# Meta-Modal Agent: Sequential Evidence Routing for Missing-Modality Candidate Reranking

arXiv: https://arxiv.org/abs/2605.25007 · HTML: https://mlsanity.com/paper/2605.25007/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Agentic recommender systems, one of 16 directions in LLM agents and tool use.
- Builds on 18 of its 29 references that are in the map.
- First posted on arXiv 24 May 2026.

## Placed in

- LLM agents and tool use / [Agentic recommender systems](https://mlsanity.com/agents/agentic-recommender-systems/)

## Builds on (most cited first, 18 of 18)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Self-Attentive Sequential Recommendation](https://mlsanity.com/paper/1808.09781/) (2018)
- [LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation](https://mlsanity.com/paper/2002.02126/) (2020)
- [Image-based Recommendations on Styles and Substitutes](https://mlsanity.com/paper/1506.04757/) (2015)
- [Recommendation as Instruction Following: A Large Language Model Empowered Recommendation Approach](https://mlsanity.com/paper/2305.07001/) (2023)
- [Representation Learning with Large Language Models for Recommendation](https://mlsanity.com/paper/2310.15950/) (2023)
- [AgentCF: Collaborative Learning with Autonomous Language Agents for Recommender Systems](https://mlsanity.com/paper/2310.09233/) (2023)
- [Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems](https://mlsanity.com/paper/2002.09102/) (2020)
- [LlamaRec: Two-Stage Recommendation using Large Language Models for Ranking](https://mlsanity.com/paper/2311.02089/) (2023)
- [Self-Supervised Reinforcement Learning for Recommender Systems](https://mlsanity.com/paper/2006.05779/) (2020)
- [Towards Unified World Models for Visual Navigation via Memory-Augmented Planning and Foresight](https://mlsanity.com/paper/2510.08713/) (2025)
- [LRMM: Learning to Recommend with Missing Modalities](https://mlsanity.com/paper/1808.06791/) (2018)
- [Do We Really Need to Drop Items with Missing Modalities in Multimodal Recommendation?](https://mlsanity.com/paper/2408.11767/) (2024)
- [Meta-learning enhanced next POI recommendation by leveraging check-ins from auxiliary cities](https://mlsanity.com/paper/2308.09309/) (2023)
- [DeepInterestGR: Mining Deep Multi-Interest Using Multi-Modal LLMs for Generative Recommendation](https://mlsanity.com/paper/2602.18907/) (2026)
- [Disentangling and Generating Modalities for Recommendation in Missing Modality Scenarios](https://mlsanity.com/paper/2504.16352/) (2025)
- [TriAlignGR: Triangular Multitask Alignment with Multimodal Deep Interest Mining for Generative Recommendation](https://mlsanity.com/paper/2605.05249/) (2026)
- [Robust Multimodal Recommendation via Graph Retrieval-Enhanced Modality Completion](https://mlsanity.com/paper/2605.00670/) (2026)

## Abstract (arXiv)

Missing-modality recommenders usually reconstruct absent representations, although the observed evidence may not determine the missing content. We formulate candidate reranking as budgeted sequential evidence acquisition. A policy queries text, image, and interaction-graph tools, incorporates textttNull returns into its observation history, and sparsely rescores a retrieved candidate pool. Our Meta-Modal Agent (MMA) uses PPO to optimize terminal NDCG and tool cost without explicit access to the route-availability mask or target identity. When only one evidence route is available, MMA-Auto improves NDCG@10 by 10.0% over the strongest completion baseline and by 9.5% over a fixed router with the same Llama scorer. It obtains the highest result in all nine reported combinations of dataset and available route against these comparators. MMA-Auto also reduces failed calls by 17.8 percentage points and uses 1.1 fewer turns than the fixed router. On the fixed candidate pools produced by full-catalog retrieval, MMA-Auto improves NDCG@10 by 19.7%. These results associate adaptive evidence routing with improved reranking under severe, constructed missingness. The code is available at: https://anonymous.4open.science/r/WSDM2027-MMA-C381.