Longchao Da

The sim-to-real problem remains a critical challenge in the real-world application of reinforcement learning (RL). The conventional sim-to-real methods heavily rely on resource-intensive re-training of the policy network to adapt to new domains, which limits the flexibility of the deployment of RL policies in ever-changing environments. Inspired by human locomotion, where individuals adjust their gait to new surface conditions without relearning the skill of walking, we introduce Latent Adaptation of Foundation Policies (Found-adapt), a framework that decouples this problem into skill acquisition and environment adaptation. Our method first pretrains a foundation policy on unlabeled offline trajectories from the source simulator, capturing diverse long-horizon behaviors as reusable skills. At deployment, instead of retraining the policy, we perform efficient latent space adaptation: a small amount of target-domain data is collected to refine a latent representation through an adapter network that incorporates parameter efficient alignment, which produces a task-ready controller under various system dynamics. This adaptation occurs entirely in the latent space, avoiding costly policy optimization while enabling robust transfer. Empirical results across multiple locomotion tasks and dynamic variations demonstrate that our method significantly reduces the sim-to-real gap. Further sensitivity analysis provides interesting insights into the requirements for data quality and applicable situations. These findings highlight how foundation policies with latent adaptation could serve as a general and efficient paradigm for real-world RL deployment.

Being able to anticipate the motion of surrounding agents is essential for the safe operation of autonomous driving systems in dynamic situations. While various methods have been proposed for trajectory prediction, the current evaluation practices still rely on error-based metrics (e.g., ADE, FDE), which reveal the accuracy from a post-hoc view but ignore the actual effect the predictor brings to the self-driving vehicles (SDVs), especially in complex interactive scenarios: a high-quality predictor not only chases accuracy, but should also captures all possible directions a neighbor agent might move, to support the SDVs’ cautious decision-making. Given that the existing metrics hardly account for this standard, in our work, we propose a comprehensive pipeline that adaptively evaluates the predictor’s performance by two dimensions: accuracy and diversity. Based on the criticality of the driving scenario, these two dimensions are dynamically combined and result in a final score for the predictor’s performance. Extensive experiments on a closed-loop benchmark using a real-world dataset show that our pipeline yields a more reasonable evaluation than traditional metrics by better reflecting the correlation of the predictors’ evaluation with the autonomous vehicles’ driving performance. This evaluation pipeline shows a robust way to select a predictor that potentially contributes most to the SDV’s driving performance. The code and dataset are provided in the supplementary material.

Large Language Models (LLMs) excel in text generation, reasoning, and decision-making, enabling their adoption in high-stakes domains such as healthcare, law, and transportation. However, their reliability is a major concern, as they often produce plausible but incorrect responses. Uncertainty quantification (UQ) enhances trustworthiness by estimating confidence in outputs, enabling risk mitigation and selective prediction. However, traditional UQ methods struggle with LLMs due to computational constraints and decoding inconsistencies. Moreover, LLMs introduce unique uncertainty sources, such as input ambiguity, reasoning path divergence, and decoding stochasticity, that extend beyond classical aleatoric and epistemic uncertainty. To address this, we introduce a new taxonomy that categorizes UQ methods based on computational efficiency and uncertainty dimensions (input, reasoning, parameter, and prediction uncertainty). We evaluate existing techniques, assess their real-world applicability, and identify open challenges, emphasizing the need for scalable, interpretable, and robust UQ approaches to enhance LLM reliability.

Large Language Models (LLMs) have demonstrated remarkable capabilities across various tasks due to large training datasets and powerful transformer architecture. However, the reliability of responses from LLMs remains a question. Uncertainty quantification (UQ) of LLMs is crucial for ensuring their reliability, especially in areas such as healthcare, finance, and decision-making. Existing UQ methods primarily focus on semantic similarity, overlooking the deeper knowledge dimensions embedded in responses. We introduce a multi-dimensional UQ framework that integrates semantic and knowledge-aware similarity analysis. By generating multiple responses and leveraging auxiliary LLMs to extract implicit knowledge, we construct separate similarity matrices and apply tensor decomposition to derive a comprehensive uncertainty representation. This approach disentangles overlapping information from both semantic and knowledge dimensions, capturing both semantic variations and factual consistency, leading to more accurate UQ. Our empirical evaluations demonstrate that our method outperforms existing techniques in identifying uncertain responses, offering a more robust framework for enhancing LLM reliability in high-stakes applications.

Deep Reinforcement Learning (RL) has been explored and verified to be effective in solving decision-making tasks in various domains, such as robotics, transportation, recommender systems, etc. It learns from the interaction with environments and updates the policy using the collected experience. However, due to the limited real-world data and unbearable consequences of taking detrimental actions, the learning of RL policy is mainly restricted within the simulators. This practice guarantees safety in learning but introduces an inevitable sim-to-real gap in terms of deployment, thus causing degraded performance and risks in execution. There are attempts to solve the sim-to-real problems from different domains with various techniques, especially in the era with emerging techniques such as large foundations or language models that have cast light on the sim-to-real. This survey paper, to the best of our knowledge, is the first taxonomy that formally frames the sim-to-real techniques from key elements of the Markov Decision Process (State, Action, Transition, and Reward). Based on the framework, we cover comprehensive literature from the classic to the most advanced methods including the sim-to-real techniques empowered by foundation models, and we also discuss the specialties that are worth attention in different domains of sim-to-real problems. Then we summarize the formal evaluation process of sim-to-real performance with accessible code or benchmarks. The challenges and opportunities are also presented to encourage future exploration of this direction. We are actively maintaining a to include the most up-to-date sim-to-real research outcomes to help the researchers in their work.

The Large language models (LLMs) have showcased superior capabilities in sophisticated tasks across various domains, stemming from basic question-answer (QA), they are nowadays used as decision assistants or explainers for unfamiliar content. However, they are not always correct due to the data sparsity in specific domain corpus, or the model’s hallucination problems. Given this, how much should we trust the responses from LLMs? This paper studies the question in two aspects, first, we find measuring directly on raw responses from LLMs is easy to be vague due to the lack of context, second this paper identifies the oversight in existing work regarding the neglect of directional relationships between sentences. To tackle the challenges, we first provide a claim-based response augmentation, by completing the fine-grained claims in raw response, and augmenting into a more explicit description, it removes the randomness, improves the completeness, and provides a response set that robustly represents the knowledge of LLMs. Then, second, methodologically we present a way to evaluate the uncertainty that captures the directional instability between sentences by constructing a directional graph using entailment probabilities, and conducting Random Walk Laplacian given the asymmetric property in the directed graph, then the uncertainty is aggregated by the derived eigenvalues from the Laplacian process. We conducted extensive experiments and empirically demonstrated the superiority of our proposed solutions.

Medical imaging is crucial for diagnosing a patient’s health condition, and accurate segmentation of these images is essential for isolating regions of interest to ensure precise diagnosis and treatment planning. Existing methods primarily rely on bounding boxes or point-based prompts, while few have explored text-related prompts, despite clinicians often describing their observations and instructions in natural language. To address this gap, we first propose a RAG-based free-form text prompt generator, that leverages the domain corpus to generate diverse and realistic descriptions. Then, we introduce FLanS, a novel medical image segmentation model that handles various free-form text prompts, including professional anatomy-informed queries, anatomy-agnostic position-driven queries, and anatomy-agnostic size-driven queries. Additionally, our model also incorporates a symmetry-aware canonicalization module to ensure consistent, accurate segmentations across varying scan orientations and reduce confusion between the anatomical position of an organ and its appearance in the scan. FLanS is trained on a large-scale dataset of over 100k medical images from 7 public datasets. Comprehensive experiments demonstrate the model’s superior language understanding and segmentation precision, along with a deep comprehension of the relationship between them, outperforming SOTA baselines on both in-domain and out-of-domain datasets.

Neural networks that can produce accurate, input-conditional uncertainty representations are critical for real-world applications. Recent progress on heteroscedastic continuous regression has shown great promise for calibrated uncertainty quantification on complex tasks, like image regression. However, when these methods are applied to discrete regression tasks, such as crowd counting, ratings prediction, or inventory estimation, they tend to produce predictive distributions with numerous pathologies. We propose to address these issues by training a neural network to output the parameters of a Double Poisson distribution, which we call the Deep Double Poisson Network (DDPN). In contrast to existing methods that are trained to minimize Gaussian negative log likelihood (NLL), DDPNs produce a proper probability mass function over discrete output. Additionally, DDPNs naturally model under-, over-, and equi-dispersion, unlike networks trained with the more rigid Poisson and Negative Binomial parameterizations. We show DDPNs 1) vastly outperform existing discrete models; 2) meet or exceed the accuracy and flexibility of networks trained with Gaussian NLL; 3) produce proper predictive distributions over discrete counts; and 4) exhibit superior out-of-distribution detection. DDPNs can easily be applied to a variety of count regression datasets including tabular, image, point cloud, and text data.

The Large Language Model has become an important assistant for many of human’s decision-making. However, a side note almost comes along with every single LLM:‘LLMs can make mistakes. Be careful with important info’. This reveals the fact that not all of the information from LLMs is trustworthy, and people still need to judge by themselves. Yet, the hallucination is so powerful that a made-up conclusion could even come with a seemingly plausible reason, which brings intricate challenges and a trust crisis among users. This paper proposes EvidenceChat, a framework that tackles this issue in a retrieval-augmented fashion, specifically, when the user uploads a material document, an indexed vector space will be constructed based on text embeddings, which helps construct a retrieval-augmented agent, enhancing the agent’s responses with additional knowledge beyond its training corpus. Then, we propose a three-step evidential retrieval method to accurately locate the evidence of a language agent’s conclusion in the source context: Chain-of-Thought (CoT) logic generation, top-k relevant chunk vector matching, and a self-reflection-based precise sentence identification. We demonstrate that our method improves the existing models’ performance in terms of identifying the exact evidence in a free-form context, providing a reliable way to examine the resources of LLM’s conclusion and help with the judgment of the trustworthiness. The dataset and code will be released.

The integration of autonomous vehicles into urban traffic has great potential to improve efficiency by reducing congestion and optimizing traffic flow systematically. In this paper, we introduce CoMAL (Collaborative Multi-Agent LLMs), a framework designed to address the mixed-autonomy traffic problem by collaboration among autonomous vehicles to optimize traffic flow. CoMAL is built upon large language models, operating in an interactive traffic simulation environment. It utilizes a Perception Module to observe surrounding agents and a Memory Module to store strategies for each agent. The overall workflow includes a Collaboration Module that encourages autonomous vehicles to discuss the effective strategy and allocate roles, a reasoning engine to determine optimal behaviors based on assigned roles, and an Execution Module that controls vehicle actions using a hybrid approach combining rule-based models. Experimental results demonstrate that CoMAL achieves superior performance on the Flow benchmark. Additionally, we evaluate the impact of different language models and compare our framework with reinforcement learning approaches. It highlights the strong cooperative capability of LLM agents and presents a promising solution to the mixed-autonomy traffic challenge

The privacy concerns associated with the use of Large Language Models (LLMs) have grown recently with the development of LLMs such as ChatGPT. Differential Privacy (DP) techniques are explored in existing work to mitigate their privacy risks at the cost of generalization degradation. Our paper reveals that the flatness of DP-trained models’ loss landscape plays an essential role in the trade-off between their privacy and generalization. We further propose a holistic framework to enforce appropriate weight flatness, which substantially improves model generalization with competitive privacy preservation. It innovates from three coarse-to-grained levels, including perturbation-aware min-max optimization on model weights within a layer, flatness-guided sparse prefix-tuning on weights across layers, and weight knowledge distillation between DP & non-DP weights copies. Comprehensive experiments of both black-box and white-box scenarios are conducted to demonstrate the effectiveness of our proposal in enhancing generalization and maintaining DP characteristics. For instance, on text classification dataset QNLI, DP-Flat achieves similar performance with non-private full fine-tuning but with DP guarantee under privacy budget, and even better performance given higher privacy budgets. Codes are provided in the supplement.

Graph regression is a fundamental task and has received increasing attention in a wide range of graph learning tasks. However, the inference process is often not interpretable. Most existing explanation techniques are limited to understanding GNN behaviors in classification tasks. In this work, we seek an explanation to interpret the graph regression models (XAIG-R). We show that existing methods overlook the distribution shifting and continuously ordered decision boundary, which hinders them away from being applied in the regression tasks. To address these challenges, we propose a novel objective based on the information bottleneck theory and introduce a new mix-up framework, which could support various GNNs in a model-agnostic manner. We further present a contrastive learning strategy to tackle the continuously ordered labels in regression task. To empirically verify the effectiveness of the proposed method, we introduce three benchmark datasets and a real-life dataset for evaluation. Extensive experiments show the effectiveness of the proposed method in interpreting GNN models in regression tasks.

Numerous solutions are proposed for the Traffic Signal Con- trol (TSC) tasks aiming to provide efficient transportation and alleviate traffic congestion. Recently, promising results have been attained by Reinforcement Learning (RL) methods through trial and error in simulators, bringing confidence in solving cities’ congestion problems. However, performance gaps still exist when simulator-trained policies are deployed to the real world. This issue is mainly introduced by the sys- tem dynamic difference between the training simulators and the real-world environments. In this work, we leverage the knowledge of Large Language Models (LLMs) to understand and profile the system dynamics by a prompt-based grounded action transformation to bridge the performance gap. Specifi- cally, this paper exploits the pre-trained LLM’s inference abil- ity to understand how traffic dynamics change with weather conditions, traffic states, and road types. Being aware of the changes, the policies’ action is taken and grounded based on realistic dynamics, thus helping the agent learn a more realistic policy. We conduct experiments on four different scenarios to show the effectiveness of the proposed PromptGAT’s ability to mitigate the performance gap of reinforcement learning from simulation to reality (sim-to-real).

In practice, it is essential to compare and rank candidate policies offline before real-world deployment for safety and reliability. Prior work seeks to solve this offline policy ranking (OPR) problem through value-based methods, such as Off-policy evaluation (OPE). However, they fail to analyze special case performance (e.g., worst or best cases), due to the lack of holistic characterization of policies’ performance. It is even more difficult to estimate precise policy values when the reward is not fully accessible under sparse settings. In this paper, we present Probabilistic Offline Policy Ranking (POPR), a framework to address OPR problems by leveraging expert data to characterize the probability of a candidate policy behaving like experts, and approximating its entire performance posterior distribution to help with ranking. POPR does not rely on value estimation, and the derived performance posterior can be used to distinguish candidates in worst-, best-, and average-cases. To estimate the posterior, we propose POPR-EABC, an Energy-based Approximate Bayesian Computation (ABC) method conducting likelihood-free inference. POPR-EABC reduces the heuristic nature of ABC by a smooth energy function, and improves the sampling efficiency by a pseudo-likelihood. We empirically demonstrate that POPR-EABC is adequate for evaluating policies in both discrete and continuous action spaces across various experiment environments, and facilitates probabilistic comparisons of candidate policies before deployment.

Transportation has greatly benefited the cities’ development in the modern civilization process. Intelligent transportation, leveraging advanced computer algorithms, could further increase people’s daily commuting efficiency. However, intelligent transportation, as a cross-discipline, often requires practitioners to comprehend complicated algorithms and obscure neural networks, bringing a challenge for the advanced techniques to be trusted and deployed in practical industries. Recognizing the expressiveness of the pre-trained large language models, especially the potential of being augmented with abilities to understand and execute intricate commands, we introduce Open-TI. Serving as a bridge to mitigate the industry-academic gap, Open-TI is an innovative model targeting the goal of Turing Indistinguishable Traffic Intelligence, it is augmented with the capability to harness external traffic analysis packages based on existing conversations. Marking its distinction, Open-TI is the first method capable of conducting exhaustive traffic analysis from scratch—spanning from map data acquisition to the eventual execution in complex simulations. Besides, Open-TI is able to conduct task-specific embodiment like training and adapting the traffic signal control policies (TSC), explore demand optimizations, etc. Furthermore, we explored the viability of LLMs directly serving as control agents, by understanding the expected intentions from Open-TI, we designed an agent-to-agent communication mode to support Open-TI conveying messages to ChatZero (control agent), and then the control agent would choose from the action space to proceed the execution. We eventually provide the formal implementation structure, and the open-ended design invites further community-driven enhancements. A demo video is provided at: https://youtu.be/pZ4-5PXz9Xs.

Traffic signal control (TSC) is a complex and important task that affects the daily lives of millions of people. Reinforcement Learning (RL) has shown promising results in optimizing traffic signal control, but current RL-based TSC methods are mainly trained in simulation and suffer from the performance gap between simulation and the real world. In this paper, we propose a simulation-to-real-world (sim-to-real) transfer approach called UGAT, which transfers a learned policy trained from a simulated environment to a real-world environment by dynamically transforming actions in the simulation with uncertainty to mitigate the domain gap of transition dynamics. We evaluate our method on a simulated traffic environment and show that it significantly improves the performance of the transferred RL policy in the real world.

Traffic simulation is an essential tool for transportation infrastructure planning, intelligent traffic control policy learning, and traffic flow analysis. Its effectiveness relies heavily on the realism of the simulators used. Traditional traffic simulators, such as SUMO and CityFlow, are often limited by their reliance on rule-based models with hyperparameters that oversimplify driving behaviors, resulting in unrealistic simulations. To enhance realism, some simulators have provided Application Programming Interfaces (APIs) to interact with Machine Learning (ML) models, which learn from observed data and offer more sophisticated driving behavior models. However, this approach faces challenges in scalability and time efficiency as vehicle numbers increase. Addressing these limitations, we introduce CityFlowER, an advancement over the existing CityFlow simulator, designed for efficient and realistic city-wide traffic simulation. CityFlowER innovatively pre-embeds ML models within the simulator, eliminating the need for external API interactions and enabling faster data computation. This approach allows for a blend of rule-based and ML behavior models for individual vehicles, offering unparalleled flexibility and efficiency, particularly in large-scale simulations. We provide detailed comparisons with existing simulators, implementation insights, and comprehensive experiments to demonstrate CityFlowER’s superiority in terms of realism, efficiency, and adaptability.

Heatwaves pose significant health risks, particularly due to prolonged exposure to high summer temperatures. Vulnerable groups, especially pedestrians and cyclists on sun-exposed sidewalks, motivate the development of a route planning method that incorporates somatosensory temperature effects through shade ratio consideration. This paper is the first to introduce a pipeline that utilizes segmentation foundation models to extract shaded areas from high-resolution satellite images. These areas are then integrated into a multi-layered road map, enabling users to customize routes based on a balance between distance and shade exposure, thereby enhancing comfort and health during outdoor activities. Specifically, we construct a graph-based representation of the road map, where links indicate connectivity and are updated with shade ratio data for dynamic route planning. This system is already implemented online, with a video demonstration, and will be specifically adapted to assist travelers during the 2024 Olympic Games in Paris.