Early to Share, Late to Save: Synchronisation-Driven Communication Gating in Bandwidth-Constrained Cooperative VLN

2026-07-09

AI Summary

This work addresses the challenge of communication efficiency in collaborative Vision-and-Language Navigation (VLN) under strict bandwidth constraints. The authors propose an efficient communication framework based on a hindsight-gated mechanism, which learns communication decisions through supervised training on failed navigation trajectories. Their analysis reveals and validates a counterintuitive communication paradigm—“early synchronization, late independence”—demonstrating that agents should communicate when they are highly confident rather than uncertain. By integrating lightweight supervised gating, recurrent hidden-state alignment analysis, and GRU-based sequence modeling, the method achieves near-unconstrained communication performance with only three communication rounds, improving communication efficiency by 260% and 320% over random and entropy-based baselines, respectively.

Abstract

Most cooperative Vision-Language Navigation (VLN) methods assume unlimited communication, not considering real-world applications where bandwidth is restricted and information efficiency is critical. We introduce \textbf{bandwidth-constrained cooperative VLN} and propose \textbf{hindsight gating}: a lightweight supervised gate that labels communication-critical steps post-hoc from navigation failures, avoiding the high variance of REINFORCE. Contrary to the intuition that agents should communicate when uncertain, we observe a consistent counter-intuitive pattern: trained gates fire predominantly in early episode steps and more often when agents are confident, across all budget levels ($B \in \{1,3,5\}$). We explain this through \textbf{recurrent hidden-state alignment}: early communication injects grounded trajectory representations that persist and compound through subsequent Gated Recurrent Unit (GRU) updates, achieving $+0.072$ cumulative alignment gain with $B{=}3$ transmissions, approaching unconstrained communication ($+0.078$) at 260\% greater alignment efficiency than random gating ($+0.020$) and 320\% greater efficiency than entropy-based gating ($+0.017$). Our results establish a new communication regime for bandwidth-limited embodied agents: synchronise representations early, navigate independently later. Our codebase is available at: https://github.com/AravG13/bandwidth-constrained-cooperative-vln

Problem

• bandwidth-constrained
• cooperative VLN
• communication gating
• vision-language navigation
• information efficiency

Innovation

• hindsight gating
• bandwidth-constrained cooperative VLN
• recurrent hidden-state alignment
• communication efficiency
• embodied AI