Large Language Models (LLMs) are increasingly used to explain, summarize, and translate real-world events, including ongoing geopolitical conflicts. Yet it remains unclear whether they reproduce conflict-specific propaganda and, if so, how this appears in their outputs. We study this question for the Russia-Ukraine war through perspectival divergence, the extent to which model outputs align with competing narratives from different information ecosystems. We construct a conflict-aware evaluation set of neutral English event statements paired with Russian (RU)- and Ukrainian (UA)-oriented reference texts drawn from news outlets and Telegram channels. We then evaluate multiple LLMs under several prompting contexts using a reference-based semantic distance metric that measures directional proximity to RU- and UA-oriented references. To explain not only which side a model is closer to but also how that alignment is expressed, we further analyze outputs using five propaganda-relevant categories: Framing Narrative, Emotional Manipulation, Source Credibility, Social Pressure Identity, and Toponymy Naming. Across models, we find stable, model-specific leanings and technique profiles that persist across prompts and are not captured by standard factuality-oriented metrics. Our findings show that models that appear neutral under conventional evaluations can still encode systematic, conflict-specific propaganda patterns, underscoring the need for conflict-aware evaluation frameworks.

Question answering (QA) systems should remain stable across meaning-preserving variations of the same problem. Earlier robustness work showed brittleness to local lexical perturbations, such as synonym replacement, word swaps, insertions, and deletions. Revisiting this robustness axis for LLMs, we find that high-performing models are now comparatively stable to such edits. However, this stability does not extend to discourse-level rewrites: answer-preserving transformations that alter a question’s structure and presentation while preserving its intent and gold answer. Across reasoning-focused QA benchmarks, these rewrites induce substantially larger performance drops than local perturbations. They also change how models fail: using a compact failure taxonomy and Total Variation distance, we show that discourse-level rewrites systematically redistribute error types, revealing distinct failure regimes. Our results suggest that local lexical perturbations remain useful, while discourse-level rewrites provide a complementary and stronger diagnostic signal for evaluating reasoning stability in QA LLMs.

In the rapidly evolving landscape of e-commerce, product returns have become a significant economic burden for businesses, where the reasons for returns may vary from wrong sizing and defective products to simply no longer needing the purchased product. This paper presents, to the best of our knowledge, the first comprehensive study of the complexities of product returns across a variety of e-commerce domains, focusing on the task of predicting the return reason. We propose a supervised approach for predicting return likelihood and the underlying return reason. We test our approach over a real-world dataset from a large e-commerce platform.

The term lot in is defined to mean an offering that contains a collection of multiple identical items for sale. In a large online marketplace, lot offerings play an important role, allowing buyers and sellers to set price levels to optimally balance supply and demand needs. In spite of their central role, platforms often struggle to identify lot offerings, since explicit lot status identification is frequently not provided by sellers. The ability to identify lot offerings plays a key role in many fundamental tasks, from matching offerings to catalog products, through ranking search results, to providing effective pricing guidance. In this work, we seek to determine the lot status (and lot size) of each offering, in order to facilitate an improved buyer experience, while reducing the friction for sellers posting new offerings. We demonstrate experimentally the ability to accurately classify offerings as lots and predict their lot size using only the offer title, by adapting state-of-the-art natural language techniques to the lot identification problem.

Large training datasets are required to achieve competitive performance in most natural language tasks. The acquisition process for these datasets is labor intensive, expensive, and time consuming. This process is also prone to human errors. In this work, we show that cross-cultural differences can be harnessed for natural language text classification. We present a transfer-learning framework that leverages widely-available unaligned bilingual corpora for classification tasks, using no task-specific data. Our empirical evaluation on two tasks – formality classification and sarcasm detection – shows that the cross-cultural difference between German and American English, as manifested in product review text, can be applied to achieve good performance for formality classification, while the difference between Japanese and American English can be applied to achieve good performance for sarcasm detection – both without any task-specific labeled data.