arXiv:2607.07045v1 Announce Type: new
Abstract: The gradual deployment of Wi-Fi 7/8 multi-link operation (MLO) will lead to long-term coexistence between legacy non-MLO stations (STAs) and MLO-capable STAs in WLANs. This mixed deployment makes throughput optimization challenging because legacy STAs follow single-link contention and transmission, whereas MLO-capable STAs can exploit multiple links with richer access opportunities. Existing learning-based methods usually treat such networks as homogeneous systems and directly map the current observation to a complete MAC action, which cannot faithfully represent both legacy single-link and MLO multi-link behaviors. To address this issue, we propose EvoOMG, an evolution-oriented multi-agent guidance framework for heterogeneous legacy-and-MLO Wi-Fi networks. EvoOMG reformulates throughput optimization as a standard-constrained staged multi-agent decision problem. Each agent encodes recent channel, queue, contention, and transmission histories, first generates contention guidance, and then produces aggregation guidance conditioned on the preceding access stage and standard-specific feasibility constraints. This autoregressive design follows the Wi-Fi MAC order of ``contention before transmission'' while preserving distinct protocol behaviors of legacy and MLO-capable STAs. NS-3 evaluations show that EvoOMG improves scheduled goodput, convergence stability, and MLO link utilization over static enhanced distributed channel access (EDCA), one-step MADDPG, and independent-learning baselines, achieving substantial performance gains in representative mixed-standard scenarios.
Science Journals
arXiv:2607.06802v1 Announce Type: cross
Abstract: Open physiological corpora are heterogeneous: they use different sensors, labels, sampling rates, recording settings, and clinical endpoints. They can support detector design, but they do not directly specify which detector rules should be built for a new contactless monitoring platform. We report a controlled four-analyst large-language-model (LLM) workflow for converting 68 public physiological corpora, screened for commercial-use compatibility, into an auditable library of candidate rule shapes for prospective validation. Four independent commercial LLM families read the corpus documentation under a controlled prompt and produced 695 candidate rule markers (top-markers). Deduplication retained 649 rule records; a threshold-bounds audit then flagged 51 sanity violations for clamping or curator review. Cross-corpus consolidation produced 436 unique rule shapes. Gate-tagging against two hard invariants, native target-hardware channel availability and no multi-night per-patient personalization, identified 94 build-now detector components across four detector-family buckets. The pipeline does not produce a validated clinical detector. It produces an auditable engineering cascade in which analyst disagreement, threshold checks, curator review, and automated continuous-integration (CI) checks route literature-derived rules toward prospective hardware validation.
arXiv:2607.06760v1 Announce Type: new
Abstract: Autonomous systems under partial observability act on beliefs, not raw sensor events. QANTIS treats the quantum processor as a calibrated belief-update service in that loop: it receives a prior and an observation model, estimates the rare-event evidence term, and returns an ordinary posterior to a classical planner. This paper asks whether that service can be reused across a sequential Tiger POMDP horizon on present IBM Heron hardware without corrupting the planner-facing posterior. We answer with a controlled hardware case study rather than an end-to-end autonomy or wall-clock speedup claim. The study compares no amplification, guarded Grover amplification, and all-step fixed-point amplification on the same trajectory, then checks whether the returned posterior would change the downstream action. All-step FPAA preserves the Tiger posterior across the reported 8-step and 12-step primary runs, and the 20-step and 32-step controls remain inside the same operating band. In every reported decision check, the hardware posterior and the exact Bayes posterior select the same immediate action. Boundary-aware BIQAE stabilizes amplitude estimation near zero and near one, while a rare-event sweep maps the logical sample-complexity envelope for one-in-a-million evidence. The result is an operating envelope for a hardware-calibrated belief-update primitive, not a standalone hardware-advantage claim.
arXiv:2607.07184v1 Announce Type: new
Abstract: Pre-deployment safety evaluations aim to inform the downstream risks of releasing a new AI model. Yet most evaluations provide limited evidence about how often undesired model behavior will occur in deployment: they generally have insufficient coverage, are unrepresentative, and are generally recognizable as tests. To address these concerns, we study a simple way to simulate a model deployment: starting from de-identified conversations from a previous model deployment, we hold fixed the initial conversation prefix and regenerate the next response using a candidate model. The resulting responses can then both be audited for novel misalignments and used to estimate the prevalence of model misbehavior before deployment. We evaluate deployment simulation across four GPT-5-series deployments, using registered, outcome-blinded predictions for GPT-5.4 and retrospective analyses of three earlier releases. We find that deployment simulation produces informative estimates of post-deployment misbehavior rates and outperforms baselines based on adversarially selected production data; its evaluation-awareness point estimates were also much closer to production traffic than those from traditional evaluations. We also identify the realism of tool resampling as a central challenge for further improving predictions and share results suggesting that this challenge is surmountable even in complex tool-use settings. Finally, we show that deployment simulation can be seeded from public chat datasets and remain informative about production misbehavior rates, suggesting a path for external researchers to run deployment-grounded evaluations without access to private production logs. Overall, deployment simulation helps evaluators forecast how language models will behave in the real world and supports more quantitative assessment of deployment risk.
arXiv:2607.06771v1 Announce Type: new
Abstract: Lane-change, a triggering of traffic disturbances to the upstream vehicles, is detrimental to traffic safety and efficiency. Coupled with car-following behavior, the joint maneuvers depict the general picture of how traffic disturbances generate and propagate through vehicle streams, especially under traffic congestion. This study proposes an integrated control framework for lane-changing and car-following for connected and automated vehicles (CAVs), where those two tasks are largely treated as independent driving tasks by prevailing methods. Utilizing the Parametrized Deep Q-Network (P-DQN) with a hybrid action space, the framework adeptly models multiple objectives in CAV control. The P-DQN's high-level control is employed for discrete lane-change decisions, while its low-level control manages continuous acceleration actions, i.e., lateral and longitudinal acceleration. These actions are interdependently determined, seamlessly integrating car-following and lane-changing control. By training to maximize cumulative rewards, the proposed control strategy ensures driving safety as well as the efficiency of car-following, lane-changing, and lane-keeping. Through numerical experiments, it is indicated that the P-DQN outperforms separated control methods, e.g., the combination of the Minimizing Overall Braking Decelerations Induced by Lane Changes (MOBIL) model and the Intelligent Driver Model (IDM), in terms of safety and comfort.
arXiv:2607.07190v1 Announce Type: new
Abstract: Graph emulators and hopsets are two fundamental concepts for distance approximation. When the multiplicative stretch is $1+\epsilon$ for arbitrarily small $\epsilon>0$, these structures are known as near-additive emulators and near-exact hopsets, respectively. Prior work showed that there is a remarkable similarity between the constructions and guarantees of these two objects. In their survey on this topic, Elkin and Neiman [Bull. EATCS 130, 2020] explicitly asked whether one can obtain a general reduction between near-additive emulators and near-exact hopsets. Following that, Kogan and Parter [FOCS, 2022] provided a general reduction from hopsets to emulators and spanners.
In this paper, we address the reverse direction and show that any construction for a near-additive emulator for undirected unweighted graphs can be leveraged as a black box to construct a hopset for an undirected weighted graph with comparable size, stretch, and a hopbound comparable to the emulator's additive stretch. Specifically, we show that any algorithm that constructs a $(1+\epsilon',\beta)$-emulator, with $0 \le \epsilon' \le 1$ and $\beta \ge 1$, of size $S_{\mathcal{A}}(n, \epsilon',\beta)$, can be used to obtain a $(1+\epsilon, O(\frac{\beta^2}{\epsilon^2} \ln(\frac{n}{\epsilon})))$-hopset of size $O((S_{\mathcal{A}}(n+m\frac{\beta}{\epsilon^2}, \frac{\epsilon}{294},\beta) \frac{1}{\epsilon} + n)\ln(\frac{n}{\epsilon}))$, for any $0 < \epsilon \le 1$. Therefore, our reduction answers the question of Elkin and Neiman [Bull. EATCS 130, 2020] for sparse graphs and further advances the understanding of the formal connection between these two structures. Designing a reduction resulting in a hopset size that does not depend on $m$ remains an intriguing open question.
arXiv:2601.06363v3 Announce Type: replace-cross
Abstract: This paper formalizes a widely used dynamical class--replicator-mutator dynamics and Price-style selection-and-transmission--and makes explicit the modeling choices (scale, atomic unit, interaction topology, transmission kernel) that determine how this class instantiates across domains. The backbone is known; we do not claim to have discovered selection. The novel contributions are threefold: (i) a scale-relative kernel parameterization where atomic units are themselves parameters, enabling systematic instantiation across physics, biology, economics, cognition, and social organization; (ii) a consent-friction instantiation for political philosophy, where friction is the primitive, legitimacy functions as survival probability, and belief-transfer functions as mutation kernel; and (iii) a derivation path from social contract theory rather than from biology or physics, arriving at the same formal structure via an independent route.
We provide a bridge principle connecting descriptive dynamics to instrumental normativity: if agents prefer lower expected friction, then "ought" claims are shorthand for policies that reduce expected friction under the specified dynamics. This conditional structure avoids the is-ought fallacy while grounding normative discourse in empirically tractable dynamics. We address pathological cases (authoritarian stability, suppressed friction) through explicit modeling of latent versus observed friction. The framework generates testable predictions through operationalization of friction, legitimacy, and belief-transfer dynamics, and is falsifiable at the level of measurement apparatus rather than formal structure.
arXiv:2603.11700v2 Announce Type: replace-cross
Abstract: This article is concerned with the inverse problem on determining the temporal component of the source term in a coupled time-fractional diffusion system by a single point observation. Under a non-degeneracy condition on the known spatial component, we establish the Lipschitz stability by observing all solution components by a series representation of the mild solution. To reduce the observation data, we prove the strict positivity of some fractional integral of the solution to the homogeneous problem by a modified Picard iteration. This, together with a coupled Duhamel's principle, lead us to the uniqueness of the inverse problem by observing any single solution component under a specific structural constraint on the unknown. Numerically, we propose an iterative regularizing ensemble Kalman method (IREKM) for the simultaneous recovery of the temporal sources. Through extensive numerical tests, we demonstrate its accuracy, robustness against noise, and scalability with respect to the number of components. Our findings highlight the essential roles of the non-degeneracy condition, measurement configuration, and fractional structural constraints in ensuring reliable reconstructions. The proposed framework provides both rigorous theoretical guarantees and a practical algorithmic approach for multi-component source identification in fractional diffusion systems.
arXiv:2606.17420v2 Announce Type: replace-cross
Abstract: Tau positron emission tomography (PET) is widely used for the in vivo characterization of disease stage and progression in Alzheimer's disease (AD). With the adoption of multiple tau PET tracers including AV-1451, PI-2620, MK-6240 with different binding behaviors in various large-scale studies, there is a great need of effective harmonization methods to enable the cross-tracer integration of tau PET datasets. While previous methods such as CenTauR were proposed to standardize scalar tau PET measures, they are limited in accounting for the heterogeneity of tau pathology. In this work, we propose Feynman-Kac Reweighted Schr\"odinger Bridge Matching (FKRSBM), a surface-based framework for cross-tracer tau PET harmonization. FKRSBM learns a direct stochastic transport between tracer domains using Schr\"odinger Bridge matching, avoiding the Gaussian-prior routing used in diffusion-based translation. To promote biologically consistent transport, FKRSBM introduces an endpoint penalty favoring bridge pairings with matched tau-pathology status and implements it through a Feynman-Kac reweighted endpoint proposal. To preserve cortical organization, FKRSBM uses a spherical convolutional network for vertex-level harmonization on cortical surface meshes. In our experiments, we demonstrate our method by harmonizing Tau PET images acquired with the AV-1451 (n=1480) and PI-2620 (n=2458) tracers from two large-scale datasets. Compared to previous methods including ComBat, CycleGAN, Diffusion Model(DF), and unregularized Schr\"odinger Bridge Model(DSBM), the proposed FKRSBM method outperforms these baselines in subgroup-level alignment, tau-positivity consistency, and diagnostic classification while preserving subject-specific cortical topography of tau pathology. The code is available at: https://github.com/jianweizhang17/FKRSBM.
arXiv:2606.23190v2 Announce Type: replace-cross
Abstract: Existing Reinforcement Learning (RL) research for Text-to-Speech (TTS) focuses on large language models (LLMs), leaving Flow-Matching (FM) under-explored. We present FlowTTS-GRPO, an online RL framework for FM-based TTS. By converting ordinary differential equation (ODE) trajectories into stochastic differential equation (SDE) paths, our method enables direct fine-tuning of open-source FM models without auxiliary models. We show that a weighted reward combination converges faster than a probabilistic scheme, and identify three practical optimizations: omitting classifier-free guidance (CFG) during training accelerates convergence; synthesizing hard cases improves robustness; and applying RL to the FM component enhances audio-detail metrics. Experiments on CosyVoice 3.0 and F5-TTS demonstrate objective and subjective preference gains in speaker similarity and perceptual quality, with F5-TTS also improving intelligibility.
arXiv:2607.07224v1 Announce Type: new
Abstract: Electromagnetic waves are intrinsically vectorial and require description via polarization, unlike scalar fields such as acoustic pressure or electronic wavefunctions. In three dimensions, the transversality constraint further prevents any globally smooth transverse-polarization frame at the $\Gamma$ point, which would apparently rule out a simple scalar band structure for three-dimensional (3D) photonic crystals. We show here that site-adapted $p$-orbitals can realize scalar-wave dispersion: the induced band representation is isomorphic to the scalar elementary band representation up to a one-dimensional character twist, so the symmetry-enforced degeneracies and compatibility relations are the same. We demonstrate this mechanism experimentally in 3D photonic meta-crystals, where the local $p$-orbital axes adapt from site to site according to symmetry. In contrast to a fixed-polarization reduction (e.g., in 2D), our construction preserves site-polarization textures while simultaneously supporting a scalar network with one amplitude per site. Thus, it offers a pathway from vectorial photonic degrees of freedom to scalar band engineering, keeping polarization as an active design knob.
arXiv:2607.07376v1 Announce Type: new
Abstract: Rare-event searches require exceptionally low background levels, motivating the development of increasingly sophisticated screening methods to push sensitivity limits. Liquid scintillators are particularly attractive detector media due to their intrinsic radiopurity and the ability to scale to large target masses. In this work, we present a screening strategy capable of measuring ultra-trace concentrations of $^{40}\text{K}$ with sensitivities below $10^{-15}$g/g. The method combines neutron activation analysis with a dedicated radiochemical treatment, followed by low-background HPGe gamma spectroscopy. Using this approach, we achieved a minimum detectable concentration of $2.9\cdot10^{-16}$g/g for $^{40}\text{K}$, placing this technique among the most sensitive currently available.
arXiv:2607.07383v1 Announce Type: new
Abstract: Knowledge-based Visual Question Answering (KB-VQA) requires models to retrieve visual entities matching the query image from large-scale encyclopedic knowledge bases and answer related questions. Existing multimodal Retrieval Augmented Generation (mRAG) methods rely on global visual features to match candidate entities, yet when the knowledge base contains numerous visually similar entities, the retriever struggles to distinguish them, populating the candidate set with visually similar but factually mismatched distractors. Since subsequent processing steps such as noise filtering are also confined to this fixed candidate set, errors from failed retrieval inevitably propagate to the final answer. To address these challenges, we propose MMAgent-R$^2$, an agentic mRAG framework that integrates visual reranking and active rejection as its internal verification mechanism. Visual reranking directly compares query and candidate images, capturing discriminative details beyond textual descriptions to precisely identify the target entity among similar candidates; active rejection discards unreliable results and retrieves additional candidates when no confident match is found, moving beyond the fixed candidate pool. We design a composite reward function with step-level verification rewards and achieve joint optimization of external retrieval, internal verification, and answer generation via GRPO training. Experiments on InfoSeek, E-VQA, and MMhops demonstrate that \ours{} achieves state-of-the-art performance, with particularly notable advantages in challenging retrieval scenarios and complex multi-image multi-hop reasoning tasks.
arXiv:2607.05127v3 Announce Type: replace
Abstract: Understanding the physics of many-body complex dynamical systems may be a non-trivial task. High-dimensional analysis approaches are often deemed necessary to prevent losing important information. Typically, these use order parameters or descriptors capturing information related to, e.g., relative positions, symmetries, etc., of the units in the studied system. However, in many cases, gaining information related to the relative positions of the constitutive units (or their velocities) alone may be insufficient, and to reach a more complete physical knowledge, one should ideally learn and correlate with each other both structure and dynamics. Here we demonstrate how to achieve such a goal efficiently by building and navigating high-dimensional Time-Derivatives (TiDe) spaces. A TiDe space can be generated for virtually any type of system/phenomenon from the time-series data collected along its observation over time. Each TiDe's dimension corresponds to a growing-order time-derivative of the extracted data, thus containing information related to different physical phenomena/events, which can be easily extracted via unsupervised approaches. We demonstrate how, by definition, TiDes can be directly analyzed without a need for prior dimensionality reduction, providing results that are intrinsically intuitive to interpret. We show the potential of the method by analyzing two prototypical example datasets extracted from molecular dynamics simulations or experimental tracking of different types of complex dynamical systems. Our results demonstrate how efficiently one can navigate and learn in information-rich TiDe spaces, which provide a robust general framework for data analysis and for studying complex dynamical systems from the data collected along their observation over time.
arXiv:2607.06748v1 Announce Type: new
Abstract: Smart home automation platforms increasingly rely on user-authored YAML configuration files to define device behaviors, but these files are prone to syntax, formatting, and semantic logic errors that can cause automation failures and safety risks. Existing YAML validators, static analysis tools, and general-purpose large language models offer limited support for end-to-end diagnosis and repair because they lack domain-specific understanding and validated correction workflows. This paper presents SmartHomeSecure, a prototype for automated detection and repair of Home Assistant configuration errors using lightweight program analysis and constraint-guided large language model generation. SmartHomeSecure parses YAML files, detects syntactic and common semantic errors, normalizes error context, applies deterministic auto-fixes for routine defects, and constructs constrained prompts that guide LLMs toward minimal and structurally valid repairs. The system is implemented as a modular web application with four layers: UI Shell, Feature Orchestrator, Domain Engine, and Integration Layer. Its repair pipeline was evaluated on 100 real-world Home Assistant YAML files with manually injected errors across five categories: syntax/parsing, indentation, mapping, sequence, and scalar quoting errors. Four models were tested: gpt-oss-20b, gpt-oss-120b, llama-3.1-8b, and llama-3.3-70b. Results show that three models achieved 100% error detection accuracy, with repair success rates ranging from 87% to 93%. Manual verification found no hallucinated or incorrect repairs among successful outputs. These findings suggest that combining domain-aware program analysis with constrained generative AI is a feasible approach for improving the reliability and usability of smart home configuration repair.
arXiv:2607.06510v2 Announce Type: replace-cross
Abstract: Spherical colloids with harsh repulsive forces have long been used as experimental analogs of the hard sphere model, with demonstrated good agreement with computer simulations for bulk and structural properties of the fluid, glass and crystal phases. However, an enigmatic discrepancy remains for the crystal-melt stiffness coefficient. Here we perform computer simulations of colloidal hard spheres under tunable buoyant mass and show that the long-standing discrepancy can be traced to a hitherto unrecognized gravity dependent contribution of the stiffness coefficient. This effect is one practical realization of a more general result for the external field dependence of stiffness coefficients of arbitrary interfaces.
arXiv:2607.06542v2 Announce Type: replace
Abstract: Dependency parsing consists of finding a tree representation for a sequence. Unsupervised dependency parsing aims to develop parsing methods without a gold standard during model training. In human languages, an unsupervised parser can be evaluated because some gold standard is usually available or can be created. For other species, a gold standard is unknown. Thus one may conclude that it is impossible to determine the accuracy of an unsupervised parser and, consequently, dependency parsing is unfeasible in other species. However, here we apply recent advances in network science to demonstrate that the proportion of correct edges retrieved by a parser must be high for the sequences of vocalizations or gestures that non-human primates produce due to the fast decay of the sequence length distribution. In contrast, human language sequences lack that property. Therefore, evaluation without a gold standard is feasible in non-human primates but a hard problem in humans.
arXiv:2607.07440v1 Announce Type: new
Abstract: In parametrized linear systems $\mathsf{P}(\mu)\mathsf{x}=\mathsf{b}$ the system matrix $\mathsf{P}$ depends nonlinearly on a parameter $\mu$ and solutions are sought for many values of this parameter. We show that the compact rational Krylov (CORK) framework, originally introduced to solve nonlinear eigenvalue problems, can be used to efficiently produce approximate solutions to such a system for many values of the parameter at once. In this approach the parametrized system is first linearized, resulting in a large shifted linear system $(\boldsymbol{\mathsf{A}}-\mu\boldsymbol{\mathsf{B}})\boldsymbol{\mathsf{y}}=\boldsymbol{\mathsf{d}}$. We formulate a left- and right-preconditioned rational Krylov GMRES method for shifted linear systems and show how the CORK framework can be used to speed up these methods in the setting of parametrized linear systems. Additionally, we show how to incorporate a right-hand side $\mathsf{b}(\mu)$ that also depends on the parameter, how to choose the shifts to steer convergence and how to allow for inexact solves at these shifts throughout the iterations. As an application we consider the 'frequency sweeping' of Helmholtz scattering problems through the Boundary Element Method (BEM), enabled via an efficient representation of the dense but data-sparse wavenumber-dependent system matrix.
arXiv:2607.07023v1 Announce Type: new
Abstract: Supervised fine-tuning (SFT) is often treated as a capability-adaptation step, while alignment is attributed to later preference optimization or reinforcement learning. This separation is incomplete: when examples are scored and kept online during fine-tuning, the choice of which data to train on already changes the model's behavioral preferences. We study online data selection as an implicit alignment mechanism. Given the same base model, optimizer, and selected-token budget, we compare random, loss-based, quality-based, and diversity-based online selectors and measure the behavioral drift they induce without any preference optimization. The proposed evaluation tracks helpfulness, refusal rate, verbosity, truthfulness, sycophancy, calibration, and jailbreak robustness, together with diagnostics for which behavioral modes are over-represented in the selected data. We formalize online selection as a reweighted SFT objective whose weights define an implicit preference over response styles and safety postures, so that an online scorer plays the role usually assigned to a reward model. This view predicts that high-scoring data can systematically favor longer, more assertive, more compliant, or more refusal-prone behaviors depending on how the online score is defined. Empirically, selectors that are statistically indistinguishable in task accuracy diverge sharply in refusal rate, verbosity, and sycophancy, and we show that the direction of the shift is predictable from the attribute mixture of the selected data. We introduce Alignment Drift Auditing (ADA), a controlled protocol for quantifying selection-induced behavioral movement, and Alignment-Aware Selection (AAS), a diagnostic online selector that retains data efficiency while constraining drift along safety and style axes.
arXiv:2607.07026v1 Announce Type: new
Abstract: Constrained decoding is essential for serving LLMs, ensuring that generated outputs follow specific structures such as JSON schema-formatted function calls. Existing systems are designed for autoregressive models and assume left-to-right generation, masking out invalid next tokens at each step. Diffusion language models, however, break this assumption: they sample multiple positions simultaneously from a fully-factorized mean-field distribution at each denoising step. In this paper, we present an exact and tractable algorithm for sampling from the constrained mean-field posterior under any constraint expressible as a finite automaton. Viewing finite automata as graphical models, we obtain tractable representations of the constrained distribution that enable efficient inference. The approach guarantees constraint satisfaction by construction, supports both greedy and sampling-based decoding, and is compatible with parallel and block-wise decoding under arbitrary remasking schedules. Applying depth-reduction techniques from arithmetic circuit theory, we further reduce sampling depth from linear to logarithmic in the sequence length. Empirical evaluations on Dream-7B and LLaDA-8B show substantial accuracy gains across various tasks including function calling (xLAM, BFCL), planning (Sudoku, Countdown), text-to-SQL (Spider), and math reasoning (GSM-Symbolic), with little inference overhead relative to unconstrained decoding. For example, on BFCL-Live, our approach improves Dream-7B's greedy decoding accuracy from 63.9% to 71.5%, and stochastic sampling accuracy from 22.3% to 69.0%, where the unconstrained baseline collapses, with under 5% wall-clock overhead.
arXiv:2607.07040v1 Announce Type: new
Abstract: How can we measure intelligence beyond human capability?
Human-authored benchmarks saturate, and above human capability, examiners may not know which tasks are both hard and verifiable. We argue that this difficulty is inherent to absolute-scale evaluation and propose a new paradigm based on relative measurement in which models generate public challenges that separate other systems. Aggregating these outcomes yields an adversarial psychometric rating system that can scale with the systems being measured. We describe practical protocols that reduce incentives for private-information attacks, support judge-free adjudication, and naturally scale with agent capabilities. We instantiate the framework across verifiable and open-ended, non-verifiable domains, illustrating how model-generated evaluation can continue to measure systems beyond the human frontier.
arXiv:2607.07373v1 Announce Type: new
Abstract: In biological circuits, sequential neural activity evolves along dynamic, low-dimensional manifolds to enable flexible behavior. Spiking network models link aspects of this sequential activity to features of manifold geometry through specific circuit mechanisms, making dynamic neural manifolds parameterizable, and thereby offering an explainable framework for neural computation. Extending this framework to neuromorphic engineering, we present an implementation on the SpiNNaker 2 chip for real-time, closed-loop control. By allowing sensory inputs to modulate heterogeneous inhibition, gain, and transient currents, our architecture drives rapid subspace rotations to switch between behaviors, as well as fine-grained trajectory control within them. We validate this via a robotic simulation where an agent uses sensory feedback to dynamically reconfigure its manifold geometry to navigate through a maze. Our results establish dynamic manifolds as a feasible approach for explainable neuromorphic architectures and a substrate for investigating biological neural dynamics.
arXiv:2607.07375v1 Announce Type: new
Abstract: Adversarial vulnerability in deep neural networks (DNNs) has been studied from the perspectives of decision-boundary geometry, feature robustness, input-output Jacobians, and the instability of inverse problems. Here, we focus on the spectral structure of intermediate linear transformations that propagate information through modern DNNs, an unexplored mechanism of adversarial vulnerability. Specifically, we investigate transformer-based vision-language models, whose linear layers admit interpretable spectral decompositions and whose widespread adoption makes understanding their robustness increasingly important. We propose a white-box spectral-subspace-guided attack (SSGRA) that aligns intermediate representations with the subspace spanned by the bottom right singular vectors. Our experiments show improved attack effectiveness over existing baselines. In addition, SSGRA offers a spectral interpretation of adversarial vulnerability in VLMs, providing insights for improving their robustness.
arXiv:2607.07481v1 Announce Type: new
Abstract: Large language models are increasingly used to evaluate and support software engineering tasks, yet the validity of these evaluations is often undermined by uncertainty about whether benchmark instances were seen during pretraining. This can lead to data contamination, which may inflate performance and result in misleading conclusions about model capability. Despite this, the training corpora of many modern models are only partially disclosed, making direct decontamination infeasible. This creates a need for practical methods that can detect a large language models' prior exposure to training data without access to the full training corpus.
To address this challenge, we organize the first Poisoned Chalice of LLM Evaluation Competition, co-located with the FSE-AIWare 2026 Competition Track. The competition frames contamination detection as a white-box membership inference task on source code and provides participants with curated datasets, target models, baseline attacks, and a final evaluation on a held-out model and dataset. This design encourages methods that generalize beyond superficial dataset artifacts and beyond a single training setting.
This paper reports the setup and results of the competition. More broadly, the competition aims to catalyze the community around trustworthy LLM evaluation for software engineering.
arXiv:2607.07483v1 Announce Type: new
Abstract: We study the problem of estimating the number of edges in an undirected, unweighted graph using sublinear query access. We consider a query model that preserves the structure of Independent Set (IS) queries, but augments their output with a random certificate: given a vertex subset, the oracle returns a uniformly random edge from the induced subgraph if one exists, and returns null otherwise.
Using this access, we give a randomized algorithm that outputs a $(1 \pm \varepsilon)$-approximation to the number of edges with constant success probability using $\widetilde{O}(\log^{2} m)$ queries. This implies an exponential separation from both standard IS queries and global random edge-sampling models: estimating the number of edges using standard IS queries require $\widetilde{\Theta}\!\left(\min\left\{\sqrt{m},\, \frac{n}{\sqrt{m}}\right\}\right)$ queries, while direct random edge-sample access requires $\widetilde{\Theta}(\sqrt{m})$ samples. Beyond separation in query complexity, our algorithm is output-sensitive: its query complexity is polylogarithmic in the number of edges in the graph. This aligns with the classical objective in group testing, where one seeks algorithms that are both worst-case optimal and instance-adaptive.
Conceptually, our model connects group testing, the decision-versus-counting dichotomy, graph property testing, and the "power of a random certificate", and can be viewed as a structured form of conditional sampling of edges in graphs.