28 April 2026 in IEEE Access
Tipping, which has evolved beyond its original role as a gesture of appreciation, has become a complex social issue that often causes friction between customers and service workers. Existing technological approaches, instead of clarifying the ambiguity in evaluation criteria, merely provide predefined tipping options that fail to improve customer satisfaction. Such practices lead to “tip fatigue” and foster a sense of coercion regardless of service quality, thereby intensifying social tension. This issue arises from the lack of objective and transparent standards for evaluating service value. To address these limitations, this paper proposes TipMate, an explainable fair tip recommender system based on Large Language Models (LLMs) that holistically evaluates service quality and transparently explains its reasoning. TipMate conducts evaluations grounded in the established service quality theory by systematically integrating (1) visual evidence from the service environment, (2) subjective textual feedback from customers, and (3) contextual data from publicly available reviews. To validate the system’s reasoning capabilities, we constructed OpenR-Tip, a benchmark dataset derived from real-world narratives with synthesized contexts. Experimental results using four state-of-the-art LLMs show that TipMate closely approximates human tipping judgments with a Mean Absolute Error (MAE) of 4.26%, achieving a high average satisfaction score of 4.43/5 in a user study.
2 October 2026 in Schizophrenia
Although the relationship between schizophrenia (SZ), formal thought disorder, and cognition has been extensively studied, comparatively less is known about which automatically extracted linguistic features are associated with cognitive status in SZ. We analyzed lexical, syntactic, semantic, and sentiment features from speech samples of 88 patients with SZ [37 cognitively impaired (Cog-D) and 51 cognitively preserved (Cog-ND)] and 69 healthy controls (HC). Features were extracted using Korean-adapted Natural Language Processing (NLP) pipelines. Medication effects were controlled for using an Extrapyramidal Symptom Rating Scale. Machine learning (ML) classification was conducted as an exploratory analysis. Patients with Cog-D showed marked reductions in lexical richness, syntactic complexity, semantic coherence and sentiment polarity compared with those with Cog-ND or HC. The performance of the ML models showed a discrepancy between the mean outer-fold performance and the performance on the hold-out test dataset. Several linguistic features were correlated with symptom severity and language performance. Linguistic disturbances in patients with SZ were heterogeneous and more pronounced in Cog-D group compared to Cog-ND group. These findings highlight the potential of NLP-based speech analysis for identifying cognitive impairment and support the cross-linguistic generalizability of linguistic markers.