Conversational stance detection, which aims to identify users’ attitudes in conversation threads, has attracted increasing attention in recent years. Yet, mainstream LLM-based approaches remain prone to biases caused by spurious correlations. Existing debiasing methods either rely on indiscriminate multi-step prompting, which can induce unnecessary reasoning and overthinking, or require expensive fine-tuning on curated data. To address these challenges, we propose ExpCSD, an experience-based debiasing framework grounded in latent thinking paths that follows a debias-only-when-needed principle. ExpCSD first assesses whether an LLM’s latent thinking path is biased and selectively retrieves and injects historically similar debiasing experiences only when necessary, without updating model parameters. Experiments on public benchmarks show that ExpCSD outperforms strong baselines in stance detection accuracy while avoiding overthinking and costly fine-tuning.

Token-level differentially private (DP) prompt optimization methods such as DP-OPT can become unstable under tight privacy budgets: on GSM8K, DP-OPT obtains 49.5 ± 28.5% across 30 runs, and a logged search trajectory reveals prompt-template drift and noise-sensitive irreversible choices. We diagnose these as structural consequences of greedy token-by-token construction over privately aggregated counts. We then propose DP-ES (Differentially Private Evolution Strategies), a structurally cleaner alternative that maintains a population of full prompts, mutates them via LLM calls that never access the private dataset, and spends privacy only on sampled-Gaussian evaluation; deterministic or Gumbel-smoothed selection is post-processing. Under a conservative (ε ≤ 1.0, δ = 10⁻⁵) guarantee, DP-ES achieves 88.1% on GSM8K (+38.6 pp over DP-OPT, approximately 9× lower standard deviation), 99.7% on MedQA, 73.5% on BANKING77, and 86.8% on Alpaca. It is also 2.5× faster in wall-clock time and uses 3.3× fewer logged private-data call groups than DP-OPT. Selection and population ablations, implementation-level noise checks, and a 200-profile exact-match memorization stress test complement the formal guarantee. Scope: Our experiments establish optimization robustness under DP noise, especially where prompt structure is critical; end-to-end validation on genuinely sensitive, non-saturated deployment data remains future work. Code: https://github.com/StephCpa/dp-es

The identification of harmful memes extends beyond a mere classification task, encompassing challenges related to multi-perspective semantic comprehension and hierarchical reasoning. Prevailing approaches predominantly depend on modal alignment or black-box classifiers, which fail to capture implicit biases and lack interpretability. In this study, we propose BPDMoE-Hate, a novel framework grounded in dual-space mixture-of-experts, which innovatively conceptualizes harmful meme detection as an integrated process of “viewpoint decoupling and hierarchical fusion”. Our approach generates adversarial binary perspectives via Visual-Language Models (VLMs) and incorporates an adaptive viewpoint gating to facilitate viewpoint selection, thereby enabling the model to autonomously discern implicit semantic inclinations. Moreover, we propose the Hyperbolic-Euclidean space expert to effectively capture the hierarchical structural relationships and semantic correlations between multimodal and viewpoint features, thereby enabling interpretable reasoning at the geometric representation level. Empirical evaluations conducted on three mainstream datasets demonstrate that BPDMoE-Hate not only substantially surpasses existing methodologies in performance but also offers visual explanations for viewpoint selection and hierarchical structuring, thereby advancing the field of interpretable multimodal content analysis.