Metraipolis › archive › dc_ad5fbd35280b
CUSP: Decomposable Collective Uncertainty for Multi-Agent Multimodal Reasoning — arXiv:2609.05708
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.05708
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T13:58:42.108Z
- SHA-256 of the archived text
3b08542bbac6191ff9795ba20d2c2f2d23cc834cd183b4795a19261d708cffed- Stable URL of this page
https://metraipolis.com/documento/dc_ad5fbd35280b
How to cite
CUSP: Decomposable Collective Uncertainty for Multi-Agent Multimodal Reasoning — arXiv:2609.05708 — arXiv (en), https://arxiv.org/abs/2609.05708 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ad5fbd35280b (Metraipolis archive dc_ad5fbd35280b, sha256 3b08542bbac6191ff9795ba20d2c2f2d23cc834cd183b4795a19261d708cffed)
BibTeX
@misc{dcad5fbd35280b,
title = {CUSP: Decomposable Collective Uncertainty for Multi-Agent Multimodal Reasoning — arXiv:2609.05708},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.05708},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ad5fbd35280b; sha256 3b08542bbac6191ff9795ba20d2c2f2d23cc834cd183b4795a19261d708cffed},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.05708. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Aggregating heterogeneous vision-language models (VLMs) can improve multimodal reasoning, but neither an individual model's confidence nor that of the aggregated answer measures reliability at the system level. We present CUSP (Collective Uncertainty through Semantic Opinion Pooling), a training-free uncertainty quantification framework that maps multiple VLM responses to a shared semantic response space, pools them into a pooled semantic opinion, and reports two complementary system-level signals: collective uncertainty, the dispersion of the pooled opinion, and Jensen-Shannon divergence (JSD), the conflict among the model-level opinions. Within this pooled semantic opinion, the unnormalized collective entropy decomposes exactly into the mean of the models' individual semantic entropies and the JSD, separating total dispersion from model conflict. Requiring neither token logits nor calibration labels, CUSP applies to open-weight and commercial VLMs alike. In static multi-VLM ensembles, collective uncertainty is the strongest signal in the small-model regime (0.764 AUROC for prediction-error detection, 0.889 AUARC for abstention), outperforming uncertainty baselines majority voting and naive selection by 4.7 to 15.8 points and widening its margin as the ensemble grows; JSD is strongest in the evaluated commercial regime (0.819 AUROC, 0.910 AUARC) and ranks hard-answer model conflict with AUROC up to 0.982. The pooled prediction also improves accuracy over the average single model by 5.6 to 13.0 points. Over the full trajectory of a multi-step, multi-agent system, subagent collective uncertainty ranks system failures above chance (0.619 AUROC) and gives the best abstention ordering among the evaluated signals (0.699 AUARC). Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.LG). Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG); Multiagent Systems (cs.MA). Pagina del lavoro: https://arxiv.org/abs/2609.05708 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-16T13:39:43.125Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).