Alioune Ngom

Federated learning (FL) enables collaborative model training without centralizing raw data, but existing defenses provide limited support for post-incident accountability and evidence preservation. We present FL-Forensics, a forensic-ready framework for malicious-client and compromised-aggregator settings. It combines client-signed update commitments, a hash-chained ledger, an independent evidence witness, trusted timestamping, key-lifecycle management, encrypted role-based storage, and permissioned quorum anchoring. We also introduce a Forensic Quality Index (FQI) with metrics for integrity, attribution, evidence availability, and audit efficiency. We implement FL-Forensics as an end-to-end Flower extension and evaluate it with 20 clients, Dirichlet non-IID data, and 15% malicious clients on MNIST, Fashion-MNIST, and CIFAR-100. It achieves F1 scores of 0.941, 0.800, and 0.521, with corresponding FQI values of 0.956, 0.900, and 0.714. Ablation results show that independent witnessing, digital signatures, trusted time, and external anchoring make distinct contributions to forensic recovery, attribution, and auditability under controlled evidence-tampering attacks.

Artificial intelligence (AI) systems are increasingly deployed in high-stakes domains, where poisoning attacks can corrupt training data, manipulate model updates, or implant covert backdoors. This survey examines poisoning attacks in federated learning (FL), using centralized learning as a baseline to explain how distributed data, client heterogeneity, privacy-preserving aggregation, and untrusted coordination expand the threat surface. It positions prior surveys and synthesizes representative primary studies through an accountability-oriented lens focused on attribution, audit evidence, traceability, and forensic readiness. The review compares major attack classes, including data poisoning, model poisoning, backdoor insertion, server-side manipulation, Sybil behavior, collusion, and multi-round poisoning. It also evaluates countermeasures such as Byzantine-robust aggregation, anomaly detection, validation-based filtering, malicious-secure aggregation, authenticated update handling, provenance mechanisms, ledger-based evidence, and verifiable aggregation protocols. The analysis shows that robustness alone is insufficient for trustworthy FL unless defenses also preserve evidence that supports independent verification, post-incident reconstruction, and governance review. Persistent gaps remain in causal forensic attribution, privacy-preserving evidence governance, malicious-server threat modeling, scalable verifiability tooling, recovery after poisoning, and deployment-ready benchmarks. The survey concludes that accountable FL should be designed as an evidence-producing system, not merely as a privacy-preserving or attack-resistant training architecture, especially for regulated, cross-silo, and high-risk real-world deployments.

Artificial intelligence (AI) systems are increasingly deployed in critical domains such as healthcare, finance, defense, and transportation. These deployments, however, face grow- ing risks from poisoning attacks that corrupt training data, manipulate model updates, or implant covert backdoors. Such attacks undermine trust, reduce transparency, and challenge the safe and accountable use of AI in high-stakes settings. This survey examines poisoning attacks in federated learning (FL), using centralized learning as a comparative baseline to clarify how the threat landscape changes when data, updates, and control are distributed. Rather than reintroducing a generic poisoning taxonomy as a standalone contri- bution, we position the paper relative to prior surveys, identify what remains insufficiently covered, and synthesize representative primary studies through an accountability-oriented lens. Because FL introduces additional vulnerabilities, including untrusted servers, non-IID heterogeneity, and limited observability of client behavior, we analyze how these properties expand the poisoning threat surface. We review state-of-the-art countermeasures, including Byzantine-robust aggregation, anomaly detection, validation-based defenses, and cryp- tographic prevention mechanisms such as malicious-secure aggregation, authenticated update handling, and verifiable aggregation protocols. Particular emphasis is placed on accountability-enabling mechanisms such as auditability, traceability, and forensic readi- ness, which are essential to responsible AI and regulatory compliance. Our analysis identi- fies persistent research gaps, including the lack of unified privacy-robustness-accountability frameworks, insufficient defenses against server-side attacks, limited verifiability tools, and scalability challenges in real-world FL systems. The paper’s contribution is therefore not to claim that accountability-oriented FL defenses are new, but to clarify the survey scope, position prior surveys directly, and integrate aggregation-based, cryptographic, and governance-oriented strands within an Accountability-Integrated Taxonomy (AIT) and an evidence-oriented discussion of trustworthy federated learning.

Inverse Kinematics (IK) is an integral part of robot manipulation. IK can be challenging to solve, and many computer-aided approaches have been proposed but each has its limitations. The emergence of Large Language Models (LLMs) has seen them applied to solving complex tasks including mathematical problems. This work proposes “LLM-IK” to utilize LLMs to solve IK problems. Relevant serial manipulator information is extracted from descriptor files, prompt engineered, and then provided to the LLMs with methods and feedback to interact with and learn about the kinematic chain. Multiple methods of breaking down kinematic chains into distinct sub-problems are implemented allowing for incremental problem solving. Experiments showed LLM-IK solves up to six Degrees-of-Freedom (DOF) and outperforms IKFast, TRAC-IK, and IKPy in both accuracy and solving time, highlighting this methodology can produce highly efficient and human-readable solutions.

Deep learning vision models are typically tailored for specific modalities and often rely on domain-specific assumptions, such as the grid structures used by nearly all existing vision models. In this work, we propose a self-supervised model based on Transformers, which we call Adaptive Superpixel Coding (ASC). The key insight of our model is to overcome the limitations of traditional Vision Transformers, which depend on fixed-size and non-adaptive patch partitioning. Instead, ASC employs adaptive superpixel layers that dynamically adjust to the underlying image content. We analyze key properties of the approach that make it effective, and find that our method outperforms widely-used alternatives on standard image downstream task benchmarks.