ProCompNav: Proactive Instance Navigation with Comparative Judgment for Ambiguous User Queries

TL;DR

ProCompNav introduces a two-stage, discriminative questioning framework for natural-language instance navigation, significantly improving success rates and reducing user response length by actively narrowing down candidate pools.

Contribution

It presents a novel pool-level discriminative questioning approach using comparative judgment, outperforming existing methods in success rate and efficiency.

Findings

ProCompNav outperforms baselines on CoIN-Bench with higher success rates.

It achieves state-of-the-art success rate on TextNav.

The method reduces response length while maintaining high accuracy.

Abstract

Natural-language instance navigation becomes challenging when the initial user request does not uniquely specify the target instance. A practical agent should reduce the user's burden by actively asking only the information needed to distinguish the target from similar distractors, rather than requiring a detailed description upfront. Existing approaches often fall short of this goal: they may stop at the first plausible candidate before sufficiently exploring alternatives, or, even after collecting multiple candidates, ask about the target's attributes derived from individual candidates rather than questions selected to distinguish candidates in the pool. As a result, despite the dialogue, the agent may still fail to distinguish the target from distractors, leading to premature decisions and lengthy user responses. We propose Proactive Instance Navigation with Comparative Judgment (ProCompNav), a two-stage framework that first constructs a candidate pool and then identifies the target through comparative judgment. At each round, ProCompNav extracts an attribute-value pair that splits the current pool, asks a binary yes/no question, and prunes all inconsistent candidates at once. This reframes disambiguation from open-ended target description to pool-level discriminative questioning, where each question is chosen to narrow the candidate set. On CoIN-Bench, ProCompNav improves Success Rate over interactive baselines with the same minimal input and non-interactive baselines with detailed descriptions, while substantially reducing Response Length. ProCompNav also achieves state-of-the-art Success Rate on TextNav, suggesting that comparative judgment is broadly useful for instance-level navigation among similar distractors. Code is available at https://github.com/tree-jhk/procompnav.