Replace arxiv pdf links with abs links (#8887)

torch_geometric/datasets/rcdd.py

@@ -16,7 +16,7 @@
 class RCDD(InMemoryDataset):
     r"""The risk commodity detection dataset (RCDD) from the
     `"Datasets and Interfaces for Benchmarking Heterogeneous Graph
-    Neural Networks" <http://shichuan.org/doc/156.pdf>`_ paper.
+    Neural Networks" <https://dl.acm.org/doi/10.1145/3583780.3615117>`_ paper.
     RCDD is an industrial-scale heterogeneous graph dataset based on a
     real risk detection scenario from Alibaba's e-commerce platform.
     It consists of 13,806,619 nodes and 157,814,864 edges across 7 node types

torch_geometric/datasets/wikidata.py

@@ -15,7 +15,7 @@
 class Wikidata5M(InMemoryDataset):
     r"""The Wikidata-5M dataset from the `"KEPLER: A Unified Model for
     Knowledge Embedding and Pre-trained Language Representation"
-    <https://arxiv.org/pdf/1911.06136.pdf>`_ paper,
+    <https://arxiv.org/abs/1911.06136>`_ paper,
     containing 4,594,485 entities, 822 relations,
     20,614,279 train triples, 5,163 validation triples, and 5,133 test triples.
 

torch_geometric/nn/conv/han_conv.py

@@ -34,7 +34,7 @@ def group(
 class HANConv(MessagePassing):
     r"""The Heterogenous Graph Attention Operator from the
     `"Heterogenous Graph Attention Network"
-    <https://arxiv.org/pdf/1903.07293.pdf>`_ paper.
+    <https://arxiv.org/abs/1903.07293>`_ paper.
 
     .. note::
 

torch_geometric/nn/conv/pdn_conv.py

@@ -12,7 +12,7 @@
 class PDNConv(MessagePassing):
     r"""The pathfinder discovery network convolutional operator from the
     `"Pathfinder Discovery Networks for Neural Message Passing"
-    <https://arxiv.org/pdf/2010.12878.pdf>`_ paper.
+    <https://arxiv.org/abs/2010.12878>`_ paper.
 
     .. math::
         \mathbf{x}^{\prime}_i = \sum_{j \in \mathcal{N}(i) \cup

torch_geometric/nn/encoding.py

@@ -6,7 +6,7 @@
 
 class PositionalEncoding(torch.nn.Module):
     r"""The positional encoding scheme from the `"Attention Is All You Need"
-    <https://arxiv.org/pdf/1706.03762.pdf>`_ paper.
+    <https://arxiv.org/abs/1706.03762>`_ paper.
 
     .. math::
 

torch_geometric/nn/models/visnet.py

@@ -340,7 +340,7 @@ def forward(
 class NeighborEmbedding(MessagePassing):
     r"""The :class:`NeighborEmbedding` module from the `"Enhancing Geometric
     Representations for Molecules with Equivariant Vector-Scalar Interactive
-    Message Passing" <https://arxiv.org/pdf/2210.16518.pdf>`_ paper.
+    Message Passing" <https://arxiv.org/abs/2210.16518>`_ paper.
 
     Args:
         hidden_channels (int): The number of hidden channels in the node
@@ -415,7 +415,7 @@ def message(self, x_j: Tensor, W: Tensor) -> Tensor:
 class EdgeEmbedding(torch.nn.Module):
     r"""The :class:`EdgeEmbedding` module from the `"Enhancing Geometric
     Representations for Molecules with Equivariant Vector-Scalar Interactive
-    Message Passing" <https://arxiv.org/pdf/2210.16518.pdf>`_ paper.
+    Message Passing" <https://arxiv.org/abs/2210.16518>`_ paper.
 
     Args:
         num_rbf (int): The number of radial basis functions.
@@ -458,7 +458,7 @@ class ViS_MP(MessagePassing):
     equivariant vector-scalar interactive graph neural network (ViSNet)
     from the `"Enhancing Geometric Representations for Molecules with
     Equivariant Vector-Scalar Interactive Message Passing"
-    <https://arxiv.org/pdf/2210.16518.pdf>`_ paper.
+    <https://arxiv.org/abs/2210.16518>`_ paper.
 
     Args:
         num_heads (int): The number of attention heads.
@@ -659,7 +659,7 @@ class ViS_MP_Vertex(ViS_MP):
     equivariant vector-scalar interactive graph neural network (ViSNet)
     from the `"Enhancing Geometric Representations for Molecules with
     Equivariant Vector-Scalar Interactive Message Passing"
-    <https://arxiv.org/pdf/2210.16518.pdf>`_ paper.
+    <https://arxiv.org/abs/2210.16518>`_ paper.
 
     Args:
         num_heads (int): The number of attention heads.
@@ -723,7 +723,7 @@ class ViSNetBlock(torch.nn.Module):
     r"""The representation module of the equivariant vector-scalar
     interactive graph neural network (ViSNet) from the `"Enhancing Geometric
     Representations for Molecules with Equivariant Vector-Scalar Interactive
-    Message Passing" <https://arxiv.org/pdf/2210.16518.pdf>`_ paper.
+    Message Passing" <https://arxiv.org/abs/2210.16518>`_ paper.
 
     Args:
         lmax (int, optional): The maximum degree of the spherical harmonics.
@@ -875,7 +875,7 @@ class GatedEquivariantBlock(torch.nn.Module):
     r"""Applies a gated equivariant operation to scalar features and vector
     features from the `"Enhancing Geometric Representations for Molecules with
     Equivariant Vector-Scalar Interactive Message Passing"
-    <https://arxiv.org/pdf/2210.16518.pdf>`_ paper.
+    <https://arxiv.org/abs/2210.16518>`_ paper.
 
     Args:
         hidden_channels (int): The number of hidden channels in the node
@@ -1039,7 +1039,7 @@ class ViSNet(torch.nn.Module):
     vector-scalar interactive graph neural network (ViSNet) from the
     `"Enhancing Geometric Representations for Molecules with Equivariant
     Vector-Scalar Interactive Message Passing"
-    <https://arxiv.org/pdf/2210.16518.pdf>`_ paper.
+    <https://arxiv.org/abs/2210.16518>`_ paper.
 
     Args:
         lmax (int, optional): The maximum degree of the spherical harmonics.

torch_geometric/nn/norm/mean_subtraction_norm.py

@@ -9,7 +9,7 @@
 class MeanSubtractionNorm(torch.nn.Module):
     r"""Applies layer normalization by subtracting the mean from the inputs
     as described in the  `"Revisiting 'Over-smoothing' in Deep GCNs"
-    <https://arxiv.org/pdf/2003.13663.pdf>`_ paper.
+    <https://arxiv.org/abs/2003.13663>`_ paper.
 
     .. math::
         \mathbf{x}_i = \mathbf{x}_i - \frac{1}{|\mathcal{V}|}