Upload README.md with huggingface_hub

README.md

@@ -0,0 +1,57 @@
+---
+license: apache-2.0
+base_model: Qwen/Qwen2.5-7B-Instruct
+pipeline_tag: text-generation
+tags:
+  - memory
+  - long-horizon
+  - reinforcement-learning
+  - grpo
+  - agent
+---
+
+# Memory-R2 7B — Memory Manager (LoGo-GRPO, 32-session champion)
+
+This is the trained **memory-management policy** from [Memory-R2: Fair Credit Assignment for Long-Horizon Memory-Augmented LLM Agents](https://arxiv.org/abs/2605.21768) (arXiv:2605.21768).
+
+It is a Qwen2.5-7B-Instruct model fine-tuned with **LoGo-GRPO** (turn-level + token-level credit assignment) via a curriculum of 8 → 16 → 32-session rollouts on the LoCoMo long-horizon dialogue dataset. Given a running conversation, it decides what to INSERT / UPDATE / DELETE in an external memory store, so that a separate answer agent can later answer questions using only the maintained memory.
+
+This checkpoint is the paper's deployed "champion" (`32sess_champion_v2`, LoGo-GRPO curriculum, global step 5) — the memory manager behind the headline `tab:main` **Memory-R2 (OSS)** result on LoCoMo:
+
+| Answer agent | F1 | BLEU-1 | LLM-judge (gpt-4o-mini) |
+| --- | ---: | ---: | ---: |
+| GPT-OSS-120B (untrained, paired at eval time) | **49.29** | **43.64** | **86.08** |
+
+See the paper's `tab:different-answer-agent` for how this same memory manager performs when paired with other answer agents (untrained Qwen-7B, RL-trained Qwen-7B, etc.).
+
+## Usage
+
+This model is one half of a two-agent pipeline (memory manager + answer agent) and is not intended as a general-purpose chat model. Full inference code, the memory-store protocol, and the answer-agent pairing are in the [project repository](https://github.com/) (see the paper for the official release).
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = AutoModelForCausalLM.from_pretrained("ahmedehabb/memory-r2-7b", subfolder="memory-manager", torch_dtype="auto", device_map="auto")
+tokenizer = AutoTokenizer.from_pretrained("ahmedehabb/memory-r2-7b", subfolder="memory-manager")
+```
+
+## Training
+
+- Base model: `Qwen/Qwen2.5-7B-Instruct`
+- Algorithm: LoGo-GRPO (turn-level + token-level advantage), curriculum-trained 8-session → 16-session → 32-session
+- Reward: per-session cumulative F1 against gold QA + a memory-compression penalty (λ=0.3)
+- Judge for reward/logging during training: GPT-OSS-120B
+
+## Citation
+
+```bibtex
+@misc{yan2026memoryr2faircreditassignment,
+      title={Memory-R2: Fair Credit Assignment for Long-Horizon Memory-Augmented LLM Agents},
+      author={Sikuan Yan and Ahmed Bahloul and Ercong Nie and Susanna Schwarzmann and Riccardo Trivisonno and Volker Tresp and Yunpu Ma},
+      year={2026},
+      eprint={2605.21768},
+      archivePrefix={arXiv},
+      primaryClass={cs.LG},
+      url={https://arxiv.org/abs/2605.21768},
+}
+```