-- We propose a novel framework for decision-making in cooperative grasping for two-robot object transport in constrained environments. The core of the framework is a Conditional Embedding (CE) model consisting of two neural networks that map grasp configuration information into an embedding space. The resulting embedding vectors are then used to identify feasible grasp configurations that allow two robots to collaboratively transport an object. T o ensure general-izability across diverse environments and object geometries, the neural networks are trained on a dataset comprising a range of environment maps and object shapes. We employ a supervised learning approach with negative sampling to ensure that the learned embeddings effectively distinguish between feasible and infeasible grasp configurations. Evaluation results across a wide range of environments and objects in simulations demonstrate the model's ability to reliably identify feasible grasp configurations.

Large language models are now routinely used in high-stakes applications where hallucinations can cause serious harm, such as medical consultations or legal advice. Existing hallucination detection methods, however, are impractical for real-world use, as they are either limited to short factual queries or require costly external verification. We present a cheap, scalable method for real-time identification of hallucinated tokens in long-form generations, and scale it effectively to 70B parameter models. Our approach targets \emph{entity-level hallucinations} -- e.g., fabricated names, dates, citations -- rather than claim-level, thereby naturally mapping to token-level labels and enabling streaming detection. We develop an annotation methodology that leverages web search to annotate model responses with grounded labels indicating which tokens correspond to fabricated entities. This dataset enables us to train effective hallucination classifiers with simple and efficient methods such as linear probes. Evaluating across four model families, our classifiers consistently outperform baselines on long-form responses, including more expensive methods such as semantic entropy (e.g., AUC 0.90 vs 0.71 for Llama-3.3-70B), and are also an improvement in short-form question-answering settings. Moreover, despite being trained only with entity-level labels, our probes effectively detect incorrect answers in mathematical reasoning tasks, indicating generalization beyond entities. While our annotation methodology is expensive, we find that annotated responses from one model can be used to train effective classifiers on other models; accordingly, we publicly release our datasets to facilitate reuse. Overall, our work suggests a promising new approach for scalable, real-world hallucination detection.

Recent studies have shown that Machine Learning (ML) models can exhibit bias in real-world scenarios, posing significant challenges in ethically sensitive domains such as healthcare. Such bias can negatively affect model fairness, model generalization abilities and further risks amplifying social discrimination. There is a need to remove biases from trained models. Existing debiasing approaches often necessitate access to original training data and need extensive model retraining; they also typically exhibit trade-offs between model fairness and discriminative performance. To address these challenges, we propose Soft-Mask Weight Fine-Tuning (SWiFT), a debiasing framework that efficiently improves fairness while preserving discriminative performance with much less debiasing costs. Notably, SWiFT requires only a small external dataset and only a few epochs of model fine-tuning. The idea behind SWiFT is to first find the relative, and yet distinct, contributions of model parameters to both bias and predictive performance. Then, a two-step fine-tuning process updates each parameter with different gradient flows defined by its contribution. Extensive experiments with three bias sensitive attributes (gender, skin tone, and age) across four dermatological and two chest X-ray datasets demonstrate that SWiFT can consistently reduce model bias while achieving competitive or even superior diagnostic accuracy under common fairness and accuracy metrics, compared to the state-of-the-art. Specifically, we demonstrate improved model generalization ability as evidenced by superior performance on several out-of-distribution (OOD) datasets.

This post - print is the paper version that was submitted to ZINC 202 5 . Abstract -- Recently, federated learning frameworks such as Python TestBed for Federated Learning Algorithms and MicroPython TestBed for Federated Learning Algorithms have emerged to tackle user privacy concerns and efficiency in embedded systems. Even more recently, an efficient federated anomaly detection algorithm, FLiForest, based on Isolation Forests has been developed, offering a low - resource, unsupervised method well - suited for edge deployment and continuous learning. In this paper, we present an appli cation of Isolation Forest - based temperature anomaly detection, developed using the previously mentioned federated learning frameworks, aimed at small edge devices and IoT systems running MicroPython. The system has been experimentally evaluated, achieving over 9 6 % accuracy in distinguishing normal from abnormal readings and above 78 % precision in detecting anomalies across all tested configurations, while maintaining a memory usage below 16 0 KB during model training.

Automatic speech recognition (ASR) systems struggle with domain-specific named entities, especially homophones. Contextual ASR improves recognition but often fails to capture fine-grained phoneme variations due to limited entity diversity. Moreover, prior methods treat entities as independent tokens, leading to incomplete multi-token biasing. To address these issues, we propose Phoneme-Augmented Robust Contextual ASR via COntrastive entity disambiguation (PARCO), which integrates phoneme-aware encoding, contrastive entity disambiguation, entity-level supervision, and hierarchical entity filtering. These components enhance phonetic discrimination, ensure complete entity retrieval, and reduce false positives under uncertainty. Experiments show that PARCO achieves CER of 4.22% on Chinese AISHELL-1 and WER of 11.14% on English DATA2 under 1,000 distractors, significantly outperforming baselines. PARCO also demonstrates robust gains on out-of-domain datasets like THCHS-30 and LibriSpeech.

Speech generation systems can produce remarkably realistic vocalisations that are often indistinguishable from human speech, posing significant authenticity challenges. Although numerous deepfake detection methods have been developed, their effectiveness in real-world environments remains unrealiable due to the domain shift between training and test samples arising from diverse human speech and fast evolving speech synthesis systems. This is not adequately addressed by current datasets, which lack real-world application challenges with diverse and up-to-date audios in both real and deep-fake categories. To fill this gap, we introduce AUDETER (AUdio DEepfake TEst Range), a large-scale, highly diverse deepfake audio dataset for comprehensive evaluation and robust development of generalised models for deepfake audio detection. It consists of over 4,500 hours of synthetic audio generated by 11 recent TTS models and 10 vocoders with a broad range of TTS/vocoder patterns, totalling 3 million audio clips, making it the largest deepfake audio dataset by scale. Through extensive experiments with AUDETER, we reveal that i) state-of-the-art (SOTA) methods trained on existing datasets struggle to generalise to novel deepfake audio samples and suffer from high false positive rates on unseen human voice, underscoring the need for a comprehensive dataset; and ii) these methods trained on AUDETER achieve highly generalised detection performance and significantly reduce detection error rate by 44.1% to 51.6%, achieving an error rate of only 4.17% on diverse cross-domain samples in the popular In-the-Wild dataset, paving the way for training generalist deepfake audio detectors. AUDETER is available on GitHub.

Dimensionality Reduction (DR) is widely used for visualizing high-dimensional data, often with the goal of revealing expected cluster structure. However, such a structure may not always appear in the projections. Existing DR quality metrics assess projection reliability (to some extent) or cluster structure quality, but do not explain why expected structures are missing. Visual Analytics solutions can help, but are often time-consuming due to the large hyperparameter space. This paper addresses this problem by leveraging a recent framework that divides the DR process into two phases: a relationship phase, where similarity relationships are modeled, and a mapping phase, where the data is projected accordingly. We introduce two supervised metrics, precision and recall, to evaluate the relationship phase. These metrics quantify how well the modeled relationships align with an expected cluster structure based on some set of labels representing this structure. We illustrate their application using t-SNE and UMAP, and validate the approach through various usage scenarios. Our approach can guide hyperparameter tuning, uncover projection artifacts, and determine if the expected structure is captured in the relationships, making the DR process faster and more reliable.

In the domain of security, vulnerabilities frequently remain undetected even after their exploitation. In this work, vulnerabilities refer to publicly disclosed flaws documented in Common Vulnerabilities and Exposures (CVE) reports. Establishing a connection between attacks and vulnerabilities is essential for enabling timely incident response, as it provides defenders with immediate, actionable insights. However, manually mapping attacks to CVEs is infeasible, thereby motivating the need for automation. This paper evaluates 14 state-of-the-art (SOTA) sentence transformers for automatically identifying vulnerabilities from textual descriptions of attacks. Our results demonstrate that the multi-qa-mpnet-base-dot-v1 (MMPNet) model achieves superior classification performance when using attack Technique descriptions, with an F1-score of 89.0, precision of 84.0, and recall of 94.7. Furthermore, it was observed that, on average, 56% of the vulnerabilities identified by the MMPNet model are also represented within the CVE repository in conjunction with an attack, while 61% of the vulnerabilities detected by the model correspond to those cataloged in the CVE repository. A manual inspection of the results revealed the existence of 275 predicted links that were not documented in the MITRE repositories. Consequently, the automation of linking attack techniques to vulnerabilities not only enhances the detection and response capabilities related to software security incidents but also diminishes the duration during which vulnerabilities remain exploitable, thereby contributing to the development of more secure systems.

We tackle atypical versus normal mitosis classification in the MIDOG 2025 challenge using EfficientViT-L2, a hybrid CNN--ViT architecture optimized for accuracy and efficiency. A unified dataset of 13,938 nuclei from seven cancer types (MIDOG++ and AMi-Br) was used, with atypical mitoses comprising ~15. To assess domain generalization, we applied leave-one-cancer-type-out cross-validation with 5-fold ensembles, using stain-deconvolution for image augmentation. For challenge submissions, we trained an ensemble with the same 5-fold split but on all cancer types. In the preliminary evaluation phase, this model achieved balanced accuracy of 0.859, ROC AUC of 0.942, and raw accuracy of 0.85, demonstrating competitive and well-balanced performance across metrics.

Large language models (LLMs) have demonstrated impressive capabilities across a variety of tasks, but their increasing autonomy in real-world applications raises concerns about their trustworthiness. While hallucinations-unintentional falsehoods-have been widely studied, the phenomenon of lying, where an LLM knowingly generates falsehoods to achieve an ulterior objective, remains underexplored. In this work, we systematically investigate the lying behavior of LLMs, differentiating it from hallucinations and testing it in practical scenarios. Through mechanistic interpretability techniques, we uncover the neural mechanisms underlying deception, employing logit lens analysis, causal interventions, and contrastive activation steering to identify and control deceptive behavior. We study real-world lying scenarios and introduce behavioral steering vectors that enable fine-grained manipulation of lying tendencies. Further, we explore the trade-offs between lying and end-task performance, establishing a Pareto frontier where dishonesty can enhance goal optimization. Our findings contribute to the broader discourse on AI ethics, shedding light on the risks and potential safeguards for deploying LLMs in high-stakes environments. Code and more illustrations are available at https://llm-liar.github.io/