We study the problem active sequential hypothesis testing, also known as pure exploration: given a new task, the learner adaptively collects data from the environment to efficiently determine an underlying correct hypothesis. A classical instance of this problem is the task of identifying the best arm in a multi-armed bandit problem (a.k.a. BAI, Best-Arm Identification), where actions index hypotheses. Another important case is generalized search, a problem of determining the correct label through a sequence of strategically selected queries that indirectly reveal information about the label. In this work, we introduce In-Context Pure Exploration (ICPE), which meta-trains Transformers to map observation histories to query actions and a predicted hypothesis, yielding a model that transfers in-context. At inference time, ICPE actively gathers evidence on new tasks and infers the true hypothesis without parameter updates. Across deterministic, stochastic, and structured benchmarks, including BAI and generalized search, ICPE is competitive with adaptive baselines while requiring no explicit modeling of information structure. Our results support Transformers as practical architectures for general sequential testing.

2048 is a stochastic single-player game involving 16 cells on a 4 by 4 grid, where a player chooses a direction among up, down, left, and right to obtain a score by merging two tiles with the same number located in neighboring cells along the chosen direction. This paper presents that a variant 2048-4x3 12 cells on a 4 by 3 board, one row smaller than the original, has been strongly solved. In this variant, the expected score achieved by an optimal strategy is about $50724.26$ for the most common initial states: ones with two tiles of number 2. The numbers of reachable states and afterstates are identified to be $1,152,817,492,752$ and $739,648,886,170$, respectively. The key technique is to partition state space by the sum of tile numbers on a board, which we call the age of a state. An age is invariant between a state and its successive afterstate after any valid action and is increased two or four by stochastic response from the environment. Therefore, we can partition state space by ages and enumerate all (after)states of an age depending only on states with the recent ages. Similarly, we can identify (after)state values by going along with ages in decreasing order.

Distortion Risk Measures (DRMs) capture risk preferences in decision-making and serve as general criteria for managing uncertainty. This paper proposes gradient descent algorithms for DRM optimization based on two dual representations: the Distortion-Measure (DM) form and Quantile-Function (QF) form. The DM-form employs a three-timescale algorithm to track quantiles, compute their gradients, and update decision variables, utilizing the Generalized Likelihood Ratio and kernel-based density estimation. The QF-form provides a simpler two-timescale approach that avoids the need for complex quantile gradient estimation. A hybrid form integrates both approaches, applying the DM-form for robust performance around distortion function jumps and the QF-form for efficiency in smooth regions. Proofs of strong convergence and convergence rates for the proposed algorithms are provided. In particular, the DM-form achieves an optimal rate of $O(k^{-4/7})$, while the QF-form attains a faster rate of $O(k^{-2/3})$. Numerical experiments confirm their effectiveness and demonstrate substantial improvements over baselines in robust portfolio selection tasks. The method's scalability is further illustrated through integration into deep reinforcement learning. Specifically, a DRM-based Proximal Policy Optimization algorithm is developed and applied to multi-echelon dynamic inventory management, showcasing its practical applicability.

The real world is inherently non-stationary, with ever-changing factors, such as weather conditions and traffic flows, making it challenging for agents to adapt to varying environmental dynamics. Non-Stationary Reinforcement Learning (NSRL) addresses this challenge by training agents to adapt rapidly to sequences of distinct Markov Decision Processes (MDPs). However, existing NSRL approaches often focus on tasks with regularly evolving patterns, leading to limited adaptability in highly dynamic settings. Inspired by the success of Wavelet analysis in time series modeling, specifically its ability to capture signal trends at multiple scales, we propose WISDOM to leverage wavelet-domain predictive task representations to enhance NSRL. WISDOM captures these multi-scale features in evolving MDP sequences by transforming task representation sequences into the wavelet domain, where wavelet coefficients represent both global trends and fine-grained variations of non-stationary changes. In addition to the auto-regressive modeling commonly employed in time series forecasting, we devise a wavelet temporal difference (TD) update operator to enhance tracking and prediction of MDP evolution. We theoretically prove the convergence of this operator and demonstrate policy improvement with wavelet task representations. Experiments on diverse benchmarks show that WISDOM significantly outperforms existing baselines in both sample efficiency and asymptotic performance, demonstrating its remarkable adaptability in complex environments characterized by non-stationary and stochastically evolving tasks.

Recent large reasoning models (LRMs) driven by reinforcement learning algorithms (e.g., GRPO) have achieved remarkable performance on challenging reasoning tasks. However, these models suffer from overthinking, generating unnecessarily long and redundant reasoning even for simple questions, which substantially increases computational cost and response latency. While existing methods incorporate length rewards to GRPO to promote concise reasoning, they incur significant performance degradation. We identify the root cause: when rewards for correct but long rollouts are penalized, GRPO's group-relative advantage function can assign them negative advantages, actively discouraging valid reasoning. To overcome this, we propose Decoupled Reward Policy Optimization (DRPO), a novel framework that decouples the length-based learning signal of correct rollouts from incorrect ones. DRPO ensures that reward signals for correct rollouts are normalized solely within the positive group, shielding them from interference by negative samples. The DRPO's objective is grounded in integrating an optimized positive data distribution, which maximizes length-based rewards under a KL regularization, into a discriminative objective. We derive a closed-form solution for this distribution, enabling efficient computation of the objective and its gradients using only on-policy data and importance weighting. Of independent interest, this formulation is general and can incorporate other preference rewards of positive data beyond length. Experiments on mathematical reasoning tasks demonstrate DRPO's significant superiority over six efficient reasoning baselines. Notably, with a 1.5B model, our method achieves 77\% length reduction with only 1.1\% performance loss on simple questions like GSM8k dataset, while the follow-up baseline sacrifices 4.3\% for 68\% length reduction.

Large Language Models (LLMs) show strong reasoning abilities, often amplified by Chain-of-Thought (CoT) prompting and reinforcement learning (RL). Although RL algorithms can substantially improve reasoning, they struggle to expand reasoning boundaries because they learn from their own reasoning trajectories rather than acquiring external knowledge. Supervised fine-tuning (SFT) offers complementary benefits but typically requires large-scale data and risks overfit-ting. Recent attempts to combine SFT and RL face three main challenges: data inefficiency, algorithm-specific designs, and catastrophic forgetting. We propose a plug-and-play framework that dynamically integrates SFT into RL by selecting challenging examples for SFT. This approach reduces SFT data requirements and remains agnostic to the choice of RL or SFT algorithm. To mitigate catastrophic forgetting of RL-acquired skills during SFT, we select high-entropy tokens for loss calculation and freeze parameters identified as critical for RL. Our method achieves state-of-the-art (SoT A) reasoning performance using only 1.5% of the SFT data and 20.4% of the RL data used by prior SoT A, providing an efficient and plug-and-play solution for combining SFT and RL in reasoning post-training. Recent Large Language Models (LLMs) has shown reasoning capability (Jaech et al., 2024; Guo et al., 2025; Anthropic, 2025). The reasoning capability are highly dependent on the use of the Chain-of-Thought (CoT) thinking pattern trained by supervise fine-tuning (SFT) or reinforcement learning (RL). Although popular RL algorithms such as PPO (Schulman et al., 2017), GRPO (Guo et al., 2025), and DAPO (Y u et al., 2025) are promising in multiple reasoning tasks, recent studies argue that RL training does not truly extend a model's reasoning boundaries.

Performative reinforcement learning is an emerging dynamical decision making framework, which extends reinforcement learning to the common applications where the agent's policy can change the environmental dynamics. Existing works on performative reinforcement learning only aim at a performatively stable (PS) policy that maximizes an approximate value function. However, there is a provably positive constant gap between the PS policy and the desired performatively optimal (PO) policy that maximizes the original value function. In contrast, this work proposes a zeroth-order Frank-Wolfe algorithm (0-FW) algorithm with a zeroth-order approximation of the performative policy gradient in the Frank-Wolfe framework, and obtains \textbf{the first polynomial-time convergence to the desired PO} policy under the standard regularizer dominance condition. For the convergence analysis, we prove two important properties of the nonconvex value function. First, when the policy regularizer dominates the environmental shift, the value function satisfies a certain gradient dominance property, so that any stationary point (not PS) of the value function is a desired PO. Second, though the value function has unbounded gradient, we prove that all the sufficiently stationary points lie in a convex and compact policy subspace $Π_Δ$, where the policy value has a constant lower bound $Δ>0$ and thus the gradient becomes bounded and Lipschitz continuous. Experimental results also demonstrate that our 0-FW algorithm is more effective than the existing algorithms in finding the desired PO policy.

Effective cross-functional coordination is essential for enhancing firm-wide profitability, particularly in the face of growing organizational complexity and scale. Recent advances in artificial intelligence, especially in reinforcement learning (RL), offer promising avenues to address this fundamental challenge. This paper proposes a unified multi-agent RL framework tailored for joint optimization across distinct functional modules, exemplified via coordinating inventory replenishment and personalized product recommendation. We first develop an integrated theoretical model to capture the intricate interplay between these functions and derive analytical benchmarks that characterize optimal coordination. The analysis reveals synchronized adjustment patterns across products and over time, highlighting the importance of coordinated decision-making. Leveraging these insights, we design a novel multi-timescale multi-agent RL architecture that decomposes policy components according to departmental functions and assigns distinct learning speeds based on task complexity and responsiveness. Our model-free multi-agent design improves scalability and deployment flexibility, while multi-timescale updates enhance convergence stability and adaptability across heterogeneous decisions. We further establish the asymptotic convergence of the proposed algorithm. Extensive simulation experiments demonstrate that the proposed approach significantly improves profitability relative to siloed decision-making frameworks, while the behaviors of the trained RL agents align closely with the managerial insights from our theoretical model. Taken together, this work provides a scalable, interpretable RL-based solution to enable effective cross-functional coordination in complex business settings.

Shared micromobility systems, such as electric scooters and bikes, have gained widespread popularity as sustainable alternatives to traditional transportation modes. However, these systems face persistent challenges due to spatio-temporal demand fluctuations, often resulting in a mismatch between vehicle supply and user demand. Existing shared micromobility vehicle scheduling methods typically redistribute vehicles once or twice per day, which makes them vulnerable to performance degradation under atypical conditions. In this work, we design to augment existing micromobility scheduling methods by integrating a small number of autonomous shared micromobility vehicles (ASMVs), which possess self-rebalancing capabilities to dynamically adapt to real-time demand. Specifically, we introduce SMART, a hierarchical reinforcement learning framework that jointly optimizes high-level initial deployment and low-level real-time rebalancing for ASMVs. We evaluate our framework based on real-world e-scooter usage data from Chicago. Our experiment results show that our framework is highly effective and possesses strong generalization capability, allowing it to seamlessly integrate with existing vehicle scheduling methods and significantly enhance overall micromobility service performance.

Recent advances in large language models (LLMs) have sparked growing interest in building generalist agents that can learn through online interactions. However, applying reinforcement learning (RL) to train LLM agents in multi-turn, multi-task settings remains challenging due to lack of scalable infrastructure and stable training algorithms. In this work, we present the AgentRL framework for scalable multi-turn, multi-task agentic RL training. On the infrastructure side, AgentRL features a fully-asynchronous generation-training pipeline for efficient multi-turn RL. To support heterogeneous environment development in multi-task RL, we design a unified function-call based API interface, containerized environment development, and a centralized controller. On the algorithm side, we propose cross-policy sampling to encourage model exploration in multi-turn settings and task advantage normalization to stabilize multi-task training. Experiments show that AgentRL, trained on open LLMs across five agentic tasks, significantly outperforms GPT-5, Clause-Sonnet-4, DeepSeek-R1, and other open-source LLM agents. Multi-task training with AgentRL matches the best results among all task-specific models. AgentRL is open-sourced at https://github.com/THUDM/AgentRL. The algorithm and framework are adopted in building \textsc{\href{https://autoglm.zhipuai.cn}{AutoGLM}}.