Papers

Open-source RISC-V GPGPUs such as Vortex ship with directed-kernel regressions but no reference-model checking, functional-coverage model, or sign-off discipline. This paper presents a UVM 1.2 environment and methodology that closes that gap. The environment wraps a bus-master SIMT DUT with role-inverted agents, integrates Vortex's functional simulator (SimX) as a per-configuration golden model over DPI-C, and renders verdicts via two injection-qualified checkers: a bidirectional end-state scoreboard and a per-instruction, per-lane lockstep comparator under five SIMT alignment rules. A two-pass load-value feed makes racy fenceless multi-core programs instruction-granularity verifiable (residual zero over 5,432 retirements), with the interrupt-timing boundary stated. A three-layer coverage model adds, to our knowledge, the first published SIMT functional-coverage layer for RTL GPU verification (divergence depth, bank conflicts, coalescing classes), closing its own layers at 98.1% covergroup-bin / 94.7% total (ISA layer separately: 83.1% bins / 89.3% weighted) under machine-generated, RTL-cited exclusions and a blocking waiver-integrity gate; an unstimulated D-extension elaboration affects no functional bin, only lowering totals. The checking depth surfaced real defects on both sides of the comparison: a JALR LSB ISA deviation, a non-scaling watchdog constant (since fixed upstream), a reset-relay X window found by restoring a silenced assertion, a missing AXI error path proven by fault injection, and a reference-model fetch bug found by the lockstep itself. FuzzGPU (USENIX Security 2026), a concurrent RTL GPU fuzzer on the same DUT, is complementary: fuzzing finds bugs, this methodology quantifies sign-off; both independently found the JALR deviation. All findings ship in an evidence-cited register; every number carries provenance and the method's boundaries are stated rather than waived.

Protocol optimization in computed tomography (CT) aims to improve diagnostic image quality while reducing radiation dose, but the interdependence of acquisition and reconstruction parameters makes exhaustive testing impractical. We propose a virtual imaging trial framework with reinforcement learning for efficient CT protocol optimization. Sixty-three computational human models with liver lesions were imaged using a validated CT simulator across 468 combinations of acquisition and reconstruction parameters, including tube voltage, tube current, reconstruction kernel, slice thickness, and pixel size. The optimization objective balanced liver lesion detectability, quantified by detectability index d-prime, against radiation dose. A Proximal Policy Optimization agent was trained and conditioned on patient-specific CT localizer embeddings derived from a pretrained vision transformer. On held-out patients, evaluating only 8 protocols per patient, about 2% of exhaustive testing, recovered 98.2% of the exhaustive-search oracle objective. With no patient-specific simulation, surrogate scoring alone achieved 89.7% recovery. Conditioning on the localizer improved zero-simulation recovery by 10.7 percentage points over the localizer-blind policy (paired 95% CI 2.9-19.5; p=0.02). These results show that the proposed framework can substantially reduce exhaustive protocol testing while enabling task-based, dose-aware protocol selection before the diagnostic scan.

A companion evaluation found that naming the target part in a manipulation prompt increased action accuracy by 0.32-0.63 across eight vision-language models, with no model outperforming a constant baseline until the part was named. However, naming the part supplies information that a real system must infer, confounding visual grounding, mechanical reasoning, and category-to-action association. We introduce GroundBench, a diagnostic benchmark that separates these explanations through six branch-and-merge conditions, each adding a controlled information bundle, and a counterfactual re-ask targeting a real alternate part visible in the same image. Across three OpenAI models and 1,068 predictions, supplying the target region without its identity leaves action accuracy at or below the 0.53 majority baseline (0.26, 0.26, and 0.53), although the models largely reproduce the supplied region. Supplying identity without location instead yields 0.74, 0.68, and 0.68. Every above-baseline gain in this curated set occurs where the supplied part category itself determines the action. A no-vision control leaves GPT-5's scores unchanged or improved, providing evidence consistent with substantial category-to-action association. GPT-4o mini declines on one condition, so this interpretation is not universal. Adding joint type and motion axis does not improve accuracy across six model-stratum comparisons. On 74 counterfactual pairs from 32 objects, GPT-5 achieves 0.86 pair-weighted compliance with a 0.07 shortcut rate but fails all observed push-to-lift-vertical cases. GroundBench identifies which supplied information changes affordance behavior and tests whether apparently grounded performance can be reproduced through textual shortcuts.

Object tracking is essential for autonomous vehicles to avoid obstacles and plan routes. Radar maintains detection performance even in adverse weather and can measure relative velocity through the Doppler effect, making it well suited for object tracking. In this paper, we propose a data-driven state PRedictor-based Interacting Multiple Model tracking method (PR-IMM) that improves nonlinear object-motion representation while preserving the stability and interpretability of physics-based motion models. The proposed method employs a transformer-based PRediction model (PR) that incorporates radar Doppler measurements to predict object displacement. The PR model is integrated into the IMM as a mode alongside the CV, CA, and CT motion models, and their prior positions are dynamically combined according to the mode probabilities. Experimental results show that PR-IMM reduces position-estimation error by 57.3% over the IMM and by 16.5% over the PR, while reducing ID switches by 25.3% and improving IDF1 by 9.6% over the IMM.

Real-time aerial small object detection is an important visual signal and image processing problem, requiring a detector to preserve fine-grained localization while avoiding redundant computation on large background regions. This paper focuses on this deployment-oriented aerial/UAV setting rather than claiming a universal detector for all object detection scenarios. Existing Transformer-based detectors provide strong global modeling, but their dense query initialization and multi-layer decoder still spend substantial computation on background tokens, which is inefficient when small objects occupy only sparse image regions. To address this problem, this paper proposes HGSQ, a Heatmap-Guided Sparse Query Detector for real-time aerial small object detection. HGSQ uses a lightweight Heatmap Budget Predictor (HBP) to predict a foreground budget map in a single forward pass. The predicted heatmap is then used by three fixed components: Heatmap-Guided Sparse Query Selection (HSQS), which initializes decoder queries from high-confidence foreground positions; Heatmap-Gated Lite Snake Convolution (HGLSConv), which performs local shape refinement only on heatmap-activated small-object regions; and Adaptive Query-Decoder Budgeting (AQDB), which adjusts the query budget and decoder depth according to the estimated object density. Unlike post-hoc heatmap generation, HGSQ treats the heatmap as a real-time computation budget rather than a visualization map during deployment. Experiments on NWPU VHR-10 and VisDrone2019 show that HGSQ achieves 95.10 mAP50 on NWPU VHR-10 and 54.8 mAP50 on VisDrone2019, while reducing GFLOPs to 48.6 and running at 96.0 FPS on an RTX 4070 under our TensorRT FP16 deployment protocol.

The primary goal of this work is to reproduce the behavior of a human trader, detailing his or her psychological processes to understand the effects of his or her decisions on the final value of an asset. The second goal is to use a formal language to detail this behavior as a tool for improving and simplifying the communication between all actors involved in the project, such as specialists from disciplines as diverse as computing, economics and psychology. As a starting point, we use a paper that shows an experiment that analyzes the influence on other trader's behavior when an agent handler and a trading robot attempt to distort the market. This work reproduces this experiment, using virtual traders that belong to a multi-agent simulation model, showing the feasibility to reproduce complex human behaviors and showing the convenience of use formal and graphical languages to simplify the understanding and the validation of the complex behaviors involved in an economic process.

Medical image classification is frequently complicated by transitional categories whose feature distributions overlap those of adjacent classes, producing ambiguous decision boundaries. Conformal prediction returns uncertainty-aware prediction sets, but these are not directly actionable in clinical screening, where a single decision is required. This work proposes adaptive conformal redistribution (AdaConRed), a label-free post-conformal decision rule that converts ambiguous prediction sets into refined class assignments. A five-stage pipeline is developed. Vision-language generative augmentation addresses minority-class scarcity; a frozen DermFoundation encoder provides embeddings; a lightweight multi-layer perceptron performs classification; an entropy-modulated, margin-aware nonconformity score constructs adaptive prediction sets; samples predicted as transitional with multi-label sets are reassigned to the most probable alternative class within the set, using only model outputs at inference. Evaluation uses the OSCC oral lesion and ISIC skin lesion benchmarks at a miscoverage level of 0.2. On the 3-class OSCC benchmark, overall accuracy improves from 73.54% to 77.38%, with oral cancer accuracy rising from 64.29% to 82.14% and benign accuracy from 56.57% to 70.20%. Reassignment of transitional samples reduces OPMD accuracy from 84.78% to 80.16%, consistent with the asymmetric cost of missed malignancy. On ISIC, overall accuracy improves from 85.83% to 87.19%, melanoma accuracy rising from 66.04% to 68.34%. AdaConRed outperforms LAC, APS and RAPS under an identical backbone and redistribution rule. Conformal prediction can be extended beyond uncertainty quantification toward actionable decision support where transitional disease categories are present, with gains concentrated in the clinically critical malignant categories. Code repository: https://github.com/saibal436ghosh/AdaConRed.

Generative artificial intelligence (AI) is increasingly used to support human decision-making, yet less attention has been paid to how AI may influence the reasoning processes that precede final judgments. This pilot study explored whether different AI-use conditions were associated with differences in reasoning breadth during ethical decision-making. Twenty-nine participants completed an ethical dilemma under one of three conditions: AI-Prohibited (n = 10), AI-Optional (n = 10), or AI-Mandatory (n = 9). Responses were evaluated by three independent blind coders using two exploratory measures: the Counterargument Diversity Score (CDS) and Perspective Diversity Index (PDI). Participants across conditions generally converged on similar ethical conclusions, most commonly favoring disclosure and customer protection. However, participants in the AI-Mandatory condition considered a broader range of perspectives, including legal, regulatory, organizational, technical, and ethical viewpoints. A statistically significant overall difference in PDI scores was observed across the three conditions, whereas differences in CDS were not statistically significant. These findings suggest that generative AI may not necessarily alter final ethical judgments but may be associated with broader exploration of perspectives prior to reaching those judgments. Given the small sample size, the findings should be interpreted cautiously and examined in larger studies.

Autonomous coding agents can remember an experiment yet carry forward a conclusion it does not justify. We reconstruct how evidence is reused in a 400-task NeuroGolf campaign, with selected wellbore-prediction records from the same operator as cross-domain comparisons. A numerical counterexample exposes an overbroad exclusion; other episodes distinguish what failed, incomplete and revised programs justify doing next. The wellbore records reveal an additional weakness: a coordinator correctly acknowledges a novelty-only rejection, then summarizes it as measured closure. Separately, a prefix-based acceptance gate at three thresholds yields worse target scores than ungated adaptation in the retained experiment. These cases motivate an inspectable handoff linking the tested proposition, its scope, candidate and evaluator identity, check statuses, and reopening condition. A dispatch tree separates investigation, repair, reopening and stopping; a promotion predicate requires that each check both passes and has evidence supporting its use for the requested decision. The practical lesson is to preserve not just experimental results, but their limits on subsequent action.

Anomaly detection in structured images is challenging in small-data settings where deep learning approaches are costly or impractical. Classical template matching is simple and interpretable but lacks robustness to geometric variations such as scale, rotation, and perspective. We propose a variational template matching framework that represents anomaly templates as a family of transformed instances and performs detection via normalized cross-correlation over this transformation space. To further improve robustness, we introduce a density-based statistical anomaly score derived from local intensity distributions using kernel density estimation (KDE). This produces a smooth representation that captures distributional concentration and tail behavior more robustly than histogram-based methods. The structural and statistical signals are integrated through a unified fusion formulation, enabling complementary modeling of geometric similarity and distributional deviation. Experiments on biological cell images demonstrate that the proposed method outperforms classical baselines and achieves competitive performance with ResNet-50 under a fully training-free setting, while providing explicit localization. The approach offers an efficient, interpretable, and practical solution for anomaly detection in structured image domains.

In this paper, we describe the Mission Performance module implemented for a fully autonomous racing car to automatically manage the longitudinal, lateral, and combined performances, aiming to speedup the laptime progression while assuring safety. Motivated by the difficulty and risks of applying the real-time estimation of the grip to critical modules like the motion planner and controller, the Mission Performance guides these modules adapting their target performance instead of changing the vehicle model parameters. The module is formed by pre-defined progressions to warm up the tires at the beginning of a run. Then, the system continuously monitors safety and vehicle dynamics metrics on a per-sector basis to adaptively reduce, maintain, or increase the performance levels for each sector, progressively converging toward the maximum allowed value. The solution's effectiveness is demonstrated on the EAV-25, a fully autonomous Dallara Superformula, at the Yas Marina Circuit during the Abu Dhabi Autonomous Racing League (A2RL) Season 2.

Automated decision-making (ADM) systems are increasingly deployed in domains such as mortgage lending, prison sentencing, health insurance coverage, and hiring. Designing a responsible ADM system in such high-stakes domains requires ensuring privacy protection, fairness across demographic groups, and robustness against adversarial manipulation. However, prioritizing one of these objectives comes at the cost of another, forcing a choice as to which tradeoff to accept in a deployment. These tradeoffs explicitly or implicitly impact the life, safety, and fundamental rights of the people in a society, and thus, the perceptions and priorities of this population are needed before we can produce appropriate solutions. To this end, we conducted a quasi-experimental study (N = 777) in which participants evaluated four decision-making scenarios with controlled tradeoffs. Participants significantly preferred human decision-making (HDM) over ADM in three of four scenarios, emphasizing the value of human judgment, contextual understanding, and the ability to incorporate non-quantifiable factors. Furthermore, in terms of tradeoffs, our findings not only show that participants' preferences are highly context-dependent, but also that their perception of a specific objective, fairness, extends beyond formal definitions. Participants interpret fairness through multiple lenses, including privacy risks and susceptibility to manipulation, and view unfair or manipulated outcomes as failures of accuracy. Overall, our findings highlight the importance of context-aware and human-centered approaches when designing and governing ADM systems in high-stakes situations. Rather than purely technical objectives, it is essential to evaluate ADM systems based on how their tradeoffs align with specific expectations within a given domain, as well as with societal values and perceptions of harm and fairness.

Industrial fault diagnosis often operates with only a handful of labeled fault examples, making few-shot learning attractive for sensor monitoring. Standard prototypical networks are simple and effective; however, their class prototypes may become unstable in the very-low-shot regime because each decision relies on a small support set. We propose Multi-Episode Prototypical Networks (MEPN), which aggregate prototypes from multiple disjoint support episodes and use their mean as the final class representative, reducing prototype variance without changing the encoder architecture. We evaluate MEPN on the DeFACTO sensor dataset using five-way fault classification with synthetic bias, drift, spike, and noise faults injected into real industrial measurements. Over 100 independent runs, MEPN reaches \SensorOneShotGcpn% in the per-episode one-shot setting ($K\!=\!1$ shot, aggregated over $N_{\text{agg}}\!=\!10$ support episodes), substantially above single-episode baselines. Under an equal 10-sample support budget, MEPN and ProtoNet at $K\!=\!10$ are statistically indistinguishable, confirming prototype accumulation as the mechanism rather than superior fixed-budget learning.

Introducing large language models (LLMs) into evolutionary computation enables semantic generation of candidates, but the classical problem of maintaining population diversity remains. Inspired by the thermodynamical genetic algorithm (TDGA), we propose T-GADE (Thermodynamical Generative-AI-Driven Evolution of LLM Artifacts), an evolutionary computation framework extended through LLMs. T-GADE keeps the notions of a population and genetic operators, takes structured artifacts generated by an LLM as individuals, and implements crossover and mutation as LLM operators. It further introduces the HoxLM, an LLM that attempts self-repair of an individual after evaluation and before survival selection; two update schemes, generational and steady-state; and Fermi-type and Bose-type occupancy rules that define solution identity and whether duplicates are permitted. Survival selection is unified as minimization of the free energy $F_T=\langle E\rangle-TH$, and the selection temperature $T$ explicitly controls the balance between quality and diversity. We tested an implementation example aligned with Evolution of Heuristics (EoH) on the online bin-packing task, where the excess is the bin count in excess of the volume lower bound, relative to that bound. The HoxLM-equipped configuration reached lower excess with fewer total LLM calls than EoH. In the preregistered comparison at 400 total calls it won 9 of 10 seed-paired runs, with a median sealed excess of 0.836% against 3.912%, and it was also better in the mean and in the worst case. Temperature controls the spread of the retained population under both update schemes, and a small positive temperature improves training quality over the collapsed $T=0$ point in the generational Bose-type scheme. The evolved heuristics are shown as readable scoring rules.

Vision-Language Navigation (VLN) in unseen indoor environments is useful in real-world robotics, where an agent must follow natural-language instructions, locate objects, and answer spatial questions without a pre-built map or fixed object vocabulary. Multimodal vision-language models (VLMs) provide strong open-vocabulary grounding and zero-shot reasoning, but struggle to emit reliable metric quantities such as range, bearing, and comparative spatial relations directly from images. Existing approaches address this by folding geometry into hand-engineered pipelines or asking models to output waypoints, requiring changes to the control stack for different robots, tasks, or vocabularies. We introduce AnchorVLN, an open-vocabulary VLN system built on a simple rule: the VLM proposes semantics; geometry decides metrics. It is realised as EMBODIED-NAV-MCP, a Model Context Protocol (MCP) server driven by a VLM agent through a compact set of callable tools. Since no tool accepts distance in metres or bearing in radians, the schema enforces the semantic-geometry boundary without modifying the downstream autonomy stack. We benchmark both tasks of the CMU Vision-Language Navigation Challenge 2026: 30 instruction-following questions over 15 scenes and a frozen 45-question object-reference set. The full system achieves 64.4 percent on instruction following, dropping by 13.3 percentage points without controller modeling (t = 2.77). On object reference, geometric anchoring clears the challenge overlap threshold on 10 of 45 questions, versus 0 of 45 for direct coordinate estimation, reducing median center error from 3.37 m to 2.48 m.

We study the selection and stable realization of local approximation spaces in Trefftz discontinuous Galerkin discretizations of the Helmholtz equation. A scaled Cauchy-trace inner product places plane waves and Fourier-Bessel functions in a common geometry: Fourier-Bessel modes are orthogonal with explicit weights, while the same weights determine the circulant spectrum of an equispaced plane-wave trace Gram matrix. This separates amplitude scaling from genuine trace-rank loss and yields an exact best-approximation identity for mixed plane-wave–Fourier–Bessel spaces. With complex plane-wave angles, the unresolved modal tail is an exponential sequence, so propagating and evanescent components can be identified by the same ESPRIT/variable-projection procedure. We prove exact recovery and a perturbation estimate for the recovered angles and, under the standard PWDG quasi-optimality bound, transfer these perturbations to the DG error. Trace-Riesz orthonormalization is then separated from a graph–Riesz normalization of the assembled operator. Numerical experiments verify the identities, recover sparse ray fields to roundoff, and resolve a propagating-to-evanescent transition without a prescribed critical angle.

World models let agents plan by predicting the consequences of their actions, but changes in the environment can make them inaccurate. We study the problem of adapting a world model to an unknown test-time environment, drawn from a known environment family, using only a few episodes of interaction. Existing approaches trade off computational cost against expressivity, i.e., the range of models a method can produce. For example, in-context learning is computationally cheap but limited in expressivity, and gradient-based adaptation is expressive but computationally expensive. We present CLAW (Context-conditioned Low-rank Adaptation of World models), which addresses this tradeoff by using a hypernetwork to generate low-rank (LoRA) adapters at test time. During pretraining, we simulate adaptation to a variety of environments and jointly train the hypernetwork and base world model. At test time, we freeze the base model and use a forward pass of the hypernetwork to generate adapters from a small batch of test-time transitions. We evaluate CLAW in locomotion and manipulation environment families that vary in dynamics, embodiment, and reward. We show that, using only seconds of test-time data, CLAW outperforms gradient-based adaptation and in-context learning during online adaptation. We also show that CLAW avoids overfitting in data-scarce regimes, that its advantage comes from the expressive adapters rather than context conditioning, and that pretraining the hypernetwork jointly with the base model outperforms training it post hoc.

Electronic health record (EHR) foundation models trained on longitudinal patient trajectories have demonstrated strong performance across diverse clinical prediction tasks. However, their clinical reasoning capabilities remain constrained by next-token prediction on limited and incomplete EHR data. To address this, we propose a reinforcement learning (RL) fine-tuning framework that treats EHR foundation models as generative policies over patient trajectories. We formulate common clinical prediction problems (e.g., hospital readmission) as event-conditioned, time-windowed reasoning tasks. We then design time-aware, rollout-sensitive rewards to account for finite rollout lengths and temporally inconclusive outcomes. We find that RL fine-tuning consistently improves over pre-trained backbones and strong baselines. Notably, it enables smaller models to match larger pre-trained models in data-limited regimes and induces positive transfer across tasks. Further analysis shows that RL fine-tuned models generate trajectories with stronger structural and semantic alignment to ground truth and greater downstream utility.

GPU-backed LLM servers multiplex interactive requests with background batch work on the same CPUs; a fixed kernel policy serves one objective and loses the other, so agentic OS control needs to switch scheduler behavior as the workload changes. The hard part is applying the switch safely and fast enough for the kernel: scheduler events occur every 1-10 $\mu$s, and any code running there must satisfy the eBPF verifier. Scalar knobs are fast but limited, while generating eBPF policy code puts compilation, verification and possible rejection on the runtime path. We present AKTS, which verifies a policy library once, at load time, and reduces the agent's runtime action to writing an integer index into an in-kernel array of preverified policies, resolved by a tail call. Because the agent emits an index rather than code, verifier failure is not a runtime outcome. On Linux 6.14, AKTS applies a policy switch in 920 ns (p50); makes an invalid index inert across 60,217 invocations on an attached scheduler; and switches policies in a vLLM workload to capture 97% of a throughput policy's batch work while matching a latency policy's burst response.

Assessing human workload for technology-mediated tasks helps prevent task failure caused by poor technology design. Traditionally, workload is assessed retrospectively using the NASA Task Load Index (TLX) after humans complete a task. What if we could forecast workload before a human attempts a task using agent simulation? We introduce Synthetic TLX, a new paradigm for proactive workload estimation that predicts NASA TLX scores for a given task, unlocking novel interaction opportunities and evaluation methods. To understand its viability, we conducted three experiments comparing human and agent-generated scores to evaluate where they align and diverge. We found agent estimates align with human scores particularly when prompted with a human persona and active task simulation. However, agents and humans diverge in the sources of workload they are sensitive to. Based on our findings, we present three applications to showcase Synthetic TLX's potential and discuss the future of workload-aware human-AI interaction.

Braille literacy among blind school-age children has fallen sharply, in part because producing illustrated braille pages still requires specialized software and trained labor. We present Tact, a browser-based pipeline that converts a spoken or typed story idea into printable braille with a matching raised tactile illustration, without an account or mandatory cost and with an offline-capable path. The paper documents the engineering history of the system: its sighted-operator ethical model; hardware rationale for consumer fused-deposition modeling; physical braille geometry and printer calibration; local, hosted, and fallback language-model paths; a deterministic Grade 1 braille translator; verified page layout and pagination; a 93-shape hand-drawn tactile illustration library; and synthesized sound design for a voice-first interface. We report engineering verification, ethical commitments, limitations, and the work required before the system is ready for real blind and low-vision readers.

First-stage recommenders in multi-stage systems produce a ranked candidate list from which a limited prefix is forwarded to downstream rankers. Because each forwarded item must be processed by more expensive ranking stages, this shortlist cannot be arbitrarily large. The first-stage objective is therefore high coverage of relevant items within the forwarded prefix, commonly measured by Recall@$k$. A relevant item may be available deeper in the retrieved list but absent from the shorter prefix that is actually consumed. This paper studies post-hoc verification for promoting such candidates into the consumed shortlist without retraining or replacing the retriever. We introduce a lightweight generative verifier for retrieval models. Given a retriever state and a candidate item, the verifier scores the item through the likelihood of its identifier tokens. It is trained post hoc with next-token cross entropy, requires no sampled negatives or candidate pool during training, and scores only the retriever's top-$K$ candidates at inference. The interface is minimal: the retriever supplies a query state and candidate items, and the item representation can use any fixed tokenization. Across Amazon product recommendation and YaMBDa music recommendation, the same verifier training recipe improves Recall@10 for SASRec, GRU4Rec, NextItNet, and MiniOneRec. Ablations show that the improvements are not explained solely by injecting item-content features into the retriever, supporting verification as a post-hoc output-side adaptation mechanism.

Retrieval-augmented generation (RAG) can improve knowledge-intensive question answering, but the first design choice is easy to overlook: how should the source corpus be partitioned into retrievable units? Fixed-size chunks often return long passages whose relation to the question is only implicit. We introduce EAR, an Entity-Aware Partitioning approach for multiple-choice question answering (MCQA). EAR extracts normalized surface anchors from the question, answer options, and corpus; retrieves local windows around matching corpus anchors; and can attach a larger parent passage through an extractive summary. We evaluate EAR on a cleaned Massive Multitask Language Understanding (MMLU)-style subset of 153 questions selected by an automatic corpus-support heuristic and using decontaminated public textbook text. Across same-protocol top-k = 3 and top-k = 8 sweeps with Mistral, Gemma, and DeepSeek, EAR entity-window reduces retrieved words by 37.5-40.2% relative to chunks. Observed accuracy changes are +5.2, +1.3, and -3.9 points at top-k = 3, and +5.9, -3.3, and -4.6 points at top-k = 8; none of the entity-window differences is statistically significant. The scoped contribution is methodological: EAR provides a compact and inspectable retrieval unit, while its rule-based anchor extractor remains domain-specific and requires separate validation before transfer.

Learning user-defined concepts as constraint networks has been extensively studied in the constraint acquisition (CA) literature. However, existing approaches typically rely on intensive interactions with a human oracle, making the learning process costly in terms of time and number of queries. In this paper, we propose a neuro-symbolic framework for automatic CA that significantly reduces user involvement by introducing neural Oracle Transformer models which learn to emulate user responses and to generalize conceptual knowledge. Trained on previously available examples, the learned oracle interacts with a dedicated CA engine, FastCA, which systematically refines the oracle's responses into a sound, consistent, and interpretable constraint network. This neuro-symbolic interaction enables the recovery of structured symbolic models from data without prior domain knowledge. Our results demonstrate that this neuro-symbolic interplay effectively aligns data-driven pattern recognition with symbolic reasoning, offering a robust approach to automating model construction in combinatorial domains.

It is known that the Sherman–Morrison (SM) formula is not numerically stable. In recent work, we introduced SMIR, an algorithm that incorporates iterative refinement to enhance the SM backward error. In this paper we take a different route: adapting an algorithm of Govaerts, originally designed for general bordered linear systems, we develop a modified Sherman–Morrison (MSM) method whose built-in self-correction makes it surprisingly resilient. Whereas standard SM requires the solution of two $n\times n$ linear systems, MSM requires three; by contrast, SMIR requires $2+k$ solves, where $k$ is the number of IR steps and can be substantially larger than three, when IR converges slowly. We then derive backward and forward error bounds for both SM and MSM. The SM backward error bound established here is stronger than the one proved in [Hashemi & Nakatsukasa 2026]: it accounts for every rounding error and holds with no conditions on the size of the capacitance. From these bounds we extract growth factors that are cheap to compute a posteriori and can be used to certify the backward and forward stability of SM and MSM on a given problem. In our experiments, MSM consistently produces backward stable solutions, and is therefore observed to be forward stable as well. A formal proof of the stability — or instability — of MSM remains an open problem.

Large Language Models (LLMs) are increasingly asked to reason over structured data such as graphs, yet how reliably they can carry out multi-step graph algorithms in language remains unclear. Existing evaluations tend to use simple tasks on small graphs, to score code generation rather than reasoning over the graph itself, or to fix a single input format. We introduce Graph Theory Bench (GT Bench), a benchmark covering 24 classical graph problems in 44 task-structure settings, with over 100,000 examples across four representations: natural language, structured language, adjacency list, and adjacency matrix. Evaluating eight LLMs on GT Bench shows that accuracy is strongly tied to the input representation, that the best representation shifts with graph density, size, and topology as well as with the model, and that this sensitivity persists, attenuated, in the strongest reasoning models. Building on these observations, we propose the Graph Theory Agent (GTA), which pairs a preference-trained representation selector with plan-and-decompose scaffolding around a frozen executor LLM. GTA lifts Phi-4 from 53.5% to 69.1% on the benchmark's easy split and from 33.0% to 41.5% on its hard split, outperforming eight prompting and agent baselines, and transfers without retraining to GraCoRe and NLGraph. Code for benchmark generation and evaluation: https://github.com/xzx34/GTA. The project homepage is available at https://xzx34.github.io/gta/.

Designing effective chemotherapy regimens is hindered by tumor heterogeneity and drug resistance, which complicate the deployment of patient-specific model-based optimal control across diverse populations. We develop and compare closed-loop deep reinforcement learning (DRL) dosing policies with continuous (TD3) and discrete (DQN) action spaces trained on a high-dimensional heterogeneous tumor model. The DRL policies are benchmarked against a Pontryagin's Maximum Principle (PMP)-derived open-loop benchmark. We assess generalization under parametric heterogeneity using a 100-patient virtual cohort with plus or minus 10 percent uniform perturbations in growth and drug-sensitivity parameters. Across this cohort, TD3 achieves higher average tumor reduction, while DQN yields tighter inter-patient dosing consistency, revealing a clear efficacy-consistency trade-off in this study. Our simulations assume full observation of all tumor subpopulations; translation to sparse and noisy clinical measurements will require partial-observability formulations and/or state estimation. Overall, the results show that simulation-trained DRL can learn state-dependent feedback dosing policies that complement open-loop optimal control benchmarks.

Graph-based network intrusion detection systems (GIDS) report strong benchmark detection metrics, but those metrics establish little about deployability. We approach the problem from first principles: rather than inheriting the preprocessing, windowing, and thresholding conventions of each published system, we ask what a controlled comparison requires and impose it uniformly. The result is GIDS-Eval, an evaluation framework that decomposes a GIDS into six interchangeable stages and turns those conventions into explicit experimental variables, so reported performance can be attributed to individual stages instead of whole pipelines. We survey nine representative GIDS, reimplement five of them within GIDS-Eval, and evaluate them on four datasets under one matched protocol. We identify nine recurring evaluation gaps and quantify the impact of each: two crafted edges achieve full evasion against three of the eight detector-dataset pairs with anything to hide; the snapshot window alone accounts for a mean 38.3% relative swing in average precision (AP); aligning preprocessing across systems moves AP by up to 61.8 percentage points for a single detector; and none of the 18 detector-dataset pairs we replay can alert as events arrive. We introduce GIDS-Lite, an encoder-free control built in the same framework, which ranks first by AP on two of the four datasets at up to 575$\times$ lower runtime. Architectural complexity is therefore not a consistent driver of detection quality under our matched protocol on current benchmarks, but it does enlarge the runtime, calibration, and attack surfaces operators must defend.

Multi-object tracking (MOT) is dominated by the tracking-by-detection paradigm, whose methods typically rely on a small set of hyperparameters that are conventionally chosen by hand. Tuning them requires repeated expert-guided experimentation, while the procedures used to select reported values are often not systematically evaluated or fully documented. Hyperparameter optimization (HPO) automates this process, yet it remains rarely used in MOT, and existing studies applying HPO to MOT predate modern deep-detector-based trackers and HOTA evaluation. We systematically apply HPO across two datasets and four tracking-by-detection methods. We also propose Multi-Fidelity Greedy Coordinate Search (MFGCS), which optimizes one hyperparameter at a time by first evaluating candidate values on a small subset of scenes and re-evaluating only promising candidates on the full dataset. Across all eight tracker-dataset combinations, the Tree-structured Parzen Estimator (TPE) and MFGCS outperform both our hand-tuned configurations and the corresponding published results, with improvements of up to 4.38 and 16.05 HOTA points, respectively. MFGCS also reaches a predefined HOTA target faster than TPE in seven of the eight combinations. Within each tracker-dataset pair, all optimizers share the same search space and evaluation pipeline, isolating the effect of the search strategy. We release the code and tuned configurations to enable future work to compare against systematically optimized rather than default or manually tuned baselines.

Large language models (LLMs) can generate biomedical hypotheses, but it remains unclear whether they truly reason from scientific evidence or simply produce convincing-sounding ideas. To study this, we combine three major biological databases: the Kyoto Encyclopedia of Genes and Genomes (KEGG), Rhea, and UniProt, into a unified biochemical knowledge graph and construct a benchmark of 550 paths connecting enzyme sources to rare disease endpoints, yielding 13,200 hypotheses from six LLMs under four conditions varying the biological information each model receives: source enzyme only, full biological path, or source and disease endpoint only. Hypotheses are scored using an expert-derived five-criterion rubric on a 1-5 scale per criterion. We find that models given both the source and disease endpoint often produce the highest-scoring hypotheses, showing that LLMs can generate compelling ideas from minimal information. However, these hypotheses are less grounded in the evidence. In contrast, models given the full biological path generate hypotheses more consistent with known mechanistic relationships. We call this evidence-disciplined reasoning. To confirm this effect, we shuffled intermediate path steps while keeping endpoints fixed. Evidence grounding dropped significantly (delta = -0.793, p < 0.001), confirming models genuinely used path structure during reasoning. Our findings show that knowledge graphs support hypothesis generation in two ways: they identify biological endpoint pairs absent from the literature, and their mechanistic paths guide how LLMs reason between them.