Biopharmaceutical manufacturing organizations operate under regulatory frameworks such as FDA guidance, EU Good Manufacturing Practice (GMP), and the EU AI Act, which can restrict the use of cloud-based artificial intelligence systems. Locally deployed large language models (LLMs) offer a privacy-preserving alternative, but their suitability for pharmaceutical manufacturing tasks remains underexplored. This study evaluates four open-source LLMs (Qwen 2.5 Coder 7B, Llama 3.1 8B, Mistral 7B, and Meditron 7B) deployed locally via Ollama for natural-language-to-SQL generation over a pharmaceutical manufacturing database. A FastAPI-based evaluation platform, PharmaBatchDB AI, was developed using a synthetic Microsoft SQL Server database containing approximately 63,000 records across Batch, Manufacturing Execution System (MES), and Clean-In-Place (CIP) modules. Models were benchmarked on 60 domain-specific natural-language questions using metrics including SQL extraction rate, SQL compliance, factual consistency, ROUGE-L, hallucination rate, throughput, and latency. Qwen 2.5 Coder 7B, Llama 3.1 8B, and Mistral 7B generated SQL for all evaluation tasks, while Meditron 7B failed on nearly all tasks due to context-window limitations and poor SQL generation capability. Llama 3.1 8B achieved the highest SQL compliance, whereas Qwen 2.5 Coder 7B achieved the strongest overall text similarity and factual consistency. Performance differences between the two leading models were not statistically significant. The results show that code-tuned general-purpose LLMs outperform a domain-specific biomedical model on structured query generation for pharmaceutical manufacturing data. Although fully local, GxP-aligned NLQ systems are feasible on consumer hardware, current performance levels still require human oversight and downstream validation for regulated use.

Scientific discovery demands intelligence, perseverance, and serendipity across vast search spaces. Today, top scientific capabilities remain siloed--one AI system for biological analysis, another for clinical reasoning, mathematical derivation, or materials simulation--and no pre-designed team can anticipate every skill a question will need. Science Earth is a planet-scale scientific runtime in which any capability--a simulation cluster, a wet-lab robot, a proof engine, a single-cell pipeline--can connect to any other, with collaboration structure emerging from the question itself. Its underlying EACN protocol lets capabilities discover one another, negotiate task ownership, and adjudicate across incompatible evidentiary standards without prior knowledge of who will meet whom. This shifts the organizing challenge from workflow design to open-ended connectivity. Two runs validate this under structurally distinct conditions. In a trans-Pacific higher-order Kuramoto synchronization study, agents identified and corrected a closure-ratio assumption in Ott-Antonsen analytic theory that fails outside the Lorentzian limit, within thirty minutes. In an eight-agent single-cell run on the 4.88M-cell Kang 2024 pan-cancer atlas, heterogeneous capabilities coupled over a 64.9-hour window with one structural external instruction, producing three new result layers and anchoring findings against an independent wet-lab study on an adjacent CCR8- TIGIT+ Treg subset. These cases are a first empirical reading, not a benchmark sweep. They show that when AI capabilities are truly connectable and coordination emerges from the problem, scientific reasoning becomes a distributed, self-correcting process--a step towards scaling AI-native discovery to the planet.

A single CPU core sustains 32 million order messages per second at sub-microsecond median end-to-end host-path response latency, 4.7-11 times faster than the best available open-source matching engines on identical hardware. Scaled out, a single 96-core commodity server (~$1,630/month) sustains ~640 million messages per second across 10,000 symbols, over 20 times the provisioned capacity of the U.S. consolidated quote feed. We reach these numbers by attacking the storage layer that sets matching latency. The dominant order-book implementation, linked lists chained through a balanced tree, imposes two costs on every operation: pointer-chased traversal to the insertion point, and root-to-leaf search to locate the target price level. Under micro-bursts these costs produce tail-latency spikes that degrade market quality precisely when liquidity is most needed. We present two data-structure contributions that eliminate them. The first is the Priority-Indicated Node (PIN), a priority queue in which entries occupy fixed-capacity, contiguously addressable slots, with indicators encoding the entry's global priority status. Unlike heaps, which require O(log n) comparisons per operation, the PIN resolves insertion position directly from the indicators without comparing entries; indicator updates are O(1), independent of queue size. A depth-aware capacity model sizes each PIN so hot entries fit within L1 residency. The second targets a broader inefficiency: balanced search trees search from root to leaf on every insertion and deletion, even when the caller already knows the key's in-order neighbors, which in electronic trading are available at zero cost. Neighbor-aware insertion and deletion use known neighbor references to attach or remove a node with O(1) reference writes, followed by single-path rebalancing, across red-black, AVL, and B+-tree variants.

Sign language translation (SLT) remains constrained by the limited availability of paired sign-video/text corpora and by the heavy-tailed vocabularies typical of real-world datasets. We study a target-side augmentation strategy in which a large language model (LLM) generates controlled paraphrase variants of the reference spoken-language sentence while the sign input remains unchanged. Concretely, we use GPT-4o to produce semantically faithful variants of the training targets and train a Signformer-style pose-based Transformer under a two-stage schedule: pre-training on the augmented corpus followed by fine-tuning on the original references. We evaluate this strategy on three datasets that span complementary challenges: PHOENIX14T (German Sign Language), a real-world corpus with moderate lexical diversity; the Greek Sign Language Dataset with highly controlled, repetitive recordings; and LSA-T (Argentinian Sign Language), a naturalistic corpus with a large vocabulary and severe long-tail sparsity. This range allows us to characterize precisely when and why target-side augmentation is beneficial. On PHOENIX14T, augmentation improves BLEU-4 from 9.56 to 10.33, demonstrating that paraphrastic exposure helps the decoder generalize beyond memorized reference phrasing. The near-saturated GSL baseline and the extremely sparse LSA-T setting reveal the limits of the approach: in both cases, single-reference lexical overlap metrics are insufficient to capture the full picture, motivating a complementary semantic evaluation. To our knowledge, this is the first study to examine LLM-generated target-side paraphrases as an augmentation mechanism for SLT, and the first to apply an LLM-as-a-Judge evaluation protocol to SLT. This complementary evaluation reveals gains in semantic fidelity that lexical overlap metrics understate.

Interactive video world models generate video chunk by chunk in response to user-controlled camera movements, enabling applications such as real-time game simulation, virtual scene navigation, and embodied AI training. However, scaling to long interactive trajectories is prohibitively expensive due to growing context memory, quadratic attention complexity, and repeated denoising steps. We present Light Interaction, a training-free inference acceleration framework for interactive video world models. Our key insight is that interaction naturally enables trajectory-dependent adaptive computation: retrieved spatial memory can be discarded during novel exploration, temporal context can be adjusted according to local latent dynamics, and early-step model outputs can be reused when the camera revisits familiar regions. Based on this insight, Light Interaction combines adaptive context management, denoising cache acceleration, and hardware-software co-designed 3D block sparse attention with fused Triton kernels. Evaluated on HY-WorldPlay and Matrix-Game-3.0, Light Interaction achieves up to 2.59x speedup without model retraining while maintaining competitive visual quality.

Many learning tasks in science and engineering are characterized by sparse datasets, which limits the effectiveness of purely data-driven approaches. At the same time, these problems are often accompanied by rich domain knowledge derived from physical laws, operational requirements, and expert heuristics. Such knowledge is frequently expressed as rules involving logical propositions and linear inequalities. Existing neuro-symbolic methods typically enforce these rules approximately through soft penalties, assume input-independent rules when designing specialized architectures, or rely on non-differentiable post-processing at inference time to achieve hard constraint satisfaction. While recent advances in differentiable optimization layers enable end-to-end feasibility enforcement within neural networks, extending these approaches to logical or mixed-integer rules remains challenging due to inherent nonconvexity. In this work, we propose a unified end-to-end framework for enforcing hard, input-dependent mixed integer linear constraints within neural networks. Our approach represents rules as disjunctive constraints and applies hierarchical convex relaxations to obtain convex hull formulations. These relaxations yield tractable linear constraints that can be embedded as differentiable optimization layers while enabling exact rule satisfaction. We demonstrate the effectiveness of the proposed framework on real-world datasets, achieving perfect rule satisfaction and strong predictive performance.

Uncrewed aerial vehicles (UAVs) are increasingly used for exploration-driven monitoring in hazardous environments such as disaster zones, contaminated sites, wildfire areas, and damaged infrastructure, where limited flight endurance must be allocated between visiting reported locations and gathering new information. In these settings, prior information regarding hazards is often incomplete, spatially imprecise, and subject to change during execution. For example, initial reports may identify a region where a hazard is likely to exist, but the actual hazard may be displaced, partially observed, or entirely unreported. We present an integrated exploration-aware UAV route optimization and path planning framework for hazard monitoring under uncertain and evolving prior information. The environment is represented as a spatial risk map, where each location has an associated belief of hazardous conditions. Reported hazards are modeled as uncertain regions of interest (ROIs) rather than confirmed target locations, requiring the UAV to inspect reported areas while also using its limited flight endurance to explore informative regions. The proposed method solves a vehicle routing problem over reported ROIs, augments the route with auxiliary pseudo-nodes to improve spatial coverage, allocates the remaining flight distance budget across route segments, and optimizes dynamically feasible B-spline trajectories for local exploration. During execution, UAV measurements update a grid-based belief map, and the remaining trajectory is replanned when new information and the remaining budget justify adaptation. Across 48 scenario configurations, online replanning improves average KL reduction by 15.9% over the offline optimized planner and 48.6% over straight-line traversal.

This paper presents a multilingual customer service self-help corpus comprising 1,122 manually validated documents in Finnish, Danish, Norwegian, and Swedish, totaling 274,599 words and 1,884,833 characters. The documents have been sourced from the public self-help pages of four Nordic telecommunications operators and subsequently filtered for person-identifiable information and relevance through a combined LLM and human annotation pipeline. Domain-specific datasets for Nordic languages remain scarce, particularly in customer service: a domain of growing importance for retrieval-augmented generation, cross-lingual transfer learning, and emerging agent-based service architectures. An analysis of the corpus reveals substantial variation in document length and structure across operators, reflecting distinct editorial strategies, as well as broad topical coverage spanning network hardware, mobile services, TV and streaming, billing, and account management. The dataset is publicly available under a CC-BY-NC-SA-4.0 license at https://zenodo.org/records/20732652, intended to support reproducible research in Nordic NLP and information retrieval.

Standard Set Representation Learning methods typically excel on curated data but often overlook the challenge of inference-time element corruption. This refers to scenarios where deployed models encounter element-level degradations, such as outliers or missing components, that may distort set representation and degrade performance. We propose SW-DRSO, a distributionally robust optimization framework tailored for sets. Rather than minimizing loss solely on observed training data, SW-DRSO optimizes a tractable surrogate of the worst-case expected loss over a family of plausible inference-time variations. We introduce a barycentric adversary that approximates the intractable search over corrupted sets by a differentiable training-time optimization over simplex weights. Extensive experiments across four tasks demonstrate that SW-DRSO effectively enhances robustness against corruption while maintaining high overall performance.

Large language models (LLMs) are increasingly applied in finance, yet most existing work emphasizes trading signals or financial NLP tasks centered on prediction. Institutional fundamental research, by contrast, requires human analysts or AI agents to gather evidence, identify business drivers, compare competing viewpoints, and generate investment memos. Its broader goal is not merely to predict outcomes, but to produce investment plans that are transparent, reusable, and verifiable, while contributing to the cumulative development of investment knowledge. We present FundaPod, a multi-persona agent platform for AI-assisted fundamental investment research. We argue that fundamental research is a human-centric decision-support task that is qualitatively distinct from trading-signal generation, and is therefore better served by an independence-preserving architecture. In FundaPod, AI agents with different personas, such as value investors or macro strategists, conduct research independently under a shared provenance contract. Their disagreements are then surfaced post hoc for adjudication by the human portfolio manager (PM) through a knowledge-graph memory system. This paper contributes five design principles for human-AI hybrid systems supporting fundamental research, grounded in design-science practice and theories of cognitive isolation and human-machine coordination. It also describes four architectural mechanisms: a persona distillation pipeline that turns public investor materials into deployable agents; a declarative skill registry that lets the planner derive typed task graphs; a grounded evidence model that links memo claims to verifiable sources; and a knowledge-graph "second brain" that connects tickers, memos, analysts, and themes. We demonstrate the architecture through a complete case study and a persona-based memo comparison.

Wearable devices enable continuous monitoring of physiological signals such as ECG and PPG, but existing mHealth systems are largely limited to task-specific prediction pipelines or reactive question answering over static summaries. They lack the ability to support temporal reasoning, persistent physiological context, and proactive monitoring over long-term signal streams. We propose VitalAgent, a tool-augmented agentic framework for ECG/PPG-based mHealth that supports both reactive question answering and proactive monitoring. VitalAgent is built on a longitudinal physiological memory and a tool-augmented reasoning interface that enables dynamic computation over raw signals. We further introduce VitalBench, a longitudinal physiological monitoring benchmark dataset comprising 1,862 QA pairs for reactive question answering and 90.2 hours of continuous ECG/PPG recordings for proactive monitoring, covering cardiac, physical activity, and stress-related tasks. Experiments demonstrate that VitalAgent achieves over 25% improvement over prompt-based and ReAct baselines in reactive evaluation and supports proactive alert monitoring over long-term physiological signals, highlighting the importance of dynamic tool use and long-term physiological monitoring.

Existing agent memory remains predominantly reactive and retrieval-based, lacking the capacity to autonomously organize experience into persistent cognitive structure. Toward genuinely autonomous agents, we introduce CogniFold, a brain-inspired "always-on" agent memory designed for the next generation of proactive assistants. CogniFold continuously folds fragmented event streams into self-emerging cognitive structures, bootstrapping progressively higher-level cognition from incoming events and accumulated knowledge. We ground this by extending Complementary Learning Systems (CLS) theory from two layers (hippocampus, neocortex) to three, adding a prefrontal intent layer. Emulating the prefrontal cortex as the locus of intentional control and decision-making, CogniFold achieves this through graph-topology self-organization: cognitive structures proactively assemble under the stream, merge when semantically similar, decay when stale, relink through associative recall, and surface intents when concept-cluster density crosses a threshold. We evaluate structural formation using CogEval-Bench, demonstrating that CogniFold uniquely produces memory structures that match cognitive expectations and concept emergence. Furthermore, across eight downstream benchmarks -- two probing long-term conversational memory (LoCoMo, LongMemEval) and six spanning other cognitive domains -- we validate that CogniFold simultaneously performs robustly on conventional memory tasks. Our code is available at https://github.com/OpenNorve/CogniFold.

GUI agents powered by large language models are advancing rapidly, creating urgent needs for evaluation and training based on realistic environments. However, directly doing so in real-world environments introduces some challenges that cannot be overlooked. Real-world environments are complex and uncontrollable, making it difficult to construct verifiable rewards and to save or reset states. Existing works prioritize reproducibility but are often limited to open-source apps or file-operation tasks for reliable reward building, leaving a persistent gap from real-world usage. Furthermore, relying on virtual machines or docker images demand high resource requirements and suffer from slow response speeds, which limit the efficiency. We present \sys, a framework that could produce high-fidelity synthesized interactive environments for GUI agents across platforms with verifiable rewards. These environments behave as backend-free webpages accessible via URL, requiring near-zero setup and low resource cost, making the approach suitable for both large-scale evaluation and downstream agent training. We support multiple GUI platforms including mobile, desktop, and automotive/in-vehicle interfaces based on the same pipeline, covering 100+ environments and 1000+ verifiable tasks. Among them, 120 challenging tasks across 63 simulated mobile applications are released as a fully synthesized mobile GUI agent benchmark. Experiment results on five state-of-the-art mobile GUI agents reveal substantial headroom -- the average success rate is only 27.92\%, dropping to 17.82\% on long-horizon subset -- while humans reach 92.08\%. A comparison against real-world sample tasks shows that assessments made in our synthetic environments generalize to real apps. The project website is at https://scalewob.github.io.

Achieving both efficient pushability (propulsion transmission) and proximally concentrated bending for steerability is challenging for magnetically actuated soft catheters: higher axial/bending stiffness improves force transmission but reduces steerability, whereas lower stiffness enables large, proximally concentrated bending yet increases kinking/buckling risk under compressive push loads. To address this trade-off, we propose a stiffness-optimized multi-segment magnetically actuated catheter (SO-MAC) that integrates a decoupled steering-advancement mechanism with a gradient-stiffness architecture. The SO-MAC concentrates bending about a stable proximal pivot during advancement while the distal section passively self-straightens to transmit propulsion, aided by the optimized stiffness distribution and elastic recovery of the spring backbone against friction-induced kinking/buckling. Over $0{-}180^{\circ}$ combined steering and advancement, the pivot remained stable and the distal tip advanced near-straight toward the target direction. A 1.5 mm-diameter SO-MAC achieved up to $180^{\circ}$ steering with a 3 mm bending radius at its 10 mm tip, with an average shape error of $1.39 \pm 0.56$ mm and a steering-pivot error of $0.35 \pm 0.10$ mm. Visual feedback control in a bronchial phantom further confirmed robust navigation through highly curved, bifurcating paths.

Whole-body tracking (WBT) models have become a key foundation for humanoid robots, enabling them to imitate diverse motions with high fidelity. Training such models from scratch requires large-scale data and computation, making rapid deployment on new humanoid platforms costly. This raises a natural question: Can pretrained WBT models transfer across embodiments with minimal adaptation? To answer this question, we propose Any2Any, a paradigm that efficiently transfers an existing WBT specialist to a new humanoid embodiment with only a small amount of data and compute. Any2Any first performs kinematic alignment between source and target humanoids, aligning their input and output spaces so that the pretrained source policy can be meaningfully reused on the target embodiment.Any2Any then performs dynamics adaptation by applying lightweight parameter-efficient fine-tuning (PEFT) components to selected dynamics-sensitive modules, preserving useful behavioral priors while enabling targeted adaptation to the target robot. Extensive experiments on multiple humanoid platforms and pretrained backbones show that Any2Any substantially accelerates convergence and reduces training cost compared with training from scratch, while achieving competitive or superior tracking performance. Notably, using only 1% of the compute and data required for full training, Any2Any successfully transfers Sonic models pre-trained on Unitree G1 to LimX Oli and LimX Luna. These results suggest that pretrained WBT specialists can be efficiently reused across embodiments, providing a scalable path toward deploying humanoid whole-body control on new robots. More results and videos are available on our project page: https://any2any.top/.

Autonomous systems have achieved superhuman performance in isolation or simulation, yet they remain brittle in shared, dynamic real-world spaces. This failure stems from the dominant single-agent paradigm for physical applications, where other actors are ignored or treated as environmental noise, preventing effective coordination. Here we show that multi-agent reinforcement learning provides the essential safety scaffolding required for real-world interaction. Using high-speed quadrotor racing as a high-stakes testbed, we train agents to navigate complex aerodynamic interactions and strategic maneuvering with a variable number of racers. Through league-based self-play, agents evolve sophisticated anticipatory behaviors, including proactive collision avoidance, overtaking, and handling multi-agent physical interactions, including aerodynamic downwash. Our agents outperform a champion-level human pilot in multi-player races at speeds exceeding 22 m/s, while simultaneously reducing collision rates by 50 % compared to state-of-the-art single-agent baselines. Crucially, training with diverse artificial agents enables zero-shot generalization to safer human interaction. These results suggest that the path to robust robotic co-existence lies not in isolated safety constraints, but in the rigorous demands of multi-agent interaction. Multimedia materials are available at: https://rpg.ifi.uzh.ch/marl

This paper shows that the concept of complex frequency, originally introduced to characterize the dynamics of signals with complex values, constitutes a generalization of eigenvalues when applied to the states of linear time-invariant (LTI) systems. Starting from the definition of geometric frequency, which provides a geometrical interpretation of frequency in electric circuits that admits a natural decomposition into symmetric and antisymmetric components associated with amplitude variation and rotational motion, respectively, we show that complex frequency arises as its restriction to the two-dimensional Euclidean plane. For LTI systems, it is shown that the complex frequencies computed from the system's states subject to a non-isometric transformation, coincide with the original system's eigenvalues. This equivalence is demonstrated for diagonalizable systems of any order. The paper provides a unified geometric interpretation of eigenvalues, bridging classical linear system theory with differential geometry of curves. The paper also highlights that this equivalence does not generally hold for nonlinear systems. On the other hand, the geometric frequency of the system can always be defined, providing a geometrical interpretation of the system flow. A variety of examples based on linear and nonlinear circuits illustrate the proposed framework.

Supervised fine-tuning (SFT) is widely used to inject new knowledge into language models, but it often degrades pretrained capabilities such as reasoning and general-domain performance. We argue this forgetting arises because fine-tuning targets from humans or external systems diverge from the model's autoregressive distribution, forcing the optimizer to imitate low-probability token sequences. To address this problem, we propose MixSD, a simple external-teacher-free method for distribution-aligned knowledge injection. Instead of training on fixed targets, MixSD constructs supervision dynamically by mixing tokens from two conditionals of the base model itself: an expert conditional that observes the injected fact in context, and a naive conditional that reflects the model's original prior. The resulting supervision sequences preserve the factual learning signal while remaining substantially closer to the base model's distribution. We evaluate MixSD on two synthetic corpora that we construct to study factual recall and arithmetic function acquisition in a controlled setting, together with established benchmarks for open-domain factual question answering and knowledge editing. Across multiple model scales and settings, MixSD consistently achieves a better memorization-retention trade-off compared to SFT and on-policy self distillation baselines, retaining up to 100% of the base model's held-out capability while maintaining near-perfect training accuracy, whereas standard SFT retains as little as 1%. We further show that MixSD produces substantially lower-NLL supervision targets under the base model and reduces harmful movement along Fisher-sensitive parameter directions. These results suggest that aligning supervision with the model's native generation distribution is a simple and effective principle for knowledge injection that mitigates catastrophic forgetting.

Estimating queue lengths at signalized intersections is a long-standing challenge in traffic management. Partial observability of vehicle flows complicates this task despite the availability of two privacy-preserving data sources: (i) aggregated vehicle counts from loop detectors near stop lines, and (ii) aggregated floating car data (aFCD) that provide segment-wise average speed measurements. However, how to integrate these sources with differing spatial and temporal resolutions for queue length estimation is rather unclear. Addressing this question, we present Q-Net: a queue estimation framework built upon a state-space formulation. This design addresses key challenges in queue modeling, such as violations of traffic conservation assumptions. Q-Net follows the Kalman predict-update structure and maintains physical interpretability in both the state evolution and measurement models. Q-Net uses an AI-augmented Kalman filter to learn time-varying gain dynamics from data. The framework supports real-time implementation and improves spatial transferability by grouping aFCD measurements into fixed-size local groups, making the number of learnable parameters independent of section length. Evaluations on urban main roads in Rotterdam, the Netherlands, show that Q-Net outperforms baseline methods, tracks queue formation and dissipation accurately, and mitigates aFCD-induced delays. By combining data efficiency, interpretability, real-time applicability, and spatial transferability, Q-Net makes accurate queue length estimation possible without costly sensing infrastructure like cameras or radar.

Reliable verification of proofs remains a bottleneck for training and evaluating AI systems on hard mathematical reasoning. Fully formal proofs, in languages like Lean, are easy to verify because they are unambiguous and modular. Most proofs, particularly those written by AI systems, have neither property, and translating them into formal languages remains challenging in many frontier math settings. We propose Pseudo-Formalization (PF), a proof format that captures the modularity and precision of formal proofs while retaining the flexibility of natural language. A Pseudo-Formal proof is decomposed into self-contained modules, each stating its premises, conclusion, and proof in natural language. To verify the correctness of a regular natural language proof, an LLM translates it to Pseudo-Formal and then verifies each module independently, an algorithm we call Block Verification (BV). We evaluate PF+BV on two benchmarks spanning olympiad and research-level mathematics, where it pareto-dominates LLM-as-judge baselines on error-finding precision and recall. To support future work, we release our research-level proof verification benchmark ArxivMathGradingBench.

Vision-language models reliably name objects in a scene, but do they represent the 3D layout those objects inhabit? We introduce a 3,034-sample human-curated benchmark targeting three components of spatial understanding: depth-ordered occlusion (probed via three independent counterfactual operationalisations), optical-geometry inference over visible reflections, and volumetric rearrangement planning. Six frontier and open-weight VLMs, scored by trained annotators on 18,204 responses with no LLM-as-judge, reveal a sharp dissociation: models that plan rearrangements over visible layouts at 53--97% accuracy and rarely violate collision constraints fall to 6--45% on occlusion and below 7% on reflections. An embodied-reasoning model reproduces the same profile. White-box analysis on Qwen3-VL-8B-Thinking localises the failure to the visual-token merger: spatial information recoverable throughout the vision encoder becomes inaccessible after token compression and only stabilises again when clean post-merger activations are patched into the language decoder.

As Large Language Models continue to grow in both capability and cost, transferring frontier capabilities into smaller, deployable students has become an important engineering problem, and knowledge distillation remains a common technique for this transfer. The prevailing recipe in industrial pipelines, static imitation of teacher-generated text, carries a structural weakness that grows more severe as tasks become longer and more reasoning-intensive. Because the student is trained on flawless teacher prefixes but generates its own at inference, small errors tend to accumulate into trajectories it has rarely been trained to recover from, and the resulting exposure bias has been shown to scale roughly with the square of sequence length. On-Policy Distillation reorganizes the training loop around this observation by having the teacher provide feedback on what the student actually produces, with the goal of reducing the compounding term toward linear and reframing distillation as an iterative correction process rather than single-pass imitation. The resulting literature has expanded along divergence design, reward-guided optimization, and self-play, yet contributions remain scattered across the knowledge distillation, RLHF, and imitation learning communities without a unified treatment. This survey provides such a treatment. We formalize OPD as f-divergence minimization over student-sampled trajectories, organize the field along three design axes (what to optimize, where the signal comes from, and how to stabilize training in practice), and consolidate success conditions, recurring failure modes, and the connection between OPD and KL-constrained reinforcement learning. We close with open problems that emerge from this synthesis, including distillation scaling laws, uncertainty-aware feedback, agent-level distillation, and the growing overlap between knowledge distillation and RL.

We analyze how automatic speech recognition (ASR) errors propagate through ASR-LLM cascades in Korean spoken question answering (SQA), focusing on downstream semantic failures that conventional ASR metrics cannot fully capture. Our analysis shows that the relative downstream degradation caused by ASR errors is consistent across LLMs with different absolute performance, suggesting that cascade degradation largely tracks ASR-stage information loss. We further identify single-character Korean ASR errors as a Korean-specific loss channel, where even a minimal transcription difference can change the intended question and degrade downstream QA performance. Finally, an auxiliary comparison shows that a large audio language model outperforms an ASR-LLM cascade with an approximately matched language backbone in noisy Korean SQA, indicating the potential of direct audio input to mitigate transcript-induced information loss.

Nonlinear finite element crash simulations are accurate but computationally expensive, limiting their use in iterative design optimisation. Machine-learning surrogate models based on graph neural networks (GNNs) offer a faster alternative. Message-passing GNNs are widely used for mesh simulation, and their shared node and edge update functions are relatively generalisable across varying graph structures. By contrast, non-shareable edge-specific aggregation layers can capture nonlinear relationships more accurately but usually require fixed graph connectivity, which limits generalisability. This paper presents Mask-Morph Graph U-Net (MMGUNet), a practical approach to addressing the limitation of hierarchical Graph U-Net architectures that use edge-specific downsampling and upsampling layers. Fixed coarse graph connectivity is required for edge-specific layers. To retain this while improving spatial correspondence, the proposed method morphs the coarsened graph hierarchy to each input mesh using feature-aligned barycentric parameterisation before constructing cross-graph edges. It further applies node masking during supervised pretraining, followed by parameter-efficient fine-tuning in which high-parameter edge-specific layers are frozen. The proposed approach is evaluated in in-distribution, out-of-distribution, and cross-component transfer settings using mean Euclidean distance and maximum intrusion percentage error. Results show that coarse-graph morphing improves test accuracy relative to a fixed-coarse-graph baseline, while masked supervised pretraining reduces the train-test discrepancy and improves data efficiency during transfer. The proposed model also achieves lower prediction error compared with external baselines. These results demonstrate a practical route toward reusable, data-efficient mesh-based surrogate modelling for crashworthiness design exploration.

Biometric facial recognition models often demonstrate significant decreases in accuracy when processing real-world images, often characterized by poor quality, non-frontal subject poses, and subject occlusions. We investigate whether targeted, AI-driven, head-pose correction and image restoration can improve recognition accuracy. Using a model-agnostic, large-scale, forensic-evaluation pipeline, we assess the impact of three restoration approaches: 3D reconstruction (NextFace), 2D frontalization (CFR-GAN), and feature enhancement (CodeFormer). We find that naive application of these techniques substantially degrades facial recognition accuracy. However, we also find that selective application of CFR-GAN combined with CodeFormer yields meaningful improvements.

As generative AI tools increasingly influence creative practice, they raise longstanding HCI questions about how creatives learn complex software and how they can be better supported. We conducted an interview study with artists and hobbyists (n=8) and a follow-up survey (n=159) to understand how this population approaches and seeks guidance for GenAI image tools. We found that creatives commonly use either self-experimentation or tutorials to explore GenAI tools, yet many struggle with confusing AI terminology. To gain further insight into creatives' learning experiences, we developed a research probe to elicit creatives' perceptions of structured guidance. Our user study with 17 creatives revealed that, even when creatives described the guidance as helpful for understanding AI, many still preferred self-experimentation, feeling that guidance could limit their creativity. Our findings highlight a central tension in supporting AI literacy for creatives: balancing guidance and promoting literacy while preserving creative freedom.

Recovering editable CAD programs from images or 3D observations is central to AI-assisted design, but progress is difficult to measure because existing evaluations are fragmented across datasets, modalities, and metrics. We introduce CADBench, a unified benchmark for multimodal CAD program generation. CADBench contains 18,000 evaluation samples spanning six benchmark families derived from DeepCAD, Fusion 360, ABC, MCB, and Objaverse; five input modalities including clean meshes, noisy meshes, single-view renders, photorealistic renders, and multi-view renders; and six metrics covering geometric fidelity, executability, and program compactness. STEP-based families are stratified by B-rep face count and all families are diversity-sampled to support controlled analysis across complexity and object variation. We benchmark eleven CAD-specialized and general-purpose vision-language systems, generating more than 1.4 million CAD programs. Under idealized inputs, specialized mesh-to-CAD models substantially outperform code-generating VLMs, which remain far from reliable CAD program reconstruction. CADBench further reveals three recurring failure modes: reconstruction quality degrades with geometric complexity, CAD-specialized models can be brittle under modality shift, and model rankings change across metrics. Together, these results position CADBench as a diagnostic testbed for measuring progress in editable 3D reconstruction and multimodal CAD understanding. The benchmark is publicly available at https://github.com/anniedoris/CADBench.

We study a class of bilevel interdiction problems in which the follower's optimization problem is computationally intractable. Motivated by network defense applications, we introduce the Randomized Max-Vertex-Coverage Interdiction (RMVCI) problem under matroid constraints. In this zero-sum Stackelberg game, the leader commits to a randomized interdiction strategy over feasible vertex subsets, while the follower, after observing the induced protection probabilities, chooses a matroid-constrained attack to maximize the expected coverage of network edges. The main challenge stems from the fact that the follower's problem is a matroid-constrained maximum vertex coverage problem and is therefore NP-hard. To address this difficulty, we first develop a general approximation framework for bilevel optimization problems with hard follower responses. The framework is based on replacing the follower's value function by a surrogate objective that approximates the follower's optimal payoff while preserving tractability of the leader's optimization problem. For the RMVCI problem, we formulate the follower's problem as an integer linear program, establish a tight integrality gap of $4/3$ for its linear relaxation, and derive a polynomial-time $4/3$-approximation algorithm via pipage rounding. We then show that a carefully designed surrogate objective admits a marginal-probability reformulation that transforms the randomized interdiction problem into a tractable optimization problem over the leader's matroid polytope. This yields a polynomial-time $2$-approximation algorithm for RMVCI under general matroid constraints. Beyond the specific application studied here, our results provide a new perspective on approximation methods for {general} bilevel optimization problems.

We provide counterexamples to the unimodal minimal filling architecture conjecture for polynomial neural networks (PNNs) with power activation functions. Fixing the input and output widths, the conjecture states that any minimal filling architecture has unimodal widths for the hidden layers. We found counterexamples via a frontier search, recursive dimension bounds on neurovarieties, and symbolic computation. Notably, several subarchitectures of our main example exhibit large defect, in contrast with the predominantly small-defect behavior observed in prior literature.

Artificial intelligence is increasingly entering digital games through diverse functions. While prior work has shown that player attitudes toward game AI are strongly context-dependent, less is known about how these attitudes are structurally combined within different groups of players. This study addresses this gap by modeling players' cross-context AI acceptance as interpretable attitude profiles. Based on questionnaire data from 771 digital game players, we apply Archetypal Analysis (AA) to centered acceptance ratings across eight representative AI application contexts in games. The analysis identifies seven distinctive profiles: AI-Skeptics, Broad AI-Supporters, Creative-Play Explorers, Experience-Oriented Supporters, Systemic Order Advocates, Emotion-Centered Supporters, and Governance-Skeptics. Exploratory one-vs-rest (OvR) logistic regressions further suggest that profile membership is associated with players' perceived AI literacy, gaming habits, disciplinary background, personality traits, and application-specific priorities. By shifting attention from isolated acceptance judgments to patterned preference structures, this study provides an exploratory empirical vocabulary for segmenting game AI audiences and offers preliminary design implications for more context-sensitive and player-sensitive AI integration in digital games.