Papers

Reinforcement Learning (RL) has emerged as a powerful paradigm for adaptive traffic signal control. However, in safety-critical infrastructure like traffic control, the opaque, black-box nature of deep RL models poses challenges for transportation agency acceptance, regulatory compliance, operational trust, troubleshooting, and fine-tuning. To bridge this gap between high-performance optimization and human-comprehensible interpretability, this effort introduces a novel, explainable entity centric RL framework for safe and transparent traffic signal control. Rather than processing traffic states through monolithic, flat vectors, the proposed architecture disaggregates real-time intersection observations into distinct, high-dimensional lane entities and phase temporal configurations to inherently preserve the structural topology and geometric configurations of the intersection. Relational dependencies and inter-lane conflicts are dynamically extracted via a dual-stage attention network featuring sequential multi-head cross-attention and self-attention blocks. This design yields a real time affinity matrix that quantifies the direct influence of signal phases on specific approach volumes and queues, providing full visual and analytical interpretability. To ensure strict operational reliability, a deterministic action-masking interface is integrated directly into the Proximal Policy Optimization pipeline, explicitly blocking invalid phase transitions to guarantee absolute compliance with established signal timing and safety constraints. Evaluated in a microscopic simulation environment, outperforms state-of-the-art baselines in delay minimization. More importantly, the emergent attention weights align precisely with established traffic engineering principles, offering an auditable, trust-enabling, and deployable architecture for next-generation adaptive traffic control systems.

Large language model (LLM) agents have shown strong decision-making capabilities in long-horizon interactive tasks, yet they still struggle to effectively leverage failed trajectories: full retries incur high interaction costs, while experience retrieval tends to dilute critical experience signals. To address this, we propose PivoARL, a self-feedback retry framework for experience exploitation in LLM agents. PivoARL identifies the pivotal erroneous turn through structured reflection and performs local retry only from the corresponding pivotal state, thereby reusing the correct prefix and reducing redundant interactions. From an information-gain perspective, we further show that pivotal retry concentrates useful experience signals near the error boundary, mitigating the signal dilution caused by state-agnostic experience utilization. Based on this insight, we design a pivotal-aware credit assignment mechanism that rewards correct prefixes while isolating erroneous suffixes, and optimize reflection quality through implicit reflection returns. We conduct a systematic evaluation on 4 agent tasks and 7 search-based QA benchmarks. Results show that PivoARL achieves significant improvements on Pass@2/3 across all tasks, with an average gain of about 11.5% over MetaRL. Moreover, benefiting from contrastive preference signals induced by pivotal turns, PivoARL also consistently improves Pass@1 on over 80% of the tasks. On Minesweeper environment, PivoARL improves over GiGPO by more than 45% and reduces interaction turns by about 42% on average compared with full-retry methods. Code is available at https://github.com/yuki-younai/PivoARL.

Large-scale Capacitated Vehicle Routing Problems (CVRPs) are commonly solved by partitioning customers into smaller routing problems that can be optimized independently. While this substantially reduces computational complexity, independently constructed routing solutions may leave some customer demand unserved even when sufficient resources exist elsewhere in the fleet. We present Collaborative Routing Constructors (CoRC), a routing framework that enables independently solved subproblems to exchange customers and vehicles during optimization rather than relying solely on a fixed partition or a subsequent global re-optimization stage. Computational experiments on AGS benchmark instances and synthetic instances containing up to 200,000 customers compare CoRC against independent routing, post-routing global re-optimization, and state-of-the-art, end-to-end routing frameworks. Across all evaluated partitioning strategies, CoRC consistently constructs feasible routing solutions where competing partition-based methods do not. Furthermore, it remains effective on problem instances for which the evaluated end-to-end routing frameworks did not produce solutions under the same computational budget. These results demonstrate that collaboration between routing subproblems provides a robust and scalable approach for feasible large-scale route construction.

Coding agents, autonomous systems that use large language models (LLMs) to resolve software engineering tasks, rely on agent harness: a middleware layer in between a developer and a large language model that orchestrates system prompts, tool execution, context management, and iterative reasoning loops. While these agent harnesses evolve at extreme velocities, no study has examined how this evolution affects agent quality (i.e., effectiveness and efficiency) over time. Practitioners regularly report quality regressions after agent harness updates, yet consistently attribute them to the underlying model rather than the harness itself. In this paper, we address this gap by conducting the first controlled longitudinal study that isolates the agent harness contribution. Unlike prior work that fixes the agent harness and varies the model, we fix the model and vary only the agent harness, evaluating 35 sequential releases to measure their impact on agent effectiveness and efficiency. We first empirically study the development and release evolution of five major open-source agent harnesses (i.e., Codex, Qwen Code, Gemini, OpenCode, and OpenHands), revealing extreme release velocities exceeding two releases per day and thousands of issues within months. We then perform a controlled deep dive into 35 sequential releases of the Qwen Code CLI, evaluating each against 50 stratified SWE-bench Verified tasks while holding the underlying LLM constant. We trace the resulting quality fluctuations to specific development patterns and architectural components, and illustrate our findings with concrete qualitative evidence linking individual pull requests to measured quality shifts.

The accurate estimation of Above-Ground Biomass (AGB) in mature tropical forests remains a critical challenge in remote sensing, primarily due to the saturation of Synthetic Aperture Radar (SAR) signals in high-density areas and persistent cloud cover affecting optical imagery. To overcome these physical limitations, we propose the Trimodal Coherent Co-attention Transformer (TCCT), a physics-informed deep learning architecture. The TCCT natively fuses optical surface reflectance (Landsat-5) with complex-valued Polarimetric SAR Interferometry (PolInSAR) data from both P and L bands. Unlike traditional fusion methods, our architecture employs complex-valued encoders to preserve spatial phase coherence, coupled with a dynamic co-attention mechanism that acts as an adaptive gating module, reducing the weight of cloud-corrupted optical pixels and shifting reliance to microwave phase data. We also derived a localized spatial allometric calibration model via Levenberg-Marquardt optimization, tailored to the specific wood density of the Paracou region in the Amazon basin. Evaluated using a two-stage protocol, the TCCT first underwent a rigorous 5-fold cross-validation to establish robust global weights (achieving a global RMSE of 4.19 m). Subsequently, following a localized spatial fine-tuning phase over 200 epochs, the model attained an absolute RMSE of 3.78 m and an $R^2$ of 0.33 for Canopy Height Models (CHM), outperforming standard Random Forest, CNN, and Vision Transformer baselines. Our ablation study confirms that preserving phase coherence mitigates deep-canopy signal saturation. When converted to AGB, the fine-tuned TCCT map yielded a Relative RMSE (rRMSE) of 4.51% in dense forest areas above 50 Mg/ha. By meeting the European Space Agency (ESA) BIOMASS mission requirement of less than 20% error, the TCCT provides a robust framework for continuous carbon stock mapping in tropical biomes.

Relational databases (RDBs) are the primary data infrastructure in many enterprises, yet recent deep learning methods designed for RDBs have been evaluated under inconsistent experimental protocols, making fair comparison difficult. We present one of the first systematic benchmarking studies of recently released deep learning methods for RDBs, evaluating them across five relational databases, with one classification and one regression task for each. We refactor all deep RDB models to allow the full range of experimental procedures to be applied consistently across all methods. Our findings indicate that the relational transformer (RT) approach delivers the strongest overall performance on both classification and regression tasks compared to the state-of-the-art graph-based modeling and learning of RDBs. Even for single-table learning tasks, deep learning methods designed for RDBs outperform the leading tabular foundation model, TabPFN 2.5. Extending learning from a single table (hop = 0) to multiple tables (hop = 1, 2) by connecting neighboring tables in relational databases enhances performance, but the additional benefit from higher hops diminishes as computational overhead grows. Deep RDB learning methods have the potential to challenge state-of-the-art tabular foundation models, especially on large-scale enterprise data. The source code for this benchmarking study is publicly available.

Gloss-free Sign Language Translation (SLT) translates sign language videos into spoken-language sentences without gloss annotations, avoiding costly labeling but requiring fine-grained modeling of hands, body, and facial cues. Existing methods often use single-modality or weakly fused features, limiting performance. We propose ViPo-MLLM, a framework that integrates spatio-temporal RGB and human pose features. Dedicated encoders model intra-modal dynamics and cross-modal attention captures long-range dependencies. The fused representation is conditioned with a structured prompt and processed by an LLM trained with contrastive and language modeling objectives. The proposed model was evaluated on the PHOENIX14T and CSL-Daily datasets and achieved new state-of-the-art results on both datasets. Moreover, the ViPo-MLLM model attained competitive performance compared to gloss-based recognition approaches, confirming the effectiveness of the proposed pose cues and cross-modal attention mechanisms.

Large Language Models are increasingly used to turn natural-language requirements into code. In access control, that shortcut is dangerous: a generated policy can compile and read correctly while granting access that no one approved. The difficulty is not only writing policy code. It is fixing what the requirements mean before code is written, and then checking that the final policy actually satisfies that intent. We present AutoCedar, a verifier-guided system that first turns natural-language access-control requirements into a reviewed, checkable target, and then synthesizes Cedar policies against that target. AutoCedar decomposes schema and policy authoring into small intent atoms: reviewable claims about vocabulary and behavior. Once those atoms pass mechanical validation and human intent review, the model proposes a candidate policy, the verifier checks it against the approved target, and each failure is turned into a repair signal that tells the model whether to broaden, narrow, or restructure the policy without changing the target. Because the model's work is split into small problems, each grounded in reviewed intent and backed by verifier feedback, end-to-end policy authoring becomes tractable. AutoCedar converges on all 221 tasks of CedarBench, our benchmark of authorization tasks paired with executable semantic boundaries. Across three requirements-corpus case studies covering healthcare, education, and conference management, AutoCedar converts noisy prose and extracted access-control fragments into reviewed schemas, formal checks, and a globally verified Cedar policy store for each scenario.

Transformer blocks are prevalent in large language model (LLM) but present deployment challenges due to their challenging computational and memory demands. While prior work has typically optimized attention mechanisms or feed-forward networks (FFNs) separately, few hardware (HW) architecture have jointly addressed both components with co-designed hardware acceleration. We present ELiTeFormer (Efficient Linear Ternary Transformer), the first Transformer model architecture that unifies hybrid linear attention with ultra-low-precision (ternary) linear projections, specifically co-designed for field-programmable gate array (FPGA) deployment. ELiTeFormer achieves 10x model weight compression and 12.8x key-value (KV) cache compression compared to LLaMA 3, while maintaining competitive accuracy (31.9% on the MMLU benchmark, within 3.0% of BitNet b1.58). Our key architectural contribution is a novel processing element (PE) micro-architecture that eliminates all multiplications in ternary linear projections through bitmasking operations, significantly reducing resource utilization by completely avoiding dedicated digital signal processing (DSP) blocks. We simulate, synthesize, and deploy ELiTeFormer targeting a Xilinx VCK5000 Versal board using high-level synthesis (HLS) flows. Block-level simulations show 9.6x speedup for FFN operations and 4.4x speedup for attention compared to standard implementations. End-to-end deployment achieves up to 3.9x lower latency and 3.2x better energy efficiency than LLaMA 3 on an NVIDIA A100 graphics processing unit (GPU) at long context lengths. This represents the first FPGA realization combining linear attention with ternary quantization, demonstrating the viability of algorithm-architecture co-design for next-generation LLM acceleration.

Extracting textual information from scanned medical documents, such as external laboratory reports and manually filled forms, has been a major challenge in modern electronic health records (EHRs). Recent advancements in vision language models (VLMs) have shown great promise over traditional OCR tools. However, at this point, most clinical OCR studies were conducted on private, institutional data. To our knowledge, there are few publicly available datasets for evaluating OCR models in the clinical domain. Furthermore, common scanning artifacts that undermine OCR performance are not reflected in those datasets, leaving a systematic evaluation unfeasible. Therefore, we release a publicly available, realistic-looking OCR benchmark dataset, ClinOCR-Bench, with 384 scanned images across 6 subsets: Normal, Handwriting, Poor Quality, Rotation, Tables, and Mix-artifacts. ClinOCR-Bench features: 1) diverse document types and layouts, 2) full coverage of common EHR scan artifacts, 3) protected health information-free, 4) template-aware train/test split, and 5) adequate sample size for OCR benchmarking. Baseline OCR performance was evaluated using state-of-the-art open-weight and proprietary VLMs. The dataset and documentation are available on GitHub (https://github.com/ClinOCR-Bench/ClinOCR-Bench).

Large language models (LLMs) are increasingly being integrated into digital health platforms to generate summaries of complex medical data. Although these models can improve patient engagement and communication, these systems also raise concerns about accuracy, faithfulness, and safety in clinical contexts. In this study, we evaluate AI-generated summaries within a cancer patient care application using a dual assessment framework. Human domain experts, including oncology clinicians and patient-facing care staff, provided ground-truth evaluations of summary quality along dimensions of accuracy, clinical relevance, and readability. In parallel, we employed LLMs serving as evaluators (LLM-as-a-judge). Some limitations were identified in the generated summaries e.g., occasional omissions and minor inaccuracies. These were systematically analyzed and used to iteratively improve prompt design, grounding, and safety guardrails.

Clinical electroencephalography (EEG) reporting remains largely manual and time-consuming, and current EEG software ecosystems do not produce the structured EEG-text supervision needed for training modern language models. Most toolboxes focus on visualization or preprocessing, providing limited support for workflows that generate high-quality datasets for AI. We introduce EEG-to-Report, a browser-based annotation and feature-text framework that links routine EEG review with the construction of AI-ready datasets. The framework integrates multi-format EEG ingestion, channel standardization, and an interactive viewer with a multimodal annotation layer that combines typed text and transcribed voice notes. For each annotated segment, a feature extraction engine computes a standardized set of spectral, temporal, entropy, Hjorth, connectivity, and spike-related descriptors, stored alongside clinical descriptions in a portable JSON schema. This yields aligned feature-text pairs designed to supervise multimodal EEG-language models. The framework also includes an auto-report module that couples an ensemble of convolutional networks with a large language model to draft clinical narratives for neurologist review. Using pilot annotations, we describe how EEG-to-Report streamlines annotation workflows and produces editable draft reports, providing a reusable foundation for automated EEG reporting systems.

Large language models (LLMs) exhibit unprecedented natural language generation and many text-based problem-solving capabilities. Indeed, in many language-based tasks, for example routine coding, these artificial intelligence models have reduced, or even eliminated, the need for human input. But rather than replacing human cognitive effort, LLMs may instead serve as cognitive tools to extend human abilities, particularly when they are engaged in a task requiring open-ended conceptual exploration and creative ideation. However, we are yet to understand how these models may enhance such generative human cognitive abilities in human–AI interactions. In this study, we explore and evaluate the ability of LLMs to follow and enhance human mental trajectories during semantic memory search. To test this, we use the semantic fluency task (SFT), a classic cognitive paradigm requiring generative semantic memory retrieval that has long served to characterize convergent and divergent thinking in humans. We demonstrate that an LLM's abilities to track and predict human memory trajectories in this task exceed those of other humans.

Understanding temporal coding in neural systems is essential for decoding brain communication and advancing knowledge of neural information processing. Neural activity often conveys information through spike sequences with stereotypical temporal structures linked to specific functions. However, these sequences are subject to variability introduced by neural dynamics. Real-time closed-loop stimulation is a powerful approach to study temporal coding through adaptive control. In this work, we evaluate how a closed-loop protocol adapts to this variability to drive neural dynamics toward a desired state. It computes the Victor-Purpura distance to quantify similarity between spike sequences generated by the neural system and a triggering pattern. If the protocol determines that a neural sequence is similar to the trigger pattern, it applies stimulation to the system. This allows for an analysis of whether the system's responses are consistent and facilitates the identification of varying spike sequences that can be considered instances of the same functional temporal code. We designed two validation experiments using the Hindmarsh-Rose model: (i) detection of a temporal code and delivery of stimulation to produce brief interspersed bursts, and (ii) detection of burst onset in chaotic activity followed by inhibitory stimulation to regularize it. Gaussian noise was progressively injected to increase variability. The protocol exhibited high degree of adaptability to variability and was effective in achieving the target dynamics. The results reported in this paper suggest that adaptive closed-loop stimulation can enhance experimental methodologies for studying neural coding under realistic variability conditions.

Heterogeneous agentic retrieval-augmented generation (RAG) systems increasingly orchestrate external APIs, internal databases, vector stores, and graph stores. Exposing all tool descriptions to an LLM agent, or selecting tools only by vector similarity, causes two costly failures: over-fetching, which increases payload size, token use, and latency, and under-fetching, which omits fields needed to answer the query. We present SchemaRouter, a lightweight routing layer that represents tools, endpoints, parameters, response fields, domain concepts, units, provenance, and license policies as a schema graph. Given a query, SchemaRouter emits an executable tool plan specifying which tools to call and which fields to retrieve. A small LLM extracts intent, concepts, and source constraints, while field selection is deterministic over the graph through intent-group projection and concept-field matching with an alias layer. On a materials-science benchmark of 110 queries, SchemaRouter achieves answer accuracy of 0.71, matching fetch-everything within overlapping confidence intervals and exceeding prompt-all's 0.66, though their intervals overlap. It uses 227 retrieved-context tokens versus 2,066 for fetch-everything and achieves 2.7x lower end-to-end latency than prompt-all. It also obtains the best tool-exact rate of 0.93 and parameter validity of 1.0. SchemaRouter grounds provenance and license information in 62 percent of answers, compared with approximately 0 percent for all baselines. We also find that minimizing selected-field count is counterproductive: it reduces answer accuracy to 0.56 with negligible token savings, while recall-preserving projection restores top accuracy. SchemaRouter improves efficiency, schema-size-independent scaling, and verifiable provenance/license-grounded answering at competitive accuracy.

Small language models in the ten to one hundred million parameter range are attractive for on device inference, rapid experimentation, and controlled scientific study, yet most of them reuse the standard transformer block without adaptation to the small scale regime. We present Wiola, a decoder only language model whose novelty is concentrated in three drop in components of every layer. First, Spiral Rotary Positional Encoding perturbs the standard rotary frequencies by a slowly growing per dimension factor so that phase trajectories fan outward, improving long range discrimination while adding no parameters. Second, Gated Spiral Attention introduces a per head, content adaptive scalar gate derived from a causal cumulative statistic of the query stream, providing an implicit and differentiable form of soft head selection at negligible cost. Third, the Butterfly feed forward block replaces the conventional expansion layer with a multiplicative interaction and an intra block bypass path, matching the parameter count of a four times gated linear unit block while improving gradient flow in shallow stacks. We formalize each component, derive exact parameter and computation budgets, and prove that the gated attention admits an exact and numerically verified equivalence between full sequence training and cached autoregressive decoding, so that no approximation is introduced at inference time. We also describe a fully reproducible training and evaluation protocol on a standard tiny story corpus. The reference implementation is released as an open source package with weights ready publishing support.

Cooperative perception enables vehicles and infrastructure to exchange sensor data via Vehicle-to-Everything (V2X) communication, extending sensing coverage beyond occlusions and mitigating blind spots. While critical for autonomous driving and safety, practical deployments often rely on bandwidth-efficient late fusion. Recently, intermediate fusion has emerged as a promising approach for an optimal bandwidth-accuracy trade-off. However, in dense urban environments, cumulative bandwidth demands can overwhelm network capacity, potentially compromising safety-critical Cooperative Intelligent Transport Systems (C-ITS) functions. To alleviate these problems, this paper proposes Hierarchical Multi-Scale Semantic-Aware Cooperative Perception (HMS-SCP), a robust noise-resilient and bandwidth-efficient framework for task-oriented semantic communication in cooperative perception. HMS-SCP employs a spatial importance predictor to identify task-relevant grid elements at each scale, which are then directly mapped into complex-valued symbols for Joint Source-Channel Coding (JSCC). Unlike prior methods that rely on high-dimensional symbol projections for robustness, HMS-SCP exploits structural semantic redundancy across multiple scales to enhance resilience against channel noise, while maintaining an ultra-low symbol rate. This design significantly reduces bandwidth consumption and mitigates network congestion in high-density vehicular environments. Extensive evaluations on the simulated OPV2V and real-world DAIR-V2X datasets demonstrate that HMS-SCP effectively prevents performance collapse under severe Rayleigh fading and extreme compression ratio, maintaining high-confidence far-field detection with a real-time latency of below 16~ms, well within the safety-critical thresholds for dynamic V2X environments.

Large Language Model (LLM) serving has become a critical cloud workload, and realistic traces are essential for motivating and benchmarking serving systems. However, existing LLM serving workload studies remain limited in scale and scope. They often observe short time periods and provide limited visibility into how users interact with models in production. As a result, they do not fully capture how LLM serving workloads evolve over time or how user-model interactions shape production traffic. In this work, we further the understanding of real-world LLM serving workloads through both a global characterization and a longitudinal study of a one-year production trace from CompanyX. Unlike prior studies, our trace captures full production behavior across many models and users, including both popular and long-tail models. We analyze the workload from aggregate, temporal, model-level, and user-level perspectives, revealing workload evolution and user-model structure that are typically hidden behind aggregate views. To support future research, we publicly release the full one-year trace, enabling downstream studies of production behavior without relying on sampled or synthetically generated workloads. The trace is available at https://github.com/HarvardMadSys/chutes_workload.

We present The Architect, a system that turns Microsoft Excel into an interactive view of deep learning mathematics. A user describes a neural network in a compact table. The system then generates a workbook that shows the full forward pass and, when requested, the backward pass and parameter updates. Computed values appear as live spreadsheet formulas, while user-controlled values such as inputs, weights, labels, and hyperparameters remain editable. Excel reactively updates the dependent computations through its recalculation engine. Most deep learning tools hide the numerical details behind library calls. Many visualization tools show architecture diagrams or training summaries, but they do not expose the full arithmetic of the model. The Architect focuses on that missing middle layer. It makes matrices, activations, losses, gradients, and updates visible as inspectable spreadsheet regions, with editable controls for values users naturally manipulate. The system also produces aligned PyTorch snippets, which helps users connect formulas to implementation. This report describes the motivation, design, implementation, and use cases of The Architect. We show how the system supports introductory arithmetic tracing, learning-rate exploration, diagnosis of dying ReLU, and inspection of vanishing gradients. The main idea is simple: spreadsheets already support formulas, direct editing, reactive recomputation, and tabular layout. These properties make them a useful medium for understanding how small educational and diagnostic neural networks compute.

AI governance frameworks can be known, used, and implemented in form without becoming governance in practice. This paper examines that problem through a role-based stress test of the NIST Artificial Intelligence Risk Management Framework (AI RMF) in consumer lending. We treat framework adoption as a governance translation problem: whether RMF language can become role-usable, cross-level, authority-connected governance over the AI system-in-use, rather than producing governance-looking artifacts. The study uses LLM-based role simulation as a structured analytic probe. We apply a 4 $\times$ 2 $\times$ 3 design across four organizational roles, two AI deployments, and three governance hard cases, producing 120 scored responses. Results show that local translation was not the main problem. Simulated actors generally understood their assigned roles and translated the RMF into local activity. The harder problem was whether that activity became governance value. Actor role was strongly associated with Cross-Level Governance Value, Authority Connection, Governance Translatability, and governance value. Deployment was strongly associated with Structural Fit: the RMF fit a bounded ML underwriting model more cleanly than a workflow-embedded LLM underwriting copilot. Risk reduction was harder still. It appeared only when governance value was present and Structural Fit was full, but neither condition was sufficient by itself. The paper contributes a diagnostic account of framework-based AI governance. Frameworks create value when they help organizations see, interpret, escalate, authorize, and correct risk in the AI system-in-use. They also create value when they reveal limits of governability under existing evidence paths, authority structures, and system boundaries.

Social service navigation requires connecting help-seeking individuals to resources that satisfy their needs and specific constraints. Although LLM agents offer a promising interface for conversational resource navigation, existing benchmarks do not capture the interaction complexity and constraint-grounding demands of this setting. We introduce HoosierHelp, an interactive benchmark grounded in 3,971 Indiana public social service resources. Agents interact with simulated users, issue structured resource-search calls, handle non-ideal interactions, and select the final resources returned by the tool. HoosierHelp enhances the realism of simulated users by varying their need structure, constraint satisfiability, and behavior patterns, including impatience, rambling, unsupported requests, and self-contradiction. Experiments on 240 samples across seven LLMs show that current LLM agents remain substantially unreliable for social service navigation. Performance drops sharply on fallback-required and self-contradictory conversations, highlighting the need for agents that are more robust to complex and non-ideal user interactions.

Legal benchmarks typically score final answers even when models also state legal authority. We test whether answer correctness can serve as a proxy for authority grounding. Under ordinary reasoning prompts that did not request statutory citations, four LLMs spontaneously produced authority markers across 238 Taiwan bar-examination items. Because each item has a verified governing provision, we automatically audit answer correctness and authority grounding jointly. The two dimensions dissociate in both directions. In criminal law, 24.0–42.4% of valid responses were answer-correct but missed the gold authority, while 15.2–21.7% were answer-incorrect but cited it. A separate statutory-retrieval probe and a permissive citation-abstention intervention further show that answer and citation behavior can move separately at the output level. Because this mismatch arises without adversarial or inconsistency-inducing prompting, answer-only scoring treats naturally occurring gold-authority misses as complete benchmark successes. Because statutory authority is structurally extractable and externally verifiable, the failure can be measured automatically. A preliminary PRC civil-law extension also observes citation-unrequested authority marking, motivating a full cross-jurisdictional joint audit. We therefore propose joint answer–authority evaluation for statute-grounded legal benchmarks.

This paper presents an information-gain-guided reinforcement-learning sensor-selection framework for single-vessel tracking in heterogeneous maritime sensor networks. The proposed approach is motivated by information-theoretic sensor management: instead of activating all sensors or repeatedly performing computationally expensive online expected-information-gain evaluation, a learned policy selects one tracking-relevant sensor at each decision epoch. A Bayesian sequential Monte Carlo tracker estimates the vessel state from noisy measurements and provides a belief representation for scheduling under nonlinear and non-Gaussian conditions. A Proximal Policy Optimization agent selects one of five sensors in a georeferenced simulation of the CMMI Smart Marina testbed at Ayia Napa Marina, Cyprus. The policy is trained on the testbed's actual five-sensor configuration. The agent observes belief-state, detection-history, coverage, sensor-geometry, and realized-information-gain features. The reward is defined as a realized-information-gain term gated by an observability mask. Final-test simulations compare the proposed framework with random single-sensor selection, always-on sensing using all sensors simultaneously, and the expected-information-gain sensor-selection baseline proposed in our previous work. Results show that the learned policy achieves tracking performance close to always-on sensing while activating only one sensor per decision time step and avoiding the computationally expensive online entropy search required by expected-information-gain selection. Additional zero-shot evaluation without retraining on ten moderately perturbed versions of actual layout configuration showed broadly stable tracking, with any increase in positional tracking error remaining below 1 meter across all perturbations.

We prove that two canonical local synaptic learning rules, the potentiation arm of spike-timing-dependent plasticity (STDP$^+$) and homeostatic plasticity (instantiated here via flashlight granule-cell-like neurons), together can implement the exact gradient of a SIGReg-like self-supervised learning objective. The equivalence requires no gradient calculations, no global error signals, no weight transport, and no label information: the only inputs are pre- and post-synaptic firing rates, local firing statistics, and the temporal contiguity of natural sensory streams. On a synthetic clustering task designed to probe whether class structure can be recovered from temporal ordering of inputs alone, ordered presentation raised cluster separation (CSR) to 2.49 while random ordering left it near baseline (0.83), a roughly threefold ($\approx 3.5\sigma$) separation attributable solely to input ordering. On temporally ordered MNIST, a two-layer network trained entirely with these rules achieved 87.3% linear-probe accuracy, showing that the mechanism functions end-to-end.

Large language models (LLMs) are increasingly used for program-aided reasoning, agentic decision making, and structured task execution, but these settings often incur substantial inference cost. Many such requests share similar computational structures while differing in variables, constraints, or contexts, creating opportunities for program-level caching. Since program caches need to reapply reusable computation logic to new requests, their key steps often involve lightweight and structured operations such as variable extraction, program binding, and generation acceleration, which are well suited for small models. We propose CacheSpec, an inference optimization framework centered on reusable program caches. The framework converts Program-of-Thoughts (PoT)-style programs from one-time reasoning artifacts into reusable cache objects, and reuses the same small model for two roles: semantic variable extraction on the cache-hit path and speculative drafting during target-LLM generation. Experiments on shopping-style request datasets, WebShop, Formula, and CodeTAT-QA show that CacheSpec reduces inference latency and improves effective cache reuse while preserving comparable or better task quality than existing caching and generation baselines, achieving up to about 3.1$\times$ latency speedup; in parallel serving experiments, it improves throughput by about 2.8$\times$ over PoT-style methods. These results suggest that the sweet spot for small models in large-model inference systems lies not in solving complex tasks independently, but in performing lightweight, structured, and verifiable auxiliary operations.

Document-based LLM systems often flatten a PDF before guardrails inspect it. That step can discard evidence that an instruction was never visible to the user. We introduce CrackedPDFs, a controlled benchmark for hidden prompt injection in PDFs. The benchmark contains 29,322 generated PDFs from 4,983 base documents. It includes 9,774 injected files and 19,548 benign or matched-confounder files. We evaluate PromptGuard and a rule baseline. We also evaluate structural-only learned models and a sanitized hybrid detector. The evaluation uses held-out provenance splits and paired benign-confounder controls. It also uses label-shuffle checks and shortcut audits. On a 2,919-document held-out test set, the hybrid detector reaches 0.960 F1. ROC-AUC is 0.998 and PR-AUC is 0.997. On a balanced subset containing 973 injected PDFs and 973 matched benign confounders, the hybrid detector achieves 95.9% classification accuracy. Using score ordering, it ranks the injected member above its matched confounder in 100% of 973 pairs. PromptGuard has low recall when given extracted text only. Structural-only learned models are weak under paired controls. A text-only TF-IDF model reaches perfect held-out scores but fails shortcut audits. These results show that document-aware hybrid detection is useful under controlled paired evaluation. They do not show broad real-world robustness or reliable cross-family generalization.

Small language models are attractive backbones for interactive agents, but direct distillation from strong teacher trajectories often turns rich multi-turn behavior into one-shot imitation targets. This is inefficient in long-horizon environments, where early decisions shape later states and rewards. We propose Prefix-GRPO, a reinforcement learning framework that decomposes teacher trajectories into replay-aligned prefix queries and online continuations. Each prefix is replayed in the environment to recover a valid intermediate state, after which the student continues online interaction and receives task reward. Unlike response-only GRPO, Prefix-GRPO also applies clipped policy updates to historical assistant tokens inside the replayed prefix, using a policy-distilled SFT checkpoint to estimate their old log-probabilities. This unifies prefix learning and continuation learning within the same policy-optimization form. Experiments on TextCraft, BabyAI, and ALFWorld show that Prefix-GRPO improves small-model agents over distillation and standard RL baselines, while ablations show that replay alone is insufficient without explicit prefix-token optimization. The implementation and reproduction scripts are available at https://github.com/HappynessI/Prefix_GRPO.

Generalizing across subjects remains challenging in invasive neural recordings because electrode configurations, anatomical structures, and neural signal patterns vary substantially across individuals. To investigate such inter-subject variability, we propose a cross-subject semantic decoding framework that aligns neural responses to speech perception from multiple subjects into a shared latent space and learns a mapping from the aligned neural representations to contextual embeddings. More specifically, using electrocorticography data collected during natural language comprehension, we estimate the shared space using the shared response model and train a decoder to predict contextual semantic embeddings from projected neural responses. For a held-out subject, we estimate a subject-specific projection into the predefined shared space, and directly apply the pretrained decoder without any retraining. Experimental results demonstrate that the proposed framework consistently outperforms baseline methods across evaluation settings and exhibits a reduced performance drop from source subject to held-out subject testing, indicating improved cross-subject generalization. These results suggest that aligning neural activity into a shared latent space, while decoding in a semantic embedding space, provides an effective strategy for improving cross-subject generalization by reducing subject-specific differences in neural responses while effectively capturing shared stimulus-related representations.

Existing approaches for reasoning about action and change provide expressive semantics for modeling dynamic systems, in most cases built on top of logic programming systems. We introduce a modular framework for transition and trajectory verification based on Tiles and implemented in Soda, which is an efficient functional programming language. The framework operationalizes action language semantics through executable verification pipelines that process states, actions, transitions, and rules as compositional functional components. Verification procedures are represented as typed functional pipelines, enabling modular specifications, reusable reasoning components, and transparent execution workflows with guaranteed pipeline termination. The framework includes an executable specification layer that allows users to define domain descriptions in YAML, which are operationalized into the underlying verification model and executable pipeline structure. We provide an open-source implementation and illustrate the framework through examples that involve misinformation and emotional reasoning.

Environmental monitoring with unmanned aerial vehicles (UAVs) requires route planning methods that maximize covered area while handling energy limits, operational constraints, and geometric complexity. This paper reports the protocol and preliminary results of an ongoing systematic literature review (SLR) on autonomous UAV route planning for coverage-oriented environmental monitoring. The review follows the PRISMA 2020 framework and searches Scopus and Web of Science for studies published between 2015 and 2026. The protocol focuses on path planning, coverage path planning, and informative path planning, with emphasis on algorithmic families, coverage and energy metrics, obstacle handling, geometric environment representations, and environmental constraints. At the current stage, 562 records have been identified, 161 duplicates have been removed, and 401 unique records have been screened by title, abstract, and keywords. From these, 247 studies were retained for full-text eligibility assessment (235 eligible and 12 borderline records to be resolved during full-text review). A preliminary analysis of the retained studies suggests strong concentration on coverage-oriented formulations, multi-UAV coordination, and energy-aware optimization, while fewer studies explicitly address weather, uncertainty, or obstacle-rich environments. Most retained studies rely on simulation-based validation, highlighting a potential simulation-to-reality gap, and recent publications show increasing interest in reinforcement learning, hybrid optimization, and geometry-aware planning. These early findings indicate an active but fragmented research landscape and support the need for a structured synthesis to identify mature techniques and unresolved gaps for realistic environmental monitoring missions.