Chronocept: Instilling a Sense of Time in Machines

TL;DR

Chronocept addresses AI's gap in temporal reasoning by modeling temporal validity as a continuous density $p_i(t)$ using a skew-normal distribution on a logarithmic time axis. It introduces two benchmarks with multi-axis annotations to capture emergence, peak relevance, and decay, and demonstrates that simple neural regressors can learn interpretable parameters $(\xi, \omega, \alpha)$, enabling applications in RAG and fact-checking. The work provides public data, baselines, and an annotation framework that emphasizes structure, showing that axis-informed representations improve model fit. Overall, Chronocept advances temporally aware AI and lays groundwork for proactive agents and knowledge lifecycle management in time-sensitive contexts.

Abstract

Human cognition is deeply intertwined with a sense of time, known as Chronoception. This sense allows us to judge how long facts remain valid and when knowledge becomes outdated. Despite progress in vision, language, and motor control, AI still struggles to reason about temporal validity. We introduce Chronocept, the first benchmark to model temporal validity as a continuous probability distribution over time. Using skew-normal curves fitted along semantically decomposed temporal axes, Chronocept captures nuanced patterns of emergence, decay, and peak relevance. It includes two datasets: Benchmark I (atomic facts) and Benchmark II (multi-sentence passages). Annotations show strong inter-annotator agreement (84% and 89%). Our baselines predict curve parameters - location, scale, and skewness - enabling interpretable, generalizable learning and outperforming classification-based approaches. Chronocept fills a foundational gap in AI's temporal reasoning, supporting applications in knowledge grounding, fact-checking, retrieval-augmented generation (RAG), and proactive agents. Code and data are publicly available.

Figures (8)

• Figure 1: Composition of samples in Chronocept benchmarks.
• Figure 2: BERT training loss curves for Benchmark I and Benchmark II. The loss flatlined after 2 epochs for both benchmarks.
• Figure 3: Annotation guidelines for Chronocept.
• Figure 4: Comparison of co-occurrence matrices before and after merging the Generic and Static axes, used to assess annotation consistency.
• Figure 5: Effect of logarithmic base choice on time axis representation. Base 1.1 preserves quasi-linear spacing; larger bases induce stronger compression.