Digital-Twin Empowered Deep Reinforcement Learning For Site-Specific Radio Resource Management in NextG Wireless Aerial Corridor

Joint base station (BS) association and beam selection in multi-UAV aerial corridors constitutes a challenging radio resource management (RRM) problem. It is driven by high-dimensional action spaces, need for substantial overhead to acquire global channel state information (CSI), rapidly varying propagation channels, and stringent latency requirements. Conventional combinatorial optimization methods, while near-optimal, are computationally prohibitive for real-time operation in such dynamic environments. While learning-based approaches can mitigate computational complexity and CSI overhead, the need for extensive site-specific (SS) datasets for model training remains a key challenge. To address these challenges, we develop a Digital Twin (DT)-enabled two-stage optimization framework that couples physics-based beam gain modeling with DRL for scalable online decision-making. In the first stage, a channel twin (CT) is constructed using a high-fidelity ray-tracing solver with geo-spatial contexts, and network information to capture SS propagation characteristics, and dual annealing algorithm is employed to precompute optimal transmission beam directions. In the second stage, a Multi-Head Proximal Policy Optimization (MH-PPO) agent, equipped with a scalable multi-head actor-critic architecture, is trained on the DT-generated channel dataset to directly map complex channel and beam states to jointly execute UAV-BS-beam association decisions. The proposed PPO agent achieves a 44%-121% improvement over DQN and 249%-807% gain over traditional heuristic based optimization schemes in a dense UAV scenario, while reducing inference latency by several orders of magnitude. These results demonstrate that DT-driven training pipelines can deliver high-performance, low-latency RRM policies tailored to SS deployments suitable for real-time resource management in next-generation aerial corridor networks.

QVLA: Not All Channels Are Equal in Vision-Language-Action Model's Quantization

The advent of Vision-Language-Action (VLA) models represents a significant leap for embodied intelligence, yet their immense computational demands critically hinder deployment on resource-constrained robotic platforms. Intuitively, low-bit quantization is a prevalent and preferred technique for large-scale model compression. However, we find that a systematic analysis of VLA model's quantization is fundamentally lacking. We argue that naively applying uniform-bit quantization from Large Language Models (LLMs) to robotics is flawed, as these methods prioritize passive data fidelity while ignoring how minor action deviations compound into catastrophic task failures. To bridge this gap, we introduce QVLA, the first action-centric quantization framework specifically designed for embodied control. In a sharp departure from the rigid, uniform-bit quantization of LLM-based methods, QVLA introduces a highly granular, channel-wise bit allocation strategy. Its core mechanism is to directly measure the final action-space sensitivity when quantizing each individual channel to various bit-widths. This process yields a precise, per-channel importance metric that guides a global optimization, which elegantly unifies quantization and pruning (0-bit) into a single, cohesive framework. Extensive evaluations on different baselines demonstrate the superiority of our approach. In the LIBERO, the quantization version of OpenVLA-OFT with our method requires only 29.2% of the original model's VRAM while maintaining 98.9% of its original performance and achieving a 1.49x speedup. This translates to a 22.6% performance improvement over the LLM-derived method SmoothQuant. Our work establishes a new, principled foundation for compressing VLA models in robotics, paving the way for deploying powerful, large-scale models on real-world hardware. Code will be released.

Edge-Optimized Vision-Language Models for Underground Infrastructure Assessment

Autonomous inspection of underground infrastructure, such as sewer and culvert systems, is critical to public safety and urban sustainability. Although robotic platforms equipped with visual sensors can efficiently detect structural deficiencies, the automated generation of human-readable summaries from these detections remains a significant challenge, especially on resource-constrained edge devices. This paper presents a novel two-stage pipeline for end-to-end summarization of underground deficiencies, combining our lightweight RAPID-SCAN segmentation model with a fine-tuned Vision-Language Model (VLM) deployed on an edge computing platform. The first stage employs RAPID-SCAN (Resource-Aware Pipeline Inspection and Defect Segmentation using Compact Adaptive Network), achieving 0.834 F1-score with only 0.64M parameters for efficient defect segmentation. The second stage utilizes a fine-tuned Phi-3.5 VLM that generates concise, domain-specific summaries in natural language from the segmentation outputs. We introduce a curated dataset of inspection images with manually verified descriptions for VLM fine-tuning and evaluation. To enable real-time performance, we employ post-training quantization with hardware-specific optimization, achieving significant reductions in model size and inference latency without compromising summarization quality. We deploy and evaluate our complete pipeline on a mobile robotic platform, demonstrating its effectiveness in real-world inspection scenarios. Our results show the potential of edge-deployable integrated AI systems to bridge the gap between automated defect detection and actionable insights for infrastructure maintenance, paving the way for more scalable and autonomous inspection solutions.

MVP-LAM: Learning Action-Centric Latent Action via Cross-Viewpoint Reconstruction

Learning \emph{latent actions} from diverse human videos enables scaling robot learning beyond embodiment-specific robot datasets, and these latent actions have recently been used as pseudo-action labels for vision-language-action (VLA) model pretraining. To make VLA pretraining effective, latent actions should contain information about the underlying agent's actions despite the absence of ground-truth labels. We propose \textbf{M}ulti-\textbf{V}iew\textbf{P}oint \textbf{L}atent \textbf{A}ction \textbf{M}odel (\textbf{MVP-LAM}), which learns discrete latent actions that are highly informative about ground-truth actions from time-synchronized multi-view videos. MVP-LAM trains latent actions with a \emph{cross-viewpoint reconstruction} objective, so that a latent action inferred from one view must explain the future in another view, reducing reliance on viewpoint-specific cues. On Bridge V2, MVP-LAM produces more action-centric latent actions, achieving higher mutual information with ground-truth actions and improved action prediction, including under out-of-distribution evaluation. Finally, pretraining VLAs with MVP-LAM latent actions improves downstream manipulation performance on the SIMPLER and LIBERO-Long benchmarks.

Mitigating Conversational Inertia in Multi-Turn Agents

Large language models excel as few-shot learners when provided with appropriate demonstrations, yet this strength becomes problematic in multiturn agent scenarios, where LLMs erroneously mimic their own previous responses as few-shot examples. Through attention analysis, we identify conversational inertia, a phenomenon where models exhibit strong diagonal attention to previous responses, which is associated with imitation bias that constrains exploration. This reveals a tension when transforming few-shot LLMs into agents: longer context enriches environmental feedback for exploitation, yet also amplifies conversational inertia that undermines exploration. Our key insight is that for identical states, actions generated with longer contexts exhibit stronger inertia than those with shorter contexts, enabling construction of preference pairs without environment rewards. Based on this, we propose Context Preference Learning to calibrate model preferences to favor low-inertia responses over highinertia ones. We further provide context management strategies at inference time to balance exploration and exploitation. Experimental results across eight agentic environments and one deep research scenario validate that our framework reduces conversational inertia and achieves performance improvements.

A Lightweight Library for Energy-Based Joint-Embedding Predictive Architectures

We present EB-JEPA, an open-source library for learning representations and world models using Joint-Embedding Predictive Architectures (JEPAs). JEPAs learn to predict in representation space rather than pixel space, avoiding the pitfalls of generative modeling while capturing semantically meaningful features suitable for downstream tasks. Our library provides modular, self-contained implementations that illustrate how representation learning techniques developed for image-level self-supervised learning can transfer to video, where temporal dynamics add complexity, and ultimately to action-conditioned world models, where the model must additionally learn to predict the effects of control inputs. Each example is designed for single-GPU training within a few hours, making energy-based self-supervised learning accessible for research and education. We provide ablations of JEA components on CIFAR-10. Probing these representations yields 91% accuracy, indicating that the model learns useful features. Extending to video, we include a multi-step prediction example on Moving MNIST that demonstrates how the same principles scale to temporal modeling. Finally, we show how these representations can drive action-conditioned world models, achieving a 97% planning success rate on the Two Rooms navigation task. Comprehensive ablations reveal the critical importance of each regularization component for preventing representation collapse. Code is available at https://github.com/facebookresearch/eb_jepa.

$V_0$: A Generalist Value Model for Any Policy at State Zero

Policy gradient methods rely on a baseline to measure the relative advantage of an action, ensuring the model reinforces behaviors that outperform its current average capability. In the training of Large Language Models (LLMs) using Actor-Critic methods (e.g., PPO), this baseline is typically estimated by a Value Model (Critic) often as large as the policy model itself. However, as the policy continuously evolves, the value model requires expensive, synchronous incremental training to accurately track the shifting capabilities of the policy. To avoid this overhead, Group Relative Policy Optimization (GRPO) eliminates the coupled value model by using the average reward of a group of rollouts as the baseline; yet, this approach necessitates extensive sampling to maintain estimation stability. In this paper, we propose $V_0$, a Generalist Value Model capable of estimating the expected performance of any model on unseen prompts without requiring parameter updates. We reframe value estimation by treating the policy's dynamic capability as an explicit context input; specifically, we leverage a history of instruction-performance pairs to dynamically profile the model, departing from the traditional paradigm that relies on parameter fitting to perceive capability shifts. Focusing on value estimation at State Zero (i.e., the initial prompt, hence $V_0$), our model serves as a critical resource scheduler. During GRPO training, $V_0$ predicts success rates prior to rollout, allowing for efficient sampling budget allocation; during deployment, it functions as a router, dispatching instructions to the most cost-effective and suitable model. Empirical results demonstrate that $V_0$ significantly outperforms heuristic budget allocation and achieves a Pareto-optimal trade-off between performance and cost in LLM routing tasks.

Live or Lie: Action-Aware Capsule Multiple Instance Learning for Risk Assessment in Live Streaming Platforms

Live streaming has become a cornerstone of today's internet, enabling massive real-time social interactions. However, it faces severe risks arising from sparse, coordinated malicious behaviors among multiple participants, which are often concealed within normal activities and challenging to detect timely and accurately. In this work, we provide a pioneering study on risk assessment in live streaming rooms, characterized by weak supervision where only room-level labels are available. We formulate the task as a Multiple Instance Learning (MIL) problem, treating each room as a bag and defining structured user-timeslot capsules as instances. These capsules represent subsequences of user actions within specific time windows, encapsulating localized behavioral patterns. Based on this formulation, we propose AC-MIL, an Action-aware Capsule MIL framework that models both individual behaviors and group-level coordination patterns. AC-MIL captures multi-granular semantics and behavioral cues through a serial and parallel architecture that jointly encodes temporal dynamics and cross-user dependencies. These signals are integrated for robust room-level risk prediction, while also offering interpretable evidence at the behavior segment level. Extensive experiments on large-scale industrial datasets from Douyin demonstrate that AC-MIL significantly outperforms MIL and sequential baselines, establishing new state-of-the-art performance in room-level risk assessment for live streaming. Moreover, AC-MIL provides capsule-level interpretability, enabling identification of risky behavior segments as actionable evidence for intervention. The project page is available at: https://qiaoyran.github.io/AC-MIL/.

Verified Critical Step Optimization for LLM Agents

As large language model agents tackle increasingly complex long-horizon tasks, effective post-training becomes critical. Prior work faces fundamental challenges: outcome-only rewards fail to precisely attribute credit to intermediate steps, estimated step-level rewards introduce systematic noise, and Monte Carlo sampling approaches for step reward estimation incur prohibitive computational cost. Inspired by findings that only a small fraction of high-entropy tokens drive effective RL for reasoning, we propose Critical Step Optimization (CSO), which focuses preference learning on verified critical steps, decision points where alternate actions demonstrably flip task outcomes from failure to success. Crucially, our method starts from failed policy trajectories rather than expert demonstrations, directly targeting the policy model's weaknesses. We use a process reward model (PRM) to identify candidate critical steps, leverage expert models to propose high-quality alternatives, then continue execution from these alternatives using the policy model itself until task completion. Only alternatives that the policy successfully executes to correct outcomes are verified and used as DPO training data, ensuring both quality and policy reachability. This yields fine-grained, verifiable supervision at critical decisions while avoiding trajectory-level coarseness and step-level noise. Experiments on GAIA-Text-103 and XBench-DeepSearch show that CSO achieves 37% and 26% relative improvement over the SFT baseline and substantially outperforms other post-training methods, while requiring supervision at only 16% of trajectory steps. This demonstrates the effectiveness of selective verification-based learning for agent post-training.

Chain-of-Goals Hierarchical Policy for Long-Horizon Offline Goal-Conditioned RL

Offline goal-conditioned reinforcement learning remains challenging for long-horizon tasks. While hierarchical approaches mitigate this issue by decomposing tasks, most existing methods rely on separate high- and low-level networks and generate only a single intermediate subgoal, making them inadequate for complex tasks that require coordinating multiple intermediate decisions. To address this limitation, we draw inspiration from the chain-of-thought paradigm and propose the Chain-of-Goals Hierarchical Policy (CoGHP), a novel framework that reformulates hierarchical decision-making as autoregressive sequence modeling within a unified architecture. Given a state and a final goal, CoGHP autoregressively generates a sequence of latent subgoals followed by the primitive action, where each latent subgoal acts as a reasoning step that conditions subsequent predictions. To implement this efficiently, we pioneer the use of an MLP-Mixer backbone, which supports cross-token communication and captures structural relationships among state, goal, latent subgoals, and action. Across challenging navigation and manipulation benchmarks, CoGHP consistently outperforms strong offline baselines, demonstrating improved performance on long-horizon tasks.