When Retrieval Fails Before It Begins: Structurally Indirect Prerequisite Eviction as a Retention Failure in Agentic Memory
arXiv cs.AIen
arXiv:2608.20400v1 Announce Type: new Abstract: Agentic memory under a fixed budget involves two stages: retention and retrieval. Existing retrieval-centered paradigms implicitly assume necessary evidence survives eviction, but we challenge this by isolating a pre-retrieval failure mode: structurally indirect prerequisite eviction, in which upstream blocks weakly aligned with the query are discarded under budget pressure. We provide an operational definition of this failure, a reproducible deterministic benchmark, and per-seed trace diagnostics. Finally, we evaluate Dependency-aware Semantic Garbage Collection (DSGC), a one-hop graph-aware rule. In our main suite, DSGC improves full-chain re
This is a short summary published by AI Global Wire. The full article is owned and hosted by arXiv cs.AI — open it there to read it in full.
Read the full story at arXiv cs.AI- Forskning
- Agenter
Related AI news
- Source: AI researcher Luke Metz, who returned to OpenAI from TML earlier this year, joins Meta's Superintelligence Labs and will report to Alexandr Wang (Ina Fried/Axios)Techmeme · August 24, 2026
- Terminal Agents: A Survey of AI Agents in Command-Line EnvironmentsarXiv cs.AI · August 24, 2026
- Difficulty-Aware Semantic-ID Optimization for Generative RecommendationarXiv cs.AI · August 24, 2026
- Environmental Slow AI: Design Principles for Generative SystemsarXiv cs.AI · August 24, 2026
- World models of environment, agent and joint agent-environment systemsarXiv cs.AI · August 24, 2026
- StateSight: Benchmarking Latent Spatial-State Reconstruction in Vision-Language ModelsarXiv cs.AI · August 24, 2026