Context Injection and Long-Horizon Reliability: Separating Duplicate Recognition, Local Processing, and Counter Tracking

Overview

arXiv:2610.20633v1 Announce Type: new Abstract: Long-horizon failures of language models are usually measured as end-task success or as a horizon length. We instead score, step by step, three components of a 400-step ledger task in which every step is easy in isolation and the model's own answers are re-injected into its context: recognising a record already seen 5 to 300 steps earlier, processing the current record locally, and writing a counter incremented at every step. On Qwen3-8B (non-think

SpinGraph analysis pending — check back after processing.