Manipulation, locomotion, sim-to-real transfer and autonomous driving: learning systems that have to survive physics.
arXiv:2609.08153v1 Announce Type: new Abstract: Generative diffusion models have emerged as a class of powerful techniques for various imaging applications, including but not limited to synthesis, re...
The article titled "The convergent laboratory: when AI reasoning, autonomous experiments, high performance and quantum computing reshape chemistry" discusses insights from the TPC26 conference, where leaders from academia, national laboratories, and industry examined how AI, autonomous agents, self-driving labs, high‑performance computing, and quantum computing converge to accelerate materials science discovery. It presents firsthand experiences from researchers at the forefront of these technologies and argues that their simultaneous maturation marks a tipping point for transformative advances and productive disruption in chemical sciences.
arXiv:2609.09137v1 Announce Type: new Abstract: Robotic Process Automation (RPA) is widely used to reduce administrative burden in United States hospitals, yet an estimated 30-50% of RPA initiatives...
The paper presents a deep active inference framework for real‑world robotic navigation that combines a diffusion policy with a multiple‑timescale recurrent state‑space model. The diffusion policy generates diverse candidate actions, while the state‑space model predicts long‑horizon outcomes, allowing the system to select actions that minimize expected free energy. Experiments show higher success rates and fewer collisions, especially in exploration‑heavy scenarios, demonstrating the effectiveness of this unified exploration and goal‑directed approach.
The paper introduces a progressive training strategy for embodied vision‑language models aimed at reducing spatio‑temporal hallucinations. It first creates a Chain‑of‑Thought dataset that breaks complex reasoning into detailed spatiotemporal steps, then uses supervised pre‑training on this dataset followed by fine‑tuning with weakly‑labeled data. Experiments show the method improves backbone accuracy and narrows the forward‑backward performance gap from over 70% to 6.53%, indicating stronger dynamic reasoning and fewer temporal biases.
arXiv:2605.12957v2 Announce Type: replace Abstract: Recent developments in generative models and large-scale datasets have substantially advanced 3D world generation, facilitating a broad range of do...
The paper "Inferring the Unspoken: Aligning Embodied Agents with Implicit Preferences" addresses the challenge of natural-language instructions that omit details needed for embodied action. It introduces the Preference-based Planning (PbP) benchmark, comprising 5,000 evaluation groups and 290 preferences across three levels, to systematically evaluate agents’ ability to infer latent user preferences from a few demonstrations. The authors propose the two-stage Inferring the Unspoken (InTU) framework, which first verbalizes inferred preferences from multimodal demonstrations and then generates action plans conditioned on that explicit representation, showing that explicit verbalization improves alignment and robustness compared to direct end-to-end planning.
PlannerForge is a unified LLM‑agent framework that covers the entire scenario‑based testing pipeline for autonomous driving systems, from scenario generation to ADS assessment, and adds ADS enhancement and benchmarking stages. It was evaluated with ten off‑the‑shelf LLMs across all tasks and five prompt conditions, achieving best‑per‑task scores between 0.88 and 1.00 and matching commercial APIs with open‑source models such as Qwen3.6:35B. The end‑to‑end chaining retains 83% of seed queries for commercial backends and 78% for open‑source, outperforming existing tools like Scenario Factory 2.0 and BM25 in natural‑language generation, attribute realization, and physically valid edits. whyItMatters":"PlannerForge demonstrates that a single LLM‑based system can streamline and improve the fragmented scenario‑based testing workflow for autonomous driving, achieving high performance without domain‑specific fine‑tuning."
RoboCousin is an extensible simulation platform that transforms user-provided observations into reusable assets, scenes, and expert trajectories for bimanual robotic manipulation. It converts object images into simulation-ready models with visual, collision, semantic, and physical metadata, automatically generates grasp candidates, and builds digital cousins that vary objects, backgrounds, layouts, and language instructions while preserving task-relevant affordances. The platform supports both tabletop and room-level scene construction, and the authors release RoboCousin-OBD with over 3,000 annotated objects and 50 backgrounds, generating more than one million expert trajectories across 50 tasks, with simulation and real-robot experiments demonstrating comparable annotation quality and effective sim-to-real transfer.
The paper introduces a multi‑modal late‑fusion perception pipeline for object detection and tracking in autonomous racing. It combines independent detections from cameras, LiDARs, and RADARs to produce timely and robust state estimates of surrounding vehicles. The tracking framework compensates for detection delays and incorporates vehicle dynamics and track layout knowledge, and its effectiveness is confirmed through real‑world experiments in diverse critical scenarios.
PV-WM is a history‑only world model that jointly predicts pedestrian root motion, 15‑joint articulation, and vehicle kinematic states in a synchronized heterogeneous state. It uses recurrent updates to generate pedestrian and vehicle motion chunks, reconstructing vehicle boxes from predicted center, heading, and observed extent, and recomputes pedestrian‑vehicle geometry after each transition. Compared to a one‑shot predictor, PV‑WM reduces Root ADE by 12.7% and MPJPE by 14.8%, and across 824 Waymo contexts it lowers Root ADE by 5.2%, MPJPE by 7.6%, P‑V distance error by 11.9%, and oriented‑box closest‑approach error by 5.8%, while using 57.1% fewer parameters, 96.5% fewer FLOPs, and 25.5% lower p95 latency.
ComplicitSplat is a novel black‑box attack that leverages 3D Gaussian Splatting (3DGS) shading to create viewpoint‑specific camouflage, embedding adversarial content into scene objects that is only visible from certain angles. The method does not require access to model architecture or weights and can successfully fool a range of popular object detectors—including single‑stage, multi‑stage, and transformer‑based models—on both real‑world physical objects and synthetic scenes. This demonstrates that downstream models using 3DGS are vulnerable to adversarial manipulation.
arXiv:2609.09945v1 Announce Type: cross Abstract: Machine learning-based credit scoring is increasingly central to Peer-to-Peer (P2P) lending, yet its resilience to adversarial manipulation, where ap...
arXiv:2609.07047v1 Announce Type: cross Abstract: Robotic manipulation often requires acting on information that is no longer visible, yet Vision-Language-Action policies are usually evaluated when t...
arXiv:2609.10522v1 Announce Type: cross Abstract: Foundation vision-language models (VLMs) exhibit broad intelligence about the world, yet translating this intelligence into robot control remains cha...
arXiv:2609.09210v1 Announce Type: cross Abstract: Teleoperated demonstrations are often multimodal even when the underlying dynamics are nearly deterministic given the executed action. We argue that...
arXiv:2609.09881v1 Announce Type: new Abstract: Semantic segmentation for autonomous driving requires reliable detection of vulnerable road users (VRUs) despite heavy class imbalance. We introduce CL...
arXiv:2609.07111v1 Announce Type: cross Abstract: Quadruped robot locomotion policies are often trained using reinforcement learning, which in turn relies heavily on hand-crafted reward functions. De...
arXiv:2609.10464v1 Announce Type: cross Abstract: Joint-Embedding Predictive Architecture (JEPA) world models learn a compact latent representation of the world that supports prediction and planning,...
RevalExo is a new benchmark for locomotion mode recognition that focuses on functional daily activities performed by older adults and clinical cohorts. It includes 27 participants from three groups—healthy older adults, stroke survivors, and older adults with probable sarcopenia—recorded with lower-body IMUs and, for a subset, synchronized egocentric video. The dataset offers 10.1 hours of frame‑level annotations across 11 locomotion modes, and the authors evaluate unimodal, multimodal, cross‑population, and cross‑modal recognition challenges, finding that sensor fusion improves performance but transitions and generalization remain difficult.