diff --git a/.github/workflows/pr.yaml b/.github/workflows/pr.yaml
index b587d0d7..eb796c12 100644
--- a/.github/workflows/pr.yaml
+++ b/.github/workflows/pr.yaml
@@ -1,11 +1,6 @@
 name: TritonBench PR Test
 on:
   pull_request:
-    paths:
-      - .ci/*
-      - test/test_gpu/*
-      - tritonbench/*
-      - .github/workflows/pr.yaml
   push:
     branches:
       - main
diff --git a/submodules/generative-recommenders b/submodules/generative-recommenders
index 53321455..77d8b947 160000
--- a/submodules/generative-recommenders
+++ b/submodules/generative-recommenders
@@ -1 +1 @@
-Subproject commit 53321455e93bef51b2622cd97ae6b28ca3c6fa68
+Subproject commit 77d8b9474e9a1acc7a892f0caf9fcec16d287aa0
diff --git a/tritonbench/operators/ragged_attention/hstu.py b/tritonbench/operators/ragged_attention/hstu.py
index 896c5ee6..bfedd369 100644
--- a/tritonbench/operators/ragged_attention/hstu.py
+++ b/tritonbench/operators/ragged_attention/hstu.py
@@ -170,7 +170,7 @@ def forward(
                 kwargs["num_targets"],
                 kwargs["ATTN_BIAS_TYPE"],  # relative_bias_type
                 kwargs["MAX_ATTN_LEN"],  # max_attn_len
-                kwargs["contextual_seq_len"],  # contextual_seq_len
+                kwargs["CONTEXTUAL_SEQ_LEN"],  # contextual_seq_len
                 kwargs["sort_by_length_indices"],  # sort_by_length
             )