Metraipolis › archive › dc_712db6b4e57b

What Does Context Compression Cost an Agent? Interaction Costs Unrevealed by Task-Completion Metrics — arXiv:2608.16370

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.16370
License (as archived)
CC0-1.0
Archived
2026-10-18T17:43:03.178Z
SHA-256 of the archived text
fd2ff7c51131f9d3a82f0d245bc4a4b0a99f270cc8daa533dca2cc42edb75c40
Stable URL of this page
https://metraipolis.com/documento/dc_712db6b4e57b

How to cite

What Does Context Compression Cost an Agent? Interaction Costs Unrevealed by Task-Completion Metrics — arXiv:2608.16370 — arXiv (en), https://arxiv.org/abs/2608.16370 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_712db6b4e57b (Metraipolis archive dc_712db6b4e57b, sha256 fd2ff7c51131f9d3a82f0d245bc4a4b0a99f270cc8daa533dca2cc42edb75c40)
BibTeX
@misc{dc712db6b4e57b,
  title = {What Does Context Compression Cost an Agent? Interaction Costs Unrevealed by Task-Completion Metrics — arXiv:2608.16370},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.16370},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_712db6b4e57b; sha256 fd2ff7c51131f9d3a82f0d245bc4a4b0a99f270cc8daa533dca2cc42edb75c40},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.16370. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Task completion is the standard metric for evaluating context compression, yet it is incomplete: compression can increase an agent's interaction cost by forcing it to reacquire dropped state while leaving completion statistically unchanged. We introduce a controlled runtime measurement protocol for reacquisition cost in a bounded-horizon tool-using agent. The agent acts in a deterministic planning environment under a fixed 24-turn horizon. We vary compression severity, compare a dropping operator with a fact-preserving operator, restore dropped state through controlled oracle interventions, and decompose tool calls into retrieval and execution. We evaluate three models across two task regimes. Retrieval calls increase in all six model-regime comparisons and account for almost all added interaction; five of six remain significant after Holm correction. At the prespecified 5x comparison point, completion changes are not significant in any cell. DeepSeek shows a significant completion drop only at 10x compression. GPT-5.5 is the clearest case: completion changes from 80% to 85% (p = 1.0) while retrieval increases from 21.0 to 63.9 calls (p = .002). Retention interventions further separate state quantity, state type, and content validity. Random selection is comparable to an offline hindsight oracle, while replacing retained D-state with semantically irrelevant content increases retrieval by 57% (p < .001) without a significant completion change. In a second environment, ALFWorld, sliding compression produces no retrieval surge, showing that the reacquisition signature is environment-dependent rather than intrinsic to shortening context. Overall, compression can impose hidden interaction costs when execution-relevant state becomes absent and must be reacquired, while completion alone may not expose those costs.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.16370 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-18T16:29:13.864Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).