ConvFill: Model Collaboration for Responsive Conversational Voice Agents
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917071361671168 |
|---|---|
| author | Srinivas, Vidya Englhardt, Zachary Powers, Maximus Patel, Shwetak Iyer, Vikram |
| author_facet | Srinivas, Vidya Englhardt, Zachary Powers, Maximus Patel, Shwetak Iyer, Vikram |
| contents | Deploying conversational voice agents with large language models faces a critical challenge: cloud-based foundation models provide deep reasoning and domain knowledge but introduce latency that disrupts natural conversation, while on-device models respond immediately but lack sophistication. We propose conversational infill, a task where a lightweight on-device model generates contextually appropriate dialogue while seamlessly incorporating streaming knowledge from a powerful backend model. This approach decouples response latency from model capability, enabling systems that feel responsive while accessing the full power of large-scale models. We present ConvFill, a 360M parameter model trained on synthetic multi-domain conversations. Evaluation across multiple backend models shows that conversational infill can be successfully learned, with ConvFill achieving accuracy improvements of 36-42% over standalone small models of the same size while consistently retaining sub-200ms response latencies. Our results demonstrate the promise of this approach for building on-device conversational agents that are both immediately responsive and knowledgeable. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_07397 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ConvFill: Model Collaboration for Responsive Conversational Voice Agents Srinivas, Vidya Englhardt, Zachary Powers, Maximus Patel, Shwetak Iyer, Vikram Computation and Language Deploying conversational voice agents with large language models faces a critical challenge: cloud-based foundation models provide deep reasoning and domain knowledge but introduce latency that disrupts natural conversation, while on-device models respond immediately but lack sophistication. We propose conversational infill, a task where a lightweight on-device model generates contextually appropriate dialogue while seamlessly incorporating streaming knowledge from a powerful backend model. This approach decouples response latency from model capability, enabling systems that feel responsive while accessing the full power of large-scale models. We present ConvFill, a 360M parameter model trained on synthetic multi-domain conversations. Evaluation across multiple backend models shows that conversational infill can be successfully learned, with ConvFill achieving accuracy improvements of 36-42% over standalone small models of the same size while consistently retaining sub-200ms response latencies. Our results demonstrate the promise of this approach for building on-device conversational agents that are both immediately responsive and knowledgeable. |
| title | ConvFill: Model Collaboration for Responsive Conversational Voice Agents |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2511.07397 |