Investigating Length Bias and Robustness in LVLMs for Multiple-Choice Question Answering
Large Vision-Language Models (LVLMs) demonstrate strong performance on multiple-choice question answering (MCQA), yet their reliability under biased conditions remains underexplored. We investigate length bias, the systematic preference for longer answer options, through controlled benchmarks spanning fine-grained visual domains. LVLMs consistently favor longer options even when visually inconsistent, with accuracy dropping 2--38\% when correct answers are shorter and increasing 1--70\% when longer. This bias persists across model scales. Through layer-wise mechanistic analysis, we reveal that length bias emerges from late-layer suppression of visual evidence in favor of language-side len...
| Authors | Md. Atabuzzaman, Hani Alomari, Chris Thomas |
|---|---|
| Primary category | cs.CV |
| Published | 17 October 2026 |
| Last updated at source | 17 October 2026 |
Popularity, last 4 days
0 now. Counts as reported by the source, snapshotted daily by Silytics.
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 23 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.