Update README.md

README.md

@@ -18,6 +18,7 @@ The core idea behind TRIM-KV is to learn the intrinsic importance of each key–
 
 The retention score is query-agnostic and captures the long-term utility of tokens. This is different from attention scores, which are query-dependent: they capture the short-term utility for predicting the next token and are recomputed at every step, making them local, myopic, and highly dependent on the transient decoding state.
 
+<a href="https://arxiv.org/pdf/2512.03324"><img src="https://img.shields.io/badge/arxiv-2512.03324-red?style=for-the-badge"></a>
 
 ### Why TRIM-KV?
 
@@ -119,7 +120,8 @@ For a runnable end-to-end example, see [`examples/test_qwen3.py`](examples/test_
 | Qwen3-4B                     | [TRIM-KV-Qwen3-4B-Math](https://huggingface.co/ngocbh/TrimKV-Qwen3-4B-Math)            | OpenR1-Math-220k          | 16K   | 512     |
 | Qwen3-8B                     | [TRIM-KV-Qwen3-8B-Math](https://huggingface.co/ngocbh/TrimKV-Qwen3-8B-Math)            |  OpenR1-Math-220k          | 16K   | 512     |
 | Qwen3-14B                    | [TRIM-KV-Qwen3-14B-Math](https://huggingface.co/ngocbh/TrimKV-Qwen3-14B-Math)           |  OpenR1-Math-220k         | 16K   | 512     |
-| Qwen3-4B-Instruct-2507 | [TrimKV-Qwen3-4B-Instruct-2507](https://huggingface.co/ngocbh/TrimKV-Qwen3-4B-Instruct-2507) | Synth-Long, BookSum, Buddhi    |   128K       | 4096     |
+| Qwen3-4B-Instruct-2507 | [TrimKV-Qwen3-4B-Instruct-2507](https://huggingface.co/ngocbh/TrimKV-Qwen3-4B-Instruct-2507) | Synth-Long, BookSum, Buddhi      |  128K      | 4096     |
 | Phi-3-mini-128k-instruct | [TrimKV-Phi-3-mini-128k-instruct](https://huggingface.co/ngocbh/TrimKV-Phi-3-mini-128k-instruct) | LongAlpaca          |  128K  | 2048 |
+| DeepSeek-R1-Distill-Llama-8B                    | [TrimKV-DeepSeek-R1-Distill-Llama-8B](https://huggingface.co/ngocbh/TrimKV-DeepSeek-R1-Distill-Llama-8B)           |  OpenR1-Math-220k         | 32K   | 512     |
 
 ---