TRAM: Benchmarking Temporal Reasoning for Large Language Models
This work addresses the problem of inconsistent evaluations for temporal reasoning in large language models, providing a standardized benchmark to facilitate progress, though it is incremental in nature.
The authors tackled the lack of standardized benchmarks for temporal reasoning in natural language by introducing TRAM, a comprehensive benchmark of ten datasets, and found that the best-performing large language models significantly lag behind human performance.
Reasoning about time is essential for understanding the nuances of events described in natural language. Previous research on this topic has been limited in scope, characterized by a lack of standardized benchmarks that would allow for consistent evaluations across different studies. In this paper, we introduce TRAM, a temporal reasoning benchmark composed of ten datasets, encompassing various temporal aspects of events such as order, arithmetic, frequency, and duration, designed to facilitate a comprehensive evaluation of the TeR capabilities of large language models (LLMs). We evaluate popular LLMs like GPT-4 and Llama2 in zero-shot and few-shot scenarios, and establish baselines with BERT-based and domain-specific models. Our findings indicate that the best-performing model lags significantly behind human performance. It is our aspiration that TRAM will spur further progress in enhancing the TeR capabilities of LLMs.