# Interpretable-by-Design Transformers via Architectural Stream Independence

arXiv: https://arxiv.org/abs/2603.07482 · HTML: https://mlsanity.com/paper/2603.07482/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 22 references that are in the map.
- First posted on arXiv 8 Mar 2026.

## Builds on (most cited first, 17 of 17)

- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://mlsanity.com/paper/2006.03654/) (2020)
- [Generating Long Sequences with Sparse Transformers](https://mlsanity.com/paper/1904.10509/) (2019)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small](https://mlsanity.com/paper/2211.00593/) (2022)
- [BERT Rediscovers the Classical NLP Pipeline](https://mlsanity.com/paper/1905.05950/) (2019)
- [Towards Understanding Sycophancy in Language Models](https://mlsanity.com/paper/2310.13548/) (2023)
- [What Does BERT Look At? An Analysis of BERT's Attention](https://mlsanity.com/paper/1906.04341/) (2019)
- [Attention is not Explanation](https://mlsanity.com/paper/1902.10186/) (2019)
- [Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Pruned](https://mlsanity.com/paper/1905.09418/) (2019)
- [Neural Module Networks](https://mlsanity.com/paper/1511.02799/) (2015)
- [Revealing the Dark Secrets of BERT](https://mlsanity.com/paper/1908.08593/) (2019)
- [What do Neural Machine Translation Models Learn about Morphology?](https://mlsanity.com/paper/1704.03471/) (2017)
- [Identifying Spurious Biases Early in Training through the Lens of Simplicity Bias](https://mlsanity.com/paper/2305.18761/) (2023)
- [Serial Position Effects of Large Language Models](https://mlsanity.com/paper/2406.15981/) (2024)

## Abstract (arXiv)

While transformers achieve strong performance, their internal decision-making processes remain opaque. We investigate whether architectural constraints can enforce interpretability by design through architectural stream independence: maintaining a token stream (carrying symbolic structure) and contextual semantics in separated streams that remain independently observable throughout processing, with integration delayed until output. We validate this principle through the Late Fusion Architecture (LFA), which demonstrates interpretable symbolic heads through all the final layers, while standard transformers show dissolution by the third of six layers; we quantify this effect by introducing the Token-Position Dependence Score (PDS), with PDSmax = 0.276 and 0.058, respectively. Crucially, intervention experiments demonstrate functional modularity: suppressing LFA's recency heads causes minimal semantic damage (Cohen's d = -0.158) versus catastrophic entanglement in baselines (d = -0.672). LFA demonstrates that architectural constraints improve underlying learning mechanisms, averaging 42% stability versus 19% and 11% for baseline comparisons, with extremes from 50% on LFA's best pairs (6 of 12 heads position-invariant) down to 0% complete collapse in over-constrained cases. By preventing premature entanglement, architectural independence steers models toward semantic understanding over positional heuristics, establishing interpretability as an architectural design criterion enforceable through structural constraints rather than post-hoc analysis.