arXiv:2511.16905v1 Announce Type: new 
Abstract: Early detection of rising talents is of paramount importance in the field of advertising. In this paper, we define a concept of talent breakout and propose a method to detect Japanese talents before their rise to stardom. The main focus of the study is to determine the effectiveness of combining Twitter and TV data on predicting time-dependent changes in social data. Although traditional time-series models are known to be robust in many applications, the success of neural network models in various fields (e.g.\ Natural Language Processing, Computer Vision, Reinforcement Learning) continues to spark an interest in the time-series community to apply new techniques in practice. Therefore, in order to find the best modeling approach, we have experimented with traditional, neural network and ensemble learning methods. We observe that ensemble learning methods outperform traditional and neural network models based on standard regression metrics. However, by utilizing the concept of talent breakout, we are able to assess the true forecasting ability of the models, where neural networks outperform traditional and ensemble learning methods in terms of precision and recall.

قدمت دراسة جديدة طريقة للتنبؤ بمعدل انطلاق المواهب اليابانية من خلال تحليل بيانات تويتر والتلفزيون. تبرز هذه البحث أهمية الكشف المبكر في مجال الإعلان وتقيّم فعالية تقنيات النمذجة المختلفة، بما في ذلك الطرق التقليدية، والشبكات العصبية، وطرق التعلم الجماعي.

Un nuevo estudio ha presentado un método para predecir la tasa de despegue de talentos japoneses mediante el análisis de datos de Twitter y televisión. La investigación destaca la importancia de la detección temprana en la publicidad y evalúa la efectividad de varias técnicas de modelado, incluidas las tradicionales, las redes neuronales y los métodos de aprendizaje por conjunto.

Une nouvelle étude a introduit une méthode pour prédire le taux d'émergence des talents japonais en analysant les données de Twitter et de la télévision. La recherche souligne l'importance de la détection précoce dans la publicité et évalue l'efficacité de diverses techniques de modélisation, y compris les méthodes traditionnelles, les réseaux de neurones et l'apprentissage par ensemble.

A new study has introduced a method for predicting the breakout rate of Japanese talents by analyzing data from Twitter and television. The research highlights the importance of early detection in advertising and evaluates the effectiveness of various modeling techniques, including traditional, neural network, and ensemble learning methods.

Predicting Talent Breakout Rate using Twitter and TV data

arXiv:2511.16885v1 Announce Type: new 
Abstract: Unlike human reasoning in abstract conceptual spaces, large language models (LLMs) typically reason by generating discrete tokens, which potentially limit their expressive power. The recent work Soft Thinking has shown that LLMs' latent reasoning via soft concepts is a promising direction, but LLMs are trained on discrete tokens. To reduce this gap between the soft concepts in reasoning and the discrete tokens in training, we propose Soft Concept Mixing (SCM), a soft concept aware training scheme that directly exposes the model to soft representations during training. Specifically, SCM constructs a soft concept vector by forming a probability-weighted average of embeddings. Then, this vector is mixed into the model's hidden states, which embody rich contextual information. Finally, the entire latent reasoning process is optimized with Reinforcement Learning (RL). Experiments on five reasoning benchmarks demonstrate that SCM improves the reasoning performance of LLMs, and simultaneously maintains a stable training dynamic.

أدت التطورات الأخيرة في نماذج اللغة الكبيرة (LLMs) إلى تقديم تقنية خلط المفاهيم الناعمة (SCM)، وهي خطة تدريب تعزز التفكير الكامن من خلال دمج تمثيلات المفاهيم الناعمة في الحالات المخفية للنموذج. تهدف هذه الطريقة إلى سد الفجوة بين تدريب الرموز المنفصلة في LLMs وقدرات التفكير الأكثر تجريدًا التي تُلاحظ في الإدراك البشري.

Los recientes avances en modelos de lenguaje de gran tamaño (LLMs) han introducido el Soft Concept Mixing (SCM), un esquema de entrenamiento que mejora el razonamiento latente al integrar representaciones de conceptos suaves en los estados ocultos del modelo. Este enfoque busca cerrar la brecha entre el entrenamiento con tokens discretos de los LLMs y las capacidades de razonamiento más abstractas observadas en la cognición humana.

Des avancées récentes dans les modèles de langage de grande taille (LLMs) ont introduit le Soft Concept Mixing (SCM), un schéma d'entraînement qui améliore le raisonnement latent en intégrant des représentations de concepts doux dans les états cachés du modèle. Cette approche vise à combler le fossé entre l'entraînement par tokens discrets des LLMs et les capacités de raisonnement plus abstraites observées dans la cognition humaine.

Recent advancements in large language models (LLMs) have introduced Soft Concept Mixing (SCM), a training scheme that enhances latent reasoning by integrating soft concept representations into the model's hidden states. This approach aims to bridge the gap between the discrete token training of LLMs and the more abstract reasoning capabilities observed in human cognition.

Predicting Talent Breakout Rate using Twitter and TV data

Was this article worth reading? Share it

Outlier Kit

Algochat

Testfeed