Early to Share, Late to Save: Synchronisation-Driven Communication Gating in Bandwidth-Constrained Cooperative VLN
For embodied agents operating under bandwidth constraints, this work provides a practical communication gating strategy that improves navigation performance, though the problem setting and method are incremental.
The paper introduces bandwidth-constrained cooperative Vision-Language Navigation and proposes hindsight gating, a supervised method that identifies communication-critical steps from navigation failures. It achieves a cumulative alignment gain of +0.072 with 3 transmissions, approaching unconstrained communication (+0.078) and outperforming random gating (+0.020) and entropy-based gating (+0.017) in alignment efficiency.
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