SFAD: Speculative Factuality-Aware Decoding

arXiv:2609.00796v1 Announce Type: new Abstract: As one of the most critical challenges in large language models, contextual faithfulness directly determines their reliability in knowledge-intensive applications. This task is particularly challenging as it requires balancing factual consistency with generation efficiency. Contrastive decoding methods require dual forward passes (with and without context) to compare model outputs, doubling inference computational overhead, while post-training alignment demands extensive reinforcement learning with substantial computational overhead. To address this challenge, we present extbf{SFAD}, a speculative decoding framework that enhances contextual faithfulness without inference degradation. We first construct extbf{ConFide}, a preference dataset with fine-grained atomic perturbations, to train a context-faithful draft model via Direct Preference Optimization. During inference, Epistemic Friction detects potential hallucinations by quantifying distributional tension weighted by specialist certainty. When friction exceeds the threshold, Asymmetric Logit Steering refines the target distribution through residual-based logit injection; otherwise, standard speculation proceeds. Extensive experiments demonstrate that SFAD substantially improves faithfulness while achieving 2.48imes speedup, offering a practical solution for efficient LLMs.

Related

• FAITH: Factuality Alignment through Integrating Trustworthiness and Honestness
• Draft-OPD: On-Policy Distillation for Speculative Draft Models
• PSD: Pushing the Pareto Frontier of Diffusion LLMs via Parallel Speculative Decoding
• ConfLayers: Adaptive Confidence-based Layer Skipping for Self-Speculative Decoding

Source: arXiv cs.CL | 2026-09-02