Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -44,3 +44,67 @@ dataset_info:
|
|
44 |
download_size: 5527713
|
45 |
dataset_size: 53565714
|
46 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
44 |
download_size: 5527713
|
45 |
dataset_size: 53565714
|
46 |
---
|
47 |
+
|
48 |
+
|
49 |
+
# UnSeenTimeQA: Time-Sensitive Question-Answering Beyond LLMs’ Memorization [ACL 2025]
|
50 |
+
|
51 |
+
<div align="center">
|
52 |
+
<img src="assets/unseentimeqa.png" width=750/>
|
53 |
+
</div><br>
|
54 |
+
|
55 |
+
We introduce **UnSeenTimeQA**, a novel data contamination-free time-sensitive question-answering (TSQA) benchmark. It differs from existing TSQA benchmarks by avoiding web-searchable queries grounded in the real world. We present a series of time-sensitive event scenarios based on synthetically generated facts. It requires large language models (LLMs) to engage in genuine temporal reasoning without depending on the factual knowledge acquired during the pre-training phase. We designed three types of time-sensitive questions to test LLMs' temporal reasoning abilities over sequential and parallel event occurrences. Please refer to our paper [here](https://arxiv.org/abs/2407.03525).
|
56 |
+
|
57 |
+
|
58 |
+
## Developed by
|
59 |
+
|
60 |
+
* Md Nayem Uddin
|
61 |
+
* Amir Saeidi
|
62 |
+
* Divij Handa
|
63 |
+
* Agastya Seth
|
64 |
+
* Tran Cao Son
|
65 |
+
* Eduardo Blanco
|
66 |
+
* Steven R. Corman
|
67 |
+
* Chitta Baral
|
68 |
+
|
69 |
+
## Dataset
|
70 |
+
|
71 |
+
Different types of events (six) from the **UnSeenTimeQA** benchmark. The benchmark is structured into **four** difficulty levels: easy, medium, hard (serial), and hard (parallel).
|
72 |
+
|
73 |
+
In the **easy level**, the start (S) and end (E) times of each event are given.
|
74 |
+
|
75 |
+
The **medium level** includes the start time (S) and duration (D) of each event.
|
76 |
+
|
77 |
+
The **hard (serial) level** presents only the duration (D) of events, assuming sequential occurrence.
|
78 |
+
|
79 |
+
The **hard (parallel) level** also includes only durations (D), but events can occur simultaneously.
|
80 |
+
|
81 |
+
Structure of **UnSeenTimeQA** dataset:
|
82 |
+
|
83 |
+
```JSON
|
84 |
+
{
|
85 |
+
"id": "int",
|
86 |
+
"data_split": "str",
|
87 |
+
"execution": "str",
|
88 |
+
"split_id": "int",
|
89 |
+
"question_type": "str",
|
90 |
+
"domain_description": "str",
|
91 |
+
"objects_description": "list",
|
92 |
+
"initial_states_description": "list",
|
93 |
+
"events": "list",
|
94 |
+
"question": "str",
|
95 |
+
"answers": "list",
|
96 |
+
"depth": "int",
|
97 |
+
"do_not_train": "bool"
|
98 |
+
}
|
99 |
+
```
|
100 |
+
|
101 |
+
## Citation
|
102 |
+
|
103 |
+
```bibtex
|
104 |
+
@article{uddin2024unseentimeqa,
|
105 |
+
title={UnSeenTimeQA: Time-Sensitive Question-Answering Beyond LLMs' Memorization},
|
106 |
+
author={Uddin, Md Nayem and Saeidi, Amir and Handa, Divij and Seth, Agastya and Son, Tran Cao and Blanco, Eduardo and Corman, Steven R and Baral, Chitta},
|
107 |
+
journal={arXiv preprint arXiv:2407.03525},
|
108 |
+
year={2024}
|
109 |
+
}
|
110 |
+
```
|