Learning Hierarchical Procedural Memory for LLM Agents through Bayesian Selection and Contrastive Refinement
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911331637002240 |
|---|---|
| author | Forouzandeh, Saman Peng, Wei Moradi, Parham Yu, Xinghuo Jalili, Mahdi |
| author_facet | Forouzandeh, Saman Peng, Wei Moradi, Parham Yu, Xinghuo Jalili, Mahdi |
| contents | We present MACLA, a framework that decouples reasoning from learning by maintaining a frozen large language model while performing all adaptation in an external hierarchical procedural memory. MACLA extracts reusable procedures from trajectories, tracks reliability via Bayesian posteriors, selects actions through expected-utility scoring, and refines procedures by contrasting successes and failures. Across four benchmarks (ALFWorld, WebShop, TravelPlanner, InterCodeSQL), MACLA achieves 78.1 percent average performance, outperforming all baselines. On ALFWorld unseen tasks, MACLA reaches 90.3 percent with 3.1 percent positive generalization. The system constructs memory in 56 seconds, 2800 times faster than the state-of-the-art LLM parameter-training baseline, compressing 2851 trajectories into 187 procedures. Experimental results demonstrate that structured external memory with Bayesian selection and contrastive refinement enables sample-efficient, interpretable, and continually improving agents without LLM parameter updates. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_18950 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Learning Hierarchical Procedural Memory for LLM Agents through Bayesian Selection and Contrastive Refinement Forouzandeh, Saman Peng, Wei Moradi, Parham Yu, Xinghuo Jalili, Mahdi Machine Learning Artificial Intelligence We present MACLA, a framework that decouples reasoning from learning by maintaining a frozen large language model while performing all adaptation in an external hierarchical procedural memory. MACLA extracts reusable procedures from trajectories, tracks reliability via Bayesian posteriors, selects actions through expected-utility scoring, and refines procedures by contrasting successes and failures. Across four benchmarks (ALFWorld, WebShop, TravelPlanner, InterCodeSQL), MACLA achieves 78.1 percent average performance, outperforming all baselines. On ALFWorld unseen tasks, MACLA reaches 90.3 percent with 3.1 percent positive generalization. The system constructs memory in 56 seconds, 2800 times faster than the state-of-the-art LLM parameter-training baseline, compressing 2851 trajectories into 187 procedures. Experimental results demonstrate that structured external memory with Bayesian selection and contrastive refinement enables sample-efficient, interpretable, and continually improving agents without LLM parameter updates. |
| title | Learning Hierarchical Procedural Memory for LLM Agents through Bayesian Selection and Contrastive Refinement |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2512.18950 |