---
license: apache-2.0
tags:
- agent
---
# The Complexity Trap: Simple Observation Masking Is as Efficient as LLM Summarization for Agent Context Management
This dataset contains our raw experimental data (ie. agent trajectories) accompanying the paper **"The Complexity Trap: Simple Observation Masking Is as Efficient as LLM Summarization for Agent Context Management"** and Tobias Lindenbauer's Master's thesis.

The data in this repository are compressed to `.tar.gz` archives. For detailed instructions on how to use these data please refer to our code repository on [Github](https://github.com/JetBrains-Research/the-complexity-trap).

If you used these data in your research, or it helped you form a hypothesis, please cite us:
```
@misc{lindenbauer2025complexitytrapsimpleobservation,
      title={The Complexity Trap: Simple Observation Masking Is as Efficient as LLM Summarization for Agent Context Management}, 
      author={Tobias Lindenbauer and Igor Slinko and Ludwig Felder and Egor Bogomolov and Yaroslav Zharov},
      year={2025},
      eprint={2508.21433},
      archivePrefix={arXiv},
      primaryClass={cs.SE},
      url={https://arxiv.org/abs/2508.21433}, 
}
```