( = Paper PDF, = Presentation slides, = Presentation video)
1.
Mohammad Reza Taesiri; Tianjun Feng; Anh Nguyen; Cor-Paul Bezemer
GlitchBench: Can large multimodal models detect video game glitches? Inproceedings
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024.
Abstract | BibTeX | Tags: Computer games, Foundation models, Game development, Gameplay videos, LLM
@inproceedings{TaesiriCVPR2024,
title = {GlitchBench: Can large multimodal models detect video game glitches?},
author = {Mohammad Reza Taesiri and Tianjun Feng and Anh Nguyen and Cor-Paul Bezemer},
year = {2024},
date = {2024-06-15},
urldate = {2024-03-15},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
abstract = {Large multimodal models (LMMs) have evolved from
large language models (LLMs) to integrate multiple input
modalities, such as visual inputs. This integration augments
the capacity of LLMs for tasks requiring visual comprehen-
sion and reasoning. However, the extent and limitations of
their enhanced abilities are not fully understood, especially
when it comes to real-world tasks. To address this gap, we
introduce GlitchBench, a novel benchmark derived from
video game quality assurance tasks, to test and evaluate the
reasoning capabilities of LMMs. Our benchmark is curated
from a variety of unusual and glitched scenarios from video
games and aims to challenge both the visual and linguis-
tic reasoning powers of LMMs in detecting and interpreting
out-of-the-ordinary events. We evaluate multiple state-of-
the-art LMMs, and we show that GlitchBench presents a
new challenge for these models. Code and data are avail-
able at: https://glitchbench.github.io/},
keywords = {Computer games, Foundation models, Game development, Gameplay videos, LLM},
pubstate = {published},
tppubtype = {inproceedings}
}
Large multimodal models (LMMs) have evolved from
large language models (LLMs) to integrate multiple input
modalities, such as visual inputs. This integration augments
the capacity of LLMs for tasks requiring visual comprehen-
sion and reasoning. However, the extent and limitations of
their enhanced abilities are not fully understood, especially
when it comes to real-world tasks. To address this gap, we
introduce GlitchBench, a novel benchmark derived from
video game quality assurance tasks, to test and evaluate the
reasoning capabilities of LMMs. Our benchmark is curated
from a variety of unusual and glitched scenarios from video
games and aims to challenge both the visual and linguis-
tic reasoning powers of LMMs in detecting and interpreting
out-of-the-ordinary events. We evaluate multiple state-of-
the-art LMMs, and we show that GlitchBench presents a
new challenge for these models. Code and data are avail-
able at: https://glitchbench.github.io/
large language models (LLMs) to integrate multiple input
modalities, such as visual inputs. This integration augments
the capacity of LLMs for tasks requiring visual comprehen-
sion and reasoning. However, the extent and limitations of
their enhanced abilities are not fully understood, especially
when it comes to real-world tasks. To address this gap, we
introduce GlitchBench, a novel benchmark derived from
video game quality assurance tasks, to test and evaluate the
reasoning capabilities of LMMs. Our benchmark is curated
from a variety of unusual and glitched scenarios from video
games and aims to challenge both the visual and linguis-
tic reasoning powers of LMMs in detecting and interpreting
out-of-the-ordinary events. We evaluate multiple state-of-
the-art LMMs, and we show that GlitchBench presents a
new challenge for these models. Code and data are avail-
able at: https://glitchbench.github.io/