Weizhe Chen

Palash Chatterjee, Ashutosh Chapagain, Weizhe Chen et al.

The paper introduces DiSProD, an online planner developed for environments with probabilistic transitions in continuous state and action spaces. DiSProD builds a symbolic graph that captures the distribution of future trajectories, conditioned on a given policy, using independence assumptions and approximate propagation of distributions. The symbolic graph provides a differentiable representation of the policy's value, enabling efficient gradient-based optimization for long-horizon search. The propagation of approximate distributions can be seen as an aggregation of many trajectories, making it well-suited for dealing with sparse rewards and stochastic environments. An extensive experimental evaluation compares DiSProD to state-of-the-art planners in discrete-time planning and real-time control of robotic systems. The proposed method improves over existing planners in handling stochastic environments, sensitivity to search depth, sparsity of rewards, and large action spaces. Additional real-world experiments demonstrate that DiSProD can control ground vehicles and surface vessels to successfully navigate around obstacles.

Weizhe Chen, Miao Zhang, Junpeng Jiang et al.

Hybrid attention architectures are becoming an increasingly important paradigm for improving LLM inference efficiency while preserving model quality, making hybrid architecture design a central problem. Existing designs often rely on manual empirical rules or proxy-based selector signals for layer-wise operator allocation. Recent NAS-style systems such as Jet-Nemotron demonstrate the promise of automated hybrid architecture search. However, Jet-Nemotron's PostNAS search stages alone use 200B tokens, making such search pipelines difficult to use as routine methods for hybrid architecture design. We introduce DASH, a fast differentiable search framework for hybrid attention architecture design, which relaxes discrete layer-wise attention operator placement into continuous architecture logits, prepares reusable teacher-aligned linear candidates, and performs architecture-only search with model and operator weights frozen to significantly enhance search efficiency. On Qwen2.5-3B-Instruct, DASH consistently outperforms a comprehensive suite of existing selector-style hybrid attention design baselines, showing that direct differentiable search can discover stronger hybrid architectures. Moreover, DASH achieves stronger RULER performance than released Jet-Nemotron models while remaining competitive on overlapping short-context and general benchmarks. Notably, each DASH search run uses only 12.3M tokens and takes about 20 minutes on a single RTX Pro 6000 GPU, corresponding to merely 0.006% of the PostNAS search tokens reported by Jet-Nemotron. These results suggest that high-quality hybrid attention architectures can be obtained through minutes-level differentiable search, providing a promising direction for hybrid architecture design.

Weizhe Chen, Sven Koenig, Bistra Dilkina

With the explosive influence caused by the success of large language models (LLM) like ChatGPT and GPT-4, there has been an extensive amount of recent work showing that foundation models can be used to solve a large variety of tasks. However, there is very limited work that shares insights on multi-agent planning. Multi-agent planning is different from other domains by combining the difficulty of multi-agent coordination and planning, and making it hard to leverage external tools to facilitate the reasoning needed. In this paper, we focus on the problem of multi-agent path finding (MAPF), which is also known as multi-robot route planning, and study the performance of solving MAPF with LLMs. We first show the motivating success on an empty room map without obstacles, then the failure to plan on the harder room map and maze map of the standard MAPF benchmark. We present our position on why directly solving MAPF with LLMs has not been successful yet, and we use various experiments to support our hypothesis. Based on our results, we discussed how researchers with different backgrounds could help with this problem from different perspectives.

Weizhe Chen, Sven Koenig, Bistra Dilkina

In the past year, large language models (LLMs) have had remarkable success in domains outside the traditional natural language processing, and their capacity is further expanded into the so-called LLM agents when connected with external tools. In all domains, the prompt to the LLMs has been shown to make a big difference in what the LLM would generate and thus affect the performance of the LLM agents. Therefore, automatic prompt engineering (APE) has become an important question for many researchers and users of LLMs. However, previous works in APE rely on a final checker to evaluate the performance of the given prompt -- a requirement that is hard to meet in the case of LLM agents, where intermediate feedback is easier to obtain, and the final evaluation could be expensive, inaccurate, or even missing. In this paper, we propose a novel method, \textsc{RePrompt}, which does a ``gradient descent"-like approach to optimize the step-by-step instructions in the prompts given to LLM agents, based on the chat history obtained from interactions and reflections with LLM agents. By leveraging intermediate feedback, \textsc{RePrompt} can optimize the prompt without the need for a final solution checker. We evaluate our approach on PDDL generation, TravelPlanner, and Meeting Planning to show that our method could generally improve performance for different reasoning tasks.

Weizhe Chen, Sven Koenig, Bistra Dilkina

Cooperative multi-agent reinforcement learning (MARL) has been an increasingly important research topic in the last half-decade because of its great potential for real-world applications. Because of the curse of dimensionality, the popular "centralized training decentralized execution" framework requires a long time in training, yet still cannot converge efficiently. In this paper, we propose a general training framework, MARL-LNS, to algorithmically address these issues by training on alternating subsets of agents using existing deep MARL algorithms as low-level trainers, while not involving any additional parameters to be trained. Based on this framework, we provide three algorithm variants based on the framework: random large neighborhood search (RLNS), batch large neighborhood search (BLNS), and adaptive large neighborhood search (ALNS), which alternate the subsets of agents differently. We test our algorithms on both the StarCraft Multi-Agent Challenge and Google Research Football, showing that our algorithms can automatically reduce at least 10% of training time while reaching the same final skill level as the original algorithm.

Weizhe Chen, Lantao Liu

In many environmental monitoring scenarios, the sampling robot needs to simultaneously explore the environment and exploit features of interest with limited time. We present an anytime multi-objective informative planning method called Pareto Monte Carlo tree search which allows the robot to handle potentially competing objectives such as exploration versus exploitation. The method produces optimized decision solutions for the robot based on its knowledge (estimation) of the environment state, leading to better adaptation to environmental dynamics. We provide algorithmic analysis on the critical tree node selection step and show that the number of times choosing sub-optimal nodes is logarithmically bounded and the search result converges to the optimal choices at a polynomial rate.

Weizhe Chen, Lantao Liu

Informative planning seeks a sequence of actions that guide the robot to collect the most informative data to build a large-scale environmental model or learn a dynamical system. Existing work in informative planning mainly focuses on proposing new planners and applying them to various robotic applications such as environmental monitoring, autonomous exploration, and system identification. The informative planners optimize an objective given by a probabilistic model, e.g., Gaussian process regression (GPR). In practice, the ubiquitous sensing outliers can easily affect the model, resulting in a misleading objective. A straightforward solution is to filter out the outliers in the sensing data stream using an off-the-shelf outlier detector. However, informative samples are also scarce by definition so they might be falsely filtered out. In this paper, we propose a method to enable the robot to re-visit the locations where outliers were sampled besides optimizing the informative planning objective. The robot can collect more samples in the vicinity of outliers and update the outlier detector to reduce the number of false alarms. We achieve this by designing a new objective for the Pareto Monte Carlo tree search (MCTS). We demonstrate that the proposed framework performs better than applying an outlier detector naively.

Zheng Chen, Weizhe Chen, Shi Bai et al.

Autonomous exploration in unknown environments using mobile robots is the pillar of many robotic applications. Existing exploration frameworks either select the nearest geometric frontier or the nearest information-theoretic frontier. However, just because a frontier itself is informative does not necessarily mean that the robot will be in an informative area after reaching that frontier. To fill this gap, we propose to use a multi-objective variant of Monte-Carlo tree search that provides a non-myopic Pareto optimal action sequence leading the robot to a frontier with the greatest extent of unknown area uncovering. We also adopted Bayesian Hilbert Map (BHM) for continuous occupancy mapping and made it more applicable to real-time tasks.