From 204a93da88510a93ccd37b8dfdc6b5df94199b72 Mon Sep 17 00:00:00 2001
From: Zakir Jiwani <108548454+JiwaniZakir@users.noreply.github.com>
Date: Thu, 16 Apr 2026 12:43:00 +0000
Subject: [PATCH] =?UTF-8?q?Add=20EMPO=C2=B2=20paper=20entry=20clarifying?=
 =?UTF-8?q?=20code=20release=20status?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---
 README.md | 1 +
 1 file changed, 1 insertion(+)

diff --git a/README.md b/README.md
index 2be94bfaf..4aa937748 100644
--- a/README.md
+++ b/README.md
@@ -46,6 +46,7 @@ To start using Agent-lightning, check out our [documentation](https://microsoft.
 
 ## ⚡ Articles
 
+- *Exploratory Memory-Augmented LLM Agent via Hybrid On- and Off-Policy Optimization* (EMPO²) arXiv paper. The corresponding implementation code is being prepared and will be released in this repository.
 - 12/17/2025 [Adopting the Trajectory Level Aggregation for Faster Training](https://agent-lightning.github.io/posts/trajectory_level_aggregation/) Agent-lightning blog.
 - 11/4/2025 [Tuning ANY AI agent with Tinker ✕ Agent-lightning](https://medium.com/@yugez/tuning-any-ai-agent-with-tinker-agent-lightning-part-1-1d8c9a397f0e) Medium. See also [Part 2](https://medium.com/@yugez/tuning-any-ai-agent-with-tinker-agent-lightning-part-2-332c5437f0dc).
 - 10/22/2025 [No More Retokenization Drift: Returning Token IDs via the OpenAI Compatible API Matters in Agent RL](https://blog.vllm.ai/2025/10/22/agent-lightning.html) vLLM blog. See also [Zhihu writeup](https://zhuanlan.zhihu.com/p/1965067274642785725).