# Route-Block Membership Selects Packed-AWQ Arithmetic: A Controlled Single-Fixture Mechanism Study

arXiv: https://arxiv.org/abs/2607.26316 · HTML: https://mlsanity.com/paper/2607.26316/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Expert routing strategies, one of 12 directions in Mixture-of-experts.
- Builds on 5 of its 10 references that are in the map.
- First posted on arXiv 28 Jul 2026.

## Placed in

- Mixture-of-experts / [Expert routing strategies](https://mlsanity.com/moe/expert-routing-strategies/)

## Builds on (most cited first, 5 of 5)

- [MegaBlocks: Efficient Sparse Training with Mixture-of-Experts](https://mlsanity.com/paper/2211.15841/) (2022)
- [MARLIN: Mixed-Precision Auto-Regressive Parallel Inference on Large Language Models](https://mlsanity.com/paper/2408.11743/) (2024)
- [UniEP: Unified Expert-Parallel MoE MegaKernel for LLM Training](https://mlsanity.com/paper/2604.19241/) (2026)
- [Bit-Exact AI Inference Verification Without Performance Tradeoffs](https://mlsanity.com/paper/2606.00279/) (2026)
- [WiSP: A Working-Set View of Mixture-of-Experts Serving on Extremely Low-Resource Hardware](https://mlsanity.com/paper/2606.21868/) (2026)

## Abstract (arXiv)

Mixture-of-experts (MoE) inference first aligns routed tokens into padded expert blocks, then executes packed quantized matrix multiplication over those blocks. This preprocessing is often treated as bookkeeping. In one pre-specified Qwen3-Coder AWQ layer-6 fixture on a pinned vLLM/Marlin build and RTX 3090 runtime, we show that the tested route-block interventions select exact packed arithmetic trajectories. Two fixed preconstruction histories produced distinct native alignments and exact trajectories. Injecting the opposite alignment transferred W13, activation, routed-W2, and final outputs. Permuting two routes within one block preserved each native trajectory, while exchanging two prior-data-selected routes across the boundary between expert-106 blocks 40 and 41 transferred the complete opposite trajectory. Source- and binary-derived schedule geometry maps those blocks to direct/full-K and split/global-reduction classes. Forcing a single-slice 200-block grid made W13 bitwise equal. Stable canonical construction made both histories converge to a third exact trajectory. The confirmatory cohort contains 70 valid cold processes and seven required perturbation rejections. This is a causal mechanism result for one fixture, not a prevalence, allocator, portability, or serving-impact claim.