Science & Mathematics

Background: Understanding individual predictions in complex tabular models remains challenging, as existing additive frameworks like SHAP and LIME often obscure structural logic by absorbing nonlinear interactions into individual feature scores. Exposing these latent interaction structures is essential for automated auditing in high-stakes environments where understanding the logic of model behavior is as essential as predictive accuracy. Methods: To address this limitation, we introduce the localized case-based feature importance (CBFI) framework. Unlike traditional numerical attribution, our method decomposes instance-specific predictions into structurally distinct components, effectively separating independent main effects (G1) from pure interactions (G4) to capture explicit feature synergies and provide a principled approach for diagnosing feature combinations. Results: Comprehensive experiments on wine quality classification and medical insurance regression demonstrate that localized CBFI successfully disentangles complex interactions hidden by additive models. Furthermore, fidelity tests and synthetic ground-truth benchmarks confirm that the identified interactions correspond precisely to the nonlinear structural dependency patterns embedded in complex architectures, including gradient boosting and deep neural networks, thereby successfully shifting the explainable AI paradigm from simple numerical attribution to the recovery of interaction-driven contributions.

Automatic red teaming increasingly relies on large language models (LLMs) to iteratively diagnose failed attacks and propose improved adversarial prompts or injections. Although prior work strengthens this loop through prompt engineering and search-harness design, the mechanisms underlying its two core roles, criticism and attack proposal, has not been explored. To further investigate the core roles and therefore improve red teaming effectiveness, we disentangle two design choices: whether criticism is generated by the target session itself (self-criticism) or by a separately instantiated critic (external criticism), and whether that criticism is translated into the next attack by the target session (self-proposal) or by a separate attacker (external proposal). Through our analysis, we find that self-criticism is more effective because it is grounded in the target's live execution context, including observations, constraints, and available reasoning; removing reasoning reduces downstream attack success by 6%-17%. In contrast, self-proposal is substantially less effective because awareness of the adversarial objective induces explicit refusal and non-refusal sandbagging. These findings motivate a role-separated red-teaming approach: the target session diagnoses why an attack failed, while a separately instantiated attacker generates the next proposal. On DTAP-BENCH, spanning Telecom, Workflow, CRM, and Code domains, under both direct misuse and indirect threat models, as well as multiple agent frameworks, this configuration achieves 63.2% overall attack success rate (ASR), much higher than baselines. Our results show that effective automatic red teaming depends not only on model capability and search strategy, but also on how execution context and adversarial responsibility are distributed across model sessions.

Repository-level code generation asks an agent to implement a task requirement inside an existing repository. To produce code that works in the repository, the agent needs repository evidence, such as target locations, helper definitions, caller usages, and configuration constraints, which determine how the code should be located, integrated, and executed. Because the context budget cannot hold the entire repository, existing retrieval, compression, and structure-aware methods select context by task similarity, context length, or structural proximity, but these criteria do not directly measure whether the selected context contains this evidence. As a result, the selected context can appear relevant to the task yet miss the evidence that the implementation depends on. We call this failure mode context utility dilution. In this paper, we propose Rivet, a context control framework that preserves repository evidence for coding agents under a fixed context budget. Rivet, first uses Signal Grounding to select task-related code anchors by fusing lexical, semantic, and structural signals. Utility Linking then expands these anchors through dependency, usage, and configuration relations to collect repository evidence. Finally, Budget Shaping packs the evidence by priority and, when the budget is tight, reduces snippets to signatures or summaries before dropping them. We evaluate Rivet, on two executable repository-level benchmarks, DevEval and RepoExec, with four backbone models, two context budgets (4K and 8K tokens), and eight baselines. Rivet, achieves the highest Pass@1 in all 16 settings and improves over the strongest baseline by 3.73--5.15 percentage points on average. Moreover, GPT-5-mini with Rivet, outperforms GPT-5 without Rivet,on both benchmarks while reducing inference cost by 73.3%--84.5%.

Wind power forecasting models are often deployed as fixed temporal backbones, but the errors they make can vary across meteorological regimes such as cold waves, heat-and-calm periods, productive typhoon, and destructive storms. We present a guarded Agent-in-the-loop residual adaptation framework that leaves the temporal backbone unchanged and adapts only a small set of scene-conditioned residual experts. An Agent layer interprets external scene information, identifies the current regime, selects residual experts, and applies a correction only when recent observed errors support it. New experts enter the pool after they outperform their initializations on a validation set, keeping model growth under explicit control. Backbone-only and Agent-enhanced forecasts are compared at lead times of 4, 8, 12, 24, and 48 across nine time-series backbones. Across these runs, the guarded framework lowers both MAE and RMSE for all 45 backbone-horizon combinations, with average MAE reductions ranging from 11.21% to 20.26% across backbones. Online-selection baselines, component ablations, and routing analyses identify the roles of residual experts, scene-aware routing, and the online acceptance rule. Candidate-scene runs evaluate the promotion path under controlled conditions; long-term unattended deployment was not evaluated.

Pretrained transformers spend a large share of their inference compute inside self-attention, yet the per-head query-key interaction is structurally low-rank by construction. We ask a diagnostic question: across layers and architectures, which heads actually use the full rank budget on real data, and which do not? We propose SVD-Attention, a post-hoc method that decomposes the per-head interaction matrix $M_h$ and scores every singular mode by both its spectral magnitude and its activation variance across tokens on a small calibration set. The score keeps the modes the head can amplify and that real data actually activates, and discards the rest; the resulting factorization is exact for the retained modes and needs no finetuning. Across seven pretrained models and five task types, the diagnostic finds a task-dependent pattern: redundancy sits in the early to middle layers for sentiment and image classification and for generation, and in the late layers for dense prediction and extractive question answering. It is also predictive: truncating the layers it flags costs at most $1.75$ F1, while truncating the same number of layers elsewhere in the same model costs up to $50$ F1. SVD-Attention is an analysis tool rather than a compression method: acting on it reduces inference FLOPs by $0.8$-$1.5\%$ on its own and by $8$-$11\%$ when composed with weight pruning, with $0$-$4\%$ change in task quality.

We present a methodology, an initial AI Skill package, and emerging results, for enhancing software engineers' acquisition and application of domain knowledge and system-wide engineering perspectives towards the development of an autonomous or semi-autonomous system in a domain that may be new to them. At the core of the approach is a process involving proactive adversarial critiquing of videos of disparate systems in related environments. Among other things, the engineers and automated AI tools look for interactions, dependencies, and environment assumptions manifested in the observed systems, which are applicable also to the target system, but that otherwise might remain unarticulated in project artifacts. The approach and its supporting tools can enhance the quality of the target system, the individual engineer's skills and the organization's engineering practices.

Camera-based bird's-eye view (BEV) perception methods such as Fast-BEV have demonstrated promising results with fast inference speed and competitive accuracy. However, their robustness to real-world corruptions remains insufficiently explored, posing risks for autonomous driving deployment where environmental variations are inevitable. In this paper, we propose Sparse Recovery Bird's-Eye View Model Soup (SR-BEVS), a sparse recovery learning approach that simultaneously improves clean accuracy and corruption robustness without requiring additional data or architectural modifications. Our method identifies pruning candidates based on layer-wise weight magnitudes, applies random pruning to create diverse sparse models, and retrains each model with a randomly sampled learning rate. By averaging these diverse recovered models into a single dense model, we achieve enhanced robustness against various corruptions while introducing zero inference overhead. We further provide a mode connectivity analysis demonstrating that our pruning-based diversification preserves the connected low-loss basin necessary for stable weight averaging. Experiments on the nuScenes dataset demonstrate up to 33.8% improvement in corruption robustness and a 1.8% gain in clean accuracy over the Fast-BEV baseline, with consistent improvements across sensor noise, weather distortions, and illumination changes. Our simple yet effective approach may be applicable to other BEV models.

Long-horizon agents continually accumulate experience, yet retaining a memory does not guarantee access when it is needed. Existing systems can adapt both memory content and access; we isolate persistent access updates as a separately controlled adaptation layer. We formulate this target as Memory Index Evolution and instantiate it with Adaptive Memory Index Evolution (AMIE), a lightweight plug-in layer that evolves per-memory access state from deployment retrievals without modifying stored content or retraining the host memory system. AMIE distils a semantic teacher's keep/suppress judgments into a retrieval-conditioned policy and accumulates its selective actions into bounded state that reweights subsequent host retrievals. Surprisingly, even a zero-call policy that simply accumulates retrieval frequency can match or outperform semantic policies in several settings, revealing retrieval traffic itself as a strong signal for access adaptation, while learned selectivity improves robustness when frequently retrieved memories remain useful. Across five retrieval-based hosts and three benchmarks, trained AMIE improves downstream F1 or accuracy in all 15 comparisons, by 0.70--3.37 percentage points (up to 25% relative), and improves retrieval in all 10 comparisons. Controlled analyses show that persistent competitor suppression can expose previously unseen evidence, while suppressing repeatedly useful memories creates the complementary failure mode. Together, these results establish persistent memory access as a practical adaptation layer, distinct from the content it serves, in retrieval-based agent memory systems.

This paper delivers a novel self-consistent non-local kinetic framework based on dose-dependent semi-Markov renewal transport and the generalized Doob's h-transform to evaluate the structural degradation of WWER-1000/1200 fuel claddings. By leveraging the Feynman-Kac formalism, we map the continuous microscopic defect path trajectories generated by the Liouville operator directly onto the evolution of the macroscopically integrated damage potential. The framework accounts for long-range memory effects in the degradation kinetics via a non-vanishing structural commutator of the path Liouville operator and Zwanzig-Mori projection operators, which mathematically links the spatial restrictions of microscopic sinks to energetic path penalties. Spatial profiling of the cladding's mean lifetime, demonstrates a distinct sigmoidal, S-shaped degradation trajectory encompassing distinct incubation, linear growth, and avalanche breakdown stages. The numerical results reveal an abrupt, localized reduction in the operational lifespan near the cladding boundaries. This marginal vulnerability is shown to be a direct physical consequence of boundary stress field singularities at the fuel cladding interface and localized embrittlement. The developed path integral architecture provides a strict thermodynamic link between non-equilibrium fluctuations and the global Griffiths macro-barrier, offering a robust predictive tool for high-burnup nuclear fuel safety assessments.

GLP-1 medications are increasingly used for weight management, and many people share their experiences online. These narratives show why people take the drugs and what they experience, but they remain unstructured and disconnected from formal pharmacovigilance. We therefore present AskGLP1, a system that transforms online patient narratives into a knowledge graph (KG) and answers questions about medication experiences. Through an LLM-assisted pipeline grounded in the MedDRA terminology, AskGLP1 screens nearly 8.5 million Reddit posts for medication experiences and distills 163,135 of them, with 2,506 WebMD reviews, into a KG of 385,790 drug-side-effect and drug-purpose relations. We find that GLP-1 experiences are source- and drug-specific. Online narratives share the most common gastrointestinal symptoms with FAERS, but they foreground everyday burdens such as fatigue and food aversion, whereas FAERS reports more often include injection-site reactions and selected serious diagnoses such as pancreatitis. We further find that posts report less nausea and vomiting for tirzepatide than for semaglutide, more fatigue for Wegovy than Ozempic and for Zepbound than Mounjaro. Evaluations show that the agent answers with verifiable evidence and declines requests for medical advice. Our findings reveal gaps between what FAERS captures and what people report online, and they call for patient-centered approaches to pharmacovigilance. Code, prompts, and evaluation questions are available at https://github.com/ZhaoqianXue/AskGLP1.

End-to-end full-duplex spoken dialogue models enable natural, real-time spoken interactions with interruptions, backchannels, and overlapping speech. Prior evaluations have assessed diverse interaction capabilities and identified some degradation patterns across conversational turns. However, how model behavior evolves over extended sessions remains insufficiently understood. In this work, we address this gap by systematically characterizing long-session degradation and developing efficient methods to mitigate it. We introduce a closed-loop evaluation framework for long-session stability and apply it to several publicly available full-duplex speech models, revealing degradation over time in responsiveness, spoken content quality, and voice consistency. To mitigate this degradation, we propose a self-distillation-based training framework for lightweight LoRA adaptation that improves long-session stability with a small amount of additional training. We further introduce a two-resolution context (TRC) architecture that preserves recent interactions in full detail while compressing older interaction history into key-value representations. We evaluate our methods on three models with distinct architectures, assessing long-session stability, information retention, and computational efficiency. Across all three models, we show that LoRA adaptation mitigates performance degradation over ten-minute sessions, while adding TRC extends these gains beyond the training session length and reduces KV-cache growth by more than 90%.

Road anomalies such as collisions, violations, and sudden hazards demand more from video understanding than detection: a practical system must verify what happened, describe the scene, and explain the cause. We find these competences pull in opposite directions for vision-language models: closed-form questions reward precise, evidence-grounded reasoning, while open-ended answers are scored by similarity to reference text, where matching the annotation style matters as much as reasoning correctly. We present Reason-or-Recite (RoR), a two-specialist framework for the Traffic Anomaly Reasoning track of the 2026 AI City Challenge. RoR (1) rewrites the official chain-of-thought annotations into a structured, temporally grounded schema and LoRA-tunes a 9B vision-language model on the enhanced traces; (2) fine-tunes a 4-bit-quantized 27B model on answers alone, absorbing the annotation style; and (3) selects between the two candidates per question with a training-free LLM judge. RoR ranks first on the in-domain leaderboard with mean score 0.6788.

Modern large language models (LLMs) often exhibit overconfident errors, assigning high confidence to incorrect predictions and thereby undermining reliability in real-world use. Existing calibration methods typically rely on post-hoc adjustment or retraining, operating at the output level without addressing the underlying causes of miscalibration. We propose CoCHI, (Class-oriented Causal Head Intervention), a framework that improves calibration by identifying and intervening on internal components responsible for distinct confidence behaviours. By analysing predictions through correctness-confidence patterns, CoCHI enables targeted intervention on model internals, reducing overconfident errors while preserving correct predictions. Our approach provides a mechanistic perspective on calibration, linking reliability to identifiable structures within the model rather than treating it as an output-level issue. Experiments across multiple LLMs show consistent improvements in calibration metrics, including Expected Calibration Error, Negative Log-Likelihood, and Brier Score, while maintaining accuracy on five multiple-choice question answering benchmarks. These results suggest that effective calibration requires not only adjusting outputs, but also addressing the underlying causal mechanisms within the model.

We study continuous-time stochastic linear--quadratic control with a scalar state and nonnegative control constraints. With quadratic running costs, randomized exploration can change the optimal mean feedback. We introduce an auxiliary problem without entropy in which the conditional policy mean is nonnegative and the covariance is prescribed in proportion to the squared state. Moment reduction and state separation yield two regional Riccati equations and characterize the finite-horizon value function and optimal conditional mean. We give a sufficient control--cost compatibility condition under which this mean coincides with the optimal feedback of the original constrained problem. For time-invariant coefficients, we characterize the undiscounted infinite-horizon optimum and establish mean square asymptotic stability under the stated assumptions. For unknown dynamics, we develop model-free actor--critic schemes for finite and infinite horizons. Numerical experiments compare the learned policies with their analytical targets and examine the effect of exploration on the optimal mean.

We study the effects of Sn-site doping on the electronic structure of nonmagnetic kagome metal YV$_6$Sn$_{6-x}$Sb$_x$ by infrared spectroscopy and density-functional theory. The results for samples with $x$ = 0 and 1.4 demonstrate that partial substitution of Sn by Sb acts as a rigid shift of the Fermi level while leaving the band dispersion largely intact. Our findings establish Sn-site substitution in YV$_6$Sn$_6$ as a route to systematically investigate band-filling-dependent electronic properties of nonmagnetic kagome metals.