Interpretability methods have recently gained significant attention, particularly in the context of large language models, enabling insights into linguistic representations, error detection, and model behaviors such as hallucinations and repetitions. However, these techniques remain underexplored in automatic speech recognition (ASR), despite their potential to advance both the performance and interpretability of ASR systems. In this work, we adapt and systematically apply established interpretability methods such as logit lens, linear probing, and activation patching, to examine how acoustic and semantic information evolves across layers in ASR systems. Our experiments reveal previously unknown internal dynamics, including specific encoder-decoder interactions responsible for repetition hallucinations and semantic biases encoded deep within acoustic representations. These insights demonstrate the benefits of extending and applying inter-pretability techniques to speech recognition, opening promising directions for future research on improving model transparency and robustness.

Formal theorem proving (FTP) has emerged as a critical foundation for evaluating the reasoning capabilities of large language models, enabling automated verification of mathematical proofs at scale. However, progress has been constrained by limited datasets due to the high cost of manual curation and the scarcity of challenging problems with verified formal-informal correspondences. We propose leveraging theoretical computer science (TCS) as a scalable source of rigorous proof problems, where algorithmic definitions enable automated generation of arbitrarily many challenging theorem-proof pairs. We demonstrate this approach on two TCS domains: Busy Beaver problems, which involve proving bounds on Turing machine halting behavior, and Mixed Boolean Arithmetic problems, which combine logical and arithmetic reasoning. Our framework automatically synthesizes problems with parallel formal (Lean4) and informal (Markdown) specifications, creating a scalable pipeline for generating verified proof challenges. Evaluation on frontier models reveals substantial gaps in automated theorem proving: while DeepSeekProver-V2-671B achieves 57.5\% success on Busy Beaver problems, it manages only 12\% on Mixed Boolean Arithmetic problems. These results highlight the difficulty of long-form proof generation even for problems that are computationally easy to verify, demonstrating the value of TCS domains for advancing automated reasoning research.

Entity linking is a fundamental task in knowledge graph (KG) construction Hofer et al. (2024), aiming to link mentions to their corresponding entities in a target knowledge base (KB). It is widely applied in downstream natural language processing (NLP) tasks, such as Question & Answering Systems Sequeda et al. (2024) and intelligent recommendation systems Chaudhari et al. (2017). Recently, the explosive growth of multimodal data on the Internet has raised challenges, as the quality of online information is often inconsistent, many mentions are ambiguous, and contextual information is frequently incomplete. Under such conditions, relying solely on a single modality (such as pure text) is often insufficient to accurately resolve reference ambiguity Gan et al. (2021). Integrating textual and visual modalities can significantly improve the precision and efficiency of disambiguation Gella et al. (2017). Consequently, multimodal entity linking, which involves combining textual and visual information to link real-world mentions to corresponding entities in a multimodal knowledge graph (MMKG), has become a critical research task. For example, as shown in Figure 1, the mention of "Apple" may be difficult to disambiguate, as it could refer to various entities, such as Apple Inc. or the apple (fruit). However, by considering both textual and visual information, it becomes easier and clearer to accurately link the mention of "Apple" to the entity "apple (fruit of the apple tree)." Currently, multimodal entity linking models are primarily based on deep learning frameworks, utilizing cross-attention mechanisms Lu and Elhamifar (2024) and visual feature encoding techniques Mokssit et al. (2023) to achieve the fusion of textual mentions and visual information.

Physics - Based Explainable AI for ECG Segmentation: A Lightweight Model Muhammad Fathur Rohman Sidiq Department of Physics, Faculty of Mathematics and Science, Brawijaya University, Malang, Indonesia Abdurrouf Department of Physics, Faculty of Mathematics and Science, Brawijaya University, Malang, Indonesia Didik Rahadi Santoso * Department of Physics, Faculty of Mathematics and Science, Brawijaya University, Malang, Indonesia * Corresponding author. E - mail: dieks@ub.ac.id Abstract The heart's electrical activity, recorded through Electrocardiography (ECG), is essential for diagnosing various cardiovascular conditions. However, many existing ECG segmentation models rely on complex, multi - layered architectures such as BiLSTM, which ar e computationally intensive and inefficient. This study introduces a streamlined architecture that combines spectral analysis with probabilistic predictions for ECG signal segmentation. Additionally, an Explainable AI (XAI) approach is applied to enhance model interpretability by explaining how temporal and frequency - based features contribute to ECG segmentation. By i ncorporating principles from physics - based AI, this method provides a clear understanding of the decision - making process, ensuring reliability and transparency in ECG analysis.

Solving financial problems demands complex reasoning, multimodal data processing, and a broad technical understanding, presenting unique challenges for current large language models (LLMs). We introduce XFinBench, a novel benchmark with 4,235 examples designed to evaluate LLM's ability in solving complex, knowledge-intensive financial problems across diverse graduate-level finance topics with multi-modal context. We identify five core capabilities of LLMs using XFinBench, i.e, terminology understanding, temporal reasoning, future forecasting, scenario planning, and numerical modelling. Upon XFinBench, we conduct extensive experiments on 18 leading models. The result shows that o1 is the best-performing text-only model with an overall accuracy of 67.3%, but still lags significantly behind human experts with 12.5%, especially in temporal reasoning and scenario planning capabilities. We further construct a knowledge bank with 3,032 finance terms for knowledge augmentation analysis, and find that relevant knowledge to the question only brings consistent accuracy improvements to small open-source model. Additionally, our error analysis reveals that rounding errors during calculation and blindness to position and intersection of curves in the image are two primary issues leading to model's poor performance in calculating and visual-context questions, respectively. Code and dataset are accessible via GitHub: https://github.com/Zhihan72/XFinBench.

As large language models (LLMs) increasingly interact with each other, most notably in multi-agent setups, we may expect (and hope) that `trust' relationships develop between them, mirroring trust relationships between human colleagues, friends, or partners. Yet, though prior work has shown LLMs to be capable of identifying emotional connections and recognizing reciprocity in trust games, little remains known about (i) how different strategies to build trust compare, (ii) how such trust can be measured implicitly, and (iii) how this relates to explicit measures of trust. We study these questions by relating implicit measures of trust, i.e. susceptibility to persuasion and propensity to collaborate financially, with explicit measures of trust, i.e. a dyadic trust questionnaire well-established in psychology. We build trust in three ways: by building rapport dynamically, by starting from a prewritten script that evidences trust, and by adapting the LLMs' system prompt. Surprisingly, we find that the measures of explicit trust are either little or highly negatively correlated with implicit trust measures. These findings suggest that measuring trust between LLMs by asking their opinion may be deceiving. Instead, context-specific and implicit measures may be more informative in understanding how LLMs trust each other.

Biased news contributes to societal polarization and is often reinforced by hostile reader comments, constituting a vital yet often overlooked aspect of news dissemination. Our study reveals that offensive comments support biased content, amplifying bias and causing harm to targeted groups or individuals. Counterspeech is an effective approach to counter such harmful speech without violating freedom of speech, helping to limit the spread of bias. To the best of our knowledge, this is the first study to explore counterspeech generation in the context of news articles. We introduce a manually annotated dataset linking media bias, offensive comments, and counterspeech. We conduct a detailed analysis showing that over 70\% offensive comments support biased articles, amplifying bias and thus highlighting the importance of counterspeech generation. Comparing counterspeech generated by humans and large language models, we find model-generated responses are more polite but lack the novelty and diversity. Finally, we improve generated counterspeech through few-shot learning and integration of news background information, enhancing both diversity and relevance.

This paper presents our approach and results for SubTask 1: Islamic Inheritance Reasoning at QIAS 2025, a shared task focused on evaluating Large Language Models (LLMs) in understanding and reasoning within Islamic inheritance knowledge. We fine-tuned the Fanar-1-9B causal language model using Low-Rank Adaptation (LoRA) and integrated it into a Retrieval-Augmented Generation (RAG) pipeline. Our system addresses the complexities of Islamic inheritance law, including comprehending inheritance scenarios, identifying eligible heirs, applying fixed-share rules, and performing precise calculations. Our system achieved an accuracy of 0.858 in the final test, outperforming other competitive models such as, GPT 4.5, LLaMA, Fanar, Mistral and ALLaM evaluated with zero-shot prompting. Our results demonstrate that QU-NLP achieves near state-of-the-art accuracy (85.8%), excelling especially on advanced reasoning (97.6%) where it outperforms Gemini 2.5 and OpenAI's o3. This highlights that domain-specific fine-tuning combined with retrieval grounding enables mid-scale Arabic LLMs to surpass frontier models in Islamic inheritance reasoning.

We introduce PGF-Net (Progressive Gated-Fusion Network), a novel deep learning framework designed for efficient and interpretable multimodal sentiment analysis. Our framework incorporates three primary innovations. Firstly, we propose a Progressive Intra-Layer Fusion paradigm, where a Cross-Attention mechanism empowers the textual representation to dynamically query and integrate non-linguistic features from audio and visual streams within the deep layers of a Transformer encoder. This enables a deeper, context-dependent fusion process. Secondly, the model incorporates an Adaptive Gated Arbitration mechanism, which acts as a dynamic controller to balance the original linguistic information against the newly fused multimodal context, ensuring stable and meaningful integration while preventing noise from overwhelming the signal. Lastly, a hybrid Parameter-Efficient Fine-Tuning (PEFT) strategy is employed, synergistically combining global adaptation via LoRA with local refinement through Post-Fusion Adapters. This significantly reduces trainable parameters, making the model lightweight and suitable for resource-limited scenarios. These innovations are integrated into a hierarchical encoder architecture, enabling PGF-Net to perform deep, dynamic, and interpretable multimodal sentiment analysis while maintaining exceptional parameter efficiency. Experimental results on MOSI dataset demonstrate that our proposed PGF-Net achieves state-of-the-art performance, with a Mean Absolute Error (MAE) of 0.691 and an F1-Score of 86.9%. Notably, our model achieves these results with only 3.09M trainable parameters, showcasing a superior balance between performance and computational efficiency.

Large language models (LLMs) have shown promise in medical question answering but often struggle with hallucinations and shallow reasoning, particularly in tasks requiring nuanced clinical understanding. Retrieval-augmented generation (RAG) offers a practical and privacy-preserving way to enhance LLMs with external medical knowledge. However, most existing approaches rely on surface-level semantic retrieval and lack the structured reasoning needed for clinical decision support. We introduce MedCoT-RAG, a domain-specific framework that combines causal-aware document retrieval with structured chain-of-thought prompting tailored to medical workflows. This design enables models to retrieve evidence aligned with diagnostic logic and generate step-by-step causal reasoning reflective of real-world clinical practice. Experiments on three diverse medical QA benchmarks show that MedCoT-RAG outperforms strong baselines by up to 10.3% over vanilla RAG and 6.4% over advanced domain-adapted methods, improving accuracy, interpretability, and consistency in complex medical tasks.