Evaluation of cultural value alignment in Large Language Models (LLMs) typically relies on direct prompting with survey-style questions, which often elicit neutral or safety-aligned responses and fail to capture the model’s underlying cultural orientation. To overcome these challenges, we propose a framework for probing and steering latent cultural representations in LLMs along the two Inglehart–Welzel axes of the World Values Survey (WVS). By translating social value questions into scenario-based dilemmas, we extract token-level probabilities to measure implicit values and apply activation steering, optionally combined with country-conditioned prompting, to shift model behavior without retraining. Across three 4B-open-weight LLMs, four countries and three social domains, we find substantial variation in steerability. Importantly, we identify latent entanglement, where interventions along one Inglehart–Welzel axis induce a shift along the other. This entanglement is consistent with correlations present in human WVS data and persists across our experimental settings. Our work identifies empirical constraints on cultural alignment in LLMs.

Self-conditioning has been central to the success of continuous diffusion language models, as it allows models to correct previous errors. Yet its ability degrades precisely in the regime where diffusion is most attractive for deployment: few-step sampling for fast inference. In this study, we show that when models only have a few denoising steps, inaccurate self-conditioning induces a substantial approximation gap; this mistake compounds across denoising steps and ultimately dominate the sample quality. To address this, we propose a novel training framework that handles these errors during learning by perturbing the self-conditioning signal to match inference noise, improving robustness to prior estimation errors. In addition, we introduce a token-level noise-awareness mechanism that prevents training from saturation, hence improving optimization. Extensive experiments across conditional generation benchmarks demonstrate that our framework surpasses standard continuous diffusion models while providing up to 400x faster inference speed, and remains competitive against other one-step diffusion frameworks.

Small Language Models (SLMs) offer computational efficiency and accessibility, yet a systematic evaluation of their performance and environmental impact remains lacking. We introduce SLM-Bench, the first benchmark specifically designed to assess SLMs across multiple dimensions, including accuracy, computational efficiency, and sustainability metrics. SLM-Bench evaluates 15 SLMs on 9 NLP tasks using 23 datasets spanning 14 domains. The evaluation is conducted on 4 hardware configurations, providing a rigorous comparison of their effectiveness. Unlike prior benchmarks, SLM-Bench quantifies 11 metrics across correctness, computation, and consumption, enabling a holistic assessment of efficiency trade-offs. Our evaluation considers controlled hardware conditions, ensuring fair comparisons across models. We develop an open-source benchmarking pipeline with standardized evaluation protocols to facilitate reproducibility and further research. Our findings highlight the diverse trade-offs among SLMs, where some models excel in accuracy while others achieve superior energy efficiency. SLM-Bench sets a new standard for SLM evaluation, bridging the gap between resource efficiency and real-world applicability.