Pop-Up Distractions Reveal Bag-of-Events Behavior in Video Large Language Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916050215370752 |
|---|---|
| author | Chew, Oscar Honcharenko, Serhii Chen, Qian-Hui Lu, Patricia Zaveri, Dishant Doan, Khoa D. Huang, Kuan-Hao |
| author_facet | Chew, Oscar Honcharenko, Serhii Chen, Qian-Hui Lu, Patricia Zaveri, Dishant Doan, Khoa D. Huang, Kuan-Hao |
| contents | A key capability for video understanding is reliably linking subjects to events across time, yet whether Video Large Language Models (VideoLLMs) actually achieve this remains unclear. In this work, we introduce DistractionBench to evaluate whether VideoLLMs can robustly link subjects and events in the presence of unrelated video segments. Through controlled interventions, such as inserting short advertisement clips into longer videos, we show that VideoLLMs frequently hallucinate interactions between entities from different segments, incorrectly attributing actions from injected advertisements to subjects in the main video. We characterize this systematic hallucination as bag-of-events (BoE) behavior, where models process videos as collections of events rather than temporally structured sequences. Evaluating 11 popular VideoLLMs, we find that all models exhibit substantial BoE behavior. Our findings suggest that VideoLLMs lack reliable mechanisms for temporal grounding and motivate the development of models with more robust subject-event association. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_27101 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Pop-Up Distractions Reveal Bag-of-Events Behavior in Video Large Language Models Chew, Oscar Honcharenko, Serhii Chen, Qian-Hui Lu, Patricia Zaveri, Dishant Doan, Khoa D. Huang, Kuan-Hao Computer Vision and Pattern Recognition Computation and Language A key capability for video understanding is reliably linking subjects to events across time, yet whether Video Large Language Models (VideoLLMs) actually achieve this remains unclear. In this work, we introduce DistractionBench to evaluate whether VideoLLMs can robustly link subjects and events in the presence of unrelated video segments. Through controlled interventions, such as inserting short advertisement clips into longer videos, we show that VideoLLMs frequently hallucinate interactions between entities from different segments, incorrectly attributing actions from injected advertisements to subjects in the main video. We characterize this systematic hallucination as bag-of-events (BoE) behavior, where models process videos as collections of events rather than temporally structured sequences. Evaluating 11 popular VideoLLMs, we find that all models exhibit substantial BoE behavior. Our findings suggest that VideoLLMs lack reliable mechanisms for temporal grounding and motivate the development of models with more robust subject-event association. |
| title | Pop-Up Distractions Reveal Bag-of-Events Behavior in Video Large Language Models |
| topic | Computer Vision and Pattern Recognition Computation and Language |
| url | https://arxiv.org/abs/2605.27101 |