add chunked loss to readme (#453)

## Summary
<!--- This is a required section; please describe the main purpose of
this proposed code change. --->

<!---
## Details
This is an optional section; is there anything specific that reviewers
should be aware of?
--->

## Testing Done
<!--- This is a required section; please describe how this change was
tested. --->

<!-- 
Replace BLANK with your device type. For example, A100-80G-PCIe

Complete the following tasks before sending your PR, and replace `[ ]`
with
`[x]` to indicate you have done them. 
-->

- Hardware Type: <BLANK>
- [ ] run `make test` to ensure correctness
- [ ] run `make checkstyle` to ensure code style
- [ ] run `make test-convergence` to ensure convergence

README.md

@@ -72,6 +72,8 @@
 
 **Liger Kernel** is a collection of Triton kernels designed specifically for LLM training. It can effectively increase multi-GPU **training throughput by 20%** and reduces **memory usage by 60%**. We have implemented **Hugging Face Compatible** `RMSNorm`, `RoPE`, `SwiGLU`, `CrossEntropy`, `FusedLinearCrossEntropy`, and more to come. The kernel works out of the box with [Flash Attention](https://github.com/Dao-AILab/flash-attention), [PyTorch FSDP](https://pytorch.org/tutorials/intermediate/FSDP_tutorial.html), and [Microsoft DeepSpeed](https://github.com/microsoft/DeepSpeed). We welcome contributions from the community to gather the best kernels for LLM training.
 
+We've also added optimized Post-Training kernels that deliver **up to 80% memory savings** for alignment and distillation tasks. We support losses like DPO, CPO, ORPO, SimPO, JSD, and many more. 
+
 ## Supercharge Your Model with Liger Kernel
 
 ![Banner](https://raw.githubusercontent.com/linkedin/Liger-Kernel/main/docs/images/banner.GIF)
@@ -95,6 +97,7 @@ With one line of code, Liger Kernel can increase throughput by more than 20% and
 | [**Lightning Trainer**](https://github.com/linkedin/Liger-Kernel/tree/main/examples/lightning)         | Increase 15% throughput and reduce memory usage by 40% with LLaMA3-8B on MMLU dataset using 8 A100s with DeepSpeed ZeRO3 |
 | [**Medusa Multi-head LLM (Retraining Phase)**](https://github.com/linkedin/Liger-Kernel/tree/main/examples/medusa)        | Reduce memory usage by 80% with 5 LM heads and improve throughput by 40% using 8 A100s with FSDP |
 | [**Vision-Language Model SFT**](https://github.com/linkedin/Liger-Kernel/tree/main/examples/huggingface/run_qwen2_vl.sh)      | Finetune Qwen2-VL on image-text data using 4 A100s with FSDP |
+| [**Liger ORPO Trainer**](https://github.com/linkedin/Liger-Kernel/blob/main/examples/alignment/run_orpo.py)      | Align Llama 3.2 using Liger ORPO Trainer with FSDP with 50% memory reduction |
 
 ## Key Features