The Flow has not summarised this story yet — read it at arXiv Computation and Language.
The paper investigates how to tune the hyperparameters of simultaneous perturbation stochastic approximation (SPSA) for training a 6‑qubit, 60‑parameter variational quantum natural language inference (QNLI) classifier. By exploring a broad grid of perturbation scales, learning rates, and gain‑decay schedules, the authors found that an AdamW‑style SPSA configuration (c₀=0.01, η=0.10, γ=0.10) achieved 55 % ± 11 % test accuracy, improving over the default SPSA but still 16–19 percentage points below parameter‑shift baselines. Classical‑gain SPSA and Bures‑preconditioned SPSA performed worse, with accuracies of 51 % and 46 % respectively, indicating that two‑sample SPSA gradients suffer from high variance when optimizing many parameters over limited epochs.
R-DEIM Net is a 76‑million‑parameter dual‑expert model designed for paraphrase detection that balances accuracy with computational efficiency. It combines an Interaction Expert, which captures token‑level similarity via multi‑scale 2D convolutions and attention, with a Reasoning Expert that generates human‑readable rationales using a Flan‑T5‑small decoder. On the Quora Question Pairs dataset, the model attains 90.07% accuracy and 90.16% F1‑score, matching strong transformer baselines while producing auxiliary rationales.
QART is a quantum‑classical hybrid architecture that augments a language model with quantum encoding, CIM‑based QUBO optimization, and quantum decoding to improve long‑horizon reasoning. The authors claim that, under certain assumptions, QART can maintain a non‑zero probability of recovering an optimal reasoning path while traditional autoregressive LLMs see their acceptance probability drop to zero as cumulative risk grows. Experiments on six benchmarks with three backbone models show that QART outperforms the baselines in 14 of 15 pairings, with relative gains up to 84.0% on SciCode.
arXiv:2607. 09113v1 Announce Type: cross Abstract: Data scarcity and class imbalance are persistent challenges in machine learning that degrade model generalization and introduce predictive bias.
arXiv:2608. 06846v1 Announce Type: cross Abstract: We test whether a parameterized quantum circuit (PQC) improves a hybrid quantum-classical model's performance on classical datasets, using an interface-matched classical map as the control while holding all other components fixed.
QTrans is a quantum transformer designed for small‑scale binary sentiment classification. It constructs query, key, and value features using parameterized quantum circuits and derives attention coefficients from Gaussian distances between quantum measurements. The model incorporates a quantum feed‑forward network, residual connections, and layer normalization, achieving higher accuracies on MR, CR, and MPQA datasets compared to classical baselines.