Add six papers on multi-agent systems and scientific discovery

[Liu-Hy]

Thank you for maintaining this excellent resource for the community. In this PR, we request to add six relevant papers from our research group that focus on LLM-based multi-agent systems and AI for science:

Papers Added:

1. GenoTEX (MLCB 2025 Oral) - An expert-curated benchmark for evaluating LLM agents on automated gene expression data analysis, featuring context-aware planning and multi-agent collaboration

   • arXiv: 2406.15341
   • GitHub: https://github.com/Liu-Hy/GenoTEX

2. GenoMAS - A multi-agent framework with guided planning for scientific discovery via code-driven gene expression analysis, featuring heterogeneous LLMs and dynamic memory reuse

   • arXiv: 2507.21035
   • GitHub: https://github.com/Liu-Hy/GenoMAS

3. aiXiv - A next-generation open-access platform with multi-agent architecture enabling collaboration between AI and human scientists

   • arXiv: 2508.15126
   • GitHub: https://github.com/aixiv-org/aiXiv

4. The Landscape of Agentic Reinforcement Learning for LLMs - A comprehensive survey on agentic RL, covering planning, tool use, memory, and reasoning capabilities

   • arXiv: 2509.02547
   • GitHub: https://github.com/xhyumiracle/Awesome-AgenticLLM-RL-Papers

5. CoMAS - Co-evolving multi-agent systems via interaction rewards, demonstrating self-improvement through inter-agent communication

   • arXiv: 2510.08529

6. Achilles Heel of Distributed Multi-Agent Systems - Studies trustworthiness challenges in distributed multi-agent systems, including communication inefficiencies

   • arXiv: 2504.07461

Relevance to Context Engineering:
All papers are highly relevant to this repository as they address core context engineering challenges including:

• Multi-agent communication and coordination
• Context-aware planning and decision making
• Memory systems and knowledge management
• Tool use and function calling in agent systems
• Benchmarking and evaluation of context-driven agent systems

The papers have been placed in appropriate sections following the repository's formatting conventions and existing paper styles.

Thank you for considering our request!