Chris Xiaoxuan Lu
YOU?
Author Swipe
InverTwin: Solving Inverse Problems via Differentiable Radio Frequency Digital Twin Open
Digital twins (DTs), virtual simulated replicas of physical scenes, are transforming various industries. However, their potential in radio frequency (RF) sensing applications has been limited by the unidirectional nature of conventional RF…
View article: VISC: mmWave Radar Scene Flow Estimation using Pervasive Visual-Inertial Supervision
VISC: mmWave Radar Scene Flow Estimation using Pervasive Visual-Inertial Supervision Open
This work proposes a mmWave radar's scene flow estimation framework supervised by data from a widespread visual-inertial (VI) sensor suite, allowing crowdsourced training data from smart vehicles. Current scene flow estimation methods for …
Fast ECoT: Efficient Embodied Chain-of-Thought via Thoughts Reuse Open
Embodied Chain-of-Thought (ECoT) reasoning enhances vision-language-action (VLA) models by improving performance and interpretability through intermediate reasoning steps. However, its sequential autoregressive token generation introduces …
Risk Controlled Image Retrieval Open
Most image retrieval research prioritizes improving predictive performance, often overlooking situations where the reliability of predictions is equally important. The gap between model performance and reliability requirements highlights t…
The Temporal Trap: Entanglement in Pre-Trained Visual Representations for Visuomotor Policy Learning Open
The integration of pre-trained visual representations (PVRs) has significantly advanced visuomotor policy learning. However, effectively leveraging these models remains a challenge. We identify temporal entanglement as a critical, inherent…
Automating the Search for Artificial Life with Foundation Models Open
With the recent Nobel Prize awarded for radical advances in protein discovery, foundation models (FMs) for exploring large combinatorial spaces promise to revolutionize many scientific fields. Artificial Life (ALife) has not yet integrated…
View article: Recent advances in machine learning for defects detection and prediction in laser cladding process
Recent advances in machine learning for defects detection and prediction in laser cladding process Open
As a fundamental component of artificial intelligence, machine learning has gained considerable prominence within the domain of laser cladding in recent years. By employing algorithms to analyze data, discern patterns and regularities, ren…
NAVIX: Scaling MiniGrid Environments with JAX Open
As Deep Reinforcement Learning (Deep RL) research moves towards solving large-scale worlds, efficient environment simulations become crucial for rapid experimentation. However, most existing environments struggle to scale to high throughpu…
Can Learned Optimization Make Reinforcement Learning Less Difficult? Open
While reinforcement learning (RL) holds great potential for decision making in the real world, it suffers from a number of unique difficulties which often need specific consideration. In particular: it is highly non-stationary; suffers fro…
Behaviour Distillation Open
Dataset distillation aims to condense large datasets into a small number of synthetic examples that can be used as drop-in replacements when training new models. It has applications to interpretability, neural architecture search, privacy,…
Discovering Minimal Reinforcement Learning Environments Open
Reinforcement learning (RL) agents are commonly trained and evaluated in the same environment. In contrast, humans often train in a specialized environment before being evaluated, such as studying a book before taking an exam. The potentia…
Artificial Generational Intelligence: Cultural Accumulation in Reinforcement Learning Open
Cultural accumulation drives the open-ended and diverse progress in capabilities spanning human history. It builds an expanding body of knowledge and skills by combining individual exploration with inter-generational information transmissi…
RadarOcc: Robust 3D Occupancy Prediction with 4D Imaging Radar Open
3D occupancy-based perception pipeline has significantly advanced autonomous driving by capturing detailed scene descriptions and demonstrating strong generalizability across various object categories and shapes. Current methods predominan…
Introduction to the Special Section on Contact-free Smart Sensing in AIoT Open
We are surrounded by a multitude of communicating sensing devices. Furnished with wearables to serve our very needs, we traverse sensor-rich environments of smart cities. Through new open standards and novel protocols this loose collection…
Click to Grasp: Zero-Shot Precise Manipulation via Visual Diffusion Descriptors Open
Precise manipulation that is generalizable across scenes and objects remains a persistent challenge in robotics. Current approaches for this task heavily depend on having a significant number of training instances to handle objects with pr…
ThermoHands: A Benchmark for 3D Hand Pose Estimation from Egocentric Thermal Images Open
Designing egocentric 3D hand pose estimation systems that can perform reliably in complex, real-world scenarios is crucial for downstream applications. Previous approaches using RGB or NIR imagery struggle in challenging conditions: RGB me…
Self-Adapting Large Visual-Language Models to Edge Devices across Visual Modalities Open
Recent advancements in Vision-Language (VL) models have sparked interest in their deployment on edge devices, yet challenges in handling diverse visual modalities, manual annotation, and computational constraints remain. We introduce EdgeV…
Recurrent Reinforcement Learning with Memoroids Open
Memory models such as Recurrent Neural Networks (RNNs) and Transformers address Partially Observable Markov Decision Processes (POMDPs) by mapping trajectories to latent Markov states. Neither model scales particularly well to long sequenc…
Discovering Temporally-Aware Reinforcement Learning Algorithms Open
Recent advancements in meta-learning have enabled the automatic discovery of novel reinforcement learning algorithms parameterized by surrogate objective functions. To improve upon manually designed algorithms, the parameterization of this…
View article: Analysing the Sample Complexity of Opponent Shaping
Analysing the Sample Complexity of Opponent Shaping Open
Learning in general-sum games often yields collectively sub-optimal results. Addressing this, opponent shaping (OS) methods actively guide the learning processes of other agents, empirically leading to improved individual and group perform…
View article: The Danger Of Arrogance: Welfare Equilibra As A Solution To Stackelberg Self-Play In Non-Coincidental Games
The Danger Of Arrogance: Welfare Equilibra As A Solution To Stackelberg Self-Play In Non-Coincidental Games Open
The increasing prevalence of multi-agent learning systems in society necessitates understanding how to learn effective and safe policies in general-sum multi-agent environments against a variety of opponents, including self-play. General-s…
View article: Orientation-Aware 3D SLAM in Alternating Magnetic Field from Powerlines
Orientation-Aware 3D SLAM in Alternating Magnetic Field from Powerlines Open
Identifying new sensing modalities for indoor localization is an interest of research. This paper studies powerline-induced alternating magnetic field (AMF) that fills the indoor space for the orientation-aware three-dimensional (3D) simul…
View article: Scaling Opponent Shaping to High Dimensional Games
Scaling Opponent Shaping to High Dimensional Games Open
In multi-agent settings with mixed incentives, methods developed for zero-sum games have been shown to lead to detrimental outcomes. To address this issue, opponent shaping (OS) methods explicitly learn to influence the learning dynamics o…
View article: Leading the Pack: N-player Opponent Shaping
Leading the Pack: N-player Opponent Shaping Open
Reinforcement learning solutions have great success in the 2-player general sum setting. In this setting, the paradigm of Opponent Shaping (OS), in which agents account for the learning of their co-players, has led to agents which are able…
Differentiable Radio Frequency Ray Tracing for Millimeter-Wave Sensing Open
Millimeter wave (mmWave) sensing is an emerging technology with applications in 3D object characterization and environment mapping. However, realizing precise 3D reconstruction from sparse mmWave signals remains challenging. Existing metho…
Multimodal Indoor Localization Using Crowdsourced Radio Maps Open
Indoor Positioning Systems (IPS) traditionally rely on odometry and building infrastructures like WiFi, often supplemented by building floor plans for increased accuracy. However, the limitation of floor plans in terms of availability and …