update torch-xla related in READMEs

PiperOrigin-RevId: 706015739

README.md

@@ -102,16 +102,6 @@ Alternately, the nightly version can be installed with:
 pip install ai-edge-torch-nightly
 ```
 
-### Update LD_LIBRARY_PATH if necessary
-
-Torch XLA builds a shared library, `_XLAC.so` that needs to link to the version of Python
-it was built with (currently 3.10 or 3.11). In order to ensure that `import _XLAC` can succeed,
-update the LD_LIBRARY_PATH to the lib directory of your Python environment:
-
-```bash
-export LD_LIBRARY_PATH=<path to Python installation>/lib:$LD_LIBRARY_PATH
-```
-
 
 * The list of versioned releases can be seen [here](https://github.com/google-ai-edge/ai-edge-torch/releases).
 * The full list of PyPi releases (including nightly builds) can be seen [here](https://pypi.org/project/ai-edge-torch/#history).

docs/pytorch_converter/README.md

@@ -327,19 +327,26 @@ pip install ai-edge-model-explorer
 model-explorer 'resnet.tflite'
 ```
 
-# Use ODML Torch Conversion Backend (Experimental)
+# Use Torch XLA Conversion Backend (Legacy)
 
-ODML Torch is an experimental conversion backend for AI Edge Torch, which enables
-conversion on macOS and generates TFLite Flatbuffer with higher quality and better
-performance. To use AI Edge Torch with ODML Torch, setup your environment these
-commands:
+AI Edge Torch has been switched to a modern conversion backend for better 
+on-device compatibility and performance. However, if you encounter compatibility
+issues, you can optionally switch to the legacy Torch XLA backend:
 
 ```bash
-# Install ODML Torch backend dependencies
-pip install -r https://raw.githubusercontent.com/google-ai-edge/ai-edge-torch/main/odmltorch-requirements.txt
-# Install ai-edge-torch with no default backend (Torch XLA) dependencies
-pip install ai-edge-torch-nightly --no-deps
-
-# Disable default AI Edge Torch backend (Torch XLA)
-export USE_TORCH_XLA=0
-```
+# Install ai-edge-torch with torch-xla dependency
+pip install ai-edge-torch-nightly[torch-xla]
+
+# Enable torch-xla as the AI Edge Torch backend
+export USE_TORCH_XLA=1
+```
+
+## Update LD_LIBRARY_PATH if necessary
+
+Torch XLA builds a shared library, `_XLAC.so` that needs to link to the version of Python
+it was built with (currently 3.10 or 3.11). In order to ensure that `import _XLAC` can succeed,
+update the LD_LIBRARY_PATH to the lib directory of your Python environment:
+
+```bash
+export LD_LIBRARY_PATH=<path to Python installation>/lib:$LD_LIBRARY_PATH
+```