LodeHQ
AI news for people who shipFree, by email

AI · arxiv.org · Friday, 20 November 2026

InflEns makes MoE routing predictable before training

InflEns links how mixture-of-experts data is allocated during training to which expert handles each prompt, so specialization can be inspected before deployment and harmful behavior traced to one expert. Its location game places experts in a prompt-trait space, then reuses those positions for routing without training a separate gate. On seven safety benchmarks, the highest-share expert had the lowest held-out NLL on about 90% of prompts, versus about 15% after shuffling allocations; retraining one expert cut HarmBench attack success from 0.51 to 0.02 while preserving instruction following.

Read the source: arxiv.org ↗In the 20 November issue →