AuroOFT attaches a zero-initialized gated low-rank nonlinear residual to QOFT layers and reports improved mathematical reasoning accuracy over matched QOFT on 1.5B/3B Qwen2.5 models, with larger gains over QLoRA at lower parameter cost.
Advances in Neural Information Processing Systems , year =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Beyond Rotations: AuroOFT for Expressive Quantized Orthogonal Fine-Tuning
AuroOFT attaches a zero-initialized gated low-rank nonlinear residual to QOFT layers and reports improved mathematical reasoning accuracy over matched QOFT on 1.5B/3B Qwen2.5 models, with larger gains over QLoRA at lower parameter cost.