arXiv:2605.11262v2 Announce Type: replace 
Abstract: Chain-of-thought and more broadly test-time compute are known to augment the expressive capabilities of language models and have led to major innovations in reasoning. Motivated by this success, this paper explores latent chain-of-thought as well as the impact of depth and looping for time-series and tabular data. We propose a recurrent scheme in which a structured-data transformer, after an initial forward pass, compresses its query-position hidden states into feedback tokens that are appended to the input and processed again, allowing multiple rounds of latent computation before prediction. We compare CoT models against a same-depth no-CoT baseline, a deeper baseline matched to the CoT model in effective depth, and a looped transformer with weight-tied recurrence but no additional chain-of-thought tokens. Across 36 datasets in time-series forecasting and tabular prediction, latent chain-of-thought improves over the baseline on 7/9 time-series datasets (+12.63\% average gain) and 23/27 tabular datasets (+3.25\% average gain), with CoT models performing best on average in both settings. We also show that the benefit of CoT extends to pretrained foundation models: applying latent CoT to nanoTabPFN, a small open-source tabular foundation model, improves its performance above the much larger TabPFN-v2 on TabArena. Together, these results demonstrate that chain-of-thought is a useful axis for scaling test-time compute for structured data.

دراسة حديثة نُشرت على arXiv تستكشف مفهوم سلسلة التفكير الكامنة في المحولات المعتمدة على البيانات المنظمة، حيث تُظهر أن هذا النهج يعزز قدرة النموذج على معالجة البيانات الزمنية والجدولية من خلال نظام متكرر يسمح بعدة جولات من الحساب قبل إجراء التنبؤات.

Un estudio reciente publicado en arXiv explora el concepto de cadena de pensamiento latente en transformadores de datos estructurados, demostrando que este enfoque mejora la capacidad del modelo para procesar datos de series temporales y tabulares a través de un esquema recurrente que permite múltiples rondas de cálculo antes de hacer predicciones.

Une étude récente publiée sur arXiv explore le concept de chaîne de pensée latente dans les transformateurs de données structurées, démontrant que cette approche améliore la capacité du modèle à traiter des données temporelles et tabulaires grâce à un schéma récurrent permettant plusieurs tours de calcul avant de faire des prédictions.

A recent study published on arXiv explores the concept of latent chain-of-thought in structured-data transformers, demonstrating that this approach enhances the model's ability to process time-series and tabular data through a recurrent scheme that allows multiple rounds of computation before making predictions.