@inproceedings{yang-etal-2026-skillnav,
title = "{S}kill{N}av: A Multi-Agent Blackboard for Hallucination-Resistant Zero-Shot Object Navigation",
author = "Yang, Zhi and
Yu, Chengpu and
Zheng, Yipin",
editor = "Wang, Yimu and
Choi, Yee Man and
Wu, Di and
Ouyang, Siqi and
Azadani, Mozhgan Nasr and
Fung, Yi R. (May)",
booktitle = "Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently ({G}round{LM} 2026)",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.groundlm-1.8/",
pages = "104--112",
ISBN = "979-8-89176-478-1",
abstract = "Embodied agents that route every high-level decision through a single language model typically ingest the entire trajectory at every step, a regime in which prompt context grows monotonically and both lost-in-the-middle degradation and trajectory-level hallucination peak. We show that multi-agent decoupling on a shared Voronoi-Frontier blackboard is a structural answer. SkillNav exposes a sparse topology over the partially-built map whose nodes carry visual evidence and two multiplicative gates (mu{\_}safe, mu{\_}mem) written by specialised Safe and Memory agents; a strategic Exploration agent reads only a compact, system-prompt-style summary of the blackboard {--} whose size is bounded by the topology rather than by trajectory length {--} and writes back only a phase decision plus an SR/IG weight pair. The three agents never see one another{'}s prompts; false-positive evidence is suppressed architecturally by mu{\_}mem rather than by asking the language model to revise its own past output. On HM3D val, SkillNav reaches 71.9 SR / 34.0 SPL with only 22.5 LLM calls per episode ({\textasciitilde}11.9K total tokens), surpassing the strongest published zero-shot baseline (RATE-Nav) by +4.1 SR / +2.7 SPL."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yang-etal-2026-skillnav">
<titleInfo>
<title>SkillNav: A Multi-Agent Blackboard for Hallucination-Resistant Zero-Shot Object Navigation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhi</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chengpu</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yipin</namePart>
<namePart type="family">Zheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yimu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yee</namePart>
<namePart type="given">Man</namePart>
<namePart type="family">Choi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Di</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Siqi</namePart>
<namePart type="family">Ouyang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mozhgan</namePart>
<namePart type="given">Nasr</namePart>
<namePart type="family">Azadani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="given">R</namePart>
<namePart type="given">(May)</namePart>
<namePart type="family">Fung</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-478-1</identifier>
</relatedItem>
<abstract>Embodied agents that route every high-level decision through a single language model typically ingest the entire trajectory at every step, a regime in which prompt context grows monotonically and both lost-in-the-middle degradation and trajectory-level hallucination peak. We show that multi-agent decoupling on a shared Voronoi-Frontier blackboard is a structural answer. SkillNav exposes a sparse topology over the partially-built map whose nodes carry visual evidence and two multiplicative gates (mu_safe, mu_mem) written by specialised Safe and Memory agents; a strategic Exploration agent reads only a compact, system-prompt-style summary of the blackboard – whose size is bounded by the topology rather than by trajectory length – and writes back only a phase decision plus an SR/IG weight pair. The three agents never see one another’s prompts; false-positive evidence is suppressed architecturally by mu_mem rather than by asking the language model to revise its own past output. On HM3D val, SkillNav reaches 71.9 SR / 34.0 SPL with only 22.5 LLM calls per episode (~11.9K total tokens), surpassing the strongest published zero-shot baseline (RATE-Nav) by +4.1 SR / +2.7 SPL.</abstract>
<identifier type="citekey">yang-etal-2026-skillnav</identifier>
<location>
<url>https://aclanthology.org/2026.groundlm-1.8/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>104</start>
<end>112</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SkillNav: A Multi-Agent Blackboard for Hallucination-Resistant Zero-Shot Object Navigation
%A Yang, Zhi
%A Yu, Chengpu
%A Zheng, Yipin
%Y Wang, Yimu
%Y Choi, Yee Man
%Y Wu, Di
%Y Ouyang, Siqi
%Y Azadani, Mozhgan Nasr
%Y Fung, Yi R. (May)
%S Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-478-1
%F yang-etal-2026-skillnav
%X Embodied agents that route every high-level decision through a single language model typically ingest the entire trajectory at every step, a regime in which prompt context grows monotonically and both lost-in-the-middle degradation and trajectory-level hallucination peak. We show that multi-agent decoupling on a shared Voronoi-Frontier blackboard is a structural answer. SkillNav exposes a sparse topology over the partially-built map whose nodes carry visual evidence and two multiplicative gates (mu_safe, mu_mem) written by specialised Safe and Memory agents; a strategic Exploration agent reads only a compact, system-prompt-style summary of the blackboard – whose size is bounded by the topology rather than by trajectory length – and writes back only a phase decision plus an SR/IG weight pair. The three agents never see one another’s prompts; false-positive evidence is suppressed architecturally by mu_mem rather than by asking the language model to revise its own past output. On HM3D val, SkillNav reaches 71.9 SR / 34.0 SPL with only 22.5 LLM calls per episode (~11.9K total tokens), surpassing the strongest published zero-shot baseline (RATE-Nav) by +4.1 SR / +2.7 SPL.
%U https://aclanthology.org/2026.groundlm-1.8/
%P 104-112
Markdown (Informal)
[SkillNav: A Multi-Agent Blackboard for Hallucination-Resistant Zero-Shot Object Navigation](https://aclanthology.org/2026.groundlm-1.8/) (Yang et al., GroundLM 2026)
ACL