Metraipolis › archive › dc_b35673d09c07

EDGE: Experience-Distillation for Guided Exploration in Agentic Reinforcement Learning — arXiv:2608.21946

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.21946
License (as archived): CC0-1.0
Archived: 2026-10-18T07:42:31.625Z
SHA-256 of the archived text: 100ffa0fb4a202c632ec965704054cbd625b7c4a4afc9296f9627391aed01cdf
Stable URL of this page: https://metraipolis.com/documento/dc_b35673d09c07

How to cite

EDGE: Experience-Distillation for Guided Exploration in Agentic Reinforcement Learning — arXiv:2608.21946 — arXiv (en), https://arxiv.org/abs/2608.21946 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b35673d09c07 (Metraipolis archive dc_b35673d09c07, sha256 100ffa0fb4a202c632ec965704054cbd625b7c4a4afc9296f9627391aed01cdf)

BibTeX

@misc{dcb35673d09c07,
  title = {EDGE: Experience-Distillation for Guided Exploration in Agentic Reinforcement Learning — arXiv:2608.21946},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.21946},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b35673d09c07; sha256 100ffa0fb4a202c632ec965704054cbd625b7c4a4afc9296f9627391aed01cdf},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.21946. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reinforcement learning with outcome-based objectives such as GRPO enables LLM-based agents to solve complex, long-horizon tasks, yet the reusable exploration patterns embedded in interaction trajectories are largely discarded after a single policy update. Existing experience-augmented approaches retrieve historical guidance at inference time, but they apply experiences without accounting for the policy's evolving capability and create persistent dependencies on external retrieval. We propose EDGE (Experience-Distillation for Guided Exploration), a framework that treats retrieved experiences as temporary training-time scaffolds and progressively internalizes their benefits into the parametric policy. Concretely, EDGE partitions each rollout group into experience-conditioned and experience-free trajectories to estimate and admit only positive marginal gains without extra sampling, then distills the induced behavior into the base policy via a reverse-KL objective on its own empirical support. A co-evolutionary experience bank further synthesizes guidance from emerging failure modes and prunes obsolete entries as the policy evolves. Across embodied, web, and search-based QA tasks, EDGE improves over strong RL baselines by up to 12.5 points and remains effective without inference-time scaffolds or a proprietary reflector. The code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-25 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.21946 — 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 2026-10-18T07:37:52.843Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).