arXiv subjects

Lili Su

Publications and source records attributed to Lili Su.

At least 19 recordsLinked to original sources

arXiv2026

Resilience Beyond Stationary Client Unavailability: Unlocking Efficient and Unbiased Federated Learning

Due to resource constraints or external and internal uncertainties, clients in real-world federated learning systems are often intermittently available edge devices. In highly dynamic environments, the parameter server lacks prior real-time knowledge of clients' availability, making it challenging to adapt traditional federated learning algorithms to be resilient to uncertainties in client availability. If not carefully addressed, complex client availability can introduce significant bias, potentially harming the performance of the trained model. Most prior work either fails to account for non-stationary client availability dynamics or demands significant memory and computational overhead. This paper aims to develop efficient federated learning algorithms that are provably resilient to heterogeneous and non-stationary stochastic client availability. We propose FedSWE, which admits novel algorithmic structures to (i) compensate for missed computations, (ii) stabilize and diffuse the global updates over rounds, and (iii) evenly mix the local updates through implicit gossiping, despite being agnostic to non-stationary dynamics. Compared with the standard FedAvg, FedSWE introduces light additional memory and computation overhead. We show that FedSWE converges to a stationary point of non-convex objectives while achieving the desired linear speedup property in certain special cases. We corroborate our analysis with numerical experiments over diversified client unavailability dynamics on real-world data sets.

cs.LG↗

arXiv2026

Rethinking Language's Role in Efficient VLA for Autonomous Vehicles: Toward Smarter, Trustworthy Driving

Vision-Language-Action (VLA) models are reshaping autonomous driving (AD) by unifying perception, reasoning, and control through language, enabling semantic grounding, interpretable decisions, and better long-tail generalization. But language is expensive onboard: latency and memory budgets are tight, and autoregressive decoding is inherently sequential. This work reframes the central question as when and where language should act at inference, since inference cost recurs at every deployed frame while training cost is paid once. We introduce the Language Residue taxonomy to organize methods by their inference-time use of language: train-time-only supervision (L1), latent non-textual reasoning (L2), conditional invocation (L3), and full per-frame generation (L4). We review representative methods and tag each across five deployment axes (latency, parameters, memory, FLOPs, tokens), analyzing them on major open- and closed-loop driving benchmarks (e.g., nuScenes, NAVSIM, Bench2Drive). We further trace how efficient methods from NLP/LLM are adapted in AD, identifying the constraints and motivations driving these adaptations. A continuously updated repository will be available at Github.

cs.RO↗

arXiv2026

Beyond Non-IID: Learner--Client Distribution Mismatch in Federated Learning

Federated learning systems are increasingly deployed to facilitate collaborative model training across a heterogeneous client population. Existing practice mostly implicitly assumes that the aggregated client data distribution is representative of the learner's target distribution or that learning from all available clients is uniformly beneficial for the learner distribution. However, such an assumption often does not hold in reality. Traditional client selection strategies in FL literature largely overlook such misalignment, while most existing work on multi-source transfer learning either requires direct access to local data or uses one-shot model/feature aggregation. In this paper, we take the initiative to understand and mitigate the impacts of such learner-client population misalignment. In particular, we consider the practical setting where the learner keeps a small proxy dataset. We observe that client contributions vary significantly across training rounds, and traditional technology is insufficient to identify beneficial sources under multi-source transfer diversity. Then, we propose a dynamic, influence-aware client selection framework that estimates each client's potential utility to the learner's optimization objective using proxy influence signals on a learner-specific proxy set. Via using leave-one-out evaluations, we prioritize the most informative sources of knowledge while controlling the negative impacts of statistical noise and data heterogeneity. Experiments on CIFAR-10 under heterogeneous data partitions demonstrate that our approach consistently outperforms static and dynamic baselines, achieving faster convergence and higher accuracy.

cs.LG↗

arXiv2026

Post-Training in End-to-End Autonomous Driving

End-to-end models that map multimodal inputs directly to future trajectories/maneuvers have emerged as an increasingly prominent research paradigm in autonomous driving. This class of models includes both Vision-Language-Action models and trajectory-generative planners. Unlike classic machine learning applications, autonomous vehicles operate in safety-critical and interaction-intensive environments where traditional open-loop imitation of expert demonstrations is not sufficient to ensure reliability. In particular, small execution errors can accumulate over time, while recovery behaviors are scarce in training data. In addition, long-horizon objectives such as safety and driving comfort are not captured by pointwise labels either. These limitations have motivated a shift toward post-training techniques, which further refine driving policies beyond pure imitation. This survey presents a unified view of post-training for autonomous driving by defining its scope and organizing the existing literature into four major families based on the form of supervision they use. For each family, we discuss its capabilities, limitations, and open challenges. We aim to facilitate a systematic understanding of this emerging area and stimulate future research on reliable and efficient post-training for autonomous driving.A collection of related papers is available at https://github.com/RYNing/Awesome-Post-Training-In-Autonomous-Driving-Papers.

cs.CV↗

arXiv2026

Collaborative Yet Personalized Policy Training: Single-Timescale Federated Actor-Critic

Despite the popularity of the actor-critic method and the practical needs of collaborative policy training, existing works typically either overlook environmental heterogeneity or give up personalization altogether by training a single shared policy across all agents. We consider a federated actor-critic framework in which agents share a common linear subspace representation while maintaining personalized local policy components, and agents iteratively estimate the common subspace, local critic heads, and local policies (i.e., actors). Under canonical single-timescale updates with Markovian sampling, we establish finite-time convergence via a novel joint linear approximation framework. Specifically, we show that the critic error converges to zero at the rate of $\tilde{\mathcal{O}}(1/((1-\gamma)^4\sqrt{TK}))$, and the policy gradient norm converges to zero at the rate of $\tilde{\mathcal{O}}(1/((1-\gamma)^6\sqrt{TK}))$, where $T$ is the number of rounds, $K$ is the number of agents, and $\gamma\in (0,1)$ is the discount factor. These results demonstrate linear speedup with respect to the number of agents $K$, despite heterogeneous Markovian trajectories under distinct transition kernels and coupled learning dynamics. To address these challenges, we develop a new perturbation analysis for the projected subspace updates and QR decomposition steps, together with conditional mixing arguments for heterogeneous Markovian noise. Furthermore, to handle the additional complications induced by policy updates and temporal dependence, we establish fine-grained characterizations of the discrepancies between function evaluations under Markovian sampling and under temporally frozen policies. Experiments instantiate the framework within PPO on federated \texttt{Hopper-v5} action-map heterogeneity, showing gains over Single PPO and FedAvg PPO and downstream transfer from the learned shared trunk.

cs.LG↗

arXiv2026

SpanVLA: Efficient Action Bridging and Learning from Negative-Recovery Samples for Vision-Language-Action Model

Vision-Language-Action (VLA) models offer a promising autonomous driving paradigm for leveraging world knowledge and reasoning capabilities, especially in long-tail scenarios. However, existing VLA models often struggle with the high latency in action generation using an autoregressive generation framework and exhibit limited robustness. In this paper, we propose SpanVLA, a novel end-to-end autonomous driving framework, integrating an autoregressive reasoning and a flow-matching action expert. First, SpanVLA introduces an efficient bridge to leverage the vision and reasoning guidance of VLM to efficiently plan future trajectories using a flow-matching policy conditioned on historical trajectory initialization, which significantly reduces inference time. Second, to further improve the performance and robustness of the SpanVLA model, we propose a GRPO-based post-training method to enable the VLA model not only to learn from positive driving samples but also to learn how to avoid the typical negative behaviors and learn recovery behaviors. We further introduce mReasoning, a new real-world driving reasoning dataset, focusing on complex, reasoning-demanding scenarios and negative-recovery samples. Extensive experiments on the NAVSIM (v1 and v2) demonstrate the competitive performance of the SpanVLA model. Additionally, the qualitative results across diverse scenarios highlight the planning performance and robustness of our model.

cs.CV↗

arXiv2026

Similarity-Aware Mixture-of-Experts for Data-Efficient Continual Learning

Machine learning models often need to adapt to new data after deployment due to structured or unstructured real-world dynamics. The Continual Learning (CL) framework enables continuous model adaptation, but most existing approaches either assume each task contains sufficiently many data samples or that the learning tasks are non-overlapping. In this paper, we address the more general setting where each task may have a limited dataset, and tasks may overlap in an arbitrary manner without a priori knowledge. This general setting is substantially more challenging for two reasons. On the one hand, data scarcity necessitates effective contextualization of general knowledge and efficient knowledge transfer across tasks. On the other hand, unstructured task overlapping can easily result in negative knowledge transfer. To address the above challenges, we propose an adaptive mixture-of-experts (MoE) framework over pre-trained models that progressively establishes similarity awareness among tasks. Our design contains two innovative algorithmic components: incremental global pooling and instance-wise prompt masking. The former mitigates prompt association noise through gradual prompt introduction over time. The latter decomposes incoming task samples into those aligning with current prompts (in-distribution) and those requiring new prompts (out-of-distribution). Together, our design strategically leverages potential task overlaps while actively preventing negative mutual interference in the presence of per-task data scarcity. Experiments across varying data volumes and inter-task similarity show that our method enhances sample efficiency and is broadly applicable.

cs.LG↗

arXiv2026

Latent Dynamics-Aware OOD Monitoring for Trajectory Prediction with Provable Guarantees

In safety-critical Cyber-Physical Systems (CPS), trajectory prediction guides downstream planning and control. Deep learning models forecast well on validation data, but their reliability drops in out-of-distribution (OOD) scenarios driven by environmental uncertainty or rare traffic behaviors [1, 2]. Such failures are often silent: forecasts stay spatially plausible while accuracy collapses, and reported uncertainty does not rise [3]. Detection is hard because traffic conditions and interaction patterns keep evolving, yet the safety-critical nature of autonomous driving (AD) demands formal guarantees on detection delay and false-alarm rate. Following [4], we reframe OOD monitoring as quickest changepoint detection (QCD), a principled statistical framework with well-established theory. We find that the evolution of prediction errors on in-distribution (ID) data is well modeled by a Hidden Markov Model (HMM). Building on this, we extend a recent cumulative Maximum Mean Discrepancy approach to our setting. The method needs no detailed prior knowledge of the post-change distribution, yet admits provable delay and false-alarm guarantees. On three real-world driving datasets, it reduces detection delay while staying robust to heavy-tailed distributions and unknown post-change conditions.

cs.RO↗

arXiv2026

Personalized Multi-Agent Average Reward TD-Learning via Joint Linear Approximation

We study personalized multi-agent average reward TD learning, in which a collection of agents interacts with different environments and jointly learns their respective value functions. We focus on the setting where there exists a shared linear representation, and the agents' optimal weights collectively lie in an unknown linear subspace. Inspired by the recent success of personalized federated learning (PFL), we study the convergence of cooperative single-timescale TD learning in which agents iteratively estimate the common subspace and local heads. We showed that this decomposition can filter out conflicting signals, effectively mitigating the negative impacts of ``misaligned'' signals, and achieving linear speedup. The main technical challenges lie in the heterogeneity, the Markovian sampling, and their intricate interplay in shaping error evolutions. Specifically, not only are the error dynamics of multiple variables closely interconnected, but there is also no direct contraction for the principal angle distance between the optimal subspace and the estimated subspace. We hope our analytical techniques can be useful to inspire research on deeper exploration into leveraging common structures. Experiments are provided to show the benefits of learning via a shared structure to the more general control problem.

cs.LG↗

arXiv2026

On the Power of Source Screening for Learning Shared Feature Extractors

Learning with shared representation is widely recognized as an effective way to separate commonalities from heterogeneity across various heterogeneous sources. Most existing work includes all related data sources via simultaneously training a common feature extractor and source-specific heads. It is well understood that data sources with low relevance or poor quality may hinder representation learning. In this paper, we further dive into the question of which data sources should be learned jointly by focusing on the traditionally deemed ``good'' collection of sources, in which individual sources have similar relevance and qualities with respect to the true underlying common structure. Towards tractability, we focus on the linear setting where sources share a low-dimensional subspace. We find that source screening can play a central role in statistically optimal subspace estimation. We show that, for a broad class of problem instances, training on a carefully selected subset of sources suffices to achieve minimax optimality, even when a substantial portion of data is discarded. We formalize the notion of an informative subpopulation, develop algorithms and practical heuristics for identifying such subsets, and validate their effectiveness through both theoretical analysis and empirical evaluations on synthetic and real-world datasets.

cs.LG↗

arXiv2025

A Trajectory Generator for High-Density Traffic and Diverse Agent-Interaction Scenarios

Accurate trajectory prediction is fundamental to autonomous driving, as it underpins safe motion planning and collision avoidance in complex environments. However, existing benchmark datasets suffer from a pronounced long-tail distribution problem, with most samples drawn from low-density scenarios and simple straight-driving behaviors. This underrepresentation of high-density scenarios and safety critical maneuvers such as lane changes, overtaking and turning is an obstacle to model generalization and leads to overly optimistic evaluations. To address these challenges, we propose a novel trajectory generation framework that simultaneously enhances scenarios density and enriches behavioral diversity. Specifically, our approach converts continuous road environments into a structured grid representation that supports fine-grained path planning, explicit conflict detection, and multi-agent coordination. Built upon this representation, we introduce behavior-aware generation mechanisms that combine rule-based decision triggers with Frenet-based trajectory smoothing and dynamic feasibility constraints. This design allows us to synthesize realistic high-density scenarios and rare behaviors with complex interactions that are often missing in real data. Extensive experiments on the large-scale Argoverse 1 and Argoverse 2 datasets demonstrate that our method significantly improves both agent density and behavior diversity, while preserving motion realism and scenario-level safety. Our synthetic data also benefits downstream trajectory prediction models and enhances performance in challenging high-density scenarios.

cs.RO↗

arXiv2025

Adaptive Multi-Mode Out-of-Distribution Detection for Trajectory Prediction in Autonomous Vehicles

Trustworthy trajectory prediction grounds autonomous vehicle (AV) safety, yet deployed models inevitably face out-of-distribution (OOD) scenes. Prior AV OOD detection targets perception, but planners act on predicted futures rather than raw scenes, so erroneous forecasts can slip past frame-level checks and corrupt control. We therefore tackle OOD detection at the trajectory-prediction level. Our analysis of real-world benchmarks reveals that prediction errors are often multi-modal, exhibiting distinct low- and high-error modes that evolve with open-world driving context. Observing this, we propose Mode-Aware CUSUM, which explicitly models multiple error modes while retaining the efficiency and general compatibility of classical CUSUM. By dynamically identifying the active error mode and adapting detection thresholds, our method enables robust monitoring across heterogeneous conditions. Experiments on large-scale trajectory benchmarks demonstrate consistent reductions in detection delay and false alarms.

cs.CV↗

arXiv2025

Trajectory Prediction Meets Large Language Models: A Survey

Recent advances in large language models (LLMs) have sparked growing interest in integrating language-driven techniques into trajectory prediction. By leveraging their semantic and reasoning capabilities, LLMs are reshaping how autonomous systems perceive, model, and predict trajectories. This survey provides a comprehensive overview of this emerging field, categorizing recent work into five directions: (1) Trajectory prediction via language modeling paradigms, (2) Direct trajectory prediction with pretrained language models, (3) Language-guided scene understanding for trajectory prediction, (4) Language-driven data generation for trajectory prediction, (5) Language-based reasoning and interpretability for trajectory prediction. For each, we analyze representative methods, highlight core design choices, and identify open challenges. This survey bridges natural language processing and trajectory prediction, offering a unified perspective on how language can enrich trajectory prediction.

cs.CL↗

arXiv2024

Probe-Me-Not: Protecting Pre-trained Encoders from Malicious Probing

Adapting pre-trained deep learning models to customized tasks has become a popular choice for developers to cope with limited computational resources and data volume. More specifically, probing--training a downstream head on a pre-trained encoder--has been widely adopted in transfer learning, which helps to prevent overfitting and catastrophic forgetting. However, such generalizability of pre-trained encoders raises concerns about the potential misuse of probing for harmful intentions, such as discriminatory speculation and warfare applications. In this work, we introduce EncoderLock, a novel applicability authorization method designed to protect pre-trained encoders from malicious probing, i.e., yielding poor performance on specified prohibited domains while maintaining their utility in authorized ones. Achieving this balance is challenging because of the opposite optimization objectives and the variety of downstream heads that adversaries can utilize adaptively. To address these challenges, EncoderLock employs two techniques: domain-aware weight selection and updating to restrict applications on prohibited domains/tasks, and self-challenging training scheme that iteratively strengthens resistance against any potential downstream classifiers that adversaries may apply. Moreover, recognizing the potential lack of data from prohibited domains in practical scenarios, we introduce three EncoderLock variants with different levels of data accessibility: supervised (prohibited domain data with labels), unsupervised (prohibited domain data without labels), and zero-shot (no data or labels available). We verify EncoderLock's effectiveness and practicality with a real-world pre-trained Vision Transformer (ViT) encoder from Facebook. These results underscore the valuable contributions EncoderLock brings to the development of responsible AI.

cs.CR↗

arXiv2024

Personalized Federated Learning via Feature Distribution Adaptation

Federated learning (FL) is a distributed learning framework that leverages commonalities between distributed client datasets to train a global model. Under heterogeneous clients, however, FL can fail to produce stable training results. Personalized federated learning (PFL) seeks to address this by learning individual models tailored to each client. One approach is to decompose model training into shared representation learning and personalized classifier training. Nonetheless, previous works struggle to navigate the bias-variance trade-off in classifier learning, relying solely on limited local datasets or introducing costly techniques to improve generalization. In this work, we frame representation learning as a generative modeling task, where representations are trained with a classifier based on the global feature distribution. We then propose an algorithm, pFedFDA, that efficiently generates personalized models by adapting global generative classifiers to their local feature distributions. Through extensive computer vision benchmarks, we demonstrate that our method can adjust to complex distribution shifts with significant improvements over current state-of-the-art in data-scarce settings.

cs.LG↗

arXiv2024

Non-transferable Pruning

Pretrained Deep Neural Networks (DNNs), developed from extensive datasets to integrate multifaceted knowledge, are increasingly recognized as valuable intellectual property (IP). To safeguard these models against IP infringement, strategies for ownership verification and usage authorization have emerged. Unlike most existing IP protection strategies that concentrate on restricting direct access to the model, our study addresses an extended DNN IP issue: applicability authorization, aiming to prevent the misuse of learned knowledge, particularly in unauthorized transfer learning scenarios. We propose Non-Transferable Pruning (NTP), a novel IP protection method that leverages model pruning to control a pretrained DNN's transferability to unauthorized data domains. Selective pruning can deliberately diminish a model's suitability on unauthorized domains, even with full fine-tuning. Specifically, our framework employs the alternating direction method of multipliers (ADMM) for optimizing both the model sparsity and an innovative non-transferable learning loss, augmented with Fisher space discriminative regularization, to constrain the model's generalizability to the target dataset. We also propose a novel effective metric to measure the model non-transferability: Area Under the Sample-wise Learning Curve (SLC-AUC). This metric facilitates consideration of full fine-tuning across various sample sizes. Experimental results demonstrate that NTP significantly surpasses the state-of-the-art non-transferable learning methods, with an average SLC-AUC at $-0.54$ across diverse pairs of source and target domains, indicating that models trained with NTP do not suit for transfer learning to unauthorized target domains. The efficacy of NTP is validated in both supervised and self-supervised learning contexts, confirming its applicability in real-world scenarios.

cs.LG↗

arXiv2024

Efficient Federated Learning against Heterogeneous and Non-stationary Client Unavailability

Addressing intermittent client availability is critical for the real-world deployment of federated learning algorithms. Most prior work either overlooks the potential non-stationarity in the dynamics of client unavailability or requires substantial memory/computation overhead. We study federated learning in the presence of heterogeneous and non-stationary client availability, which may occur when the deployment environments are uncertain, or the clients are mobile. The impacts of heterogeneity and non-stationarity on client unavailability can be significant, as we illustrate using FedAvg, the most widely adopted federated learning algorithm. We propose FedAPM, which includes novel algorithmic structures that (i) compensate for missed computations due to unavailability with only $O(1)$ additional memory and computation with respect to standard FedAvg, and (ii) evenly diffuse local updates within the federated learning system through implicit gossiping, despite being agnostic to non-stationary dynamics. We show that FedAPM converges to a stationary point of even non-convex objectives while achieving the desired linear speedup property. We corroborate our analysis with numerical experiments over diversified client unavailability dynamics on real-world data sets.

cs.DC↗

arXiv2024

Reactive Multi-Robot Navigation in Outdoor Environments Through Uncertainty-Aware Active Learning of Human Preference Landscape

Compared with single robots, Multi-Robot Systems (MRS) can perform missions more efficiently due to the presence of multiple members with diverse capabilities. However, deploying an MRS in wide real-world environments is still challenging due to uncertain and various obstacles (e.g., building clusters and trees). With a limited understanding of environmental uncertainty on performance, an MRS cannot flexibly adjust its behaviors (e.g., teaming, load sharing, trajectory planning) to ensure both environment adaptation and task accomplishments. In this work, a novel joint preference landscape learning and behavior adjusting framework (PLBA) is designed. PLBA efficiently integrates real-time human guidance to MRS coordination and utilizes Sparse Variational Gaussian Processes with Varying Output Noise to quickly assess human preferences by leveraging spatial correlations between environment characteristics. An optimization-based behavior-adjusting method then safely adapts MRS behaviors to environments. To validate PLBA's effectiveness in MRS behavior adaption, a flood disaster search and rescue task was designed. 20 human users provided 1764 feedback based on human preferences obtained from MRS behaviors related to "task quality", "task progress", "robot safety". The prediction accuracy and adaptation speed results show the effectiveness of PLBA in preference learning and MRS behavior adaption.

cs.RO↗