A Born-measured transformer gives exact word-level evidence to within 10^-11, with class probabilities written as polynomials in the word amplitudes and integrated gradients computed by finite quadrature. On the historical side, HERIT tackles Hanja-to-English translation by using retrieval-augmented generation over Hanja-Korean corpora to build pseudo-labelled data and reduce temporal bias in scarce historical material.

Reading Why: Born-Measured Attention with Exact Word Evidence on Light and Qubits

This is a quantum-inspired interpretability model, so MT practitioners can ignore the optics but note the exact integrated-gradient evidence and deletion tests as a reminder that attribution quality and faithfulness are not the same thing.

When a language model answers, we can usually only estimate why. We build a transformer in which every step is a Born measurement - attention |langle q|krangle|^4 normalised over each row, a measured product in place of the feed-forward nonlinearity, and a Born readout. With its denominators held at their values for the input, each class probability is a polynomial in the word amplitudes, so its integrated-gradient evidence is computed by a finite quadrature with no quadrature error: the words' contributions sum to the class probability above its no-information share, to 10^-11 or better. On e-SNLI natural language inference, with about 98k parameters outside the embedding, the model reaches 76.98 +/- 0.17% test accuracy, level with a softmax transformer of the same size (76.62 +/- 0.36%). Deleting the word its evidence ranks first changes the answer in 45% of test pairs, out of the 53% any single deletion can change, against 32% for the softmax model explained the same way and 40% by its attention x gradient. Attention can look more like the words annotators highlighted, yet accounts for the answer less well: deleting the word it ranks first changes the Born Transformer's answer in 32% of pairs. The same weights compile exactly to optoelectronic optics and to qubits, and a coherent attention layer becomes one circuit equal to the model (852 CNOTs on 23 qubits). On IBM's ibm_kingston, reading only the attention overlaps by Hadamard tests, the device gave the model's answers on two short test pairs and kept or changed them under deletion as the model does; the evidence itself did not yet survive the device, and we work out the shots and calibration that measuring it on hardware would need.

HERIT: Enabling Global Access to Hanja-to-English Historical Translation by Mitigating Temporal Bias

HERIT adapts Hanja-to-English translation with RAG-generated pseudo-labels to reduce temporal bias, so practitioners working on historical text should expect data augmentation and period balancing to matter as much as model choice.

Historical archives are invaluable for multidisciplinary research but remain inaccessible globally due to severe language barriers. While manual translation is prohibitively expensive and time-consuming, simply applying existing machine translation models is inadequate due to unique linguistic nuances and the temporal bias inherent in scarce human-labeled corpora. To address these challenges, we propose HERIT, a systematic domain-adaptation pipeline that leverages retrieval-augmented generation (RAG) to generate high-quality pseudo-labeled data from abundant Hanja-Korean corpora. This approach substantially expands the training dataset, mitigating both data scarcity and temporal overfitting across different historical periods. Extensive automated and human evaluations demonstrate that HERIT consistently outperforms baselines. Finally, we translate previously untranslated portions of the archives using our model, thereby enabling global access to these resources for researchers.

Papers announced by arXiv on Tuesday, 3 November 2026, covering submissions from 2 November to 3 November 2026.