node.has_input("EndsTensorList")) {
+        ends = handle_minus_index(node.get_ng_inputs("EndsTensorList"), spec_dim_node);
+    } else if (node.has_attribute("ends")) {
+        auto ends_vec = node.get_attribute<std::vector<int64_t>>("ends");
+        if (is_contain_minus(ends_vec)) {
+            PADDLE_OP_CHECK(node, (false), "Currently not support minus ends!");
+        }
         ends = handle_minus_index(ends_vec, spec_dim_node);
-        steps = default_opset::Constant::create(element::i64, {step_vec.size()}, step_vec);
+    } else
+        PADDLE_OP_CHECK(node, (false), "Invalid arguments!");
+
+    if (node.has_input("StepsTensorList")) {
+        steps = handle_minus_index(node.get_ng_inputs("StepsTensorList"), spec_dim_node);
+    } else if (node.has_attribute("steps")) {
+        auto step_vec = node.get_attribute<std::vector<int64_t>>("steps");
+        if (is_contain_minus(step_vec)) {
+            PADDLE_OP_CHECK(node, (false), "Currently not support minus steps!");
+        }
+        steps = handle_minus_index(step_vec, spec_dim_node);
     } else
         PADDLE_OP_CHECK(node, (false), "Invalid arguments!");
 
     // for unsepcified end: x[::2], end will be 2147483647
-    // ends = handle_maximum_index(ends, spec_dim_node);
+    ends = handle_maximum_index(ends, spec_dim_node);
 
     // 3.1 get starts node
     starts_node =
@@ -142,7 +160,12 @@ NamedOutputs set_value(const NodeContext& node) {
         std::make_shared<default_opset::ScatterNDUpdate>(input_shape, axes_node, value_shape_update_node);
 
     // 4.5 broadcast
-    value_node = std::make_shared<default_opset::Broadcast>(value_node, value_target_shape);
+    auto value_shape = std::make_shared<default_opset::ShapeOf>(value_node);
+    auto value_rank = std::make_shared<default_opset::ShapeOf>(value_shape);
+    auto value_rank_scalar = std::make_shared<default_opset::Squeeze>(value_rank);
+    Output<Node> broadcast_axes =
+        std::make_shared<default_opset::Range>(zero_node, value_rank_scalar, one_node, element::i64);
+    value_node = std::make_shared<default_opset::Broadcast>(value_node, value_target_shape, broadcast_axes);
 
     // get total number of elements
     const auto numel_node = std::make_shared<default_opset::ReduceProd>(input_shape, zero_node);
diff --git a/src/frontends/paddle/src/op_table.cpp b/src/frontends/paddle/src/op_table.cpp
index c22441c72d85cd..3030f140996de2 100644
--- a/src/frontends/paddle/src/op_table.cpp
+++ b/src/frontends/paddle/src/op_table.cpp
@@ -98,6 +98,7 @@ OP_CONVERTER(reshape2);
 OP_CONVERTER(reverse);
 OP_CONVERTER(rnn);
 OP_CONVERTER(roi_align);
+OP_CONVERTER(round);
 OP_CONVERTER(scale);
 OP_CONVERTER(select_input);
 OP_CONVERTER(set_value);
@@ -233,6 +234,7 @@ std::map<std::string, CreatorFunction> get_supported_ops() {
             {"reverse", op::reverse},
             {"rnn", op::rnn},
             {"roi_align", op::roi_align},
+            {"round", op::round},
             {"scale", op::scale},
             {"select_input", op::select_input},
             {"set_value", op::set_value},
diff --git a/src/frontends/paddle/tests/op_fuzzy.cpp b/src/frontends/paddle/tests/op_fuzzy.cpp
index d1d9d913004a3a..ecc7fe619195b6 100644
--- a/src/frontends/paddle/tests/op_fuzzy.cpp
+++ b/src/frontends/paddle/tests/op_fuzzy.cpp
@@ -490,6 +490,7 @@ static const std::vector<std::string> models{
     std::string("rnn_lstm_layer_2_bidirectional_seq_len_4/rnn_lstm_layer_2_bidirectional_seq_len_4.pdmodel"),
     std::string("roi_align_test"),
     std::string("roi_align_test2"),
+    std::string("round"),
     std::string("scale_bias_after_float32"),
     std::string("scale_bias_after_int32"),
     std::string("scale_bias_after_int64"),
@@ -505,6 +506,7 @@ static const std::vector<std::string> models{
     std::string("set_value5"),
     // std::string("set_value6"),
     // std::string("set_value7"),
+    // std::string("set_value8"),
     // std::string("set_value_dynamic1"),
     std::string("set_value_dynamic2"),
     std::string("shape"),
diff --git a/src/frontends/paddle/tests/test_models/gen_scripts/generate_round.py b/src/frontends/paddle/tests/test_models/gen_scripts/generate_round.py
new file mode 100644
index 00000000000000..18d93ded10d7bf
--- /dev/null
+++ b/src/frontends/paddle/tests/test_models/gen_scripts/generate_round.py
@@ -0,0 +1,40 @@
+# Copyright (C) 2018-2023 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+#
+# round paddle model generator
+#
+import numpy as np
+from save_model import saveModel
+import paddle
+import sys
+
+data_type = 'float32'
+
+def paddle_round(name:str, x):
+    paddle.enable_static()
+
+    with paddle.static.program_guard(paddle.static.Program(), paddle.static.Program()):
+        data = paddle.static.data(name='x', shape=x.shape, dtype = data_type)
+        out = paddle.round(data)
+
+        cpu = paddle.static.cpu_places(1)
+        exe = paddle.static.Executor(cpu[0])
+        # startup program will call initializer to initialize the parameters.
+        exe.run(paddle.static.default_startup_program())
+
+        outs = exe.run(
+            feed={'x': x},
+            fetch_list=[out])
+
+        saveModel(name, exe, feedkeys=['x'], fetchlist=[out], inputs=[x], outputs=[outs[0]], target_dir=sys.argv[1])
+
+    return outs[0]
+
+def main():
+    x = np.random.uniform(-1000,1000, (8, 24, 32)).astype(data_type)
+
+    paddle_round("round", x)
+
+if __name__ == "__main__":
+    main()
diff --git a/src/frontends/paddle/tests/test_models/gen_scripts/generate_set_value.py b/src/frontends/paddle/tests/test_models/gen_scripts/generate_set_value.py
index 4a9be9b3017e78..3230b09dffc3d3 100644
--- a/src/frontends/paddle/tests/test_models/gen_scripts/generate_set_value.py
+++ b/src/frontends/paddle/tests/test_models/gen_scripts/generate_set_value.py
@@ -10,6 +10,7 @@
 import paddle
 from save_model import saveModel
 
+maxint32 = np.iinfo(np.int32).max
 
 def concat(data):
     data = [np.expand_dims(d, 0) for d in data]
@@ -141,6 +142,21 @@ def set_value5(x, value, *slice):
 
     # paddle_set_value("set_value7", data, value, set_value_step1, dtype)
 
+    shape = (7, 9)
+    dtype = "int32"
+    data = np.random.randint(0, 5, shape).astype(dtype)
+    value = np.random.randint(-100, -1, (3, 1)).astype(dtype)
+
+    starts = generate_data([4], np.int64)
+    ends = generate_data([maxint32], np.int64)
+    steps = generate_data([1], np.int64)
+
+    def set_value8(x, value, *slice):
+        x[build_slice(*slice)] = value
+        return x
+
+    paddle_set_value("set_value8", data, value, set_value8, dtype, starts, ends, steps)
+
     # shape = (10, 5)
     # dtype = "float32"
     # data = np.random.randint(0, 5, shape).astype(dtype)
@@ -167,4 +183,4 @@ def set_value7(x, value, *slice):
     paddle_set_value("set_value_dynamic2", data, value, set_value7, dtype, starts, ends, steps, is_dynamic=True)
 
 if __name__ == "__main__":
-    main()
\ No newline at end of file
+    main()