[PERFORMANCE OPTIMIZATION] Flash Sparse Attention

.github/ISSUE_TEMPLATE/bug_report.md

@@ -20,7 +20,7 @@ A clear and concise description of what the bug is.
 
 **To Reproduce**
 Steps to reproduce the behavior:
-1. Import flash_dmattn
+1. Import flash_sparse_attn
 2. Run the following code:
 ```python
 # Paste your code here

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -22,7 +22,7 @@ body:
     attributes:
       label: Describe the bug
       description: Provide a concise description of the incorrect behaviour.
-      placeholder: Unexpected error when calling flash_dmattn(...)
+      placeholder: Unexpected error when calling flash_sparse_attn(...)
     validations:
       required: true
   - type: textarea
@@ -31,7 +31,7 @@ body:
       label: Steps to reproduce
       description: Share the minimal steps or code necessary for us to see the failure.
       placeholder: |
-        1. Import flash_dmattn
+        1. Import flash_sparse_attn
         2. Run the snippet below
         3. Observe the error
       render: python

.github/ISSUE_TEMPLATE/feature_request.yml

@@ -44,7 +44,7 @@ body:
     attributes:
       label: Implementation details
       description: Call out potential CUDA/Python changes, performance implications, or compatibility considerations.
-      placeholder: Requires updates to flash_dmattn_interface and CUDA op...
+      placeholder: Requires updates to flash_sparse_attn_interface and CUDA op...
   - type: textarea
     id: use-case
     attributes:

.github/PULL_REQUEST_TEMPLATE/bug_fix.yml

@@ -27,7 +27,7 @@ body:
     attributes:
       label: Changes
       description: Highlight the notable code-level modifications.
-      placeholder: Updated flash_dmattn_interface to...
+      placeholder: Updated flash_sparse_attn_interface to...
     validations:
       required: true
   - type: textarea

.github/PULL_REQUEST_TEMPLATE/feature_support.yml

@@ -27,7 +27,7 @@ body:
     attributes:
       label: Changes
       description: Describe new or changed public APIs, configuration, or CLI behaviour.
-      placeholder: Adds flash_dmattn.feature_flag...
+      placeholder: Adds flash_sparse_attn.feature_flag...
     validations:
       required: true
   - type: textarea

.github/workflows/publish.yml

@@ -93,7 +93,7 @@ jobs:
           pip install torch --index-url https://download.pytorch.org/whl/cpu
       - name: Build core package
         env:
-          FLASH_DMATTN_SKIP_CUDA_BUILD: "TRUE"
+          FLASH_SPARSE_ATTENTION_SKIP_CUDA_BUILD: "TRUE"
         run: |
           python setup.py sdist --dist-dir=dist
       - name: Deploy

CITATION.cff

@@ -1,7 +1,7 @@
 cff-version: "1.2.0"
 date-released: 2025-06
 message: "If you use this software, please cite it using these metadata."
-title: "Flash Sparse Attention: Trainable Dynamic Mask Sparse Attention"
+title: "Trainable Flash Sparse Attention"
 url: "https://github.com/flash-algo/flash-sparse-attention"
 authors:
   - family-names: Shi
@@ -42,7 +42,7 @@ preferred-citation:
       given-names: Guang
     - family-names: Luo
       given-names: Yuyu
-  title: "Trainable Dynamic Mask Sparse Attention"
+  title: "Trainable Flash Sparse Attention"
   year: 2025
   url: "https://arxiv.org/abs/2508.02124"
   doi: "10.48550/arXiv.2508.02124"

CONTRIBUTING.md

@@ -1,4 +1,4 @@
-# Contributing to Flash Dynamic Mask Attention
+# Contributing to Flash Sparse Attention
 
 Everyone is welcome to contribute, and we value everybody's contribution. Code contributions are not the only way to help the community. Answering questions, helping others, and improving the documentation are also immensely valuable.
 
@@ -8,7 +8,7 @@ However you choose to contribute, please be mindful and respect our [code of con
 
 ## Ways to contribute
 
-There are several ways you can contribute to Flash-DMA:
+There are several ways you can contribute to FSA:
 
 * Fix outstanding issues with the existing code.
 * Submit issues related to bugs or desired new features.
@@ -30,7 +30,7 @@ Do your best to follow these guidelines when submitting a bug-related issue or a
 
 ### Did you find a bug?
 
-The Flash-DMA library is robust and reliable thanks to users who report the problems they encounter.
+The FSA library is robust and reliable thanks to users who report the problems they encounter.
 
 Before you report an issue, we would really appreciate it if you could **make sure the bug was not already reported** (use the search bar on GitHub under Issues). Your issue should also be related to bugs in the library itself, and not your code.
 
@@ -50,7 +50,7 @@ python -c "import torch; print(f'PyTorch: {torch.__version__}'); print(f'CUDA: {
 
 ### Do you want a new feature?
 
-If there is a new feature you'd like to see in Flash-DMA, please open an issue and describe:
+If there is a new feature you'd like to see in FSA, please open an issue and describe:
 
 1. What is the *motivation* behind this feature? Is it related to performance optimization, memory efficiency, or new attention mechanisms?
 
@@ -77,7 +77,7 @@ We're always looking for improvements to the documentation that make it more cle
 
 Before writing any code, we strongly advise you to search through the existing PRs or issues to make sure nobody is already working on the same thing.
 
-You will need basic `git` proficiency to contribute to Flash-DMA. You'll need **Python 3.8+** and **CUDA 11.8+** to contribute.
+You will need basic `git` proficiency to contribute to FSA. You'll need **Python 3.8+** and **CUDA 11.8+** to contribute.
 
 ### Development Setup
 
@@ -120,7 +120,7 @@ You will need basic `git` proficiency to contribute to Flash-DMA. You'll need **
    python -m pytest tests/ -v
    ```
 
-   Flash-DMA also includes performance benchmarks. Run them to ensure your changes don't regress performance:
+   FSA also includes performance benchmarks. Run them to ensure your changes don't regress performance:
 
    ```bash
    python benchmarks/forward_performance.py