# MESA:Task-Adaptive Multi-Structure Evidence Selection for Long-Horizon Agent Memory

arXiv: https://arxiv.org/abs/2608.10108 · HTML: https://mlsanity.com/paper/2608.10108/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 37 references that are in the map.
- First posted on arXiv 10 Aug 2026.

## Builds on (most cited first, 25 of 25)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Dense Passage Retrieval for Open-Domain Question Answering](https://mlsanity.com/paper/2004.04906/) (2020)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models](https://mlsanity.com/paper/2506.05176/) (2025)
- [A-MEM: Agentic Memory for LLM Agents](https://mlsanity.com/paper/2502.12110/) (2025)
- [Evaluating Very Long-Term Conversational Memory of LLM Agents](https://mlsanity.com/paper/2402.17753/) (2024)
- [LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory](https://mlsanity.com/paper/2410.10813/) (2024)
- [Meta-Harness: End-to-End Optimization of Model Harnesses](https://mlsanity.com/paper/2603.28052/) (2026)
- [Gemma 4 Technical Report](https://mlsanity.com/paper/2607.02770/) (2026)
- [Auto-GPT for Online Decision Making: Benchmarks and Additional Opinions](https://mlsanity.com/paper/2306.02224/) (2023)
- [From Human Memory to AI Memory: A Survey on Memory Mechanisms in the Era of LLMs](https://mlsanity.com/paper/2504.15965/) (2025)
- [MemoRAG: Boosting Long Context Processing with Global Memory-Enhanced Retrieval Augmentation](https://mlsanity.com/paper/2409.05591/) (2024)
- [StructRAG: Boosting Knowledge Intensive Reasoning of LLMs via Inference-time Hybrid Information Structurization](https://mlsanity.com/paper/2410.08815/) (2024)
- [AMA-Bench: Evaluating Long-Horizon Memory for Agentic Applications](https://mlsanity.com/paper/2602.22769/) (2026)
- [PlugMem: A Task-Agnostic Plugin Memory Module for LLM Agents](https://mlsanity.com/paper/2603.03296/) (2026)
- [From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms](https://mlsanity.com/paper/2605.06716/) (2026)
- [Synthesizing Multi-Agent Harnesses for Vulnerability Discovery](https://mlsanity.com/paper/2604.20801/) (2026)
- [Choosing How to Remember: Adaptive Memory Structures for LLM Agents](https://mlsanity.com/paper/2602.14038/) (2026)
- [E-mem: Multi-agent based Episodic Context Reconstruction for LLM Agent Memory](https://mlsanity.com/paper/2601.21714/) (2026)
- [S3Mem: Structured Spatiotemporal Scene-Event Memory for Long-Horizon Interactive Question Answering](https://mlsanity.com/paper/2605.28831/) (2026)
- [Diagnosing and Mitigating Compounding Failures in Agentic Persuasion via Taxonomic Strategy Retrieval](https://mlsanity.com/paper/2606.24976/) (2026)

## Built on by (most cited first, 1 of 1)

- [The Missing Complement: State-Conditioned Minimal Sufficient Evidence for Coding Agents](https://mlsanity.com/paper/2609.20050/) (2026)

## Abstract (arXiv)

Long-horizon agents accumulate trajectories spanning hundreds of interleaved reasoning, action, and observation steps, where answering a query may depend on evidence buried far back in the history. External memory stores such trajectories as structured representations, yet each structure provides a distinct and incomplete view. Existing multi-memory systems either read a fixed set of structures for every query, inflating context and introducing noise, or route each query to a single structure, preventing the composition of complementary evidence. A controlled analysis on AMA-Bench shows that the optimal memory configuration is typically neither a single structure nor the full union, but a tailored composition of multiple structural memories that varies with query and task demands. Motivated by these findings, we formulate structure-level dynamic selection: selecting and fusing a query-adaptive subset from a library of specialized memory structures. We propose MESA (a Multi-structure Evidence Selection framework for long-horizon Agent), which builds five complementary structure views of each trajectory and learns from end-to-end answer-level feedback to select and fuse a query-specific subset for a frozen answer model. To learn under this weak supervision, MESA employs harness optimization with prior-guided search and UCB-guided scheduling to balance exploration and exploitation. On AMA-Bench, MESA outperforms the strongest baseline by 8.5% while using 41% fewer evidence tokens than the all-structure alternative.