Navigating Connected Memories with a Task-oriented Dialog System

TL;DR

This paper introduces a new dataset and task for multi-turn, interactive dialog systems to improve personal media retrieval, enabling users to have more natural and connected conversations about their memories.

Contribution

It presents COMET, a large, grounded dialog dataset for connected memories, and proposes benchmark tasks to advance multi-turn, multimodal dialog systems for personal media retrieval.

Findings

State-of-the-art models show challenges in multimodal understanding.

The dataset enables benchmarking of connected memory dialog tasks.

Analysis highlights key difficulties in multi-turn, multimodal conversations.

Abstract

Recent years have seen an increasing trend in the volume of personal media captured by users, thanks to the advent of smartphones and smart glasses, resulting in large media collections. Despite conversation being an intuitive human-computer interface, current efforts focus mostly on single-shot natural language based media retrieval to aid users query their media and re-live their memories. This severely limits the search functionality as users can neither ask follow-up queries nor obtain information without first formulating a single-turn query. In this work, we propose dialogs for connected memories as a powerful tool to empower users to search their media collection through a multi-turn, interactive conversation. Towards this, we collect a new task-oriented dialog dataset COMET, which contains $11.5k$ user<->assistant dialogs (totaling $103k$ utterances), grounded in simulated personal memory graphs. We employ a resource-efficient, two-phase data collection pipeline that uses: (1) a novel multimodal dialog simulator that generates synthetic dialog flows grounded in memory graphs, and, (2) manual paraphrasing to obtain natural language utterances. We analyze COMET, formulate four main tasks to benchmark meaningful progress, and adopt state-of-the-art language models as strong baselines, in order to highlight the multimodal challenges captured by our dataset.