Ford has rehired over 300 veteran quality inspectors after AI-driven quality checks underperformed compared to experienced human engineers. The company acknowledges the limitations of AI without proper training from experts, and notes a return to the top of an industry quality study.

WebBrain is a free, MIT-licensed AI browser agent for Chrome and Firefox. It reads pages, extracts data, and automates multi-step tasks through Ask and Act modes. Run it on local models like llama.cpp or Ollama for privacy, or connect any cloud API.

The AI Engineer World’s Fair ended with a debate about loops, a report on the state of AI engineering, and closing keynotes focused on what to build next. The debate highlighted tensions between optimism and caution over autonomous coding loops. A survey found 95% use agents but 59% fear long-term liabilities. Keynotes encouraged building AI-native companies.

The author is not against LLMs but worries that AI-generated websites lack soul, making handcrafted sites more valuable than ever.

Deep Agents is an open-source agent harness by LangChain, designed for long-horizon, multi-step tasks. It includes built-in features such as sub-agents, filesystem access, context management, shell access, persistent memory, and human-in-the-loop approval. Model-agnostic and built on LangGraph, it is production-ready with LangSmith integration.

AI visibility tools claim to measure brand presence in AI assistants but suffer from deep methodological flaws: frontend scraping captures synthetic sessions, API calls differ from consumer apps, prompt sets bias results, geography changes everything, and model drift invalidates trend lines. An honest dashboard would show distributions and methodology, not false precision.

This paper presents novel approaches for fixed and moving obstacle avoidance for unmanned surface vehicles (USVs) using a combination of global and local path planners. The global planner integrates Grassfire, a modified Grassfire, and a new variant of Probabilistic Roadmap. The local planner employs high-level decision logic based on the obstacle's motion direction relative to the USV's path, systematically routing the vehicle behind the obstacle. Simulations validate the method against the D* algorithm.

A novel layered planning and control framework using multi-rate nonlinear model predictive control (MR-NMPC) enables quadrupedal robots to perform wall-assisted bipedal locomotion in constrained environments. The high-level MR-NMPC simultaneously plans discrete contact points and continuous CoM/orientation trajectories, while a low-level whole-body controller tracks references. Simulations on Unitree A1 show a 2.9x success rate improvement over heuristic-based MPC. Accepted to IEEE/RSJ IROS 2026.

This study proposes a reconfigurable rocker-bogie mechanism that achieves efficient turning with few actuators while maintaining high step-climbing capability. A prototype demonstrated zero-radius turning speeds over five times faster than conventional designs, with only 17% of the average wheel torque, and successfully climbed a 40 cm step.

This paper introduces the SE(2) Navigation Mesh (SE(2) NavMesh), a polygonal representation that encodes yaw-dependent traversability for ground robots in complex multi-level environments. It uses footprint masks for traversability evaluation, builds a graph over yaw-specific layers with explicit translational and rotational connectivity, and proposes an A*-String Pulling-A* (ASA) path planning strategy. Simulations show over 50% more traversable area captured than classic NavMeshes, and real-world experiments on a physical robot validate real-time online generation and successful navigation.

BIFROST is a new method for sim2real transfer in robotics that learns invariant features from raw observations using a cross-domain bisimulation objective, enabling zero-shot policy transfer from simulation to reality. It outperforms existing approaches in tasks with both visual and dynamics domain gaps.

A lightweight human-in-the-loop simulation framework for autonomous driving, tightly integrated with the CommonRoad platform, enabling real-time interaction between human drivers and autonomous vehicles.

This paper proposes a neuro-symbolic safety guidance mechanism for flow matching based Vision-Language-Action (VLA) models, enabling predictive collision avoidance. It formulates safety enforcement as a minimum-norm constrained optimization problem that corrects safety violations during the denoising process. On the SafeLIBERO benchmark, it achieves 82.8% collision avoidance and 81.6% task success, improvements of 6.3% and 19.8% over single-step methods, with largest gains on long-horizon tasks.

This paper systematically surveys ROS 2 middleware and introduces a conceptual framework examining its architectural limits through three dimensions: Space, Time, and State. It highlights trade-offs under constrained wireless conditions and outlines a research roadmap for robust middleware.

This paper proposes an adaptive group-accompaniment method for social robots based on Vision-Language Models (VLMs). It leverages VLM semantic reasoning to infer companion positions, maintain social distances, and understand group dynamics, combined with a Model Predictive Path Integral (MPPI) controller for stability. Experiments show a 15% improvement in success rate and a 25% reduction in collision rate, with user study perceiving behaviors as natural and socially appropriate.

This paper proposes WaveLander, a hierarchical reinforcement learning framework for autonomous UAV landing on wave-disturbed marine platforms. It decouples vertical landing decision-making from low-level flight stabilization, using an RL policy to output a vertical velocity reference while a conventional controller handles attitude and lateral tracking. Simulations show robust performance and generalization to unseen disturbances.

Researchers propose 'Sign in the Air to Unlock', an in-air signature interface for VR/AR authentication using a Point-Voxel Cross-Attention Network (PV-Net). It enables natural signing in 3D space without breaking immersion or requiring specialized sensors. PV-Net achieves 2.5% Equal Error Rate on DeepAirSig and 76% classification accuracy on a new Meta Quest 2 dataset, demonstrating the potential of 3D behavioral authentication.

G-CBM, a Graph-based Concept Bottleneck Model, performs unsupervised concept discovery and uses graph attention networks for interpretable visual explanations, achieving 3.7% average relative AUC improvement over ResNet-50 baselines across multiple benchmarks.

This study applies the YOLOv10 computer vision framework to automatically detect brown howler monkeys in camera trap videos, aiming to monitor canopy bridge usage and reduce the time conservationists spend reviewing false-positive images.

This dissertation aims to narrow the gap between machine visual tracking and human perception by enhancing target discrimination, robust adaptation, and geometric reasoning in tracking models.

Multi-subject personalized image generation requires the precise rendering of all requested reference identities and their specified interactions based on a guiding prompt. However, state-of-the-art models still struggle with this process, frequently omitting subjects, failing to preserve reference appearances, or misattributing interactions. Furthermore, existing metrics designed primarily for single-subject fidelity cannot reliably capture these errors, suffering severe degradation in ranking separability and failing to align with human preference as the subject count increases. To address this gap, we introduce Multi-subject Interaction Benchmark and Evaluator (MIBE), a unified framework comprising a Multi-subject Interaction Benchmark (MIB) and a Multi-subject Interaction Evaluator (MIE). MIB systematically covers diverse relation types and scene complexities through a decoupled data regime. This consists of a 60K-pair VLM-labeled Silver Set for scalable metric training and a 4K-pair double-blind Human Evaluation Gold Set covering a diverse range of state-of-the-art generators, with the Silver Set reaching 95.1% cross-VLM preference agreement. To demonstrate the utility of this benchmark, we present MIE, a lightweight, reference-conditioned evaluator trained exclusively on the Silver Set with a dual-head ranking and diagnosis objective. MIE exhibits strong cross-generator generalization on the Gold Set, achieving 0.922 overall pairwise accuracy against human preference, including 0.982 on seen generators and 0.884 on unseen generators. By outperforming a broad spectrum of baseline metrics, including CLIP and DINO variants, MIE demonstrates that diagnostic supervision can preserve ranking separability and human alignment where traditional evaluators collapse.

MapDreamer is a generative diffusion model that creates lane-level vector maps from a single aerial image, using latent diffusion and transformer-based graph prediction. It introduces a lane cardinality module and sliding-window aggregation for scalable map generation. Experiments show improved fidelity over baselines.

This paper introduces an input-aware extendable expert module to enhance fine-grained feature extraction for video-based person re-identification. Using input-aware expert selection and spatial-temporal selection mechanisms, the method achieves state-of-the-art performance on large-scale datasets.

KathaTrace is a generator-agnostic protocol for diagnosing semantic trajectory collapse, defined as the loss of transition meaning between scenes in visual narratives. The authors introduce KathaBench-25K, a dataset of 5,000 narratives from classical collections, and define the Semantic Trajectory Gap (STG) metric. Experiments show substantial STG (23.5±1.3) across state-of-the-art generators. Semantic Compass, an actionability probe, uses KathaTrace signals for post-generation repair and improves storyboard selection.

Proposes CPG-PAD, a framework that introduces model-level concept guidance into prompt learning for Presentation Attack Detection. It uses XAI to discover attack-relevant visual concepts and injects them into prompts, achieving state-of-the-art cross-domain performance on nine benchmarks.

AnchorSplat is a novel 3D-native refinement paradigm that operates directly on 3D structures, avoiding expensive optimization overhead of traditional pipelines. It enforces geometric consistency via a Point Anchor Mechanism and replaces iterative densification with single-pass multiplication, requiring no original multi-view images. Experiments show throughput up to 10^5 times faster than optimization methods with robust zero-shot generalization.

TurnNat is a likelihood-based framework for automatic evaluation of turn-taking naturalness in dyadic spoken dialogue. It uses a causal prediction model to compute negative log-likelihood of future voice activity, quantifying timing atypicality, and demonstrates effectiveness on a perturbation benchmark.

RuleChef is a framework that uses LLMs to generate executable rules for NLP tasks such as text classification, NER, and relation extraction. Rules are generated from task descriptions and labeled examples, then iteratively improved via additional examples and human feedback. LLMs are used only at learning time, resulting in a fast, deterministic, and inspectable rule system. Preliminary evaluation on classification and NER tasks shows promise, and the system is open-sourced under Apache 2.0.

The Office Comprehension Bench (OCB) is the first public benchmark to jointly evaluate LLM systems on Word, Excel, and PowerPoint comprehension over native file formats. It comprises two tracks: File Fidelity Q&A (structural/visual perception) and Domain Q&A (expert reasoning across 12 domains). Even the strongest frontier system achieves only 59.3% on Domain Q&A; increasing thinking depth within a tier yields no material improvement, while moving to a higher product tier offers modest gains. The dataset, evaluation tooling, judge prompt, and leaderboard are released.

Researchers propose RAGP, a novel prompt compression method that models text as a multiplex graph and uses Lévy walks for efficient pruning. It achieves a 49.3 average score on LongBench at 4x compression, outperforming existing LLM-based methods.