A recurring challenge in preference fine-tuning (PFT) is handling $\textit{intransitive}$ (i.e., cyclic) preferences. Intransitive preferences often stem from either $\textit{(i)}$ inconsistent rankings along a single objective or $\textit{(ii)}$ scalarizing multiple objectives into a single metric. Regardless of their source, the downstream implication of intransitive preferences is the same: there is no well-defined optimal policy, breaking a core assumption of the standard PFT pipeline. In response, we propose a novel, game-theoretic solution concept -- the $\textit{Maximum Entropy Blackwell Winner}$ ($\textit{MaxEntBW}$) -- that is well-defined under multi-objective intransitive preferences. To enable computing MaxEntBWs at scale, we derive $\texttt{PROSPER}$: a provably efficient PFT algorithm. Unlike prior self-play techniques, $\texttt{PROSPER}$ directly handles multiple objectives without requiring scalarization. We then apply $\texttt{PROSPER}$ to the problem of fine-tuning large language models (LLMs) from multi-objective LLM-as-a-Judge feedback (e.g., rubric-based judges), a setting where both sources of intransitivity arise. We find that $\texttt{PROSPER}$ outperforms all baselines considered across both instruction following and general chat benchmarks, releasing trained model checkpoints at the 7B and 3B parameter scales.

Many of the challenges encountered in in-the-wild public deployments of robots remain undocumented despite sharing many common pitfalls. This creates a high barrier of entry and results in repetition of avoidable mistakes. To articulate the tacit knowledge in the HRI community, this paper presents a guideline in the form of a checklist to support researchers in preparing for robot deployments in public. Drawing on their own experience with public robot deployments, the research team collected essential topics to consider in public HRI research. These topics are represented as modular flip cards in a hierarchical table, structured into deployment phases and important domains. We interviewed six interdisciplinary researchers with expertise in public HRI and show how including community input refines the checklist. We further show the checklist in action in context of real public studies. Finally, we contribute the checklist as an open-source, customizable community resource that both collects joint expertise for continual evolution and is usable as a list, set of cards, and an interactive web tool.

AI training datasets will inevitably contain AI-generated examples, leading to ``feedback'' in which the output of one model impacts the training of another. It is known that such iterative feedback can lead to model collapse, yet the mechanisms underlying this degeneration remain poorly understood. Here we show that a broad class of feedback processes converges to a low-dimensional invariant structure in latent space, a phenomenon we call neural resonance. By modeling iterative feedback as a Markov Chain, we show that two conditions are needed for this resonance to occur: ergodicity of the feedback process and directional contraction of the latent representation. By studying diffusion models on MNIST and ImageNet, as well as CycleGAN and an audio feedback experiment, we map how local and global manifold geometry evolve, and we introduce an eight-pattern taxonomy of collapse behaviors. Neural resonance provides a unified explanation for long-term degenerate behavior in generative models and provides practical diagnostics for identifying, characterizing, and eventually mitigating collapse.

Inverse lithography (ILT) is critical for modern semiconductor manufacturing but suffers from highly non-convex objectives that often trap optimization in poor local minima. Generative AI has been explored to warm-start ILT, yet most approaches train deterministic image-to-image translators to mimic sub-optimal datasets, providing limited guidance for escaping non-convex traps during refinement. We reformulate mask synthesis as conditional sampling: a generator learns a distribution over masks conditioned on the design and proposes multiple candidates. The generator is first pretrained with WGAN plus a reconstruction loss, then fine-tuned using Group Relative Policy Optimization (GRPO) with an ILT-guided imitation loss. At inference, we sample a small batch of masks, run fast batched ILT refinement, evaluate lithography metrics (e.g., EPE, process window), and select the best candidate. On \texttt{LithoBench} dataset, the proposed hybrid framework reduces EPE violations under a 3\,nm tolerance and roughly doubles throughput versus a strong numerical ILT baseline, while improving final mask quality. We also present over 20\% EPE improvement on \texttt{ICCAD13} contest cases with 3$\times$ speedup over the SOTA numerical ILT solver. By learning to propose ILT-friendly initializations, our approach mitigates non-convexity and advances beyond what traditional solvers or GenAI can achieve.

Prompt tuning of large-scale vision-language models such as CLIP enables efficient task adaptation without updating model weights. However, it often leads to poor confidence calibration and unreliable predictive uncertainty. We address this problem by proposing a calibration framework that enhances predictive reliability while preserving the geometry of the pretrained CLIP embedding space, which is required for robust generalization. Our approach extends the standard cross-entropy loss with two complementary regularizers: (1) a mean-variance margin penalty that stabilizes inter-class logit margins by maximizing their average while minimizing dispersion, mitigating underconfidence and overconfidence spikes; and (2) a text moment-matching loss that aligns the first and second moments of tuned text embeddings with their frozen CLIP counterparts, preserving semantic dispersion crucial for generalization. Through extensive experiments across 7 prompt-tuning methods and 11 diverse datasets, we demonstrate that our approach significantly reduces the Expected Calibration Error (ECE) compared to competitive calibration techniques on both base and novel classes

Accurate sex identification in fish is vital for optimizing breeding and management strategies in aquaculture, particularly for species at the risk of extinction. However, most existing methods are invasive or stressful and may cause additional mortality, posing severe risks to threatened or endangered fish populations. To address these challenges, we propose FishProtoNet, a robust, non-invasive computer vision-based framework for sex identification of delta smelt (Hypomesus transpacificus), an endangered fish species native to California, across its full life cycle. Unlike the traditional deep learning methods, FishProtoNet provides interpretability through learned prototype representations while improving robustness by leveraging foundation models to reduce the influence of background noise. Specifically, the FishProtoNet framework consists of three key components: fish regions of interest (ROIs) extraction using visual foundation model, feature extraction from fish ROIs and fish sex identification based on an interpretable prototype network. FishProtoNet demonstrates strong performance in delta smelt sex identification during early spawning and post-spawning stages, achieving the accuracies of 74.40% and 81.16% and corresponding F1 scores of 74.27% and 79.43% respectively. In contrast, delta smelt sex identification at the subadult stage remains challenging for current computer vision methods, likely due to less pronounced morphological differences in immature fish. The source code of FishProtoNet is publicly available at: https://github.com/zhengmiao1/Fish_sex_identification

Detecting LLM training data is generally framed as a membership inference attack (MIA) problem. However, conventional MIAs operate passively on fixed model weights, using log-likelihoods or text generations. In this work, we introduce \textbf{Active Data Reconstruction Attack} (ADRA), a family of MIA that actively induces a model to reconstruct a given text through training. We hypothesize that training data are \textit{more reconstructible} than non-members, and the difference in their reconstructibility can be exploited for membership inference. Motivated by findings that reinforcement learning (RL) sharpens behaviors already encoded in weights, we leverage on-policy RL to actively elicit data reconstruction by finetuning a policy initialized from the target model. To effectively use RL for MIA, we design reconstruction metrics and contrastive rewards. The resulting algorithms, \textsc{ADRA} and its adaptive variant \textsc{ADRA+}, improve both reconstruction and detection given a pool of candidate data. Experiments show that our methods consistently outperform existing MIAs in detecting pre-training, post-training, and distillation data, with an average improvement of 10.7\% over the previous runner-up. In particular, \MethodPlus~improves over Min-K\%++ by 18.8\% on BookMIA for pre-training detection and by 7.6\% on AIME for post-training detection.

Theoretical analyses of Empirical Risk Minimization (ERM) are standardly framed within the Real-RAM model of computation. In this setting, training even simple neural networks is known to be $\exists \mathbb{R}$-complete -- a complexity class believed to be harder than NP, that characterizes the difficulty of solving systems of polynomial inequalities over the real numbers. However, this algebraic framework diverges from the reality of digital computation with finite-precision hardware. In this work, we analyze the theoretical complexity of ERM under a realistic bit-level model ($\mathsf{ERM}_{\text{bit}}$), where network parameters and inputs are constrained to be rational numbers with polynomially bounded bit-lengths. Under this model, we reveal a sharp dichotomy in tractability governed by the network's activation function. We prove that for deep networks with {\em any} polynomial activations with rational coefficients and degree at least $2$, the bit-complexity of training is severe: deciding $\mathsf{ERM}_{\text{bit}}$ is $\#P$-Hard, hence believed to be strictly harder than NP-complete problems. Furthermore, we show that determining the sign of a single partial derivative of the empirical loss function is intractable (unlikely in BPP), and deciding a specific bit in the gradient is $\#P$-Hard. This provides a complexity-theoretic perspective for the phenomenon of exploding and vanishing gradients. In contrast, we show that for piecewise-linear activations such as ReLU, the precision requirements remain manageable: $\mathsf{ERM}_{\text{bit}}$ is contained within NP (specifically NP-complete), and standard backpropagation runs in polynomial time. Our results demonstrate that finite-precision constraints are not merely implementation details but fundamental determinants of learnability.

Why do language agents fail on tasks they are capable of solving? We argue that many such failures are reliability failures caused by stochastic drift from a task's latent solution structure, not capability failures. Every well-defined tool-use task imposes a canonical solution path (i.e., a convergent set of tool invocations shared across successful runs) and agent success depends critically on whether a trajectory stays within this path's operating envelope. We establish this causally using a natural experiment that holds model capability and task difficulty fixed by construction. We analyze trajectories from the Toolathlon benchmark: 22 frontier models each attempt 108 real-world tool-use tasks across 3 independent runs, yielding 515 model$\times$task units where the same model succeeds on some runs and fails on others due to LLM sampling stochasticity alone. Within these units, successful runs adhere significantly more closely to the canonical solution path than failed runs ($+$0.060 Jaccard, $p<0.0001$, $n=488$ units, 95% CI [+0.043, +0.077]). This result survives six robustness checks including cross-model-family leave-one-out validation. Critically, the causal mechanism is gradual and self-reinforcing: the adherence gap is statistically indistinguishable from zero through the first 50% of the trajectory, ruling out early-branching selection bias, and each off-canonical tool call raises the probability that the next call is also off-canonical by 22.7 percentage points ($\hatβ=+0.227$, $p<0.0001$), more than doubling the baseline rate. These findings imply that agent reliability cannot be improved by capability scaling alone, but offer a highly actionable intervention: a simple monitor that restarts the bottom tercile of runs based on mid-trajectory canonical adherence lifts success rates by $+$8.8 percentage points among intervened runs.

We present a systematic evaluation of large language models on quantum mechanics problem-solving. Our study evaluates 15 models from five providers (OpenAI, Anthropic, Google, Alibaba, DeepSeek) spanning three capability tiers on 20 tasks covering derivations, creative problems, non-standard concepts, and numerical computation, comprising 900 baseline and 75 tool-augmented assessments. Results reveal clear tier stratification: flagship models achieve 81\% average accuracy, outperforming mid-tier (77\%) and fast models (67\%) by 4pp and 14pp respectively. Task difficulty patterns emerge distinctly: derivations show highest performance (92\% average, 100\% for flagship models), while numerical computation remains most challenging (42\%). Tool augmentation on numerical tasks yields task-dependent effects: modest overall improvement (+4.4pp) at 3x token cost masks dramatic heterogeneity ranging from +29pp gains to -16pp degradation. Reproducibility analysis across three runs quantifies 6.3pp average variance, with flagship models demonstrating exceptional stability (GPT-5 achieves zero variance) while specialized models require multi-run evaluation. This work contributes: (i) a benchmark for quantum mechanics with automatic verification, (ii) systematic evaluation quantifying tier-based performance hierarchies, (iii) empirical analysis of tool augmentation trade-offs, and (iv) reproducibility characterization. All tasks, verifiers, and results are publicly released.

Purpose: Non-invasive grading of prostate cancer (PCa) from micro-ultrasound (micro-US) could expedite triage and guide biopsies toward the most aggressive regions, yet current models struggle to infer tissue micro-structure at coarse imaging resolutions. Methods: We introduce an unpaired histopathology knowledge-distillation strategy that trains a micro-US encoder to emulate the embedding distribution of a pretrained histopathology foundation model, conditioned on International Society of Urological Pathology (ISUP) grades. Training requires no patient-level pairing or image registration, and histopathology inputs are not used at inference. Results: Compared to the current state of the art, our approach increases sensitivity to clinically significant PCa (csPCa) at 60% specificity by 3.5% and improves overall sensitivity at 60% specificity by 1.2%. Conclusion: By enabling earlier and more dependable cancer risk stratification solely from imaging, our method advances clinical feasibility. Source code will be publicly released upon publication.

We aim to learn a joint representation between inertial measurement unit (IMU) signals and 2D pose sequences extracted from video, enabling accurate cross-modal retrieval, temporal synchronization, subject and body-part localization, and action recognition. To this end, we introduce MoBind, a hierarchical contrastive learning framework designed to address three challenges: (1) filtering out irrelevant visual background, (2) modeling structured multi-sensor IMU configurations, and (3) achieving fine-grained, sub-second temporal alignment. To isolate motion-relevant cues, MoBind aligns IMU signals with skeletal motion sequences rather than raw pixels. We further decompose full-body motion into local body-part trajectories, pairing each with its corresponding IMU to enable semantically grounded multi-sensor alignment. To capture detailed temporal correspondence, MoBind employs a hierarchical contrastive strategy that first aligns token-level temporal segments, then fuses local (body-part) alignment with global (body-wide) motion aggregation. Evaluated on mRi, TotalCapture, and EgoHumans, MoBind consistently outperforms strong baselines across all four tasks, demonstrating robust fine-grained temporal alignment while preserving coarse semantic consistency across modalities. Code is available at https://github.com/bbvisual/ MoBind.

The powerful reasoning of modern Vision Language Models open a new frontier for advanced personalization study. However, progress in this area is critically hampered by the lack of suitable benchmarks. To address this gap, we introduce Life-Bench, a comprehensive, synthetically generated multimodal benchmark built on simulated user digital footprints. Life-Bench features over questions evaluating a wide spectrum of capabilities, from persona understanding to complex reasoning over historical data. These capabilities expand far beyond prior benchmarks, reflecting the critical demands essential for real-world applications. Furthermore, we propose LifeGraph, an end-to-end framework that organizes personal context into a knowledge graph to facilitate structured retrieval and reasoning. Our experiments on Life-Bench reveal that existing methods falter significantly on complex personalized tasks, exposing a large performance headroom, especially in relational, temporal and aggregative reasoning. While LifeGraph closes this gap by leveraging structured knowledge and demonstrates a promising direction, these advanced personalization tasks remain a critical open challenge, motivating new research in this area.

LLM agents are increasingly expected to function as general-purpose systems capable of resolving open-ended user requests. While existing benchmarks focus on domain-aware environments for developing specialized agents, evaluating general-purpose agents requires more realistic settings that challenge them to operate across multiple skills and tools within a unified environment. We introduce General AgentBench, a benchmark that provides such a unified framework for evaluating general LLM agents across search, coding, reasoning, and tool-use domains. Using General AgentBench, we systematically study test-time scaling behaviors under sequential scaling (iterative interaction) and parallel scaling (sampling multiple trajectories). Evaluation of ten leading LLM agents reveals a substantial performance degradation when moving from domain-specific evaluations to this general-agent setting. Moreover, we find that neither scaling methodology yields effective performance improvements in practice, due to two fundamental limitations: context ceiling in sequential scaling and verification gap in parallel scaling. Code is publicly available at https://github.com/cxcscmu/General-AgentBench.

The automation of fruit harvesting has gained increasing significance in response to rising labor shortages. A sensorized gripper is a key component of this process, which must be compact enough for confined spaces, able to stably grasp diverse fruits, and provide reliable feedback on fruit conditions for efficient harvesting. To address this need, we propose FruitTouch, a compact gripper that integrates high-resolution, vision-based tactile sensing through an optimized optical design. This configuration accommodates a wide range of fruit sizes while maintaining low cost and mechanical simplicity. Tactile images captured by an embedded camera provide rich information for real-time force estimation, slip detection, and softness prediction. We validate the gripper in real-world fruit harvesting experiments, demonstrating robust grasp stability and effective damage prevention.

Organizations across finance, healthcare, transportation, content moderation, and critical infrastructure are rapidly deploying highly automated AI systems, yet they lack principled methods to quantify how increasing automation amplifies harm when failures occur. We propose a parsimonious Bayesian risk decomposition expressing expected loss as the product of three terms: the probability of system failure, the conditional probability that a failure propagates into harm given the automation level, and the expected severity of harm. This framework isolates a critical quantity -- the conditional probability that failures propagate into harm -- which captures execution and oversight risk rather than model accuracy alone. We develop complete theoretical foundations: formal proofs of the decomposition, a harm propagation equivalence theorem linking the harm propagation probability to observable execution controls, risk elasticity measures, efficient frontier analysis for automation policy, and optimal resource allocation principles with second-order conditions. We motivate the framework with an illustrative case study of the 2012 Knight Capital incident ($440M loss) as one instantiation of a broadly applicable failure pattern, and characterize the research design required to empirically validate the framework at scale across deployment domains. This work provides the theoretical foundations for a new class of deployment-focused risk governance tools for agentic and automated AI systems.

Infrared radiation computing underpins advances in climate science, remote sensing and spectroscopy but remains constrained by manual workflows. We introduce InfEngine, an autonomous intelligent computational engine designed to drive a paradigm shift from human-led orchestration to collaborative automation. It integrates four specialized agents through two core innovations: self-verification, enabled by joint solver-evaluator debugging, improves functional correctness and scientific plausibility; self-optimization, realized via evolutionary algorithms with self-discovered fitness functions, facilitates autonomous performance optimization. Evaluated on InfBench with 200 infrared-specific tasks and powered by InfTools with 270 curated tools, InfEngine achieves a 92.7% pass rate and delivers workflows 21x faster than manual expert effort. More fundamentally, it illustrates how researchers can transition from manual coding to collaborating with self-verifying, self-optimizing computational partners. By generating reusable, verified and optimized code, InfEngine transforms computational workflows into persistent scientific assets, accelerating the cycle of scientific discovery. Code: https://github.com/kding1225/infengine

Modern 3D game levels rely heavily on visual guidance, yet the navigability of level layouts remains difficult to quantify. Prior work either simulates play in simplified environments or analyzes static screenshots for visual affordances, but neither setting faithfully captures how players explore complex, real-world game levels. In this paper, we build on an existing open-source visual affordance detector and instantiate a screen-only exploration and navigation agent that operates purely from visual affordances. Our agent consumes live game frames, identifies salient interest points, and drives a simple finite-state controller over a minimal action space to explore Dark Souls-style linear levels and attempt to reach expected goal regions. Pilot experiments show that the agent can traverse most required segments and exhibits meaningful visual navigation behavior, but also highlight that limitations of the underlying visual model prevent truly comprehensive and reliable auto-navigation. We argue that this system provides a concrete, shared baseline and evaluation protocol for visual navigation in complex games, and we call for more attention to this necessary task. Our results suggest that purely vision-based sense-making models, with discrete single-modality inputs and without explicit reasoning, can effectively support navigation and environment understanding in idealized settings, but are unlikely to be a general solution on their own.

Adapting image-pretrained backbones to video typically relies on time-domain adapters tuned to a single temporal scale. Our experiments show that these modules pick up static image cues and very fast flicker changes, while overlooking medium-speed motion. Capturing dynamics across multiple time-scales is, however, crucial for fine-grained temporal analysis (i.e., opening vs. closing bottle). To address this, we introduce Frame2Freq -- a family of frequency-aware adapters that perform spectral encoding during image-to-video adaptation of pretrained Vision Foundation Models (VFMs), improving fine-grained action recognition. Frame2Freq uses Fast Fourier Transform (FFT) along time and learns frequency-band specific embeddings that adaptively highlight the most discriminative frequency ranges. Across five fine-grained activity recognition datasets, Frame2Freq outperforms prior PEFT methods and even surpasses fully fine-tuned models on four of them. These results provide encouraging evidence that frequency analysis methods are a powerful tool for modeling temporal dynamics in image-to-video transfer. Code is available at https://github.com/th-nesh/Frame2Freq.

Aerial robots are evolving from avoiding obstacles to exploiting the environmental contact interactions for navigation, exploration and manipulation. A key challenge in such aerial physical interactions lies in handling uncertain contact forces on unknown targets, which typically demand accurate sensing and active control. We present a drone platform with elastic horns that enables touch-and-go manoeuvres - a self-regulated, consecutive bumping motion that allows the drone to maintain proximity to a wall without relying on active obstacle avoidance. It leverages environmental interaction as a form of embodied control, where low-level stabilisation and near-obstacle navigation emerge from the passive dynamic responses of the drone-obstacle system that resembles a mass-spring-damper system. Experiments show that the elastic horn can absorb impact energy while maintaining vehicle stability, reducing pitch oscillations by 38% compared to the rigid horn configuration. The lower horn arrangement was found to reduce pitch oscillations by approximately 54%. In addition to intermittent contact, the platform equipped with elastic horns also demonstrates stable, sustained contact with static objects, relying on a standard attitude PID controller.

Preference-driven behavior in LLMs may be a necessary precondition for AI misalignment such as sandbagging: models cannot strategically pursue misaligned goals unless their behavior is influenced by their preferences. Yet prior work has typically prompted models explicitly to act in specific ways, leaving unclear whether observed behaviors reflect instruction-following capabilities vs underlying model preferences. Here we test whether this precondition for misalignment is present. Using entity preferences as a behavioral probe, we measure whether stated preferences predict downstream behavior in five frontier LLMs across three domains: donation advice, refusal behavior, and task performance. Conceptually replicating prior work, we first confirm that all five models show highly consistent preferences across two independent measurement methods. We then test behavioral consequences in a simulated user environment. We find that all five models give preference-aligned donation advice. All five models also show preference-correlated refusal patterns when asked to recommend donations, refusing more often for less-preferred entities. All preference-related behaviors that we observe here emerge without instructions to act on preferences. Results for task performance are mixed: on a question-answering benchmark (BoolQ), two models show small but significant accuracy differences favoring preferred entities; one model shows the opposite pattern; and two models show no significant relationship. On complex agentic tasks, we find no evidence of preference-driven performance differences. While LLMs have consistent preferences that reliably predict advice-giving behavior, these preferences do not consistently translate into downstream task performance.

Accurate perception of object hardness is essential for safe and dexterous contact-rich robotic manipulation. Here, we present TactEx, an explainable multimodal robotic interaction framework that unifies vision, touch, and language for human-like hardness estimation and interactive guidance. We evaluate TactEx on fruit-ripeness assessment, a representative task that requires both tactile sensing and contextual understanding. The system fuses GelSight-Mini tactile streams with RGB observations and language prompts. A ResNet50+LSTM model estimates hardness from sequential tactile data, while a cross-modal alignment module combines visual cues with guidance from a large language model (LLM). This explainable multimodal interface allows users to distinguish ripeness levels with statistically significant class separation (p < 0.01 for all fruit pairs). For touch placement, we compare YOLO with Grounded-SAM (GSAM) and find GSAM to be more robust for fine-grained segmentation and contact-site selection. A lightweight LLM parses user instructions and produces grounded natural-language explanations linked to the tactile outputs. In end-to-end evaluations, TactEx attains 90% task success on simple user queries and generalises to novel tasks without large-scale tuning. These results highlight the promise of combining pretrained visual and tactile models with language grounding to advance explainable, human-like touch perception and decision-making in robotics.

Domain-specific speech remains a persistent challenge for automatic speech recognition (ASR), even for state-of-the-art systems like OpenAI's Whisper. We introduce Whisper: Courtside Edition, a novel multi-agent large language model (LLM) pipeline that enhances Whisper transcriptions without retraining. The pipeline intercepts Whisper's initial transcript, applies specialized LLM agents for domain context identification, named entity recognition, and jargon detection, and generates compact prompts that guide Whisper's decoder. Evaluated on 421 NBA basketball commentary segments (a domain characterized by dense proper nouns and technical terminology) our best pipeline achieves a statistically significant 17.0% relative reduction in word error rate (WER; from 0.217 to 0.180, p<0.001). Improvements are observed in 40.1% of segments with degradation in only 7.1%, substantially outperforming direct transcript post-editing. These results demonstrate that prompt-based augmentation can deliver scalable domain adaptation for ASR, offering a practical alternative to costly model fine-tuning.

Face presentation attack detection (FacePAD) is critical for securing facial authentication against print, replay, and mask-based spoofing. This paper proposes CASO-PAD, an RGB-only, single-frame model that enhances MobileNetV3 with content-adaptive spatial operators (involution) to better capture localized spoof cues. Unlike spatially shared convolution kernels, the proposed operator generates location-specific, channel-shared kernels conditioned on the input, improving spatial selectivity with minimal overhead. CASO-PAD remains lightweight (3.6M parameters; 0.64 GFLOPs at $256\times256$) and is trained end-to-end using a standard binary cross-entropy objective. Extensive experiments on Replay-Attack, Replay-Mobile, ROSE-Youtu, and OULU-NPU demonstrate strong performance, achieving 100/100/98.9/99.7\% test accuracy, AUC of 1.00/1.00/0.9995/0.9999, and HTER of 0.00/0.00/0.82/0.44\%, respectively. On the large-scale SiW-Mv2 Protocol-1 benchmark, CASO-PAD further attains 95.45\% accuracy with 3.11\% HTER and 3.13\% EER, indicating improved robustness under diverse real-world attacks. Ablation studies show that placing the adaptive operator near the network head and using moderate group sharing yields the best accuracy--efficiency balance. Overall, CASO-PAD provides a practical pathway for robust, on-device FacePAD with mobile-class compute and without auxiliary sensors or temporal stacks.

Sarcasm detection poses a fundamental challenge in computational semantics, requiring models to resolve disparities between literal and intended meaning. The challenge is amplified in low-resource languages where annotated datasets are scarce or nonexistent. We present \textbf{Yor-Sarc}, the first gold-standard dataset for sarcasm detection in Yorùbá, a tonal Niger-Congo language spoken by over $50$ million people. The dataset comprises 436 instances annotated by three native speakers from diverse dialectal backgrounds using an annotation protocol specifically designed for Yorùbá sarcasm by taking culture into account. This protocol incorporates context-sensitive interpretation and community-informed guidelines and is accompanied by a comprehensive analysis of inter-annotator agreement to support replication in other African languages. Substantial to almost perfect agreement was achieved (Fleiss' $κ= 0.7660$; pairwise Cohen's $κ= 0.6732$--$0.8743$), with $83.3\%$ unanimous consensus. One annotator pair achieved almost perfect agreement ($κ= 0.8743$; $93.8\%$ raw agreement), exceeding a number of reported benchmarks for English sarcasm research works. The remaining $16.7\%$ majority-agreement cases are preserved as soft labels for uncertainty-aware modelling. Yor-Sarc\footnote{https://github.com/toheebadura/yor-sarc} is expected to facilitate research on semantic interpretation and culturally informed NLP for low-resource African languages.

This paper presents a depth-enhanced YOLO-SAM2 framework for detecting ballast insufficiency in railway tracks using RGB-D data. Although YOLOv8 provides reliable localization, the RGB-only model shows limited safety performance, achieving high precision (0.99) but low recall (0.49) due to insufficient ballast, as it tends to over-predict the sufficient class. To improve reliability, we incorporate depth-based geometric analysis enabled by a sleeper-aligned depth-correction pipeline that compensates for RealSense spatial distortion using polynomial modeling, RANSAC, and temporal smoothing. SAM2 segmentation further refines region-of-interest masks, enabling accurate extraction of sleeper and ballast profiles for geometric classification. Experiments on field-collected top-down RGB-D data show that depth-enhanced configurations substantially improve the detection of insufficient ballast. Depending on bounding-box sampling (AABB or RBB) and geometric criteria, recall increases from 0.49 to as high as 0.80, and F1-score improves from 0.66 to over 0.80. These results demonstrate that integrating depth correction with YOLO-SAM2 yields a more robust and reliable approach for automated railway ballast inspection, particularly in visually ambiguous or safety-critical scenarios.

The remarkable performance of modern AI systems has been driven by unprecedented scales of data, computation, and energy -- far exceeding the resources required by human intelligence. This disparity highlights the need for new guiding principles and motivates drawing inspiration from the fundamental organizational principles of brain computation. Among these principles, modularity has been shown to be critical for supporting the efficient learning and strong generalization abilities consistently exhibited by humans. Furthermore, modularity aligns well with the No Free Lunch Theorem, which highlights the need for problem-specific inductive biases and motivates architectures composed of specialized components that solve subproblems. However, despite its fundamental role in natural intelligence and its demonstrated benefits across a range of seemingly disparate AI subfields, modularity remains relatively underappreciated in mainstream AI research. In this work, we review several research threads in artificial intelligence and neuroscience through a conceptual framework that highlights the central role of modularity in supporting both artificial and natural intelligence. In particular, we examine what computational advantages modularity provides, how it has emerged as a solution across several AI research areas, which modularity principles the brain exploits, and how modularity can help bridge the gap between natural and artificial intelligence.

Real-time hand tracking in trauma surgery is essential for supporting rapid and precise intraoperative decisions. We propose a YOLOv10-based framework that simultaneously localizes hands and classifies their laterality (left or right) in complex surgical scenes. The model is trained on the Trauma THOMPSON Challenge 2025 Task 2 dataset, consisting of first-person surgical videos with annotated hand bounding boxes. Extensive data augmentation and a multi-task detection design improve robustness against motion blur, lighting variations, and diverse hand appearances. Evaluation demonstrates accurate left-hand (67\%) and right-hand (71\%) classification, while distinguishing hands from the background remains challenging. The model achieves an $mAP_{[0.5:0.95]}$ of 0.33 and maintains real-time inference, highlighting its potential for intraoperative deployment. This work establishes a foundation for advanced hand-instrument interaction analysis in emergency surgical procedures.

This paper addresses the problem of temporal logic motion planning for an autonomous robot operating in an unknown environment. The objective is to enable the robot to satisfy a syntactically co-safe Linear Temporal Logic (scLTL) specification when the exact locations of the desired labels are not known a priori. We introduce a new type of automaton state, referred to as commit states. These states capture intermediate task progress resulting from actions whose consequences are irreversible. In other words, certain future paths to satisfaction become not feasible after taking those actions that lead to the commit states. By leveraging commit states, we propose a sound and complete frontier-based exploration algorithm that strategically guides the robot to make progress toward the task while preserving all possible ways of satisfying it. The efficacy of the proposed method is validated through simulations.

Transformer models contain substantial internal redundancy arising from coordinate-dependent representations and continuous symmetries, in model space and in head space, respectively. While recent approaches address this by explicitly breaking symmetry, we propose a complementary framework based on symmetry reduction. We reformulate representations, attention mechanisms, and optimization dynamics in terms of invariant relational quantities, eliminating redundant degrees of freedom by construction. This perspective yields architectures that operate directly on relational structures, providing a principled geometric framework for reducing parameter redundancy and analyzing optimization.