Metraipolis › archive › dc_ff5f3e4e41bb

In-Model Context Compression with Agentic Soft Tokens for Long-Horizon LLM Agents — arXiv:2610.41291

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.41291
License (as archived): CC0-1.0
Archived: 2026-11-03T07:42:43.194Z
SHA-256 of the archived text: 78a9e2885d173735203eb2862c36a5c0904a974adfef6a12534f6466f92fac03
Stable URL of this page: https://metraipolis.com/documento/dc_ff5f3e4e41bb

How to cite

In-Model Context Compression with Agentic Soft Tokens for Long-Horizon LLM Agents — arXiv:2610.41291 — arXiv (en), https://arxiv.org/abs/2610.41291 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ff5f3e4e41bb (Metraipolis archive dc_ff5f3e4e41bb, sha256 78a9e2885d173735203eb2862c36a5c0904a974adfef6a12534f6466f92fac03)

BibTeX

@misc{dcff5f3e4e41bb,
  title = {In-Model Context Compression with Agentic Soft Tokens for Long-Horizon LLM Agents — arXiv:2610.41291},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41291},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ff5f3e4e41bb; sha256 78a9e2885d173735203eb2862c36a5c0904a974adfef6a12534f6466f92fac03},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41291. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Long-horizon LLM agents play an important role in solving complex and sustained tasks, with ever-growing interaction contexts. However, these accumulated contexts will raise substantial computational cost and obscure key decision-relevant evidence. Existing methods typically train external compressors or context-management agents to rewrite context as shorter hard-token sequences, but such external selection may not align with the LLM's evolving representation and is necessarily lossy. In this paper, we introduce Agentic Soft-Token Compression (ASTC), a training-free compression method using soft tokens within LLM agent for long-horizon tasks. Specifically, ASTC uses the model's native recurrent dynamics to determine the already absorbed information, leveraging the commonly-used hybrid Gated Delta Network (GDN)-attention. It then carries this information forward as agentic soft tokens while retaining exact values as hard tokens. The resulting compact sequence reduces later-layer computation and yields reusable KV caches, extending the savings across the agent's trajectory. Extensive experiments on long-horizon tasks show that ASTC matches or exceeds the uncompressed agent's accuracy with up to 50.68% fewer end-to-end Attention Pairs, while trained compressors require 15.4 to 103.7 times more Attention Pairs per successful task. These findings highlight agentic soft tokens as a model-native way to carry absorbed context forward as the task representation evolves.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE).
Pagina del lavoro: https://arxiv.org/abs/2610.41291 — 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 2026-11-03T07:39:37.372Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).