SportsMetrics: Blending Text and Numerical Data to Understand Information Fusion in LLMs
Paper • 2402.10979 • Published
id stringlengths 13 15 | domain stringclasses 2
values | task stringclasses 17
values | system stringclasses 2
values | user stringlengths 12.3k 34.8k | label stringlengths 35 709 | name_map stringlengths 67 749 ⌀ |
|---|---|---|---|---|---|---|
400489111_0-0 | NBA | conflict-swap_two_players | You are a helpful assistant tasked with analyzing sports games. You have been given a play-by-play breakdown of a basketball game between two teams.
The "Time" column shows the exact time on the game clock when each play took place. The game clock counts down, so this column displays times in a descending order.
The ... | Initially, you are given a JSON object where all values are set to 0. Based on the provided play-by-play breakdown and team-player data, you will update the game points in the provided JSON format.
{"Denver Nuggets": 0, "New York Knicks": 0}
Team-Player Table:
New York Knicks: wilson chandler, kenneth faried, andrea ... | {'Denver Nuggets': 111, 'New York Knicks': 81} | {'New York Knicks': ['kenyon martin', 'carmelo anthony'], 'Denver Nuggets': ['wilson chandler', 'kenneth faried']} |
400900227_0-1 | NBA | conflict-swap_two_players | You are a helpful assistant tasked with analyzing sports games. You have been given a play-by-play breakdown of a basketball game between two teams.
The "Time" column shows the exact time on the game clock when each play took place. The game clock counts down, so this column displays times in a descending order.
The ... | Initially, you are given a JSON object where all values are set to 0. Based on the provided play-by-play breakdown and team-player data, you will update the game points in the provided JSON format.
{"Brooklyn Nets": 0, "Memphis Grizzlies": 0}
Team-Player Table:
Memphis Grizzlies: bojan bogdanovic, rondae hollis jeffe... | {'Brooklyn Nets': 94, 'Memphis Grizzlies': 121} | {'Memphis Grizzlies': ['jamychal green', 'james ennis iii'], 'Brooklyn Nets': ['bojan bogdanovic', 'rondae hollis jefferson']} |
401468026_0-2 | NBA | conflict-swap_two_players | You are a helpful assistant tasked with analyzing sports games. You have been given a play-by-play breakdown of a basketball game between two teams.
The "Time" column shows the exact time on the game clock when each play took place. The game clock counts down, so this column displays times in a descending order.
The ... | Initially, you are given a JSON object where all values are set to 0. Based on the provided play-by-play breakdown and team-player data, you will update the game points in the provided JSON format.
{"San Antonio Spurs": 0, "Charlotte Hornets": 0}
Team-Player Table:
Charlotte Hornets: jeremy sochan, keldon johnson, ma... | {'San Antonio Spurs': 115, 'Charlotte Hornets': 116} | {'Charlotte Hornets': ['pj washington', 'gordon hayward'], 'San Antonio Spurs': ['jeremy sochan', 'keldon johnson']} |
400489856_0-3 | NBA | conflict-swap_two_players | You are a helpful assistant tasked with analyzing sports games. You have been given a play-by-play breakdown of a basketball game between two teams.
The "Time" column shows the exact time on the game clock when each play took place. The game clock counts down, so this column displays times in a descending order.
The ... | Initially, you are given a JSON object where all values are set to 0. Based on the provided play-by-play breakdown and team-player data, you will update the game points in the provided JSON format.
{"Atlanta Hawks": 0, "Denver Nuggets": 0}
Team-Player Table:
Denver Nuggets: paul millsap, pero antic, timofey mozgov, r... | {'Atlanta Hawks': 94, 'Denver Nuggets': 95} | {'Denver Nuggets': ['wilson chandler', 'kenneth faried'], 'Atlanta Hawks': ['paul millsap', 'pero antic']} |
400827984_0-4 | NBA | conflict-swap_two_players | You are a helpful assistant tasked with analyzing sports games. You have been given a play-by-play breakdown of a basketball game between two teams.
The "Time" column shows the exact time on the game clock when each play took place. The game clock counts down, so this column displays times in a descending order.
The ... | Initially, you are given a JSON object where all values are set to 0. Based on the provided play-by-play breakdown and team-player data, you will update the game points in the provided JSON format.
{"Oklahoma City Thunder": 0, "Phoenix Suns": 0}
Team-Player Table:
Phoenix Suns: kevin durant, serge ibaka, tyson chandl... | {'Oklahoma City Thunder': 104, 'Phoenix Suns': 123} | {'Phoenix Suns': ['pj tucker', 'markieff morris'], 'Oklahoma City Thunder': ['kevin durant', 'serge ibaka']} |
401161286_0-5 | NBA | conflict-swap_two_players | You are a helpful assistant tasked with analyzing sports games. You have been given a play-by-play breakdown of a basketball game between two teams.
The "Time" column shows the exact time on the game clock when each play took place. The game clock counts down, so this column displays times in a descending order.
The ... | Initially, you are given a JSON object where all values are set to 0. Based on the provided play-by-play breakdown and team-player data, you will update the game points in the provided JSON format.
{"Miami Heat": 0, "Sacramento Kings": 0}
Team-Player Table:
Sacramento Kings: duncan robinson, derrick jones jr, harriso... | {'Miami Heat': 136, 'Sacramento Kings': 95} | {'Sacramento Kings': ['nemanja bjelica', 'marvin bagley iii'], 'Miami Heat': ['duncan robinson', 'derrick jones jr']} |
401071544_0-6 | NBA | conflict-swap_two_players | You are a helpful assistant tasked with analyzing sports games. You have been given a play-by-play breakdown of a basketball game between two teams.
The "Time" column shows the exact time on the game clock when each play took place. The game clock counts down, so this column displays times in a descending order.
The ... | Initially, you are given a JSON object where all values are set to 0. Based on the provided play-by-play breakdown and team-player data, you will update the game points in the provided JSON format.
{"Golden State Warriors": 0, "Sacramento Kings": 0}
Team-Player Table:
Sacramento Kings: kevin durant, draymond green, d... | {'Golden State Warriors': 108, 'Sacramento Kings': 140} | {'Sacramento Kings': ['harrison barnes', 'willie cauley stein'], 'Golden State Warriors': ['kevin durant', 'draymond green']} |
400579452_0-7 | NBA | conflict-swap_two_players | "You are a helpful assistant tasked with analyzing sports games. You have been given a play-by-play (...TRUNCATED) | "Initially, you are given a JSON object where all values are set to 0. Based on the provided play-by(...TRUNCATED) | {'Charlotte Hornets': 95, 'Toronto Raptors': 71} | "{'Toronto Raptors': ['tyler hansbrough', 'demar derozan'], 'Charlotte Hornets': ['marvin williams',(...TRUNCATED) |
401468441_0-8 | NBA | conflict-swap_two_players | "You are a helpful assistant tasked with analyzing sports games. You have been given a play-by-play (...TRUNCATED) | "Initially, you are given a JSON object where all values are set to 0. Based on the provided play-by(...TRUNCATED) | {'Toronto Raptors': 111, 'Dallas Mavericks': 94} | "{'Dallas Mavericks': ['dorian finney smith', 'reggie bullock'], 'Toronto Raptors': ['thaddeus young(...TRUNCATED) |
400489548_0-9 | NBA | conflict-swap_two_players | "You are a helpful assistant tasked with analyzing sports games. You have been given a play-by-play (...TRUNCATED) | "Initially, you are given a JSON object where all values are set to 0. Based on the provided play-by(...TRUNCATED) | {'Miami Heat': 116, 'Oklahoma City Thunder': 91} | "{'Oklahoma City Thunder': ['kevin durant', 'thabo sefolosha'], 'Miami Heat': ['chris bosh', 'shane (...TRUNCATED) |
End of preview. Expand in Data Studio
SportsMetrics
Benchmark data to evaluate numerical reasoning and information fusion of LLMs.
SportsMetrics: Blending Text and Numerical Data to Understand Information Fusion in LLMs
Yebowen Hu, Kaiqiang Song, Sangwoo Cho, Xiaoyang Wang, Hassan Foroosh, Dong Yu, Fei Liu
In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL'24), Bangkok, Thailand.
Arxiv Paper
Usage
from datasets import load_dataset
def get_task(domain, task):
bench_data = []
dataset = load_dataset("huuuyeah/SportsMetrics",split="test")
for instance in dataset:
if instance["domain"]==domain and instance["task"]==task:
bench_data.append(instance)
return bench_data
def message_iter(domain, task):
bench_data = get_task(domain, task)
if len(bench_data) == 0:
print("No data loaded.")
return
for instance in bench_data:
messages = [
{"role": "system", "content": instance["system"]},
{"role": "user", "content": instance["user"]}
]
yield messages
return
Benchmark Tasks
The LLM is mandatorily required to generate responses in JSON format.
Reasoning Task
- reasoning-team_points_tracking: (NBA) Tracking team points in one match.
- reasoning-key_stats_tracking: (NBA, NFL) Tracking the key statistics for sports analytics.
Conflicts Task
- conflict-one_point_rule: (NBA) All scoring actions in the competition are set to be worth only one point.
- conflict-swap_{num}_players: (NBA) Swap {num} of spalyer between two teams.
Robustness Task
- robustness-duplicate_{prob}: (NBA) Replicate the non-scoring move with a probability of {prob}.
- robustness-remove_{prob}: (NBA) Remove the non-scoring move with a probability of {prob}.
- robustness-shuffled_pbp: (NBA) Shuffle the order of all moves in play-by-play descriptions while maintain the original order of timestamps.
- robustness-{num}_fiction_names: (NFL) Randomly select {num} of players from both teams and replace them with names from fiction movies.
Bibtex
@misc{hu2024sportsmetricsblendingtextnumerical,
title={SportsMetrics: Blending Text and Numerical Data to Understand Information Fusion in LLMs},
author={Yebowen Hu and Kaiqiang Song and Sangwoo Cho and Xiaoyang Wang and Hassan Foroosh and Dong Yu and Fei Liu},
year={2024},
eprint={2402.10979},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2402.10979},
}
- Downloads last month
- 37