Large Language Models (LLMs) have demonstrated remarkable performance across various tasks but remain prone to hallucinations. Detecting hallucinations is essential for safety-critical applications, and recent methods leverage attention map properties to this end, though their effectiveness remains limited. In this work, we investigate the spectral features of attention maps by interpreting them as adjacency matrices of graph structures. We propose the $\text{LapEigvals}$ method, which utilises the top-$k$ eigenvalues of the Laplacian matrix derived from the attention maps as an input to hallucination detection probes. Empirical evaluations demonstrate that our approach achieves state-of-the-art hallucination detection performance among attention-based methods. Extensive ablation studies further highlight the robustness and generalisation of $\text{LapEigvals}$, paving the way for future advancements in the hallucination detection domain.

Large language models have shown remarkable reasoning abilities and scaling laws suggest that large parameter count, especially along the depth axis, is the primary driver. In this work, we make a stronger claim -- many reasoning problems require a large depth but not necessarily many parameters. This unlocks a novel application of looped models for reasoning. Firstly, we show that for many synthetic reasoning problems like addition, $p$-hop induction, and math problems, a $k$-layer transformer looped $L$ times nearly matches the performance of a $kL$-layer non-looped model, and is significantly better than a $k$-layer model. This is further corroborated by theoretical results showing that many such reasoning problems can be solved via iterative algorithms, and thus, can be solved effectively using looped models with nearly optimal depth. Perhaps surprisingly, these benefits also translate to practical settings of language modeling -- on many downstream reasoning tasks, a language model with $k$-layers looped $L$ times can be competitive to, if not better than, a $kL$-layer language model. In fact, our empirical analysis reveals an intriguing phenomenon: looped and non-looped models exhibit scaling behavior that depends on their effective depth, akin to the inference-time scaling of chain-of-thought (CoT) reasoning. We further elucidate the connection to CoT reasoning by proving that looped models implicitly generate latent thoughts and can simulate $T$ steps of CoT with $T$ loops. Inspired by these findings, we also present an interesting dichotomy between reasoning and memorization, and design a looping-based regularization that is effective on both fronts.

Event Detection (ED) is the task of identifying typed event mentions of interest from natural language text, which benefits domain-specific reasoning in biomedical, legal, and epidemiological domains. However, procuring supervised data for thousands of events for various domains is a laborious and expensive task. To this end, existing works have explored synthetic data generation via forward (generating labels for unlabeled sentences) and inverse (generating sentences from generated labels) generations. However, forward generation often produces noisy labels, while inverse generation struggles with domain drift and incomplete event annotations. To address these challenges, we introduce FIG, a hybrid approach that leverages inverse generation for high-quality data synthesis while anchoring it to domain-specific cues extracted via forward generation on unlabeled target data. FIG further enhances its synthetic data by adding missing annotations through forward generation-based refinement. Experimentation on three ED datasets from diverse domains reveals that FIG outperforms the best baseline achieving average gains of 3.3% F1 and 5.4% F1 in the zero-shot and few-shot settings respectively. Analyzing the generated trigger hit rate and human evaluation substantiates FIG's superior domain alignment and data quality compared to existing baselines.

Asking questions is a fundamental aspect of learning that facilitates deeper understanding. However, characterizing and crafting questions that effectively improve learning remains elusive. To address this gap, we propose QUEST (Question Utility Estimation with Simulated Tests). QUEST simulates a learning environment that enables the quantification of a question's utility based on its direct impact on improving learning outcomes. Furthermore, we can identify high-utility questions and use them to fine-tune question generation models with rejection sampling. We find that questions generated by models trained with rejection sampling based on question utility result in exam scores that are higher by at least 20% than those from specialized prompting grounded on educational objectives literature and models fine-tuned with indirect measures of question quality, such as saliency and expected information gain.

Continuous Integration (CI) is a cornerstone of modern software development. However, while widely adopted in traditional software projects, applying CI practices to Machine Learning (ML) projects presents distinctive characteristics. For example, our previous work revealed that ML projects often experience longer build durations and lower test coverage rates compared to their non-ML counterparts. Building on these quantitative findings, this study surveys 155 practitioners from 47 ML projects to investigate the underlying reasons for these distinctive characteristics through a qualitative perspective. Practitioners highlighted eight key differences, including test complexity, infrastructure requirements, and build duration and stability. Common challenges mentioned by practitioners include higher project complexity, model training demands, extensive data handling, increased computational resource needs, and dependency management, all contributing to extended build durations. Furthermore, ML systems' non-deterministic nature, data dependencies, and computational constraints were identified as significant barriers to effective testing. The key takeaway from this study is that while foundational CI principles remain valuable, ML projects require tailored approaches to address their unique challenges. To bridge this gap, we propose a set of ML-specific CI practices, including tracking model performance metrics and prioritizing test execution within CI pipelines. Additionally, our findings highlight the importance of fostering interdisciplinary collaboration to strengthen the testing culture in ML projects. By bridging quantitative findings with practitioners' insights, this study provides a deeper understanding of the interplay between CI practices and the unique demands of ML projects, laying the groundwork for more efficient and robust CI strategies in this domain.

Tabular datasets are inherently heterogeneous, posing significant challenges for developing pre-trained foundation models. The recently introduced transformer-based Tabular Prior-data Fitted Network v2 (TabPFN v2) achieves unprecedented in-context learning accuracy across multiple tabular datasets, marking a pivotal advancement in tabular foundation models. In this paper, we comprehensively evaluate TabPFN v2 on over 300 datasets, confirming its exceptional generalization capabilities on small- to medium-scale tasks. Our analysis identifies randomized feature tokens as a key factor behind TabPFN v2's success, as they unify heterogeneous datasets into a fixed-dimensional representation, enabling more effective training and inference. To further understand TabPFN v2's predictions, we propose a leave-one-fold-out approach, transforming TabPFN v2 into a feature extractor and revealing its capability to simplify data distributions and boost accuracy. Lastly, to address TabPFN v2's limitations in high-dimensional, large-scale, and many-category tasks, we introduce a divide-and-conquer mechanism inspired by Chain-of-Thought prompting, enabling scalable inference. By uncovering the mechanisms behind TabPFN v2's success and introducing strategies to expand its applicability, this study provides key insights into the future of tabular foundation models.

Neurological disorders represent significant global health challenges, driving the advancement of brain signal analysis methods. Scalp electroencephalography (EEG) and intracranial electroencephalography (iEEG) are widely used to diagnose and monitor neurological conditions. However, dataset heterogeneity and task variations pose challenges in developing robust deep learning solutions. This review systematically examines recent advances in deep learning approaches for EEG/iEEG-based neurological diagnostics, focusing on applications across 7 neurological conditions using 46 datasets. We explore trends in data utilization, model design, and task-specific adaptations, highlighting the importance of pre-trained multi-task models for scalable, generalizable solutions. To advance research, we propose a standardized benchmark for evaluating models across diverse datasets to enhance reproducibility. This survey emphasizes how recent innovations can transform neurological diagnostics and enable the development of intelligent, adaptable healthcare solutions.

Majorities of distillation methods on pre-trained diffusion models or on pre-trained rectified flow, focus on either the distillation outputs or the trajectories between random noises and clean images to speed up sample generations from pre-trained models. In those trajectory-based distillation methods, consistency distillation requires the self-consistent trajectory projection to regulate the trajectory, which might avoid the common ODE approximation error {while still be concerning about sampling efficiencies}. At the same time, rectified flow distillations enforce straight trajectory for fast sampling, although an ODE solver is still required. In this work, we propose a trajectory distillation method, \modelname, that enjoys the benefits of both and enables few-step generations. TraFlow adopts the settings of consistency trajectory models, and further enforces the properties of self-consistency and straightness throughout the entire trajectory. These two properties are pursued by reaching a balance with following three targets: (1) reconstruct the output from pre-trained models; (2) learn the amount of changes by pre-trained models; (3) satisfy the self-consistency over its trajectory. Extensive experimental results have shown the effectiveness of our proposed method.

Robots are prone to making errors, which can negatively impact their credibility as teammates during collaborative tasks with human users. Detecting and recovering from these failures is crucial for maintaining effective level of trust from users. However, robots may fail without being aware of it. One way to detect such failures could be by analysing humans' non-verbal behaviours and reactions to failures. This study investigates how human gaze dynamics can signal a robot's failure and examines how different types of failures affect people's perception of robot. We conducted a user study with 27 participants collaborating with a robotic mobile manipulator to solve tangram puzzles. The robot was programmed to experience two types of failures -- executional and decisional -- occurring either at the beginning or end of the task, with or without acknowledgement of the failure. Our findings reveal that the type and timing of the robot's failure significantly affect participants' gaze behaviour and perception of the robot. Specifically, executional failures led to more gaze shifts and increased focus on the robot, while decisional failures resulted in lower entropy in gaze transitions among areas of interest, particularly when the failure occurred at the end of the task. These results highlight that gaze can serve as a reliable indicator of robot failures and their types, and could also be used to predict the appropriate recovery actions.

Large Language Models (LLMs) have recently demonstrated remarkable performance in tasks such as Retrieval-Augmented Generation (RAG) and autonomous AI agent workflows. Yet, when faced with large sets of unstructured documents requiring progressive exploration, analysis, and synthesis, such as conducting literature survey, existing approaches often fall short. We address this challenge -- termed Progressive Document Investigation -- by introducing Graphy, an end-to-end platform that automates data modeling, exploration and high-quality report generation in a user-friendly manner. Graphy comprises an offline Scrapper that transforms raw documents into a structured graph of Fact and Dimension nodes, and an online Surveyor that enables iterative exploration and LLM-driven report generation. We showcase a pre-scrapped graph of over 50,000 papers -- complete with their references -- demonstrating how Graphy facilitates the literature-survey scenario. The demonstration video can be found at https://youtu.be/uM4nzkAdGlM.