Router-Suggest: Dynamic Routing for Multimodal Auto-Completion in Visually-Grounded Dialogs

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Mishra, Sandeep, Budagam, Devichand, Mandal, Anubhab, Santra, Bishal, Goyal, Pawan, Gupta, Manish
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911363939434496
author Mishra, Sandeep
Budagam, Devichand
Mandal, Anubhab
Santra, Bishal
Goyal, Pawan
Gupta, Manish
author_facet Mishra, Sandeep
Budagam, Devichand
Mandal, Anubhab
Santra, Bishal
Goyal, Pawan
Gupta, Manish
contents Real-time multimodal auto-completion is essential for digital assistants, chatbots, design tools, and healthcare consultations, where user inputs rely on shared visual context. We introduce Multimodal Auto-Completion (MAC), a task that predicts upcoming characters in live chats using partially typed text and visual cues. Unlike traditional text-only auto-completion (TAC), MAC grounds predictions in multimodal context to better capture user intent. To enable this task, we adapt MMDialog and ImageChat to create benchmark datasets. We evaluate leading vision-language models (VLMs) against strong textual baselines, highlighting trade-offs in accuracy and efficiency. We present Router-Suggest, a router framework that dynamically selects between textual models and VLMs based on dialog context, along with a lightweight variant for resource-constrained environments. Router-Suggest achieves a 2.3x to 10x speedup over the best-performing VLM. A user study shows that VLMs significantly excel over textual models on user satisfaction, notably saving user typing effort and improving the quality of completions in multi-turn conversations. These findings underscore the need for multimodal context in auto-completions, leading to smarter, user-aware assistants.
format Preprint
id arxiv_https___arxiv_org_abs_2601_05851
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Router-Suggest: Dynamic Routing for Multimodal Auto-Completion in Visually-Grounded Dialogs
Mishra, Sandeep
Budagam, Devichand
Mandal, Anubhab
Santra, Bishal
Goyal, Pawan
Gupta, Manish
Computation and Language
Artificial Intelligence
Computer Vision and Pattern Recognition
Real-time multimodal auto-completion is essential for digital assistants, chatbots, design tools, and healthcare consultations, where user inputs rely on shared visual context. We introduce Multimodal Auto-Completion (MAC), a task that predicts upcoming characters in live chats using partially typed text and visual cues. Unlike traditional text-only auto-completion (TAC), MAC grounds predictions in multimodal context to better capture user intent. To enable this task, we adapt MMDialog and ImageChat to create benchmark datasets. We evaluate leading vision-language models (VLMs) against strong textual baselines, highlighting trade-offs in accuracy and efficiency. We present Router-Suggest, a router framework that dynamically selects between textual models and VLMs based on dialog context, along with a lightweight variant for resource-constrained environments. Router-Suggest achieves a 2.3x to 10x speedup over the best-performing VLM. A user study shows that VLMs significantly excel over textual models on user satisfaction, notably saving user typing effort and improving the quality of completions in multi-turn conversations. These findings underscore the need for multimodal context in auto-completions, leading to smarter, user-aware assistants.
title Router-Suggest: Dynamic Routing for Multimodal Auto-Completion in Visually-Grounded Dialogs
topic Computation and Language
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.05851