Skip to main content

# Routing Gate

2 related articles

Subscribe to updates

arXiv Oct 31, 2026
PRISM-Net splits product retrieval into line and color routes, lifting SwinV2 line mAP@10 by 5.74 points
PRISM-Net splits a product image's holistic features into a line pathway and a color pathway, supervises the two routers zero-shot with frozen SigLIP text prompts, and fuses both with an image-adaptive meta-router. In 5-fold cross-validation over 275,454 e-commerce images in 18 categories and full-scale retrieval of 55,091 queries against a 220,363-image gallery, the SwinV2 backbone gains 5.74 points in line mAP@10, while the color-only configuration gains 9.12 points in color mAP@10; on ConvNeXt-V2 and DINOv3, product retrieval and attribute alignment trade off.
PRISM-Net splits a product image's holistic features into a line pathway and a color pathway, supervises the two routers zero-shot with frozen SigLIP text prompts, and fuses both with an image-adaptive meta-router. In 5-fold cross-validation over 275,454 e-commerce images in 18 categories and full-scale retrieval of 55,091 queries against a 220,363-image gallery, the SwinV2 backbone gains 5.74 points in line mAP@10, while the color-only configuration gains 9.12 points in color mAP@10; on ConvNeXt-V2 and DINOv3, product retrieval and attribute alignment trade off.
National Yang Ming Chiao Tung University National Chengchi University
Chung‐Ming Lo（First author, Corresponding author）
PRISM-Net Fine-Grained Product Retrieval Mixture of Experts Knowledge Distillation Routing Gate
arXiv Nov 1, 2026
Knowledge-graph prompts don't need the whole graph: relation knowledge picks a dozen facts, and seven open LLMs answer better
Knowledge-graph question answering usually ranks facts by relevance and packs them into the prompt, but relevance guarantees neither sufficiency nor consistency. This method uses relation knowledge, either hand-written rules for what relations mean and how they compose or a mapping learned from training questions annotated with relation paths, to search the graph for a compact fact set that derives an answer, then applies a derivability check for whether an answer follows and a consistency check for whether facts conflict, with a validation-chosen gate deciding when the derived answer replaces the LLM's. On WebQSP, seven open LLMs read these contexts, 14.7 facts on average, at 10.2 exact-match points above SubgraphRAG's 100 facts.
Knowledge-graph question answering usually ranks facts by relevance and packs them into the prompt, but relevance guarantees neither sufficiency nor consistency. This method uses relation knowledge, either hand-written rules for what relations mean and how they compose or a mapping learned from training questions annotated with relation paths, to search the graph for a compact fact set that derives an answer, then applies a derivability check for whether an answer follows and a consistency check for whether facts conflict, with a validation-chosen gate deciding when the derived answer replaces the LLM's. On WebQSP, seven open LLMs read these contexts, 14.7 facts on average, at 10.2 exact-match points above SubgraphRAG's 100 facts.
Seunghui Jwa（First author）
Compact Sufficient Context Selection Derivability Check Consistency Check Knowledge Graph Routing Gate

You’re all caught up