GRIP Cuts RAG Hallucination 73% by Deliberately Starving the Evidence Channel of Capacity

arXiv 2608.16776·medium signalXBlueskyLinkedInCopy link

The paper names a RAG failure mode — 'query dominance', where a high-capacity encoder lets the query dominate the latent state and renders retrieved evidence functionally irrelevant. GRIP imposes capacity asymmetry: the decoder keeps full-dimensional access to the query while retrieved evidence passes through a severe stochastic bottleneck, forcing that channel to encode only residual information the query cannot supply. Across five reasoning benchmarks it beats strong iterative baselines, cuts a query-latent mutual-information diagnostic roughly 30x (14.8 → 0.47 bits), and reduces hallucination by 73%.