| title | Self-Compacting Language Model Agents |
|---|---|
| authors | Tianjian Li, Jingyu Zhang, William Jurayj, Xi Wang, Chuanyang Jin, Mehrdad Farajtabar, Eric Nalisnick, Daniel Khashabi |
| date | 2026-06-22 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2606.23525 |
| upvotes | 19 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf |
| abstract | Long agent traces composed of chains of thought and tool calls accumulate stale content that anchor subsequent generations, and eventually outgrow the context window. Existing scaffolds mitigate it with fixed-interval compaction triggered at a token threshold. Such triggers pay no heed to trajectory structure, risking discard of partial results mid-derivation or mid-search. We propose SelfCompact, a scaffold that allows the model itself to decide when and how to compact. Specifically, it pairs two inference-time elements: (i) a compaction tool the model invokes to summarize the accumulated context, and (ii) a lightweight rubric specifying when to fire (a sub-task has resolved, or the trajectory is converging) and when to suppress (mid-derivation, or when stuck). Both are needed. The tool alone is unevenly used across open-weight models, often invoked at unhelpful moments or not at all; the rubric alone cannot act. Together, they elicit effective adaptive compaction without any fine-tuning or external supervision. We present empirical results on six benchmarks (competitive math and agentic search) and seven models. Our results show that SelfCompact matches or exceeds fixed-interval summarization at a fraction of the token cost, improving over a no-summarization baseline by up to 18.1 points on math and 5-9 points on agentic search at 30-70% lower per-question cost. Our results expose a meta-cognitive gap: although unprompted models cannot reliably tell when their own context is rotting, a lightweight rubric closes this gap, reframing when to compact as a capability that scaffolds can supply without training. |
cite
@misc{li2026self,
title={Self-Compacting Language Model Agents},
author={Tianjian Li and Jingyu Zhang and William Jurayj and Xi Wang and Chuanyang Jin and Mehrdad Farajtabar and Eric Nalisnick and Daniel Khashabi},
year={2026},
eprint={2606.23525},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2606.23525}
}