2026-10-13T23:50:52Z https://api.figshare.com/v2/oai

oai:figshare.com:article/34017416 2026-09-28T17:48:33Z category_734 category_931 category_106 portal_5 item_type_3 month_year_09_2026

<p>Five-run results on CMU-MOSI (mean sample SD).</p> Wenfei Cao (19510369) Jian Xu (31545) Yinghao Li (2850890) Huizi Yan (25132805) Zhuowei Hu (316112) Wenxu Chen (12043916) Biological Sciences not elsewhere classified Information Systems not elsewhere classified Science Policy strongest overall regression shared latent space noisy temporal observations insufficient global guidance gated residual connections dynamic hypernode injection multimodal sentiment analysis learnable static prior propagation hypernode injection mae checkpoint selection div >< p findings support pre graph attention fusion gat injection provided injection showed graph propagation dependent hypernode multimodal models attention pooling vocal behavior visual sequences slightly higher run comparison oriented balance modal interaction modal inputs level representations language use injected prior corresponding results aligned tri affective state <div><p>Language use, vocal behavior, and facial activity provide complementary indicators of affective state, but multimodal models can be affected by noisy temporal observations and insufficient global guidance before cross-modal interaction. We developed a framework combining bidirectional gated recurrent unit encoders, attention pooling, dynamic hypernode injection, and graph attention fusion. Textual, acoustic, and visual sequences were mapped into a shared latent space and compressed into modality-level representations. A sample-dependent hypernode and a learnable static prior were then injected through gated residual connections before graph propagation. The model was evaluated on CMU-MOSI and CMU-MOSEI using five random seeds and validation-MAE checkpoint selection. On CMU-MOSI, the model obtained an MAE of , a correlation of , and an Acc-7 of . On CMU-MOSEI, the corresponding results were , , and . A five-run comparison of pre-GAT, post-GAT, and no injection showed that pre-GAT injection provided the strongest overall regression-oriented balance, although post-GAT was slightly higher on MOSEI Acc-7. Gate, graph-attention, representation, and balanced case diagnostics showed that the injected prior was sample dependent and modality selective. These findings support pre-propagation hypernode injection under complete, word-aligned tri-modal inputs, without establishing universal superiority or robustness to missing and corrupted modalities.</p></div> 2026-09-28T17:48:21Z Dataset Dataset 10.1371/journal.pone.0359425.t005 https://figshare.com/articles/dataset/_p_Five-run_results_on_CMU-MOSI_mean_sample_SD_p_/34017416 CC BY 4.0