--- format: "aidr-story-markdown/v1" id: "d59c4716fcf993c9d04698c70ca6657cc3ef07acf526a2cf6e243a1a0c726877" canonical_url: "https://aidr.today/d59c4716?lang=en" title: "Heavy-Tailed Memory Traces in Long-Horizon Language Agents" lang: "en" requested_lang: "en" available_langs: ["en","vi"] translation_fallback: null fallback_fields: [] published_at: "2026-10-02T04:00:00.000Z" category: "Research" topics: ["agent"] source_urls: ["https://arxiv.org/abs/2610.00010"] summary: "arXiv:2610.00010v1 Announce Type: new Abstract: Long-horizon language agents increasingly rely on external memory as a frozen world model, yet current memory systems are usually judged only by task success or token cost. We argue that the missing object is the shape of memory use: under finite context and repeated retrieval, agent memory can concentrate on a small core while leaving rare states in a long tail where prediction errors accumulate. We study this effect through a conservative tail audit and find that concentration is reproducible but policy-dependent. Random-walk agents produce log-normal-compatible retrieval artifacts, whereas semantic LLM policies yield the strongest truncated-power-law-compatible core--tail traces. Motivated by this audit, we propose Core--Tail World Model (CTWM), a rank-based memory controller that allocates prompt budget with a single exponent $\\tau$ while retaining a summarized tail. On Synthetic Graph World, CTWM preserves full state and transition coverage, reduces prompt tokens by 5.9%, and lowers bottom-half tail prediction error by 13.6% relative to a graph-memory baseline. The same paired comparison gives consistent token savings on ALFWorld" --- # Heavy\-Tailed Memory Traces in Long\-Horizon Language Agents > [Open the canonical story]() **Published:** 2026-10-02T04:00:00.000Z **Category:** Research **Topics:** agent ## Summary arXiv:2610\.00010v1 Announce Type: new Abstract: Long\-horizon language agents increasingly rely on external memory as a frozen world model, yet current memory systems are usually judged only by task success or token cost\. We argue that the missing object is the shape of memory use: under finite context and repeated retrieval, agent memory can concentrate on a small core while leaving rare states in a long tail where prediction errors accumulate\. We study this effect through a conservative tail audit and find that concentration is reproducible but policy\-dependent\. Random\-walk agents produce log\-normal\-compatible retrieval artifacts, whereas semantic LLM policies yield the strongest truncated\-power\-law\-compatible core\-\-tail traces\. Motivated by this audit, we propose Core\-\-Tail World Model \(CTWM\), a rank\-based memory controller that allocates prompt budget with a single exponent $\\tau$ while retaining a summarized tail\. On Synthetic Graph World, CTWM preserves full state and transition coverage, reduces prompt tokens by 5\.9%, and lowers bottom\-half tail prediction error by 13\.6% relative to a graph\-memory baseline\. The same paired comparison gives consistent token savings on ALFWorld ## Sources - [Story source]()