arXiv:2607.15751v1 Announce Type: new
Abstract: This work introduces Physics-Informed Splines (PI-Splines), a structured spline-based architecture for physics-informed learning. Instead of representing the solution of a differential equation with a neural network, PI-Splines directly parametrize the unknown field through a tensor-product B-spline expansion with trainable control coefficients. This formulation preserves the residual-based training paradigm of Physics-Informed Neural Networks while providing compact support, explicit smoothness control, analytical derivatives, and a direct geometric interpretation of the trainable parameters. When compatible with the spline representation, boundary conditions can be imposed strongly by fixing suitable boundary control coefficients. The proposed method is evaluated on several benchmark problems of increasing difficulty and compared with standard physics-informed frameworks under matched governing equations, collocation sets, loss terms, and optimization procedures, so as to isolate the effect of the approximation architecture. Numerical experiments show that PI-Splines provide a competitive and stable alternative to neural physics-informed architectures, particularly in settings where structured representations, locality, and parameter efficiency are desirable.
Science Journals
arXiv:2607.15730v1 Announce Type: new
Abstract: Language-based user profiles convert long behavioral histories into explicit semantic representations for recommendation. However, most profile generators are optimized in an open loop: they may summarize past behavior fluently, but are not directly trained to improve future recommendation. We study this problem in real-world short-video recommendation, where user behaviors continuously arrive as streams and profiles must be incrementally updated under limited capacity. This requires maintaining a consistent bounded profile state and constructing profile-targeted semantic feedback from industrial implicit behavior logs. We propose RECAP, an offline closed-loop framework for optimizing streaming structured semantic profiles with historical recommendation feedback. RECAP maintains each profile as a bounded structured memory by combining LLM-based semantic updates with deterministic lifecycle and capacity control. RECAP constructs profile-targeted semantic feedback by filtering label-consistent behavior pairs with an LLM judge and training a dual-tower evaluator whose matching score serves as a GRPO reward. Experiments on Kuaishou short-video data show that RECAP improves uAUC by 0.0084 and Recall@2000 by about 4.9% over the base generator. Further analyses confirm the benefits of feedback construction and policy optimization, and show more grounded refinement and user-level abstraction in profile updates. A seven-day online A/B test further shows a statistically significant 0.139% improvement in average application usage time per user.
arXiv:2607.15736v1 Announce Type: new
Abstract: Large reasoning models often solve problems through long chain-of-thought (CoT) traces, yet much of this computation is spent on redundant derivations, repeated self-verification, and detours that do not improve the final answer. Existing on-policy self-distillation methods reduce this cost by matching a student model to a concise copy of itself on prefixes sampled from the student's own rollouts. We show that this objective has an initialization bottleneck. Since supervision is applied only to visited prefixes, training from a verbose base model places the KL loss on contexts that are often noisy, redundant, or already off track. In such regions, a concise teacher can provide only local corrections, while the student continues to explore trajectories that an efficient reasoner should avoid. In this paper, we propose BIRD(Bootstrapped Iterative Self-Reasoning Distillation), a two-stage self-reasoning distillation method that improves the rollout distribution before on-policy training. BIRD first samples concise solutions from the base model under a brevity instruction, keeps only answer-correct traces, and performs a lightweight prompt-switch SFT step. The traces are generated with the brevity instruction but learned under the original task prompt, turning instruction-induced conciseness into a default reasoning behavior. Starting from this warm model, BIRD then applies on-policy reverse-KL distillation with a concise self-teacher, now on cleaner and more informative prefixes. Across Qwen3 series models, BIRD achieves a stronger accuracy-efficiency trade-off than prompting and cold-start on-policy distillation on MATH-500 and AIME benchmarks. On Qwen3-8B, it improves MATH-500 accuracy from 86.2% to 92.0% while reducing the average response length from 3,099 to 1,115 tokens. These results highlight prefix support as a central factor in efficient reasoning distillation.
arXiv:2607.15513v1 Announce Type: new
Abstract: In this paper we examine the emergent structures of random networks that have undergone bond percolation an arbitrary, but finite, number of times. We define two types of sequential branching processes: a competitive branching process - in which each iteration performs bond percolation on the residual graph (RG) resulting from previous generations; and, a collaborative branching process - where percolation is performed on the giant connected component (GCC) instead. We investigate the behaviour of these models, including the expected size of the GCC for a given generation, the critical percolation probability and other topological properties of the resulting graph structures using the analytically exact method of generating functions. We explore this model for Erdos-Renyi and scale free random graphs. This model can be interpreted as a seasonal N-strain model of disease spreading.
arXiv:2508.21516v3 Announce Type: replace
Abstract: A digital twin (DT) contains a set of virtual models of real systems and processes that are synchronized with their physical counterparts. In a setup in which contact with the physical world is maintained through sensors and actuators that are wirelessly connected to the DT's computing engine, DT alignment requires periodic status updates, while safety-critical messages and fault conditions call for low-latency anomaly reporting, creating a fundamental trade-off in how wireless resources are used. We present a medium access framework combining pull-based updates, centrally scheduled according to goal-oriented principles, with urgent push-based updates, for which transmission decisions are made directly by the sensors. This enables the system to quickly detect and recover from anomalies while maintaining DT alignment. We thus design a push-pull scheduler (PPS) that strikes a balance in the trade-off between DT alignment in normal conditions and anomaly reporting, optimizing resource usage and reducing DT drift by 20 - 30% with respect to state-of-the-art solutions while maintaining the same anomaly detection guarantees, or reducing worst-case anomaly detection times by 30 - 70% while meeting the same DT alignment conditions.
arXiv:2509.13026v2 Announce Type: replace
Abstract: Strong functors and monads are ubiquitous in Computer Science. More recently, (strong) comonads have demonstrated their use in structuring context-dependent notions of computation. However, the dualisation of ``being strong'' property passed somehow unobserved so far. We argue that ``being costrong'' gives a different understanding of how functors can interact with monoidal structures. We shall see that the well-known correspondence between distributive laws $F T \to T F$ of an endofunctor $F$ over a monad $T$, on one hand, and extensions of $F$ to the Kleisli category of that monad, on the other hand, generalises from ordinary monads to graded ones. The gist here is to recognise that the costrength of a costrong functor is nothing but a ``graded'' distributive law. As such, ``being costrong'' is a structure that a functor may have. Examples of costrong functors with respect to different graded monads are provided, with emphasis to the cartesian case, and applications to optics and coalgebras are given.
arXiv:2509.13186v2 Announce Type: replace
Abstract: Phishers achieve large-scale attacks by using ready-to-deploy phishing websites (phishing kits) to rapidly launch campaigns that leverage specific data exfiltration, evasion, or mimicry techniques. In contrast, researchers and defenders continue to rely on manual analysis to identify features for kit fingerprinting. In this paper, we examine the link between a page's client-side behavior and the underlying phishing kit used, enabling automated aggregation of phishing pages. Our key insight is that client-side techniques make heavy use of browser APIs, which, in turn, differentiate underlying kits based on their feature sets. Using an instrumented browser and a URL fuzzing utility, we collected traces from 1,328,917 pages and recovered kit archives for 4,180 pages between August 2023 and January 2025. For the labeled subset, we find that clustering based on the set of browser APIs executed yields 98% accuracy in grouping them by the underlying kit. We also find that 434,495 phishing pages execute enough browser APIs to cluster into 9,306 clusters, compressing multi-lingual phishing pages across various domains into a single cluster. Our findings show that analysts and researchers can leverage the complexity of client-side phishing code to track phishers' kit deployments in the wild.
arXiv:2607.15487v1 Announce Type: new
Abstract: Shared autonomy enables humans and robots to collaboratively perform tasks by combining human input with autonomous assistance. Most prior work focuses on improving intent inference under a fixed environment, overlooking how workspace design itself affects inference difficulty. We observe that the physical arrangement of objects directly influences the separability of candidate goals under noisy user inputs. We formulate workspace design as an optimization problem and derive a probabilistic correctness guarantee under a bounded noise model. Through simulation experiments across multiple tabletop scenarios, we show that optimized layouts improve goal inference reliability and reduce ambiguity compared to baseline arrangements. We further demonstrate a real-world shared autonomy system that integrates the proposed inference framework. This highlights the role of environment design as a complementary axis for improving shared autonomy systems.
arXiv:2607.15498v1 Announce Type: new
Abstract: The key-value (KV) cache is the main memory bottleneck in long-context large language model (LLM) inference. Two leading training-free families are both structurally limited: token-selection methods (SnapKV, Ada-KV) score importance from an observation window and evict low-scoring tokens, but eviction is irreversible -- so when the importance signal degrades under query-agnostic reuse, accuracy collapses by 11-15 points; uniform low-rank coding keeps every token but spends equal rank everywhere, wasting budget. We observe that both failures share one cure: rank should be allocated, not evicted. We present VarRate, a training-free KV codec that assigns each token a variable low-rank budget by its query salience, keeping every token at a nonzero rank. Comparable adaptive-rank codecs reach this allocation only through training; VarRate requires none. Because no token is dropped, it degrades by only 3.5-5.5 points where query-aware selection collapses. At a matched 20% budget on LongBench (16 tasks), VarRate stays within 0.8 points of the uncompressed model on both Llama-3.1-8B and Qwen2.5-7B. Averaged over the two, it is the strongest matched-memory compressor. It significantly beats its uniform-rank ablation on both models. Against KVzip, a method purpose-built for query-agnostic reuse, it is accuracy-equivalent in three of four settings and within a point overall, at about one-eighth the prefill overhead.
arXiv:2607.15508v1 Announce Type: new
Abstract: In this paper, we address the challenge of multi-objective motion planning for systems under kinodynamic constraints. We consider three problem classes: (i) lexicographic optimization, in which objectives are minimized according to a strict priority ordering, (ii) constrained optimization, in which a primary objective is minimized subject to bounds on the remaining costs, and (iii) Pareto front optimization, in which the goal is to approximate the full set of optimal trade-offs among competing objectives. We first show that established cost scalarization methods for multi-objective problems cannot be extended to continuous-domain systems with correctness guarantees. Then, we propose a unified algorithmic framework built upon the Stable Sparse-RRT (SST) algorithm, in which the single representative maintained at each witness neighborhood is replaced by a representative set of locally Pareto-optimal nodes. This structure gives rise to three distinct algorithms: lexSST for lexicographic minimization, coSST for constrained optimization, and poSST for Pareto-front approximation. We provide theoretical guarantees for the completeness and optimality of our algorithms and demonstrate their effectiveness through extensive empirical evaluations.
arXiv:2607.15875v1 Announce Type: new
Abstract: Scientific claims shared on social media are often difficult to verify and may contribute to the spread of misinformation. To address this challenge, automated fact verification systems require scientific claim-source retrieval, the task of identifying the source publication underlying a given claim. However, claims often differ substantially from their source publications in language, style, and specificity, making retrieval challenging. We present a comparative study of scientific claim-source retrieval on the CheckThat! 2026 benchmark across sparse and dense retrieval models. Our results show that translating claims into English outperforms both original and bilingual claim representations, while incorporating publication metadata provides additional retrieval gains by capturing indirect source references. In addition, we analyze four style transfer approaches and find that they improve retrieval performance for most models, although the optimal style depends on the underlying retrieval objective. Finally, we investigate similarity- and signal-based re-ranking approaches, introducing three novel re-ranking models based on attribution, entity overlap, and verification-based reasoning. Verification-based re-ranking yields additional gains beyond semantic similarity and achieves the best overall performance with an MRR@5 of 0.758.
arXiv:2510.00182v2 Announce Type: replace
Abstract: While we know that large language models (LLMs) can solve some planning problems, we do not understand the extent of these capabilities for robotics. One promising direction is to integrate the semantic knowledge of LLMs with the formal reasoning of task and motion planning (TAMP). However, designing such systems is complicated by the myriad of choices for how to integrate LLMs within TAMP. We develop 16 algorithms that use LLMs to substitute key TAMP components. Our zero-shot experiments across 13750 evaluations and three domains reveal that LLM-based planners exhibit lower success rates and higher planning times than engineered systems. Providing geometric details increases the number of task-planning errors compared to pure PDDL descriptions, and (faster) direct LLM variants outperform (slower) reasoning variants in most cases. Code and results are available at https://github.com/jorge-a-mendez/llm-pddlstream.
arXiv:2607.15422v1 Announce Type: new
Abstract: Soft robots are well-suited for applications such as rehabilitation and surgery that require adaptable and safe interaction with their environment. However, the challenges of reproducible and scalable fabrication of soft robots limit their real-world deployment. Various fabrication methods have been introduced, but many are labor-intensive and prone to human error. Therefore, traditional two-part pour casting remains an attractive option. This paper presents procedures for robust, repeatable, and scalable fabrication of soft pneumatic actuators using two-part pour casting. The presented methods prevent internal cavity clogging and ensure air-tight sealing. Additionally, a robust sensor embedding procedure for thin-film flex sensors is presented, which allows for accurate and repeatable data acquisition. Finite Element Modeling (FEM) of the soft actuator is performed to analyze stress and deformation from internal pressure loadings. Pneumatic actuation experiments with PID pressure control are performed. Automated image processing is used to calibrate the embedded flex sensor to bending angle measurements. Staircase and sinusoidal profile actuation experiments validate the performance of the fabricated actuator. Angle response experiments for the staircase input show repeatable performance, and the sinusoidal input shows a small amount of hysteresis consistent with viscoelastic response to pneumatic actuation of soft actuators. Simulated and real-world bending angles show comparable response. These methods provide a repeatable and robust fabrication procedure, validated across two operators and 24 successful fabrications, along with benchmark simulations and experimental testing. These benchmarks will enable more widespread adoption of soft robotics.
arXiv:2607.15334v1 Announce Type: new
Abstract: Numerical breakdown at high Weissenberg number is often attributed to loss of symmetric positive definiteness (SPD) of the conformation tensor. That conclusion follows from Maxwell-type models without solvent viscosity. With solvent fraction $\beta>0$, the initial-value problem is locally well posed for arbitrary symmetric stress. We derive the missing quantitative theory for indefinite states and test its computational consequences. Frozen-coefficient analysis gives a growth rate uniformly bounded in wavenumber and the direction-resolved instability threshold $\lambda_{\min}(A)<-\beta/(1-\beta)$; stress diffusion supplies a closed-form cutoff, while the classical $\sigma\propto k$ catastrophe is recovered as solvent viscosity vanishes. A determinant identity shows that violations self-heal on the timescale $\lambda/2$, so persistent violations measure the truncation error that recreates them. Spectral and lattice Boltzmann tests reproduce the threshold, solvent-fraction reversal, and resolution independence. In four-roll-mill interventions, enforcing SPD delays blow-up by 15 convective times but reduces the stagnation-point Weissenberg number by 30%. Across five coupling schemes, a local second-moment stress source remains stable through the full budget at $\mathrm{Wi}=50$ while carrying $\det A\approx-8.5\times10^5$; the divergence-coupled variant fails at $t^*=47$. The surviving scheme matches published benchmarks within 0.05% and 0.18% at $\mathrm{Wi}=10$ and 20. Thus loss of positive definiteness is neither necessary nor sufficient for breakdown: the discrete coupling route decides, and the violation is a resolution gauge for which we provide run-time monitors.
arXiv:2607.15793v1 Announce Type: new
Abstract: Accurate experimentally verified models of molecular hot-band transitions are essential for interpreting high-temperature spectra in environments ranging from exoplanetary atmospheres to combustion systems. However, line-resolved measurements of infrared hot-band transitions reaching highly excited vibrational states above 10000 cm$^{-1}$ are missing because these transitions are too weak to observe at room temperature and become spectrally congested at elevated temperatures. Here, we introduce a nonlinear spectroscopic approach that enables simultaneous measurement of individual infrared hot-band transitions between four vibrational bands up to 12000 cm$^{-1}$ over a broad spectral range with sub-Doppler resolution and sub-MHz frequency accuracy (10$^{-10}$ relative line position accuracy). The methods is based on an all-optical triple-resonance (AOTR) scheme that combines stepwise mid-infrared pumping using an optical-frequency-comb-stabilized, double-seeded continuous-wave optical parametric oscillator with broadband highly sensitive near-infrared probing using a cavity-enhanced optical frequency comb. As a proof of principle, we measure transitions between high polyads (P) of methane - groups of strongly interacting, near-degenerate vibrational energy states arising from couplings between the C-H stretching and bending modes. In a single probe spectrum, we simultaneously resolve sub-Doppler P4$\leftarrow$P0, P6$\leftarrow$P2 and P8$\leftarrow$P4 transitions, reaching the poorly understood polyad P8 near 12000 cm$^{-1}$ and providing the first set of 41 experimentally observed lines in the P8$\leftarrow$P4 spectral region.Comb-based AOTR spectroscopy opens a new route for broadband exploration of highly excited molecular states, delivering extensive high-accuracy spectroscopic data needed to refine molecular models and improve predictions of high-temperature spectra.
arXiv:2607.15876v1 Announce Type: new
Abstract: We present a new ML-like programming language Yarrow with algebraic effects and region-based memory management. Reconciling these programming language features into one language is challenging: the non-local control flow of algebraic effects break the stack discipline of function calls and returns that region-based memory management relies on, and multi-shot effect handlers break the invariant that regions can be exited at most once. We present a program logic, called Yarrow Logic (YL), that supports safe and modular reasoning about regions in the presence of one-shot and multi-shot effect handlers. We prove the logic sound w.r.t. the operational semantics of Yarrow which is inspired by the runtime of OCaml but refined for regions. We use YL to prove correctness of a number of case studies with algebraic effects, including checkpointing, asynchronous computation and a LIFO data structure implementation. Since all memory locations used in these case studies are allocated in regions, these case studies avoid using the less efficient garbage collected heap memory. We have formalized Yarrow's operational semantics, the Yarrow program logic, and all our case studies using the Iris separation logic framework on top of the Rocq Prover.
arXiv:2607.15374v1 Announce Type: new
Abstract: Multimodal large language models (MLLMs) ground whole objects well from free-form language queries, but they struggle when the query names a part rather than the object. We trace this to a missing object-part hierarchy, since parts are localized in the same single step used for objects. We propose Object-Part Hierarchical Reflective Grounding (OP-HRG), a coarse-to-fine reasoning-guided grounding strategy that first localizes the parent object and then the part within it. A self-check then reflects on the result, with an extension to re-encode the predicted crop to inspect the region it is correcting. We introduce a part-aware GRPO framework to train our pipeline with stage-wise rewards. A 4B model trained this way outperforms 7B grounding LLMs and SAM3 across PascalPart, PartImageNet, and InstructPart, and transfers to reasoning segmentation.
arXiv:2607.16073v1 Announce Type: new
Abstract: We consider the $hp$-FEM applied to the Helmholtz scattering problem with wavenumber $k$, truncated with a perfectly-matched layer. The scatterer consists of a combination of Dirichlet, Neumann, and penetrable obstacles together with variable coefficients. Provided that the Helmholtz solution operator is polynomially bounded in $k$, all coefficients are piecewise smooth, all boundary surfaces are Gevrey and all coefficients restricted to boundary surfaces are Gevrey together with all their normal derivatives, we show that the $hp$-FEM is quasioptimal when $p\geq 1+\varepsilon \log k$ and $hk/p$ is sufficiently small; i.e., the $hp$-FEM does not suffer from the pollution effect. This result generalises the analogous results in both [Bernkopf, Chaumont-Frelet, Melenk 2025] (proved for piecewise analytic coefficients and analytic boundaries) and [Galkowski, Lafontaine, Spence, Wunsch 2024] (proved for smooth coefficients that are analytic near analytic obstacles) to a much larger class of scatterers.
arXiv:2607.15347v1 Announce Type: cross
Abstract: We introduce a geometric framework that organizes the second-order statistics of the Elasser fields into polarization states on generalized Poincar\'e spheres. In this representation, energy, cross-helicity, residual energy, and the phase lag between counter-propagating wave packets emerge as complementary polarization parameters. We derive a Bloch-analogue equation governing the evolution of polarization states and show that distinct polarization geometries are associated with different cascade dynamics. The framework predicts that transitions in the turbulent spectral scaling coincide with changes in the polarization state of the interacting modes.
arXiv:2607.16075v1 Announce Type: new
Abstract: Monitoring changes of process-relevant plasma parameters, such as the electron density and ion flux to the wafer, is essential for the development and control of plasma processes. However, invasive plasma diagnostics, such as probe measurements, typically cannot be applied to commercial reactors. At the example of a low pressure capacitive radio frequency discharge operated in different mixtures of CF$_4$ and Ar, we demonstrate that changes of such plasma parameters can be monitored non-invasively by phase resolved optical emission spectroscopy via optical fibers (Fiber PROES), for which the ports are usually available at industrial plasma sources. In this way, the spatio-temporally resolved dynamics of energetic electrons are tracked by observing a selected emission line. By measuring the electron density and ion flux directly via probe and retarding field energy analyzer diagnostics as a function of driving voltage and pressure, changes of these plasma parameters, including hysteresis effects, are found to be correlated with transitions of the electron power absorption mode revealed by Fiber PROES. Such mode transitions cause the electron energy distribution function (EEDF) to change and, thus, affect such plasma parameters. Based on these findings, Fiber PROES can be used as a non-invasive diagnostic for the monitoring and knowledge-based development of plasma processes.
arXiv:2607.15439v1 Announce Type: new
Abstract: Our previous ARC-AGI-3 agent bundled executable world modeling, scheduled simplification, and exact replay verification, leaving unclear which idea accounted for its performance. We address this attribution question with four nested Codex-based agents: a textual baseline; a flexible-interface executable world model without replay verification; the same executable model with scheduled simplification; and a fixed-interface verification treatment that retains simplification and requires exact reproduction of recorded observations. The main study evaluates all four agents with gpt-5.4 and gpt-5.5 at high and xhigh reasoning effort on the public ARC-AGI-3 games. Exploratory follow-ups evaluate the textual and verification variants with gpt-5.6-sol at xhigh and max. The most robust result is that every agent variant improves with a stronger model and with greater reasoning effort. Within each model-effort setting, differences among variants are smaller than anticipated, while the effects of individual components vary across settings. Requiring a persistent executable deliverable is not universally beneficial: the textual variant outperforms the flexible-interface executable variant in both gpt-5.5 settings. Simplification improves performance in three of the four model-effort settings, with the weakest setting as the only exception. The complete verification treatment ranks first in all four settings, although it uses substantially more resources. In the gpt-5.6-sol follow-up, the verification variant fully solves every public game at both reasoning efforts, achieves about 99% RHAE, and uses fewer than half the total actions of the human baseline. Because the model postdates these games and held-out performance remains untested, this result should be interpreted as saturation of the public set only.
arXiv:2607.15765v1 Announce Type: new
Abstract: Lensless imaging is a technique that recovers object information computationally from diffraction patterns recorded without imaging optics, making its performance strongly dependent on the forward model used during image reconstruction. A common source of reconstruction error is the assumption of fully coherent propagation, even when the illumination exhibits partial spatial coherence in practice. Here, the role of spatial coherence is examined for assorted object classes using a coherence-aware forward model based on generalised van Cittert-Zernike Schell propagation. Simulated measurements are generated over a controlled range of effective source spatial coherence lengths and reconstructed using either a partially coherent forward model or a conventional coherent propagation model. Our results show that decreasing spatial coherence progressively degrades reconstructions under the coherent assumption, whereas incorporating partial coherence can preserve object structure and improve image quality, particularly for dense high-spatial-frequency features. Reconstructions from experimental measurements further confirm the practical need to use coherence-aware inversion under partially spatially coherent illumination. These findings establish spatial coherence as a defining component of the inverse problem in lensless imaging and provide a route to optimization of partially coherent lensless imaging systems.
arXiv:2607.16077v1 Announce Type: new
Abstract: Palm-vein recognition is a highly secure biometric modality due to the uniqueness and subcutaneous nature of vein patterns. However, low contrast in palm-vein images, caused by NIR light scattering and sensor limitations, remains a significant challenge. To address this, we propose the Intensity-Limited Adaptive Contrast Stretching with Bidirectional Gaussian-weighted Overlapping Tiles (ILACS-BGOT) method, an enhancement of the previously developed ILACS with Layered Gaussian-weighted Overlapping Tiles (ILACS-LGOT) technique. ILACS enhances local contrast, while BGOT mitigates blocky artefacts. This study further integrates RootSIFT features with KNN+RT and incorporates the previously introduced Mean and Median Distance (MMD) filter to investigate the parameter variations of both MMD and RT, and their impact on recognition performance. A comprehensive analysis was conducted across three benchmark datasets (CASIA, PolyU, and PUT), using 42 combinations of MMD filter thresholds and RT values. Results were evaluated using EER and Accuracy. Findings reveal that higher template sizes improve performance, while varying MMD thresholds reflect dataset-specific rotational variations. The proposed system demonstrates superior generalisability, achieving significant improvements in both EER and Accuracy over existing methods. Furthermore, the underlying ILACS-BGOT mechanism suggests potential applicability beyond palm vein recognition to other biometric modalities such as finger vein and palmprint recognition, and more generally to low-contrast image enhancement across computer vision applications.
arXiv:2607.15589v1 Announce Type: new
Abstract: Communication-limited robots in mission-critical scenarios such as disaster inspection and search-and-rescue must make reliable onboard decisions without access to remote operators or high-capacity reasoning services. Episodic memory reuse is an attractive low-cost fallback, but retrieval similarity does not guarantee execution validity, i.e., a retrieved action may match the current context yet be unsafe due to changed topology, insufficient battery margin, or unreliable prior outcomes. We call such high-similarity but execution-invalid episodes memory traps. This creates a safety-efficiency design space where similarity only reuse minimizes fallback cost but can be unsafe, while always invoking local reasoning improves safety at high computational and energy cost. This paper presents MemoGuard, a lightweight adaptive runtime that validates episodic memories against topology, resource, and outcome contracts before reuse, invoking fallback only when validation fails. In a graph-based corridor-inspection simulator, MemoGuard reduces battery safety violations by 76.6% over similarity-only top-1 reuse while reducing fallback calls by 21.4% over always reasoning. On an NVIDIA Jetson AGX Xavier with local llama3.2:3b fallback reasoning, this corresponds to 3.67 s and 36.97 J of avoided fallback-reasoning overhead per trial. We open-source MemoGuard at https://github.com/hetheiin/memoguard.
arXiv:2607.15590v1 Announce Type: new
Abstract: This paper presents our solution to the KDD Cup 2026 Tencent UNIREC Challenge. The task requires joint modeling of multi-domain user behavior sequences and non-sequential multi-field features for target-ad pCVR prediction. We develop a Field-Aware RankMixer (FA-RankMixer) with dual-stream bilinear fusion. The model first applies target-aware DIN modules to extract user interests from multiple behavior domains. It also models recent and earlier interests separately for the longest behavior sequence. The model then forms semantic tokens based on feature fields and behavior domains and uses RankMixer blocks for cross-token interaction. A shallow MLP stream complements the deep RankMixer stream, and a group-wise bilinear module fuses their representations. Our final solution ranks ninth on the official leaderboard. Our code is available at https://github.com/PixelCookie-zyf/TAAC-2026-SeRankMixer.