SyncAI.news, a Varaisys broadcasting
Support-Guided Triadic Decoding for Hallucination Mitigation in Multimodal Dialogue with LVLMs
YL

Yiheng Li, Wenhao Wang, Xingyi Li, Zhilin Tang, Zichang Tan, Zhen Lei, Yang Yang

· 1 min read

ResearcharXiv cs.CV

Support-Guided Triadic Decoding for Hallucination Mitigation in Multimodal Dialogue with LVLMs

arXiv:2611.25535v1 Announce Type: new Abstract: In multimodal multi-turn dialogue, dialogue history plays a dual role: reliable responses provide useful context for subsequent turns, whereas hallucinated responses can propagate errors across turns. Existing training-free methods globally adjust the effect of dialogue history on generation, making it difficult to suppress misleading information without weakening useful context. To address this trade-off, we introduce Support-Guided Triadic Decoding (SGTD), a training-free framework that selectively controls the influence of unreliable history and fuses complementary predictions during decoding. Specifically, Historical Answer Support Estimation measures the visual support of historical responses against type-matched alternatives and identifies low-support content that may mislead later generation. Support-Targeted Attention Intervention then weakens the influence of this content while preserving the surrounding dialogue, producing a support-refined prediction. Triadic Decoding combines this prediction with the full-history prediction and answer-free visual evidence, while the Support-Adaptive Plausibility Constraint filters implausible tokens using visual agreement and historical support. Experiments on MMHalSnowball and VisDiaHalBench across representative LVLMs show that SGTD consistently achieves the highest accuracy under hallucinated histories while preserving the benefits of factual context.

Original source

This story was published by arXiv cs.CV and written by Yiheng Li, Wenhao Wang, Xingyi Li, Zhilin Tang, Zichang Tan, Zhen Lei, Yang Yang. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News