Sampling methods for large language models select candidate tokens based on logit statistics, implicitly assuming that high log-its indicate desirable outputs. We identify the Logit Conflation Problem , where a to-ken’s logit aggregates prompt-independent factors, including linguistic fluency and parametric associations, with prompt-relevance. However, only prompt-relevance determines instruction-following quality. We propose SEAL-Sampling ( S ignal E xtraction for A ctive Re L evance) to isolate this component through attention-weighted attribution. Our framework defines prompt-relevance as the causal effect of prompt content on token logits and establishes attention patterns as an efficient proxy. Experiments on LLaMA-3 demonstrate significant improvements over top-nσ , with gains of 1.8% on AlpacaEval 2.0 and 2.2% on IFEval. Furthermore, attribution scores correlate weakly with raw logits, confirming the extraction of an orthogonal signal. The method is training-free and introduces minimal latency, adding less than 12ms overhead per token.
Pinlong Zhao, Huijun Tang, Pengfei Jiao et al.· Annual Meeting of the Associ...· 0 citations
Multimodal Large Language Models (MLLMs) often suffer from modality bias, where the model disproportionately relies on one modality while neglecting critical information from others. Existing debiasing methods via modality masking create biased responses by completely removing an entire modality, forming an extreme and static training environment. However, real-world multimodal bias often emerges under subtle perturbations (e.g., mild occlusion, noisy instructions), where both modalities are present but the model is tempted to rely on spurious shortcuts. We propose B-APO (Bias-Targeted Adversarial Preference Optimization), which casts debiasing as a bias-targeted min-max game: we generate hard negatives by applying small adversarial perturbations in the latent space to maximally induce language-vision-prior reliance, and then perform preference alignment to enlarge the margin between clean and adversarial responses. This encourages the model to anchor on true cross-modal evidence even under the most adversarial conditions. Extensive experiments on bias and hallu-cination benchmarks demonstrate that B-APO achieves superior debiasing performance while maintaining general capabilities.
Pinlong Zhao, Zike Ding, Zengshu Ye et al.· Annual Meeting of the Associ...· 0 citations