Metraipolis › archive › dc_80817eea8801
Jaxolotl: A Unified High-Performance Benchmark Suite for LTL-Based Multi-Task RL — arXiv:2609.38065
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.38065
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T16:28:05.692Z
- SHA-256 of the archived text
ef44176b9a159f6b71524e93f9875532676cc8aaaf035f2b2606b62898643fda- Stable URL of this page
https://metraipolis.com/documento/dc_80817eea8801
How to cite
Jaxolotl: A Unified High-Performance Benchmark Suite for LTL-Based Multi-Task RL — arXiv:2609.38065 — arXiv (en), https://arxiv.org/abs/2609.38065 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_80817eea8801 (Metraipolis archive dc_80817eea8801, sha256 ef44176b9a159f6b71524e93f9875532676cc8aaaf035f2b2606b62898643fda)
BibTeX
@misc{dc80817eea8801,
title = {Jaxolotl: A Unified High-Performance Benchmark Suite for LTL-Based Multi-Task RL — arXiv:2609.38065},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.38065},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_80817eea8801; sha256 ef44176b9a159f6b71524e93f9875532676cc8aaaf035f2b2606b62898643fda},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38065. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Training agents to follow arbitrary instructions is an important goal of multi-task reinforcement learning (RL). Linear temporal logic (LTL) provides a precise and structured formalism for specifying instructions to agents, and has been successfully adopted for training generalist multi-task policies. However, differences in implementations, task distributions, and evaluation protocols make existing methods difficult to compare, while high computational costs limit the scale and statistical reliability of experiments. We introduce Jaxolotl, a unified high-performance benchmark suite for multi-task LTL-RL to address these concerns. Jaxolotl provides a modular, end-to-end JAX implementation of six representative algorithms and four environments, together with newly curated task suites and a standardised, statistically robust evaluation protocol. By precompiling symbolic task representations into static arrays, Jaxolotl enables fully JIT-compiled training and evaluation, achieving end-to-end speedups of up to $220\times$ and supporting controlled comparisons at substantially greater experimental scale. We use this framework to systematically evaluate existing approaches, revealing complementary strengths and limitations: general methods capable of non-myopic reasoning struggle as the number of propositions grows, while methods with stronger scaling rely on environment-specific assumptions and suffer from myopia. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.38065 — 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 2026-10-07T15:54:54.066Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).