The paper introduces LP‑BTS, a learning‑guided planning framework for mobile charging in large, dynamic action spaces. It uses a graph proposal policy to narrow candidate stops, a value critic to evaluate leaf nodes, and edge‑budgeted PUCT to compare short simulated futures before action selection. Experiments on a 30‑scenario battery‑life benchmark show LP‑BTS achieving the highest survival and alive‑AUC, outperforming domain‑engineered baselines and heuristic policies.
arXiv:2609. 03241v1 Announce Type: cross Abstract: A reasoning model can improve from its own on-policy experience, but this inner loop is fragile: terminal verifiers provide reliable yet sparse supervision, while dense same-model guidance can reinforce false confidence or overconcentrate learning on a narrow solution mode.
arXiv:2609.13845v1 Announce Type: cross Abstract: World models trained with joint-embedding predictive architectures learn compact, structured latent representations from physical interaction, yet pl...
arXiv:2608.30406v1 Announce Type: new Abstract: Goal-conditioned reinforcement learning struggles with long horizons when rewards are sparse. While a planner can provide subgoals to guide a low-level...
arXiv:2608.21946v1 Announce Type: cross Abstract: Reinforcement learning with outcome-based objectives such as GRPO enables LLM-based agents to solve complex, long-horizon tasks, yet the reusable exp...
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:2605. 08732v2 Announce Type: replace-cross Abstract: Modern vision-based world models can represent observations as compact yet expressive latent manifolds, but fast goal-oriented planning in these spaces remains challenging.
arXiv:2608. 14125v1 Announce Type: new Abstract: LeWM is a lightweight visual world model that learns latent dynamics end-to-end from pixels and ranks candidate action sequences by the distance between their predicted endpoints and the goal.
arXiv:2607. 15610v1 Announce Type: cross Abstract: Reinforcement learning (RL) has become a key approach for training LLM agents, yet popular methods such as GRPO/RLOO rely on multiple independently sampled complete trajectories for advantage estimation.
Reinforced Planning with Latent World Models (RP1) is a novel method that learns to evaluate imagined outcomes via a critic and to improve multi‑step plans through an optimizer trained offline on world‑model roll‑outs. It is the first approach to fully learn plan improvement and can be attached to any pretrained latent world model. In experiments on visual navigation, arm reaching, and robotic manipulation, RP1 outperforms hand‑designed search algorithms, achieving near‑perfect success while using far fewer roll‑outs and running up to 67× faster than the strongest alternative.
arXiv:2608. 07809v1 Announce Type: new Abstract: A world model is only useful for physical AI if it changes what the agent does, and only safe if it declines to do so when it is wrong.
The paper proposes CANOPY, a minimalist reinforcement learning protocol that addresses two common pitfalls—signal starvation and policy drift—in outcome‑only RL for long‑horizon interactive tasks. By scaling same‑task exploration, keeping updates on‑policy, and anchoring updates with KL divergence, CANOPY enables a Qwen3‑14B agent to achieve top leaderboard results on the AppWorld coding benchmark without auxiliary supervision or elaborate scaffolding. The approach also improves performance on SWE‑bench for a Qwen3.5‑9B model.
arXiv:2607. 08925v1 Announce Type: new Abstract: Training reinforcement-learning agents directly on physical robots makes every fall costly, since a fall can damage the platform and cannot be undone like a simulator reset; the goal is therefore to minimize falls during training rather than trade them off against return, as constrained Markov decision process (MDP) formulations do.