Metraipolis › archive › dc_281b790751b3
What Does Multi-Turn Reinforcement Learning Really Improve in LLM Agents? — arXiv:2611.06483
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.06483
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-05T18:18:59.436Z
- SHA-256 of the archived text
666f969befc8078886a73549f000bd49e0da45866a87752151e09488fa493aa4- Stable URL of this page
https://metraipolis.com/documento/dc_281b790751b3
How to cite
What Does Multi-Turn Reinforcement Learning Really Improve in LLM Agents? — arXiv:2611.06483 — arXiv (en), https://arxiv.org/abs/2611.06483 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_281b790751b3 (Metraipolis archive dc_281b790751b3, sha256 666f969befc8078886a73549f000bd49e0da45866a87752151e09488fa493aa4)
BibTeX
@misc{dc281b790751b3,
title = {What Does Multi-Turn Reinforcement Learning Really Improve in LLM Agents? — arXiv:2611.06483},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.06483},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_281b790751b3; sha256 666f969befc8078886a73549f000bd49e0da45866a87752151e09488fa493aa4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.06483. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
While multi-turn agentic reinforcement learning (RL) excels on long-horizon tasks, its performance gains remain highly sensitive to interaction constraints. Evaluating across sampling breadth $K$ and interaction depth $T$ via $Pass@(K,T)$, we show that trajectory-level RL advantages depend on sufficient turn budgets, diminishing significantly when budgets tighten. Besides, agentic RL does not uniformly elevate single-step decision quality; rather, its strength concentrates in error recovery and long-horizon resilience. To mitigate the deployment risks caused by this budget dependence, we provide a preliminary discussion on the benefits and limitations of process supervision signals and introduce BudgetMix RL as an alternative that varies interaction budgets during training. Experiments demonstrate that BudgetMix RL achieves robust performance across diverse execution settings. Project Page: this https URL Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2611.06483 — 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 2026-11-05T18:13:03.955Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).