Computers and Society
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
Tuesday 11 August
The question of whether artificial intelligence will "destroy jobs" is too coarse to guide economic analysis or institutional design. A job is not an indivisible object, and machine cognition is not a uniform substitute for human labor. This paper develops a task-based and institutionally grounded framework for analyzing generative AI as cheap, scalable, and fallible cognition. The relevant margins are exposure, adoption, verification, question selection, workflow redesign, demand elasticity, apprenticeship, and rent allocation. We distinguish the technical reach of large language models from equilibrium labor-market displacement by introducing a task vulnerability index and an adoption condition that makes verification, liability, trust, and governance explicit. We then model occupations as governance bundles rather than task lists, firms as architectures of distributed intelligence, and labor-market effects as a balance among task compression, scale expansion, new human work, and institutional bargaining. A further implication is that when answer generation becomes abundant, the scarce human capital shifts upstream and downstream: toward asking economically meaningful questions, framing problems, generating hypotheses, interpreting results, and bearing responsibility for consequential use. The central dynamic concern is expertise formation: Junior tasks jointly produce current output, question sense, and future judgment, so their automation can raise short-run productivity while weakening the pipeline into accountable expertise unless AI is designed to teach rather than merely bypass. The paper concludes that AI's labor-market destiny is neither mechanical unemployment nor automatic abundance. It is an institutional equilibrium shaped by workflow design, apprenticeship systems, liability rules, competition policy, worker voice, and the distribution of rents from cheap cognition.
Machine Learning
Let it Cook: Learning to Wait in Sequential Decision Making
In sequential decision making, an agent typically observes its environment and acts at every timestep. However, such active participation may not always be necessary; tasks such as brewing coffee include periods that are served equally well by letting the environment evolve without constant monitoring and control. During such periods, the agent could simply wait to conserve its resources, or redirect its attention to another task. We capitalize on these opportunities by training a "waiting policy" that decides where and how long to wait. This involves forgoing sensing to commit to a wait action, representing a deliberate pause for a set number of timesteps. We formalize "learning to wait" as minimizing the frequency of sensing and decision making without sacrificing task performance (e.g., the total amount of time to complete a task). To train a waiting policy, we propose an approach that employs reinforcement learning with lexicographically ordered objectives. In experiments across 4 discrete-state household tasks and 3 continuous-state environments, we show that our approach successfully learns waiting behaviors, and can adapt pre-trained policies to wait where appropriate. While different tasks permit different amounts of waiting without sacrificing task performance, our approach consistently finds solutions with significant waiting, sometimes waiting for over 50 percent of the task duration.
Read the full abstract
Artificial Intelligence · Cross-listed from q-bio.NC
Conflict and Congruency Effects in Large Language Models: In-Weight and In-Context Competition in a Verbal Conflict Task
Congruency effects, observed in conflict tasks such as Stroop and flanker tasks, have been investigated for nearly a century in psychology and neuroscience, but their mechanistic basis is not fully understood. We introduce a verbal-only LLM conflict task in which a prompt stem elicits a default same-color completion and an explicit rule either agrees with (congruent condition) or conflicts with (incongruent condition) the completion. Gemma-2-2B and six Pythia models ranging from 410M to 12B parameters showed strong default same-color tendencies, and six of seven models showed strong congruency effects. Using causal attribution analysis, attention analysis, and attention ablations, we identified distinct processing pathways in these LLMs: a pathway involving short-range attention to a superficial color cue that is preferentially activated in the congruent condition, and a pathway involving long-range attention to the rule prefix that is preferentially activated in the incongruent condition. Fine-tuning that strengthened the default same-color tendency had divergent effects on task conditions, reducing incongruent performance while increasing congruent performance. In contrast, increasing rule set size selectively impaired incongruent performance. These converging findings support an account in which congruency effects in this task arise from competition between an in-weight default mapping and an in-context rule-based mapping. More broadly, our findings illustrate how LLMs can serve as model systems for mechanistic analysis of competition between default and rule-governed response tendencies within a single learned network.
Read the full abstract
Machine Learning
RelShap: Relationally Consistent Shapley Explanations
Machine learning pipelines commonly flatten relational data into single-table representations, discarding structural constraints. Widely used Shapley value-based feature attributions then rely on feature independence, evaluating the model on combinations that could never arise in the underlying data, producing misleading explanations. We propose RelShap, a framework that incorporates relational constraints and data provenance into Shapley value computation, restricting both background data and coalition evaluation to relationally valid configurations. The framework is estimator-agnostic and composes with Kernel SHAP, Monte Carlo, and Leverage SHAP without altering their sampling or weighting properties. Functional dependencies further induce equivalence classes over feature coalitions, which RelShap exploits to reduce runtime without changing Shapley values; we provide a combinatorial characterization of the expected speedup. Experiments across multiple datasets, models, and estimators show that RelShap produces explanations that are more faithful to the data-generating process, correctly identifying the dominant feature in controlled settings where existing methods, including Conditional SHAP and ManifoldShap, do not. Our code is available at: https://github.com/duneag2/relshap.
Read the full abstract
Neural and Evolutionary Computing
Predictive Allostatic Organization in Recurrent and Spiking Agents Under Partial Observability
Adaptive behavior under partial observability depends on internal organization that carries information beyond the current observation. Drawing on Barrett and Miller's account of categorization as predictive, compressive, functionally organized, and allostatically constrained, we test whether recurrent and spiking agents develop internal states with corresponding computational properties. Agents operate in an energy-constrained foraging task requiring resource acquisition, threat avoidance, contact-dependent consumption, and regulation of an internal energy variable. In a frozen benchmark, learned agents outperform random and heuristic baselines; the trace-augmented recurrent policy is strongest overall, while spiking variants show stress-specific differences. Early internal dynamics predict later full-safe-efficient success above permutation baseline, reaching a maximum ROC-AUC of 0.802. Reduced PCA subspaces retain behaviorally relevant information. Feature-family controls show that predictive signal is distributed across trace, policy-head, internal-dynamics, observation, and allostatic variables, and low-energy state remains strongly decodable after explicit energy-related features are removed. Evaluation-time perturbations to temporal state, sensory information, operating conditions, and allostatic mechanisms alter behavior and/or internal prediction. Seed-balanced event probes show weaker but measurable information about future contact, successful consumption, and threat events, alongside strong low-energy decoding. We interpret this pattern as a computational analogue of predictive allostatic organization: distributed control regimes that are predictive, energy-sensitive, action-relevant, and partly causally involved, without claiming biological validation or discrete symbolic categories.
Read the full abstract
Networking
In QKD, Key Metadata is Key
Federated interconnected QKD networks are becoming the norm, and with several standards already in place, interoperability is now technically feasible. Many aspects related to cross-domain key distribution, however, are left unresolved: breach announcement and mitigation, integration with the space segment, extension to a future QKD-as-a-Service commercial model, custom requests referring to key freshness, jurisdictions, security levels, etc. We argue that, to move from syntactic to semantic interoperability, a common language must be developed to ensure transfer of meaning along with the keys. We posit this is possible via QKD key metadata, and we propose a 6-step process to develop it as a standard.
Read the full abstract
Networking
Multi-Pair Fidelity-Aware Rate Allocation in a Quantum Network: Approximation Schemes
Entanglement distribution in quantum networks must jointly account for limited link capacities, probabilistic entanglement swapping, and heterogeneous link fidelities. In this paper, we study multi-pair fidelity-aware rate allocation in quantum networks. We formulate three rate-allocation problems: rate sum, rate sum subject to minimum-rate constraints, and max-min fairness. Prior work has studied a special case of the rate sum problem, where all links have identical fidelity. This special case admits a polynomial-time algorithm. We prove that all three problems are NP-hard. We then study optimization versions of these problems which maximize the minimum end-to-end fidelity subject to throughput or fairness requirements. We present fully polynomial-time approximation schemes (FPTAS) for solving these optimization problems. Experiments on randomly generated networks demonstrate the computational effectiveness of the proposed schemes.
Read the full abstract
Game Theory
Strengthening Full Justified Representation: Efficient Verification and Computation
Full justified representation (FJR) is among the strongest known satisfiable proportionality axioms for approval-based committee elections. Recent work has shown that an FJR committee can be found in polynomial time, but verifying whether a given committee satisfies FJR remains coNP-complete. We introduce FJR+, a strict strengthening of FJR and EJR+ that can be verified and satisfied in polynomial time. We then analyze the Residual-Budget Greedy (RBG) algorithm and prove that it selects a partial committee such that every size-$k$ completion satisfies FJR+. This freedom allows us to use sequential Phragmén to obtain a priceable completion. The resulting rule always satisfies FJR+ and the sub-core, and it is priceable whenever at least $k$ candidates receive an approval. We also obtain a Droop-quota version of FJR+. Finally, we extend FJR+ to approval-based participatory budgeting with arbitrary project costs. A project-specific version of RBG computes this property in polynomial time and can be continued to a priceable outcome satisfying a cost-based version of the sub-core.
Read the full abstract
Machine Learning
HyperFix: Combinatorial Nonlinear Correction for Task Vector Merging
Task vectors enable model merging without joint retraining. In practice, the subset of task vectors to be merged may vary, but many existing methods use scalar tuning for a particular subset, requiring repeated tuning across subsets and restricting task vector merging to linear rescaling. We therefore formulate merging across varying task subsets as a combinatorial correction problem and introduce HyperFix, a lightweight hypernetwork that predicts subset-conditioned nonlinear corrections in weight space. Trained once on singleton, pair, and triple subsets from a task bank, HyperFix generalizes to larger subsets without per-subset optimization. Our local perturbation analysis bounds the residual correction beyond linear merging and motivates learning it from small task updates. Experiments across diverse benchmarks show that HyperFix outperforms existing task vector merging methods while reducing tuning cost.
Read the full abstract
Computer Vision
Language-Structured Relational Q-Learning for Threat-Aware Control in Safety-Critical Driving
Natural-language-based scenario generation offers an intuitive means of describing rare and complex driving interactions, yet it is still uncertain whether training with language-structured data leads to truly adaptive control policies. We propose Language-Structured Relational Q-Learning, instantiated through an Ego-Centric Relational Q-Network (ERQ-Net), which jointly learns inter-vehicle relevance and action values from dynamic traffic graphs. Language descriptions define surrounding-vehicle behaviours during training, while prompts and semantic actor roles are hidden from the policy. ERQ-Net must therefore infer threat relevance solely from observable kinematics and interactions. Across 2,500 safety-critical scenarios, language-structured training improves test success from 49-52% to 55-58% and increases adversary-focused attention from 1.2x to 2.1x, demonstrating emergent threat awareness. However, this representational gain does not consistently translate into adaptive control: trained policies perform similarly to the best constant action, while a portfolio of simple policies solves 76% of scenarios. We formalise this discrepancy as a recognition-control gap and show that reward reweighting and margin shaping do not eliminate the resulting policy collapse. Evaluations of realism, criticality, semantic accuracy, and transfer of state-interface representations to CARLA further highlight both the strengths and the constraints of language-structured relational policy learning in safety-critical driving scenarios.
Read the full abstract
Logic in Computer Science
Discrete Linear Ensemble Logic
We study the discrete point-based fragment of Ensemble Logic $\EL(\Nat)$ over the natural numbers, a logic combining displacement $\varphi_u$, bounded metric modalities $\boldBox_t$ and $\mdiamond_t$ with additive bounds, Boolean connectives, and first-order quantification over $\Nat$. Motivated by the need for a unified symbolic layer for biomedical knowledge with temporal, spatial, genomic, and multimodal metric content, we develop the foundational discrete theory of the formalism. We give syntax and semantics, and prove a forward embedding of $\EL(\Nat)$ over a finite proposition set $\mathcal{P}$ into first-order monadic Presburger arithmetic $\FO(\Nat,<,+;\mathcal{P})$. This embedding yields the analytical upper bounds, while a reduction from nondeterministic two-counter machines with recurring control states proves that satisfiability is $\Sigma^1_1$-complete and validity is dually $\Pi^1_1$-complete. Expressively, $\EL(\Nat)$ strictly extends the star-free $\omega$-languages and is incomparable with the $\omega$-regular languages: it defines the non-$\omega$-regular counting language $\{a^mb^mc^md^m\mid m\geq 1\}\cdot\Sigma^\omega$, whereas a delimited parity language remains outside the logic by classical Presburger-arithmetic lower bounds. On the proof-theoretic side, we present a sound Hilbert system $\HEL$ and establish completeness relative to monadic Presburger validity as oracle, noting that completeness relative to plain Presburger arithmetic is impossible.
Read the full abstract
Machine Learning
Defending against Model Extraction for GNNs with Model Reprogramming
Graph Neural Networks (GNNs) serve as the backbone for high-stakes applications in Machine-Learning-as-a-Service (MLaaS). Still, their black-box deployment exposes them to Model Extraction (ME) attacks, in which adversaries steal intellectual property by querying APIs. Existing defenses suffer from a critical ”Euclidean bias”: they transfer image-based strategies (e.g., random noise) to graphs, ignoring the complex topological dependencies between nodes, which often results in severe utility degradation. Passive methods like watermarking also fail to prevent theft in real time. To bridge this gap, we propose GraphRP (Graph Reprogramming Protection), a proactive defense framework that repurposes Model Reprogramming for security. Unlike static perturbations, GraphRP introduces a Structure-Aware Gating Mechanism driven by learnable topological prototypes. This creates a dynamic ”structural firewall” that selectively modulates the model's decision boundary: it preserves fidelity for benign queries residing on the training manifold, while maximizing the Fisher Information along the perturbation direction for adversarial queries. Under standard assumptions (bounded loss, optimal attacker, and local second-order approximation), we prove a lower bound on the attacker's estimation error that increases with the structural sensitivity of the reprogramming noise. Extensive experiments on both hard-label and soft-label ME attacks demonstrate that GraphRP significantly degrades attack effectiveness while preserving benign utility.
Read the full abstract
Artificial Intelligence
From Prompting to Behavioral Alignment: Personalized LLM Judges for Recommendation Evaluation
Traditional offline recommendation evaluation relies heavily on complex, manually maintained feature pipelines that are difficult to scale. While Large Language Models (LLMs) offer a promising alternative by predicting user engagement directly from raw text logs, empirical analysis in this study identifies a critical failure mode termed bidirectional rationalization. In a zero-shot setting, LLMs are found to convincingly argue for both positive and negative user engagement outcomes on the exact same item with identical evidence, highlighting the unreliability of off-the-shelf LLMs in predicting user engagement. To resolve this, we develop and apply a sequential behavioral alignment framework pairing fine-tuning with preference optimization over paired correct and counterfactual rationales. Evaluated on real-world homepage interaction logs, this aligned reasoning approach achieves a 32.19% lift in Macro-F1 score over the zero-shot baseline and matches the production feature-engineered baseline. The results demonstrate that behavioral alignment mitigates bidirectional rationalization while delivering human-interpretable reasoning traces without manual pipeline overhead.
Read the full abstract
Cryptography and Security
Cross-Corpus Evaluation of Generalizable Vulnerability Detection in IoT Firmware
IoT firmware vulnerability detection is constrained by ecosystem heterogeneity, resource-limited platforms, and benchmark quality limitations. Existing datasets are often synthetic or general-purpose and lack human-verified, contamination-screened annotations, leaving cross-corpus generalization across training sources, architectures, and curriculum design underexplored. In this study, we have introduced IoTVulBench, a human-verified benchmark for cross-corpus firmware vulnerability detection. IoTVulBench was built from GitHub repositories, validated by three expert reviewers, and evaluated on a contamination-screened held-out target across five architectures, two tuning methods, and three curriculum strategies, with ensemble, distillation, and robustness analyses. Models trained on IoTVulBench reached the highest Matthews Correlation Coefficient (MCC) among undersampling-matched single-source datasets, at 0.58 versus 0.44 for PrimeVul and 0.39 for D2A. Staged curriculum learning raised MCC to 0.69, and a diversity-optimized ensemble reached 0.73. This gain represents a 0.42 MCC improvement over the strongest reference comparator, a static analyzer with an MCC of 0.31, and a 0.29 MCC improvement over the strongest single-source dataset, PrimeVul. At a 0.5% false-positive rate, the model missed only 21% of vulnerabilities versus 71% for the comparator. The model also retained 86% of its performance under identifier renaming, with strong calibration. These results indicate that domain-matched training data and curriculum design, rather than model scale alone, drive generalization in firmware vulnerability detection, and yield both a benchmark and deployment-ready configurations for IoT security.
Read the full abstract
Computers and Society
The Accuracy Trap: Structural Scarcity Amplifies Relative Inequality in Algorithmic Allocation
Algorithmic systems increasingly rank individuals for access to scarce public resources, from child welfare interventions to cancer treatment referrals. The prevailing fairness frame treats disparity as a property of biased data or deficient models, with remedies through calibration and debiasing. Under structural scarcity, where demand exceeds supply by an order of magnitude, allocation becomes a rationing problem, and the statistical properties of ranking diverge sharply from those of classification. We derive a scaling law $D \propto \exp(t \cdot \rho \cdot \Delta)$, in which relative disparity between two groups separated by a structural gap $\Delta$ grows in the product of the scarcity-induced threshold $t$ and rank-discrimination fidelity $\rho$. Scarcity and accuracy interact multiplicatively, producing exponentially larger between-group disparities. We term this dynamic the Accuracy Trap. We validate this Accuracy Trap through Monte Carlo simulation and two independent public-sector systems in Canadian child welfare and U.S. cancer care. Debiasing alone cannot dissolve the trap.
Read the full abstract
Systems and Control
Matched Disturbance Rejection for Port-Hamiltonian Systems with Coupled Dynamics
This paper investigates the rejection of matched disturbances generated by coupled port-Hamiltonian (PH) dynamics in previously stabilized PH systems. The disturbance dynamics are incorporated into a unified PH representation, allowing the disturbance to affect the plant through both the matched input channel and an interconnection structure. Unlike existing results that typically impose restrictive structures on the disturbance dynamics, the proposed framework accommodates a more general class of coupled PH disturbances, including nonzero interconnection and damping terms. A baseline disturbance rejection scheme is first established for known disturbance storage parameters. The framework is then extended to the case of an unknown symmetric storage matrix through online parameter estimation. Two control designs are developed under different structural conditions, with the latter relaxing the dimensional restriction imposed by the first design. The proposed methods guarantee asymptotic convergence of the plant state to the desired equilibrium, while preserving a port-Hamiltonian representation of the closed-loop dynamics under the corresponding conditions. The results generalize existing disturbance rejection approaches and broaden their applicability to coupled PH disturbance models.
Read the full abstract
Cryptography and Security
A Runtime Decentralized Attestation and Coordinated Repair Framework for Securing Automotive ECUs
The evolution of automotive technology increasingly integrates components, transforming vehicles into interconnected systems of systems. Modern vehicles are controlled by a distributed system of computing devices, known as electronic control units (ECUs). However, this interconnectedness means that any error poses significant risks to the vehicle operator. In particular, malware can be injected into ECUs, threatening vehicle safety. To address this, we need mechanisms to detect compromised ECUs then repair them to a benign state. Existing approaches mainly focus on detection and do not address the challenge of integrating detection with runtime ECU repair. This integration is nontrivial because runtime repair involves both local rollback and reboot with timing determined from global vehicle context to avoid unsafe behavior. In this work, we have designed DACER, a runtime decentralized attestation and coordinated repair framework for automotive ECUs. DACER is the first approach that co-designs attestation and repair to unify the “local” nature of firmware rollback with the “global” nature of ECU reboot. In DACER, each ECU performs efficient local self-attestation and self-repair functions, enabling low-overhead coordination for distributed operations. In addition, DACER takes advantage of the hierarchical vehicle computing architecture. Our resulting DACER design checks the entire state of the vehicle, resists single points of failure, conforms to real-time constraints, and enables firmware restoration during runtime. The key functions are enabled by the ARM TrustZone equipped within each ECU and the secure flash memory controller embedded in the storage device. We implemented DACER on real-world hardware and experimentally demonstrated its low overhead.
Read the full abstract
Numerical Analysis
Stability of Finite-Batch Particle Mean-Field Variational Inference Beyond Strong Convexity
We study the implementable finite-batch particle algorithm for mean-field variational inference as a fully discrete stochastic approximation of the projected Wasserstein dynamics. The target potential is globally smooth but need not be strongly convex. The departure from contractivity is quantified by the curvature defect \[ \mathfrak d_\alpha(x,y) = \bigl[\alpha\|x-y\|^2- \langle\nabla V(x)-\nabla V(y),x-y\rangle\bigr]_+, \] which is the additive loss in the one-step Euler contraction estimate. We prove a non-asymptotic Wasserstein stability bound that separates initialization, product-empirical approximation, finite-batch drift error, time discretization, and the defects accumulated along the coupled trajectories. Under the uniform bound $\mathfrak d_\alpha\leq\beta$, the particle iterates remain within $O(\sqrt{\beta/\alpha})$ of any MFVI minimizer, up to explicit errors in the particle number, batch size, and step size. The proof uses a stationary comparison array whose population law is an MFVI minimizer but whose particle-level law is a random product empirical measure, and it controls the resulting projected-drift discrepancy explicitly. We also give coordinatewise defect estimates and structural conditions for dimension-independent projected-drift sensitivity, construct an arbitrary-dimensional smooth nonconvex benchmark with a closed-form MFVI minimizer, and explain why polynomially growing drifts require a modification of the untamed explicit scheme.
Read the full abstract
Digital Libraries
Patterns of Research Funding Across Research Subjects: The Case of NSERC
Funding agencies aim to maximize scientific output from the resources they allocate, a goal that may inadvertently reinforce cumulative advantage within funding systems. Focusing on research areas within Natural Sciences and Engineering (NSE), this study examines the distribution of Canadian research funding over time, with particular attention to the extent to which certain subjects gain prominence as emerging or high-growth fields, while others experience relative stagnation or decline. The results indicate that many research subject groups receiving decreased funding in recent years are concentrated in the pure sciences and mathematics. In contrast, social sciences and medical sciences, which initially had relatively low average funding per grant application, have demonstrated increasing funding trends over time. These findings offer valuable insights for funding agencies seeking to design policies that promote more effective and equitable allocation of resources.
Read the full abstract
Artificial Intelligence
A Modular Agentic Framework for Synthetically Constrained Multi-Objective Hit-to-Lead Optimization
Hit-to-lead optimization requires iterative design of hit analogs across competing potency, selectivity, physicochemical, pharmacokinetic, safety, and synthetic constraints. We present SABLE (Synthetically-accessible Agentic Bayesian Ligand Exploration), an open-source framework that employs natural-language orchestration to guide chemical structure optimization. SABLE uses an LLM to interpret user-defined goals and route tasks, while specialized tools perform reaction-templated analog enumeration, physicochemical and ADMET property prediction, structure-based affinity scoring, and Bayesian optimization. The resulting workflow is a computational twin of the analytical and prioritization stages of the design-make-test-analyze cycle, providing provenance of each numerical output. Across single, and multi-objective optimization studies, SABLE enriches candidate sets for user-defined computational objectives while evaluating only a subset of the enumerated search space. Its modular architecture allows tools and characterization backends to be replaced by editing a simple config file, without modifying operational logic. SABLE provides an extensible decision-support framework for prioritizing synthetically constrained analogs in early-stage drug discovery.
Read the full abstract
Systems and Control
Forward Trajectory Steering for Hamilton-Jacobi Reachability Analysis
Hamilton-Jacobi (HJ) reachability provides a mathematically rigorous framework for safe control of dynamical systems, but its practical application is bottlenecked by the computational complexity of solving Hamilton-Jacobi-Isaacs variational inequality PDEs in high dimensions. Physics-informed neural networks (PINNs) have recently emerged as a promising alternative to classical mesh-based solvers, yet their performance is highly sensitive to the choice of collocation sampling. In order to learn accurate safety value functions, existing PINNs-based HJ reachability solvers must rely on complex training pipelines and auxiliary supervision. In this work, we propose STEER2REACH (S2R), a PINNs-based HJ reachability solver that requires minimal modification on top of standard PINNs training. S2R's key contribution is a lightweight, low-overhead adaptive collocation sampling distribution constructed by steering forward trajectories using a combination of the optimal control and disturbance signals induced by the current value function, with injected stochastic exploration noise. We demonstrate that despite its simplicity, S2R achieves competitive–and in some cases improved–performance on safety metrics while reducing relative L2 error across a range of reachability benchmarks compared with SoTA MPC-guided HJ reachability solvers, all without requiring multi-stage training or MPC-based supervision.
Read the full abstract
Machine Learning
Convergence Guarantees of Gradient Descent for Neural Networks via Generalized Lipschitz Smoothness
We establish the first convergence guarantees of gradient descent for general feedforward neural networks of any width or depth, with any initialization or dataset. We only assume that the activation functions are linearly bounded, Lipschitz continuous, and Lipschitz smooth—properties that hold for linear, tanh, softplus, sigmoid, and smoothed ReLU functions—and that the loss function is Lipschitz smooth in the model outputs, a mild condition satisfied by mean squared error and binary cross-entropy loss. By relating the Lipschitz properties of one layer to the next, we obtain a novel generalized Lipschitz smoothness condition for an $L$-layer neural network where the change in gradient is upper bounded by the change in the parameter space, multiplied by a polynomial of the parameter norms at both endpoints of degree $2L - 2$. This yields a descent lemma where the loss decreases as long as the learning rate is small enough with respect to the parameter norms. By ensuring that the leading term of the polynomial grows at a controlled rate, we prove that the minimum squared gradient norm converges to zero in $T$ iterations at rate $O(1/T^{1/L})$.
Read the full abstract
Digital Libraries
Characterizing Peace Through Scientific Keywords
This study aims to show that peace, as reflected in scientific research, is not a neutral or uniform concept but is constructed differently across disciplines and geopolitical contexts. To do this, this study analyzes differences in the use of keywords among scholarly publications aligned with SDG 16 across various disciplinary, societal, and national contexts, and identifies which keywords are most strongly associated with countries exhibiting higher versus lower levels of societal peace. Results show that while peace-related research is strongly centered on violence, it is interpreted through distinct lenses: technological and forensic approaches in biomedical and NSE fields, and institutional, legal, and societal perspectives in the social sciences. At the country level, violence remains a central anchor, but its framing varies with levels of peace, with higher-peace countries embedding it within welfare, health, and social systems, and lower-peace countries emphasizing conflict, security, and geopolitics. These results could contribute to shaping both research priorities and science policy initiatives.
Read the full abstract
Computational Engineering
Stochastic Corridor Time Network Capacity Planning for Low Altitude Airspace Systems
Regulators in China, the United States, and the European Union now provide low-altitude airspace access as priced, time-windowed corridor authorizations, booked in advance and forfeited if unused. We ask how much capacity a UAV logistics planner should reserve on each corridor–time unit before demand is realized, to maximize expected profit net of reservation cost. Reserved capacity cannot be transferred across corridors or time windows and is consumed jointly along time-respecting paths, so reservations are coupled through the network in ways that models with exogenous airspace capacity cannot capture. We formulate a two-stage stochastic program whose recourse selects and routes accepted requests on a time-expanded network, prove its arc-based and path-packing forms equivalent, and solve it by Benders decomposition with column-generated subproblems. The decomposition operates on the LP relaxation, and all reported reservation and routing decisions are recovered as integer plans. Computational experiments achieve single-digit LP-Benders gaps on moderate-sized networks and extend to much larger instances through a truncated-path approach. A Shenzhen case study shows reservations concentrating on structurally central corridors, with demand level and reservation price having more influence on the quantity of capacity reserved than the selection of corridors.
Read the full abstract
Machine Learning · Cross-listed from q-bio.QM
Probing and steering biology across Boltz-1s trunk-diffusion boundary
AlphaFold3-class structure predictors pair a representational trunk, which processes sequence and context, with a diffusion module, which generates atomic coordinates. How biological information changes as it crosses this architectural boundary remains poorly understood. We analyze per-residue activations from the Pairformer trunk and diffusion module of Boltz-1 using linear probes, sparse autoencoders (SAEs), and causal interventions. From the trunk, both geometry (secondary structure, disorder) and sequence chemistry (amino-acid identity, signal peptides, disulfide-bond annotations) are linearly decodable. In the diffusion module, the two diverge. Secondary structure transfers essentially unchanged, whereas sequence chemistry is strongly attenuated. We then test whether decodable directions can steer the model, intervening on the final trunk single representation that conditions the diffusion module. Helix and coil directions change predicted structure dose-dependently against matched-norm random controls, but a beta-strand direction that is highly predictive (F1 =0.82) produces no measurable increase in strand content: linear decodability does not imply causal influence at the site we tested. The same probes also score markedly lower against sparse SwissProt annotations than against dense DSSP labels, because unannotated residues that the model gets right are charged as false positives; such scores are therefore lower bounds. Finally, supervised probes outscore single SAE features wherever a label already exists. We release the trained trunk and diffusion SAEs, Boltz-1 per-residue activations, and the analysis code.
Read the full abstract
Computer Vision
Test-Time Hallucination Control in Large Vision-Language Models
Object Hallucination in large vision-language models (LVLMs), where models generate non-factual content about input images, remains a critical barrier to their reliability in real-world applications. Existing mitigation strategies can be categorized into training-based and training-free methods. Training-based methods often achieve strong performance but are costly, requiring extensive computational resources, large-scale data, and time-consuming fine-tuning. Training-free approaches are particularly appealing due to their efficiency. However, existing training-free methods either require multiple decoding rounds, which adds computational overhead, or modify internal states in a model-specific way that risks degrading pretrained knowledge. We propose Test-Time Hallucination Mitigation (TTH) method, a novel training-free method that addresses both limitations. TTH introduces a token-validator module, implemented as a zero-shot Multi-Modal Classifier (MMC), to generate auxiliary logits grounded in the input image. These logits are fused with the original LVLM outputs at the token level for object tokens selected from a candidate pool. An entropy-based weighting scheme is then applied to enable robust and accurate predictions. Extensive experiments across multiple LVLM families and diverse benchmarks demonstrate that TTH consistently improves accuracy and robustness, underscoring its generalizability and practical effectiveness. Code is released at https://github.com/Mehran-TAM/TTH
Read the full abstract
Machine Learning
Dual-Primal Graph VAEs for Noisy Label Aggregation
Inferring the ground-truth from noisy crowdsourced labels is an important theoretical and practical problem. Neural network-based methods offer an alternative to classical Bayesian models which require specifying a family of generative models used for inference. However, current models either still rely on fairly simple generative models for inference or require pseudo-labels or synthetic data to train the aggregate classifier. We propose a graph VAE architecture in which the decoder and encoder use GAT-based message passing on the adjacency graph of a crowdsourced dataset and its dual, respectively. The ground-truth labels are treated as latent variables, enabling unsupervised representation learning without needing to train a separate classifier. We show our model achieves state of the art performance on crowdsourcing benchmarks. We then demonstrate the generality of our approach by showing how the original crowdsourcing graph can be augmented to incorporate side information such as representations from neural network classifiers trained on the noisy labels to substantially boost their classification performance at test time.
Read the full abstract
Computer Vision
Gaussian Meta-Space Augmentation for Stacking Ensembles in Multimodal IPMN Risk Stratification
Pancreatic cancer is among the most lethal malignancies; risk stratification of intraductal papillary mucinous neoplasms (IPMNs) offers a crucial opportunity for early intervention but typically requires invasive tissue biopsy. Dominant vision-based approaches, including radiomics and deep learning, provide promising but initially separate discrimination opportunities. Similarly, multisequence MRI (T1W/T2W) and anatomically decomposed (head, body and tail) analysis of the pancreas provide additional and potentially complementary signals. Effective fusion of this information is crucial in ordinal IPMN dysplasia risk prediction and can be accomplished via a meticulously regularized and calibrated ensemble stacking combiner. We present cUPMI, a class-conditional Gaussian augmentation of a combiner's log-probability meta-features, and test it on various prediction paradigms. In our multi-center analysis, we find cUPMI adds limited value to properly regularized L2-logistic binary classification stacks, but consistently regularizes higher-capacity tree combiners in the binary and radiomics-only setting (RF +0.015 and XGBoost +0.024 binary AUC, positive in all seeds). Its cleanest ordinal benefit appears for XGBoost on an 8-stream radiomics task (3-class no < low < high, +0.022 QWK in all seeds). Separately, fold-locked fusion of radiomics and 2.5D CNN streams yields the strongest overall model, an RF stack reaching QWK 0.595 (95% CI [0.54, 0.64]) and binary AUC 0.839, surpassing radiomics, 2.5D ResNet, and 3D DenseNet-121 baselines.
Read the full abstract
Cryptography and Security
The Next Challenge for Agentic Cybersecurity: A Realistic, Contamination-Free Reverse Engineering Benchmark
AI agents are rapidly improving in cybersecurity when source code is available, yet much of the software most consequential to security, including malware, firmware, and proprietary applications, exists only as binaries. Analyzing such software requires reverse engineering (RE): recovering program semantics before analysis can proceed. Evaluating agentic RE poses a fundamental challenge: realistic benchmark instances must (1) be absent from LLMs' training data to prevent shortcuts by memorization, and (2) reflect the scale and anti-analysis protections of real-world binaries. We introduce SRE-Bench, the first realistic, contamination-free RE benchmark. Built from scratch by RE experts with over 5,000 expert hours, SRE-Bench comprises 19 private, real-world-scale programs averaging 16.9K lines of code. We further developed 44 in-house anti-analysis mechanisms, yielding 262 binary instances and 1,572 deterministically graded tasks. We evaluated 13 agentic settings across 11 models: eight in public-facing settings and five in internal unconstrained settings with cyber safeguards disabled and no budget cap. Realistic RE remains challenging for frontier agents: GPT-5.6-Sol and Claude-Fable-5.1, despite strong source-code security capabilities, fully solve only 31.5% and 26.9% of graded instances, suggesting that success in source-code security does not translate into effective binary analysis. Without a budget cap and safety guard, GPT-6-Astra achieves a near-perfect pass@4 score, yet reliably identifying the correct candidate remains difficult. Agents are largely insensitive to compiler optimization and static linking, and ablations confirm that both contamination control and realistic scale are essential to understanding agents' RE capability. These findings highlight RE as a distinct frontier for agentic cybersecurity and establish SRE-Bench as a rigorous testbed for measuring progress.
Read the full abstract
Machine Learning
PAC-Bayes Beyond Parameter Space: Behavioral Equivalence, Z-Information, and Exact Complexity Decomposition
PAC-Bayes theory provides generalization guarantees by controlling the Kullback–Leibler (KL) divergence between posterior and prior distributions over a chosen hypothesis representation. However, predictive risk depends only on the predictive behavior induced by a hypothesis, not on the particular internal realization that implements that behavior. In over-parameterized systems, many distinct configurations induce identical predictive behavior, yet the classical PAC-Bayes KL divergence does not distinguish uncertainty over predictive behavior from variation among behaviorally equivalent realizations. We show that this distinction induces an exact structural decomposition of classical PAC-Bayes complexity. We formalize behavioral equivalence through a measurable behavior map and use measure disintegration to decompose probability measures on the configuration space into a distribution over predictive behaviors and conditional distributions over behavioral fibers. This yields an exact decomposition of the classical PAC-Bayes KL divergence into a behavior-selection term and a realization-level term given by an expected conditional KL within fibers. We define Z-information as the negative of this realization-level contribution: the exact gap between the KL divergence and the complexity of uncertainty over predictive behavior alone. We further show that the behavior-selection term admits an exact variational characterization: it is the minimum KL divergence among all posteriors inducing the same distribution over predictive behaviors, attained by a canonical fiber-symmetrized representative. Finally, we show that symmetry, behavior-preserving directions, fiber geometry, and invariance under fiber-preserving perturbations arise naturally from the same behavior-map structure. Together, these results identify predictive behavior as the natural object of PAC-Bayes complexity.