Faithful reasoning in medical vision-language models (VLMs) requires not only accurate predictions but also transparent alignment between textual rationales and visual evidence. While Chain-of-Thought (CoT) prompting has shown promise in medical visual question answering (VQA), no large-scale expert-level dataset has captured stepwise reasoning with precise visual grounding. We introduce S-Chain, the first large-scale dataset of 12,000 expert-annotated medical images with bounding boxes and structured visual CoT (SV-CoT), explicitly linking visual regions to reasoning steps. The dataset further supports 16 languages, totaling over 700k VQA pairs for broad multilingual applicability. Using S-Chain, we benchmark state-of-the-art medical VLMs (ExGra-Med, LLaVA-Med) and general-purpose VLMs (Qwen2.5-VL, InternVL2.5), showing that SV-CoT supervision significantly improves interpretability, grounding fidelity, and robustness. Beyond benchmarking, we study its synergy with retrieval-augmented generation, revealing how domain knowledge and visual grounding interact during autoregressive reasoning. Finally, we propose a new mechanism that strengthens the alignment between visual evidence and reasoning, improving both reliability and efficiency. S-Chain establishes a new benchmark for grounded medical reasoning and paves the way toward more trustworthy and explainable medical VLMs.

Real-time, scalable, and accurate decoding is a critical component for realizing a fault-tolerant quantum computer. While Transformer-based neural decoders such as \textit{AlphaQubit} have demonstrated high accuracy, the computational complexity of their core attention mechanism, which scales as $\mathcal{O}(d^4)$ with code distance $d$, results in decoding speeds insufficient for practical real-time applications. In this work, we introduce and evaluate a \textit{Mamba}-based decoder, a state-space model with $\mathcal{O}(d^2)$ complexity. In memory experiments using Sycamore hardware data, our Mamba decoder matches the performance of its Transformer-based counterpart, providing that its superior efficiency does not come at the cost of performance. Crucially, in simulated real-time scenarios that account for decoder-induced noise, the Mamba decoder significantly outperforms the Transformer, exhibiting a higher error threshold of $0.0104$ compared to $0.0097$. These results demonstrate that Mamba decoders offer a compelling balance between speed and accuracy, making them a promising architecture for scalable, real-time quantum error correction.

Human motion is fundamentally driven by continuous physical interaction with the environment. Whether walking, running, or simply standing, the forces exchanged between our feet and the ground provide crucial insights for understanding and reconstructing human movement. Recent advances in wearable insole devices offer a compelling solution for capturing these forces in diverse, real-world scenarios. Sensor insoles pose no constraint on the users' motion (unlike mocap suits) and are unaffected by line-of-sight limitations (in contrast to optical systems). These qualities make sensor insoles an ideal choice for robust, unconstrained motion capture, particularly in outdoor environments. Surprisingly, leveraging these devices with recent motion reconstruction methods remains largely unexplored. Aiming to fill this gap, we present Step2Motion, the first approach to reconstruct human locomotion from multi-modal insole sensors. Our method utilizes pressure and inertial data-accelerations and angular rates-captured by the insoles to reconstruct human motion. We evaluate the effectiveness of our approach across a range of experiments to show its versatility for diverse locomotion styles, from simple ones like walking or jogging up to moving sideways, on tiptoes, slightly crouching, or dancing.

Retrieval-Augmented Generation (RAG) faces a core bottleneck with knowledge-sparse and semantically ambiguous long-tail queries, where retrieval noise distorts reasoning and necessitates costly post-processing. To tackle this, we propose RaCoT (Retrieval-aware Contrastive-of-Thought), a novel framework that shifts contrastive thinking to the pre-retrieval stage. By automatically generating a semantically adjacent yet differently answered contrastive question and extracting a $Δ$-Prompt to capture their key differences, RaCoT guides the model to proactively focus on the ``critical details that determine answer divergence." This approach allows it to suppress semantic interference within a single retrieval pass, overcoming the theoretical bottleneck of single-vector queries that struggle to simultaneously encode signals for what to attend to and what to ignore. On six authoritative benchmarks, including PopQA and TriviaQA-unfiltered, RaCoT outperforms strong baselines like RankRAG and Self-RAG by 0.9-2.4 percentage points. It exhibits superior robustness, with a performance drop of only 8.6\% in adversarial tests, far surpassing the over 15\% degradation in other methods. Furthermore, its low latency (3.12s) and token overhead (11.54) place it on the accuracy-efficiency Pareto frontier, while ablation studies validate the necessity of each component. Ultimately, RaCoT reframes the RAG paradigm from ``post-hoc context cleaning" to ``a priori shaping of discriminative reasoning", offering an efficient and robust path toward reliable AI systems for real-time, resource-constrained deployments.

Multimodal Retrieval-Augmented Generation (MRAG) has emerged as a promising method to generate factual and up-to-date responses of Multimodal Large Language Models (MLLMs) by incorporating non-parametric knowledge from external knowledge bases. However, existing MRAG approaches suffer from static retrieval strategies, inflexible modality selection, and suboptimal utilization of retrieved information, leading to three critical challenges: determining when to retrieve, what modality to incorporate, and how to utilize retrieved information effectively. To address these challenges, we introduce Windsock, a query-dependent module making decisions on retrieval necessity and modality selection, effectively reducing computational overhead and improving response quality. Additionally, we propose Dynamic Noise-Resistance (DANCE) Instruction Tuning, an adaptive training strategy that enhances MLLMs' ability to utilize retrieved information while maintaining robustness against noise. Moreover, we adopt a self-assessment approach leveraging knowledge within MLLMs to convert question-answering datasets to MRAG training datasets. Extensive experiments demonstrate that our proposed method significantly improves the generation quality by 17.07% while reducing 8.95% retrieval times.

Despite their impressive generalization capabilities, instruction-tuned Large Language Models often underperform on text classification benchmarks. We introduce SALSA, a coherent pipeline that combines structured prompting, class-to-token mapping, and parameter-efficient fine-tuning, thereby avoiding cold-start training. Each class label is mapped to a distinct output token, and prompts are constructed to elicit a single-token response. During inference, the model's output is projected only onto the logits of the relevant class tokens, enabling efficient and accurate classification in a single forward pass. SALSA achieves state-of-the-art results across diverse benchmarks, demonstrating its robustness and scalability for LLM-based classification applications.

If-then rules are widely used to explain machine learning models; e.g., "if employed = no, then loan application = rejected." We present the first proposal to apply rules to explain the emerging class of large language models (LLMs) with retrieval-augmented generation (RAG). Since RAG enables LLM systems to incorporate retrieved information sources at inference time, rules linking the presence or absence of sources can explain output provenance; e.g., "if a Times Higher Education ranking article is retrieved, then the LLM ranks Oxford first." To generate such rules, a brute force approach would probe the LLM with all source combinations and check if the presence or absence of any sources leads to the same output. We propose optimizations to speed up rule generation, inspired by Apriori-like pruning from frequent itemset mining but redefined within the scope of our novel problem. We conclude with qualitative and quantitative experiments demonstrating our solutions' value and efficiency.

The recent application of deep learning models to financial trading has heightened the need for high fidelity financial time series data. This synthetic data can be used to supplement historical data to train large trading models. The state-of-the-art models for the generative application often rely on huge amounts of historical data and large, complicated models. These models range from autoregres-sive and diffusion-based models through to architecturally simpler models such as the temporal-attention bilinear layer. Agent-based approaches to modelling limit order book dynamics can also recreate trading activity through mechanistic models of trader behaviours. In this work, we demonstrate how a popular agent-based framework for simulating intraday trading activity, the Chiarella model, can be combined with one of the most performant deep learning models for forecasting multi-variate time series, the T ABL model. This forecasting model is coupled to a simulation of a matching engine with a novel method for simulating deleted order flow. Our simulator gives us the ability to test the generative abilities of the forecasting model using stylised facts. Our results show that this methodology generates realistic price dynamics however, when analysing deeper, parts of the markets microstructure are not accurately recreated, highlighting the necessity for including more sophisticated agent behaviors into the modeling framework to help account for tail events.

Large Language Models (LLMs) often expend significant computational resources generating boilerplate responses, such as refusals, simple acknowledgements and casual greetings, which adds unnecessary cost and latency. To address this inefficiency, we propose a simple yet highly effective method for detecting such responses after only a single generation step. We demonstrate that the log-probability distribution of the first generated token serves as a powerful signal for classifying the nature of the entire subsequent response. Our experiments, conducted across a diverse range of small, large, and reasoning-specialized models, show that the first-token log-probability vectors form distinctly separable clusters for different response types. Using a lightweight k-NN classifier, we achieve high accuracy in predicting whether a response will be a substantive answer or a form of boilerplate response, including user-specified refusals. The primary implication is a practical, computationally trivial technique, optimizing LLM inference by enabling early termination or redirection to a smaller model, thereby yielding significant savings in computational cost. This work presents a direct path toward more efficient and sustainable LLM deployment.

Graph Neural Networks (GNNs) have demonstrated remarkable performance across a spectrum of graph-related tasks, however concerns persist regarding their vulnerability to adversarial perturbations. While prevailing defense strategies focus primarily on pre-processing techniques and adaptive message-passing schemes, this study delves into an under-explored dimension: the impact of weight initialization and associated hyper-parameters, such as training epochs, on a model's robustness. We introduce a theoretical framework bridging the connection between initialization strategies and a network's resilience to adversarial perturbations. Our analysis reveals a direct relationship between initial weights, number of training epochs and the model's vulnerability, offering new insights into adversarial robustness beyond conventional defense mechanisms. While our primary focus is on GNNs, we extend our theoretical framework, providing a general upper-bound applicable to Deep Neural Networks. Extensive experiments, spanning diverse models and real-world datasets subjected to various adversarial attacks, validate our findings. We illustrate that selecting appropriate initialization not only ensures performance on clean datasets but also enhances model robustness against adversarial perturbations, with observed gaps of up to 50\% compared to alternative initialization approaches.