Cov[gi(s,a),gj(s,a)]. (9) The n optimal baselinesare given by the values that minimise Equation 9; i.e.b?i(s, σπi (a)) . Note that whileyi depends on the full action,xi depends only on the actions influencing the targets in [KΣψ(s,a)]i. Ingeneral,thereareveryfewmethods that can solve these type of systems, and those that can are limited to bounds of approximately d|Σ| 20. These explore the impact of the factor baseline across aset of dimensionalities and learning rates. This implies that the performance observed in the search bandit itlikely totell usabout the performance infull MDPs.

Surgeries may become safer and more precise than ever before. A new brain implant now lets people control Apple devices, such as iPads, iPhones and the Vision Pro, using only their thoughts. Synchron, an endovascular brain-computer interface (BCI) company based in New York, demonstrated the first wireless BCI that works with Apple's official protocol. Ten patients have received the implant: six in the U.S. and four in Australia. With this technology, users living with severe paralysis can navigate apps, send messages and operate devices hands-free.

Large Language Models (LLMs) fine-tuned to align with human values often exhibit alignment drift, producing unsafe or policy-violating completions when exposed to adversarial prompts, decoding perturbations, or paraphrased jailbreaks. While prior work has behaviorally characterized alignment failure, little is known about the training-time belief sources underlying these failures. We introduce TraceAlign, a unified framework for tracing unsafe completions back to their root causes in the model's training corpus. Central to our approach is the Belief Conflict Index (BCI), which quantifies semantic inconsistency between generated spans and aligned policies, based on retrieved training documents using suffix-array matching. We propose three complementary interventions: (i) TraceShield, an inference-time safety filter that refuses completions with high-BCI spans, (ii) Contrastive Belief Deconfliction Loss, a contrastive fine-tuning objective penalizing high-BCI continuations during DPO, and (iii) Prov-Decode, a provenance-aware decoding strategy that vetoes beam expansions predicted to yield high-BCI spans. Together, these defenses reduce alignment drift by up to 85% on our curated Alignment Drift Benchmark (ADB) while preserving utility on standard tasks, with delta less than 0.2 and improved refusal quality. We further derive a theoretical upper bound on drift likelihood via suffix-array span statistics, linking memorization frequency and length to adversarial reactivation risk. TraceAlign thus provides the first scalable, traceable, and grounded toolkit for understanding and mitigating alignment failures at source. To encourage further exploration and development, we open-source our implementation at: https://anonymous.4open.science/r/tracealign-2DA7

-- Emotion recognition in dynamic social contexts requires an understanding of the complex interaction between facial expressions and situational cues. This paper presents a salience-adjusted framework for context-aware emotion recognition with Bayesian Cue Integration (BCI) and Visual-Language Models (VLMs) to dynamically weight facial and contextual information based on the expressivity of facial cues. We evaluate this approach using human annotations and automatic emotion recognition systems in prisoner's dilemma scenarios, which are designed to evoke emotional reactions. Our findings demonstrate that incorporating salience adjustment enhances emotion recognition performance, offering promising directions for future research to extend this framework to broader social contexts and multimodal applications. I. INTRODUCTION Automatic expression recognition traditionally treats facial expressions as signifying the emotional state of the expresser. Recently, there has been growing appreciation that observer perceptions differ from self-reported emotions [6] (people can seem happy when experiencing negative emotions [2], [22]), yet these (mis)perceptions are crucial for explaining human social behavior.

Neurotech company Synchron has unveiled the latest version of its brain-computer interface, which uses Nvidia technology and the Apple Vision Pro to enable individuals with paralysis to control digital and physical environments with their thoughts. In a video demonstration at the Nvidia GTC conference this week in San Jose, California, Synchron showed off how its system allows one of its trial participants, Rodney Gorham, who is paralyzed, to control multiple devices in his home. From his sun-filled living room in Melbourne, Australia, Gorham is able to play music from a smart speaker, adjust the lighting, turn on a fan, activate an automatic pet feeder, and run a robotic vacuum. Gorham has lost the use of his voice and much of his body due to having amyotrophic lateral sclerosis, or ALS. The degenerative disease weakens muscles over time and eventually leads to paralysis.

For years now, brain-computer interfaces (BCI) have incrementally advanced, giving people with spinal injuries or lost limbs the ability to control prosthetics and computer cursors using their signals. But even though the tech has made strides, the replicating subtle, delicate, nuanced sensations of touch has remained just out of reach. Now, however, a team of researchers from the Cortical Bionics Research Group believe they have made a major breakthrough. A pair of patients wearing a BCI was able to control a bionic arm and "feel" tactile edges, shapes, and curvatures along its fingers. The researchers' findings were published today in the journal Science.

Abstract: This study offers a revolutionary strategy to developing wheelchairs based on the Brain-Computer Interface (BCI) that incorporates Artificial Intelligence (AI) using a The device uses electroencephalogram (EEG) data to mimic wheelchair navigation. Five different models were trained on a pre-filtered dataset that was divided into fixed-length windows using a sliding window technique. Each window contained statistical measurements, FFT coefficients for different frequency bands, and a label identifying the activity carried out during that window that was taken from an open-source Kaggle repository. The XGBoost model outperformed the other models, CatBoost, GRU, SVC, and XGBoost, with an accuracy of 60%. The CatBoost model with a major difference between training and testing accuracy shows overfitting, and similarly, the bestperforming model, with SVC, was implemented in a tkinter GUI. The wheelchair movement could be simulated in various directions, and a Raspberry Pi-powered wheelchair system for braincomputer interface is proposed here. Keywords: Brain Computer Interfacing, FFT (Fast Fourier Transform), Raspberry-pi, electroencephalogram 1. Introduction Brain-Computer Interfaces (BCIs) represent a cutting-edge technology that facilitates direct communication between the human brain and external devices. In recent years, BCIs have been widely explored for assisting individuals with mobility impairments. This paper focuses on a novel BCI-based wheelchair control system that leverages EEG signals associated with control using various movements related dataset. The system incorporates various machine learning models with various optimization techniques for hyper-parameter tuning and finally, shows an attention mechanism for enhancing the performance of Bi-directional Long Short-Term Memory (Bi-LSTM) networks, which are employed for EEG signal classification. To integrate the braincomputer interface (BCI) for the wheelchair, an analysis of brain activity is necessary-based on modern technology. The signs of brain activity can be obtained using a variety of techniques [1]. In order to help people with severe disabilities live their daily lives, new aids, gadgets, and assistive technologies are required, as demonstrated by the pandemic emergency of the coronavirus illness 2019 (COVID-19). Brain-Computer Interfaces (BCIs) that use electroencephalography (EEG) can help people who experience major health issues become more independent and participate in activities more easily. This can improve their general well-being and prevent deficits [2].

Elon Musk's brain implant company Neuralink announced on Tuesday that it is launching a study to test its implant for a new use: allowing a person to control a robotic arm using just their thoughts. "We're excited to announce the approval and launch of a new feasibility trial to extend BCI control using the N1 implant to an investigational assistive robotic arm," Neuralink said in a post on Musk's social media platform X. A BCI, or brain-computer interface, is a system that allows a person to directly control outside devices with their brain waves. It works by reading and decoding intended movement signals from neurons. Neuralink's BCI involves a coin-sized device dubbed N1 that is surgically implanted in the brain by a robot.

Brain computer interface (BCI) applications in robotics are becoming more famous and famous. People with disabilities are facing a real-time problem of doing simple activities such as grasping, handshaking etc. in order to aid with this problem, the use of brain signals to control actuators is showing a great importance. The Emotive Insight, a Brain-Computer Interface (BCI) device, is utilized in this project to collect brain signals and transform them into commands for controlling a robotic arm using an Arduino controller. The Emotive Insight captures brain signals, which are subsequently analyzed using Emotive software and connected with Arduino code. The HITI Brain software integrates these devices, allowing for smooth communication between brain activity and the robotic arm. This system demonstrates how brain impulses may be utilized to control external devices directly. The results showed that the system is applicable efficiently to robotic arms and also for prosthetic arms with Multi Degree of Freedom. In addition to that, the system can be used for other actuators such as bikes, mobile robots, wheelchairs etc.

Mark, a 64-year-old with amyotrophic lateral sclerosis, or ALS, uses Amazon Alexa all the time using his voice. But now, thanks to a brain implant, he can also control the virtual assistant with his mind. ALS affects the nerve cells in the brain and spinal cord, causing loss of muscle control over time. Mark, who asked that his last name not be used, has limited mobility as a result of his condition. He can walk and talk but has no use of his arms and hands.