Cayley unitary adapters executed on real quantum hardware improve LLM perplexity by 1.4% on Llama 3.1 8B with 6000 parameters and recover 83% of compression-induced degradation on SmolLM2.
Preprint at arXiv:2505.13205 (2025)
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
quant-ph 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
QNNs retain most hidden-task signals through public-task interfaces while classical networks transmit little, with transmission governed by teacher drift magnitude and the visible fraction of hidden drift in a unified geometric model.
citing papers explorer
-
Quantum-enhanced Large Language Models on Quantum Hardware via Cayley Unitary Adapters
Cayley unitary adapters executed on real quantum hardware improve LLM perplexity by 1.4% on Llama 3.1 8B with 6000 parameters and recover 83% of compression-induced degradation on SmolLM2.
-
Quantum Subliminal Learning
QNNs retain most hidden-task signals through public-task interfaces while classical networks transmit little, with transmission governed by teacher drift magnitude and the visible fraction of hidden drift in a unified geometric model.