Model Releases
The Thousand-Graph Hypothesis: A Testable Hypothesis of Task-Conditioned Relation Materialization in Repository-Level Code Reasoning
arXiv:2608.26602v1 Announce Type: cross Abstract: Large software repositories are often beyond model context limits. Training repository knowledge into models is costly and quickly stale, while local
arXiv:2608.26602v1 Announce Type: cross Abstract: Large software repositories are often beyond model context limits. Training repository knowledge into models is costly and quickly stale, while local retrieval can miss scattered requirements, and explicit relation graphs add ongoing maintenance burden. We propose an entity-only external interface with task-conditioned relation materialization during inference. A two-layer index separates global routing from local entity focus and is evaluated on DeepSeek-V4-Flash and SWE-bench Verified. The base, one-layer, and two-layer conditions achieve 92.1%, 94.2%, and 95.6% success, respectively, under zero pre-built entity-relation edges.
Related
- Code2LoRA: Hypernetwork-Generated Adapters for Code Language Models under Software Evolution
- A Storage-Retrieval Gap in Parametric Knowledge Graph Memory
- Embodied Task Planning via Graph-Informed Action Generation with Large Language Models
Source: arXiv cs.CL | 2026-08-28