The paper introduces TSBPO, a tree‑structured sampling‑to‑update framework for training interactive language agents. It tackles three intertwined challenges—sampling task and state continuations, assigning delayed feedback, and aggregating token losses—by using a likelihood‑fitted Root/Branch allocator, trajectory and observation‑group advantages, and a clipped token surrogate aggregation. Experiments with a Qwen2.5‑1.5B actor show high success rates on ALFWorld and WebShop benchmarks, demonstrating the effectiveness of the proposed training configuration.
arXiv:2608. 06861v1 Announce Type: new Abstract: Training large language model agents in long-horizon environments requires assigning credit from sparse terminal outcomes to individual actions.
arXiv:2606. 11119v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) is a promising approach for enhancing reasoning and agentic behavior in large language models.
arXiv:2610.17666v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) for multi-turn agents relies on sparse terminal rewards that say little about individual steps....
arXiv:2610.20334v1 Announce Type: cross Abstract: A ReAct-style agent accumulates observations, intermediate results, and failed attempts in a single context. As this context grows, stale and off-pat...
SPO++ is a new reinforcement learning method that eliminates the need for waiting on sibling rollouts by using a persistent prompt-level value estimate. It corrects a mismatch in the original SPO approach by standardizing terminal-outcome advantages under an action-token measure and reorganizes prompt evidence based on the policy event that generated it. Experiments on ALFWorld and Math‑TIR show that SPO++ improves online learning efficiency, with action-token-measure normalization identified as the key component.
arXiv:2610.14737v1 Announce Type: cross Abstract: Training language agents on long-horizon interactive tasks is challenging because sparse terminal rewards provide weak credit signals for intermediat...
A ReAct-style agent accumulates observations, intermediate results, and failed attempts in a single context. As this context grows, stale and off-path information consumes its limited capacity and obs...
The paper introduces CoPE, a method that trains language agents by adding multi‑step action targets to the policy backbone. By supervising each action forecasted from preceding histories, CoPE enables acting and forecasting to co‑evolve without extra inference cost. Across five benchmarks, CoPE outperforms existing RL baselines, achieving up to twice the sample efficiency on ScienceWorld and improving overall performance on realistic applications.
The paper introduces T1, a 122‑billion‑parameter Mixture‑of‑Experts model trained with reinforcement learning to perform long‑horizon terminal tasks such as coding and scientific discovery. T1 operates a real shell in a cloud sandbox, making over 300 tool‑call turns per task and receiving rewards from task‑specific verifiers. The authors detail a training recipe that includes aggressive warm‑starting, TITO construction with drift repair, and rollout‑routing replay, achieving significant performance gains on Terminal‑Bench 2.1 and surpassing GPT‑5.4 and GLM‑5.1 on the Long‑Horizon Terminal Bench.
arXiv:2610.00388v1 Announce Type: cross Abstract: Reinforcement learning enables large language model (LLM) agents to learn multi-step behaviors through interaction with their environments. However,...
Terminal-agent capability depends jointly on model weights and the runtime harness that formats prompts, binds tools, and handles error recovery. Existing harness-model co-evolution approaches improve...
Long-horizon large language model (LLM) agents are typically optimized with sparse terminal outcomes, making fine-grained credit assignment across multi-step interactions difficult. Existing approache...