Agent4POI: Agentic Context-Conditioned Affordance Reasoning for Multimodal Point-of-Interest Recommendation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Jinze, Zeng, Yangchen, Zhang, Tiehua, Zhang, Lu, Liu, Yuze, Liu, Yongchao, Ma, Xingjun, Sun, Zhu
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916013548765184
author Wang, Jinze
Zeng, Yangchen
Zhang, Tiehua
Zhang, Lu
Liu, Yuze
Liu, Yongchao
Ma, Xingjun
Sun, Zhu
author_facet Wang, Jinze
Zeng, Yangchen
Zhang, Tiehua
Zhang, Lu
Liu, Yuze
Liu, Yongchao
Ma, Xingjun
Sun, Zhu
contents We introduce Agent4POI, the first POI recommendation framework that generates context-conditioned multimodal representations at recommendation time, rather than relying on static POI embeddings pre-computed independently of context. Existing multimodal systems encode each POI once as a static embedding, a design that precludes reasoning about why the same cafe affords solo work on Monday but group celebration on Friday evening. We formally prove that no pre-computed encoder can satisfy context-sensitive ranking under standard bilinear scoring, motivating inference-time item-side representation. Agent4POI inverts this computation: given a situational context, a four-phase LLM agent generates dynamic, context-specific affordance queries (Phase 1) and executes a five-step cross-modal chain-of-thought over image, review, and metadata evidence (Phase 2). The resulting uncertainty-aware affordance representation is grounded in Gibsonian affordance theory. These cross-modal verdicts form a structured, uncertainty-adjusted affordance representation (Phase 3), which is aligned with user preferences via a semantic caching system for low-latency ranking (Phase 4). On three POI benchmarks and three evaluation configurations (standard, cold-start, context-shift), Agent4POI achieves a 23.2% relative gain over the strongest baseline and degrades by only 7.5% under context-shift versus 16--17\% for the strongest baselines. In cold-start scenarios, Agent4POI outperforms the best content-based baseline by up to 2.4x, whereas ID-based methods fail to generalize.
format Preprint
id arxiv_https___arxiv_org_abs_2605_15203
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Agent4POI: Agentic Context-Conditioned Affordance Reasoning for Multimodal Point-of-Interest Recommendation
Wang, Jinze
Zeng, Yangchen
Zhang, Tiehua
Zhang, Lu
Liu, Yuze
Liu, Yongchao
Ma, Xingjun
Sun, Zhu
Information Retrieval
Artificial Intelligence
Multiagent Systems
We introduce Agent4POI, the first POI recommendation framework that generates context-conditioned multimodal representations at recommendation time, rather than relying on static POI embeddings pre-computed independently of context. Existing multimodal systems encode each POI once as a static embedding, a design that precludes reasoning about why the same cafe affords solo work on Monday but group celebration on Friday evening. We formally prove that no pre-computed encoder can satisfy context-sensitive ranking under standard bilinear scoring, motivating inference-time item-side representation. Agent4POI inverts this computation: given a situational context, a four-phase LLM agent generates dynamic, context-specific affordance queries (Phase 1) and executes a five-step cross-modal chain-of-thought over image, review, and metadata evidence (Phase 2). The resulting uncertainty-aware affordance representation is grounded in Gibsonian affordance theory. These cross-modal verdicts form a structured, uncertainty-adjusted affordance representation (Phase 3), which is aligned with user preferences via a semantic caching system for low-latency ranking (Phase 4). On three POI benchmarks and three evaluation configurations (standard, cold-start, context-shift), Agent4POI achieves a 23.2% relative gain over the strongest baseline and degrades by only 7.5% under context-shift versus 16--17\% for the strongest baselines. In cold-start scenarios, Agent4POI outperforms the best content-based baseline by up to 2.4x, whereas ID-based methods fail to generalize.
title Agent4POI: Agentic Context-Conditioned Affordance Reasoning for Multimodal Point-of-Interest Recommendation
topic Information Retrieval
Artificial Intelligence
Multiagent Systems
url https://arxiv.org/abs/2605.15203