Add pipeline tag, library name, and paper/code links

README.md

@@ -1,16 +1,20 @@
 ---
-license: mit
+base_model:
+- Qwen/Qwen3-4B-Thinking-2507
 language:
 - en
 - zh
-base_model:
-- Qwen/Qwen3-4B-Thinking-2507
+license: mit
+pipeline_tag: text-generation
+library_name: transformers
 ---
 
 # MemSifter-4B-Thinking
 
 **MemSifter** is a lightweight generative session ranker trained with DAPO reinforcement learning. It serves as the core retrieval component of the [MemSifter](https://github.com/plageon/MemSifter) system—an LLM memory retrieval offloading framework based on outcome-driven proxy reasoning.
 
+This model is presented in the paper [MemSifter: Offloading LLM Memory Retrieval via Outcome-Driven Proxy Reasoning](https://huggingface.co/papers/2603.03379).
+
 Given a user query and a set of candidate conversation sessions (pre-filtered by a dense embedding model), MemSifter-4B-Thinking performs fine-grained reranking to surface the sessions most relevant to the query, which are then passed as context to a downstream chat LLM.
 
 ## System Pipeline
@@ -24,6 +28,11 @@ Session Embedding  →  Session Ranking (MemSifter)  →  Chat LLM
 2. **Session Ranking** — MemSifter (this model) performs fine-grained reranking of the pre-filtered candidates.
 3. **Chat LLM** — the top-ranked sessions are assembled into a context window and passed to any OpenAI-compatible chat model.
 
+## Resources
+
+- **GitHub:** [https://github.com/plageon/MemSifter](https://github.com/plageon/MemSifter)
+- **Paper:** [https://arxiv.org/abs/2603.03379](https://arxiv.org/abs/2603.03379)
+
 ## How to Use
 
 Install the required packages: