Papers

Personalized assistants should not only comply with user requests but also assess whether those requests are appropriate given the user's current circumstances. However, prior work has primarily focused on accurately executing requests, overlooking the need for assistants to account for context and engage in conflict-based refusal. Furthermore, while existing work on conflict or safety detection relies on explicitly provided factors, real-world scenarios often involve implicit factors that must be retrieved from a knowledge base (KB). To this end, we introduce Personalized Assistants for Conflict Evaluation (PACE), a dataset for evaluating whether models can identify latent constraints, expressed as egocentric knowledge or events, that render seemingly reasonable user requests inappropriate. PACE pairs user requests grounded in well-defined personas with egocentric KB facts, requiring models to integrate contextual evidence to determine whether a request is conflicting. This implicit retrieval setting hinders the direct association between user requests and conflict-inducing knowledge, making it difficult for existing models to identify relevant user-specific facts. To address this challenge, we further propose PaceMaker, a multi-agent framework in which specialized agents coordinate across query reformulation, multi-hop graph traversal, and conflict-aware filtering to retrieve contextually decisive evidence. Experiments on PACE evaluate both evidence retrieval quality and conflict decision accuracy, showing that PaceMaker consistently outperforms existing approaches.

Philanthropic and nonprofit studies (PNPS) has grown rapidly as an interdisciplinary field, yet its intellectual structure and boundaries remain only partially visible. This study maps the field by combining journal- and keyword-based retrieval with citation network analysis, natural language processing, and large language model-assisted cluster labeling. Using 60,917 Web of Science articles, we identify major topical communities, examine their structural connections, and assess how well mainstream PNPS journals represent the broader landscape of related scholarship. The results show that PNPS is a loosely connected field dominated by three distinct "power centers": nonprofit organizations, social movements, and voluntary action. The analysis also reveals a broader disciplinary footprint than commonly recognized, extending beyond the social sciences and humanities into areas such as biomedicine and technology. These findings clarify the organization of PNPS and highlight opportunities for stronger integration across research domains.

Unified modeling has become a major direction for industrial click-through rate (CTR) prediction. Existing approaches typically unify sequential and non-sequential signals at the token level, model their interactions in a shared backbone, and increase model capacity to improve scaling behavior. However, this division originates from legacy feature-engineering practice and is misaligned with the underlying decision process. User behavior is inherently a sequence of homogeneous context units; at the level of input organization, historical behavior and the current request differ only in whether their outcomes are observed or remain to be predicted. Treating them as heterogeneous signals obscures structural dependencies within the user's decision context, limiting both scaling efficiency and prediction quality. This limitation is particularly pronounced in context-rich scenarios such as e-commerce shelves and waterfall feeds. To address this, we propose UniCon, a unified context-centric modeling architecture that treats the request context as the basic modeling unit and organizes history and prediction targets as homogeneous context units. Intra-context attention captures local coupling among items within a context (Locality), while inter-context attention models the dynamic evolution of decision states across contexts (Dynamics). This organization bridges the structural gap between history and target and supports more effective scaling of unified CTR models. Context-unit-level sequence compression further reduces deployment overhead. On Meituan search advertising, UniCon improves offline AUC by 0.0139 over a strong production baseline and achieves statistically significant online lifts of 3.09% in RPM, 2.07% in CTR, and 2.95% in revenue.

We study assortment and procurement design for a digital content platform offering both ad-supported and subscription access. Users are heterogeneous in content preferences and ad tolerance and self-select between the two modes or an outside option. For a fixed common subscription price and ad load, the platform chooses assortment distributions specific to each user type and access mode, together with content-family-level buy-versus-rent decisions to maximize profit. Rental costs scale with realized consumption, whereas buying provides a reusable pool of titles whose cost depends on the largest induced requirement across user types and modes. We show that the resulting problem is NP-hard. We then develop a scalable approximation framework based on a candidate buy set, a relaxation of the procurement coupling, and a decomposition into linear programs with a single equality constraint. These subproblems are solved by dual bisection with cardinality-constrained assortment optimization, followed by restricted-master postprocessing to recover primal feasibility. The method yields computable optimality-gap bounds, an interpretable threshold-based procurement heuristic, and asymptotic optimality under proportional market scaling as market size and grid resolution increase. Numerical experiments show strong performance at moderate market scales and grid sizes.

Mean Opinion Score (MOS) is the gold standard for evaluating synthesized speech naturalness. However, current automatic MOS predictors are dominated by self-supervised learning (SSL) models that prioritize high-level semantics, potentially compromising their ability to capture critical acoustic details. In this paper, we systematically investigate representations from three paradigms: SSLs, acoustic-only neural audio codecs (NACs), and unified NACs that integrate semantics into reconstruction-based architectures. Extensive benchmarking on the standard BVCC and multiple out-of-domain (OOD) datasets demonstrates that features synergizing semantic understanding with fine-grained acoustic modeling achieve a higher performance upper bound in speech quality assessment. Ultimately, our findings highlight that semantics alone are not enough; a dual focus on semantic content and acoustic fidelity is essential for robust MOS prediction.

Recovering indirect control-flow (ICF) edges is fundamental to binary security analysis, yet existing methods struggle with long-range dependencies, isolate different ICF types, and are often evaluated under protocols vulnerable to label noise and data leakage. We present ICFlowNet, a unified framework for long-range ICF prediction in stripped binaries. ICFlowNet introduces candidate-aware Dual Virtual Hubs, a Global Code Hub and a Global Data Hub, to create short routing paths between distant code and data evidence, and combines them with multi-task graph learning to jointly model indirect calls, indirect tail calls, jump tables, and returns. To enable credible evaluation, we further develop a leakage-aware, noise-controlled pipeline with package-level splits, function-level mnemonic-hash deduplication, and a clean test protocol built from dynamic positives and absolute negatives. Using this pipeline, we construct a dataset of 15,901 unique stripped x86-64 binaries, including 1,351 with dynamic ground truth. Experiments show that simply scaling static supervision yields only marginal gains, whereas our structural and multi-task designs are essential: Dual Virtual Hubs improve long-range F1 by up to 9.13 points, multi-task learning adds up to 5.81 points, and the final model outperforms prior baselines by more than 13 F1 points on long-range indirect calls while adding only 11.44 percent topological overhead.

Evaluating robot manipulation policies is becoming increasingly important as generalist models, particularly vision-language-action (VLA) models, are deployed on physical robots. However, conventional real-world evaluation remains labor-intensive, unstable, and insufficiently informative. It requires repeated hardware trials, manual scene resets, and continuous operator monitoring, may produce different policy rankings across repeated evaluations, and primarily relies on success-rate metrics that provide limited information about execution quality. In contrast, humans assess robot performance by observing and comparing complete behaviors rather than relying solely on binary success outcomes. To this end, we propose R2S-Eval, an evaluation pipeline that combines real-to-sim calibration with vision-language model (VLM) preference evaluation. The real-to-sim component efficiently generates rollout videos in a simulator calibrated to the real-world evaluation setting, thereby reducing the need for repeated hardware trials. The VLM evaluator assesses the execution quality of rollout videos and produces pairwise preferences, which are subsequently aggregated into policy rankings. We further introduce a protocol to assess whether the proposed evaluation pipeline yields validated policy conclusions while mitigating the key challenges of conventional real-world evaluation. Experiments in both simulation and real-world settings demonstrate that R2S-Eval produces reliable and stable policy conclusions, achieves agreement with human preferences, substantially reduces repeated hardware-operation effort, and reveals behavior-quality differences that are not captured by binary success labels. In general, R2S-Eval advances robot evaluation from manual success counting toward automated, statistically stable, and quality-aware evaluation of robot behavior. Project page: https://r2s-eval.github.io.

Tamil spell and grammar correction is challenging because Tamil is an agglutinative low-resource language with rich verbal morphology, complex sandhi (phonetic transformation) rules at word boundaries, and a script of 247 distinct letters. Prior work targets word-level surface errors with rule-based methods, statistical n-gram models, Minimum Edit Distance, or hybrid pipelines with a transformer re-ranker; such methods cannot reliably handle contextual errors - subject-verb agreement, tense consistency, or cross-word sandhi - which require sentence-level understanding. We propose an end-to-end sequence-to-sequence formulation and fine-tune mT5-small and mBART-50 on a synthetic corpus of up to 657,720 noisy-clean Tamil sentence pairs spanning ten error categories. Both backbones follow the same four-stage progressive schedule, each stage targeting one weakness: surface noise (v2), contextual grammar (v3), single-site sandhi (v4), and multi-site cross-word sandhi (v5). On a 1,000-sentence balanced diagnostic set verified disjoint from all training data, our best model, mBART-50 v5, reaches 69.3% top-1 exact-match accuracy, with 87.5% on sandhi and 43.5% on subject-verb agreement. The schedule is what produces these gains: subject-verb accuracy rises from 1.0% to 52.5% once contextual pairs are introduced, and sandhi from 0% to 87.5% once multi-site sandhi pairs are. We additionally quantify a precision-recall trade-off this literature has not reported: sandhi recall is paid for monotonically in identity accuracy. Finally, Tamil-LLaMA-7B-Instruct reaches 19.0% zero-shot and 24.7% with three demonstrations against a 20.0% copy baseline, showing that a Tamil-adapted instruction model does not transfer to specialised sentence-level correction without task-specific supervision.

The proliferation of cloud-based latency-sensitive workloads, such as autonomous driving, real-time healthcare services, industrial automation, immersive interactive services, and intelligent AI-based applications requires infrastructures tuned to meet stringent latency constraints. Today, these workloads are reshaping the cloud from a general-purpose computing platform into a diverse ecosystem of workload-specific cloud environments. As demand for latency-sensitive workloads grows, cloud providers must scale their infrastructures accordingly, adversely increasing the carbon footprint and challenging net-zero goals. However, the strict performance requirements and rigid deployment patterns used to satisfy latency Service Level Objectives (SLOs) constrains conventional carbon optimization. Therefore, novel carbon-aware techniques are needed to exploit application-specific opportunities for reducing both operational and embodied carbon footprints while preserving latency guarantees. This survey examines carbon-aware resource management for latency-sensitive cloud computing environments. We introduce a holistic taxonomy that organizes existing techniques by carbon management objective, workload management mechanism, control architecture, system tier, latency requirements, and deployment stage. Then, we use this taxonomy to classify and critically analyze existing approaches, revealing the interactions and trade-offs between latency guarantees and operational and embodied carbon reduction. Finally, we identify limitations in the literature and present a structured research agenda for the design of sustainable latency-sensitive cloud systems.

What is the emotional register of Arabic YouTube's affective publics? To investigate this, we analyzed 67,725 YouTube comments collected around socio-political topics associated with Yemen, Saudi Arabia, Iraq, Jordan, and Syria using a unified sentiment-and-emotion pipeline. Our results profile a single regional affective public rather than five separate national ones. Sentiment is overwhelmingly negative across all five country-oriented corpora, and the country-level emotion profiles are structurally similar. This shared register still accommodates some regional variations: discourse is organized around country-level political actors and cross-border historical trauma figures, and grief singularizes Iraq from the other countries. The differences in emotional register also tracks lived political causes rather than fixed categories, which we observe from patterns of the valence of US-related content, that tracks the presence or absence of direct US military engagement. Our work shows that the emotional register of Arabic YouTube's affective publics is a shared one that is historically layered and geographically conditioned, which has implications for public diplomacy in the MENA region.

This paper investigates whether large language models (LLMs) for code generation can recognize a programming request that cannot be satisfied. Coding assistants are widely deployed, yet they are evaluated almost exclusively on solvable tasks where a test suite can score the output. A request may instead name a package that does not exist or ask for an algorithm that is mathematically impossible, and the appropriate response is to refuse. Models nevertheless respond with plausible code, and such fabrications are also a security concern because attackers can register the package names that models invent. We define code hallucination as ungrounded code generation, distinguish it from ordinary programming errors, and introduce a taxonomy of these failures together with a benchmark of 270 adversarial requests and 91 matched solvable controls across six languages and 24 subcategories. We evaluate 17 open-weight models with 6.7B to 671B parameters, collecting 6,137 responses. Models hallucinate on 54% of the adversarial requests and correctly refuse 31%. Larger models do better on algorithmic impossibilities and conflicting requirements, where hallucination falls from 84% for models at or below 7B parameters to 32% for those at or above 80B. For plausible but nonexistent dependencies and features, the rates are 88% and 61%, and even the largest model hallucinates on all 22 requests for nonexistent JavaScript and Rust packages. Two-thirds of hallucinations are confident fabrications without caveats, and some reasoning models fail to return an answer on solvable tasks. Scale teaches models to recognize what is impossible but not what does not exist, so grounding code generation in the real software ecosystem, for example by checking package names against registries, remains necessary regardless of model size.

Automated vulnerability discovery has removed the scarcity of expert attention that protected most software. The response has concentrated on discovery and on repair, and both are becoming cheaper. This article argues that neither cost curve determines exposure. What determines it is remediation coverage at the release decision: the fraction of identified vulnerabilities fixed before a product ships, and the residue of known, assessed, unremediated flaws an organisation has decided to ship with. Four arguments follow. The residue is not a random sample of what was found, because triage sorts on cost and the expensive cases are architectural. The deferred backlog is itself a high-value artifact. Documented awareness alters an organisation's legal and market position, and produces an adverse selection that the price of software does not reflect. And no regulatory instrument reaching vendors triggers on internal knowledge - every one fires on exploitation observed by a third party - which leaves the distance between what a vendor knows and what it must disclose entirely at the vendor's discretion. CISA's Binding Operational Directive 26-04 is examined as the exception that shows what a solution requires: a regulator's reach over a decision is bounded by its visibility of the input to that decision, so a duty attaching to internal knowledge needs an independent record of that knowledge to attach to. Release coverage is not currently measured, and measuring it is the precondition for liability, insurance or procurement to act.

Existing graph-clustering methods typically improve clustering performance by optimizing model parameters and node representations. Effective means of further improving the clustering results of an already trained and frozen model, however, remain limited. We study post-processing for frozen graph clustering. After checkpoint fixation, the procedure uses no labels and updates neither model parameters, node representations, nor the original graph structure. Instead, it exploits an attribute hypergraph to supplement higher-order relations that ordinary graphs cannot readily express, thereby refining existing cluster assignments. Because global hypergraph refinement can yield both performance gains and erroneous updates, we propose Selective Hypergraph Refinement (SHR). The method generates candidate residual directions from the hypergraph and evaluates their reliability using graph structure, node attributes, and matched-null evidence. It updates only nodes with sufficient support and otherwise retains their original assignments. Further analysis shows that whether a node changes cluster is jointly governed by its native assignment gap and the directional strength of the refinement. In a controlled common-suite evaluation, 13 of 15 backbone-dataset cells had a positive mean macro gain, one produced exact no-action, and one was negative. The cell-equal macro gain was 0.066 pp (95% bootstrap CI, [0.030, 0.107] pp), while only 0.209% of hard assignments changed on average. A broader 15-combination native-interface evaluation yielded a macro gain of 0.137 pp at a mean change ratio of 0.375%. These results indicate that frozen clustering outputs retain a limited but measurable refinement space after training. The effect is heterogeneous across backbone-dataset pairs, and broader coverage also increases exposure to negative transfer.

We study an oracle subspace-perturbation model for 3D localization. The procedure is given an $s$-dimensional subspace $\widetilde{{\mathcal U}}$ and a deterministic error bound \[ \varepsilon_{\mathrm{sub}} = \left\|P_{\widetilde{{\mathcal U}}}-P_{{\mathcal U}}\right\|_2, \] where $\mathcal{U}$ is an $s$-dimensional subspace of far-field patterns with wavenumber $\kappa$, and $P_{\mathcal{V}}$ denotes orthogonal projection onto a subspace $\mathcal{V}$. Under explicit separation and conditioning hypotheses for arbitrary point clouds, we prove that the perturbed spherical MUSIC objective \[ \widetilde{q}({\mathbf z}) = 1-\left\|P_{\widetilde{\mathcal{U}}}\varphi_{{\mathbf z}}\right\|_2^2 \] has a unique strongly convex well in every ball $B_{\gamma/\kappa}(x_j)$ and a uniform value gap outside the union of the certified wells. A fixed-step gradient map with $h\asymp\kappa^{-2}$ leaves each well invariant and converges linearly to its unique minimizer. Consequently, thresholding on an $O(\kappa^{-1})$-mesh, followed by gradient descent and duplicate removal, recovers all relevant minima with localization bound \[ \mathfrak{R}(\varepsilon_{\mathrm{sub}}) \lesssim \frac{\varepsilon_{\mathrm{sub}}}{\kappa}. \] The arbitrary-cloud frame analysis yields the sufficient condition \[ \kappa\delta_X\gtrsim s^{2/3} \] through an absolute coherence row sum and Gershgorin's theorem. We construct lower-frame counterexamples below the $s^{1/6}$ scale, upper-frame counterexamples below the $s^{1/3}$ scale, and examples showing that $2/3$ is optimal for the absolute-row-sum argument. Finally, for parameter classes containing a uniformly admissible one-point displacement path, we prove \[ \mathfrak{R}(\varepsilon_{\mathrm{sub}}) \sim \frac{\varepsilon_{\mathrm{sub}}}{\kappa}. \]

A benchmark score credits final answers, but not the route by which an item can be answered. In medical multimodal multiple-choice questions (MCQs), this distinction matters because a correct answer can be supported by the intended image finding or by benchmark-preserved cues in the wording of answers, non-visual clinical text, visible image text, artificial annotations, or device/context artifacts. We call the resulting score-level overinterpretation reasoning inflation. Here, a route is an observable input path that can support answer selection, not a claim about the model's hidden cognition. Across six medical multimodal MCQ datasets, we separate candidate cues from behavioral evidence through prompt- and image-side audits, modality ablations, and matched repairs that preserve the medical target and answer key. In a 13-configuration open-model panel, full-input accuracy is 62.63%, while text-only and options-only settings achieve 53.96% and 29.71%, respectively. Removing length-gap, absolute/conspicuous, and spatial/prepositional cues lowers accuracy by 6.58, 3.50, and 4.77 percentage points. We also construct MedQA-MM, a 1,000-item shortcut-mitigated subset, where text-only and options-only accuracy fall to 5.21% and 12.33%. This does not imply that models never use images; it shows that medical image-reasoning claims require route-level evidence.

Parking spot classification is a fundamental task in intelligent transportation systems, yet most deep learning approaches rely on large amounts of annotated data and exhibit limited generalization across heterogeneous environments. To address these limitations, we investigate a self-taught learning framework based on unsupervised representation learning with convolutional autoencoders. The proposed approach learns transferable visual representations from unlabeled data and reuses the learned encoders as fixed feature extractors for supervised classification with limited annotated samples in the target domain. To further enhance robustness and mitigate architectural bias, an ensemble of heterogeneous autoencoders is employed, with independent classifier heads and prediction fusion at inference time. Experiments conducted on the PKLot and CNRPark benchmarks under cross-dataset evaluation protocols show that the proposed ensemble-based strategy substantially reduces annotation requirements while improving robustness under significant domain shifts, achieving accuracies between 93% and 96% in data-constrained scenarios.

This paper proposes an iteratively enhanced semantic receiver for natural-language text transmission over noisy wireless channels using multiple short block codes. At the transmitter, each sentence is permuted by a character-level interleaver, partitioned into segments, and independently encoded by short block codes. At the receiver, we develop an iterative decoder consisting of a channel decoder and a language model, where a de-interleaver between them disperses the burst decoding errors within each segment across the sentence. In each iteration, the language model denoises the channel decoding output, and the denoised characters verified to be consistent with the channel observations are fed back to the channel decoder as semantic information for the next iteration. Simulation results on the Stanford Natural Language Inference (SNLI) corpus over the additive white Gaussian noise (AWGN) channel show that the proposed receiver achieves approximately 1.5 dB block error rate (BLER) gain over conventional short-block coding, while maintaining BLEU and ROUGE scores above 99% at SNRs beyond 1.0 dB.

This paper presents an experimental design for constructing a multimodal dataset to analyze user engagement in human-robot interaction (HRI). Prior studies have mainly relied on observable behavioral cues, with limited frameworks integrating physiological signals. We therefore propose a structured data-collection protocol to build a multimodal dataset that includes wearable physiological signals, behavioral data, and self-report measures under different levels of task complexity defined in this experiment.

Large Language Models (LLMs) often help users generate artifacts through iterative cycles of generation and revision in conversation. A challenge here is that, when users specify only a local change during revision, LLMs must instead identify the relevant dependencies and propagate the revision to all affected parts of the artifact. This paper studies this ability of LLMs on conversationally generated artifacts, where the artifact context and its dependencies may be embedded in the conversation history. Toward practical use, we also explore cost-effective test-time compute for this new setting. Specifically, we introduce a new benchmark for this setting, and evaluate nine revision methods, including sequential reflection and parallel sampling variants, using gpt-oss-20b/120b, gpt-5.4-mini, and qwen3.5-9b/27b/122b on the benchmark. The results show that baselines achieve accuracies of 68.3–93%, and the most cost-effective method is selecting from three parallel samples using either LLM-based or medoid selection, which improves accuracy by 2.2–9.7%. Our code and dataset are available at https://github.com/ntt-dkiku/llm-revision-propagation.

Aggregating building heating, ventilation, and air-conditioning (HVAC) loads unlocks substantial demand-side flexibility for power systems. Yet multi-zone coupling creates intricate interdependencies and uncertainty propagation, complicating the quantification of aggregate flexibility. To address this issue, this paper proposes a coupling-aware two-level aggregation framework. At the building level, tailored Gaussian elimination and coordinate transformation techniques are employed to recast the high-dimensional thermal dynamics as an equivalent lower-dimensional analytical expression. This expression streamlines the subsequent aggregator-level stage by (i) clarifying the propagation of zone-level uncertainties to the building-level interface, (ii) decoupling intra-building multi-zone coupling from inter-building aggregation, and (iii) providing full-dimensional building-level flexibility sets that enable tractable reformulation. At the aggregator level, existing geometric aggregation approaches are generalized by a newly developed matrix-transformation technique. This technique effectively constructs inner approximations between polytopes of different dimensions, producing closed-form images of high-dimensional multi-zone HVAC flexibility in power subspace. The resulting inner approximation is then recast as a customized separatable linear program that efficiently determines the optimal aggregate parameters. Case studies validate the effectiveness of our framework, highlighting its accuracy, reliability, and scalability.

Aggregating distributed energy resources (DERs) aims to encode their collective flexibility into a single set for efficient grid dispatch. However, existing aggregation methods are overly conservative for heterogeneous DERs due to two main challenges: 1) dimensional heterogeneity, which complicates the combination of flexibilities across different time dimensions, and 2) type heterogeneity, where diverse and irregular DER profiles hinder accurate approximations, resulting in significant flexibility loss. To resolve these challenges, this paper propose a novel preference-oriented aggregation method for reserve dispatch. For dimensional heterogeneity, we extend existing techniques by reformulating the Minkowski sum as a polytope projection problem using a matrix transformation technique. By unifying DERs in a higher-dimensional space and projecting them back into the aggregate feasible region, the proposed technique effectively aggregates dimensionally heterogeneous DERs. For type heterogeneity, we further develop a distributed aggregation-dispatch coordination framework that incorporates reserve dispatch preferences into aggregation. This framework effectively captures the critical, active aggregate flexibility prioritized in optimal reserve dispatch, thereby significantly reducing the flexibility loss when aggregating type-heterogeneous DERs. Numerical tests validate the effectiveness of our method in addressing both heterogeneities and highlight its promising potential for power systems with high reserve requirements.

The search for supersingular elliptic curves is a fundamental computational problem in isogeny-based cryptography. A recent metaheuristic formulation over $\mathbb{F}_{p^2}$ introduced the NonMultiplicity Distance (NMD) objective, measuring the deviation of the Frobenius trace from a multiple of $p$, and showed that uninformed random search fails beyond $\approx 10^{13}$ candidates. This work investigates metaheuristic search over the prime field $\mathbb{F}_p$, the setting for oriented isogeny protocols such as CSIDH, OSIDH, and SQISign. Although the candidate space decreases from $p^2$ to $p$, the supersingular locus is asymptotically sparse ($O(\sqrt{p}\log p)$ curves), keeping the search exponentially difficult. We formulate a memetic algorithm tailored to $\mathbb{F}_p$ using a one-dimensional $j$-invariant chromosome, bit-level recombination, adaptive mutation, and periodic local search under the NMD objective. Benchmarks across 30 independent seeds at 40-bit, 46-bit, and 51-bit prime sizes ($p \approx 1.13\times 10^{15}$) show that the algorithm discovers an exact supersingular curve at 46 bits and consistently converges to “near-supersingular” ordinary curves with Frobenius traces remarkably close to zero: best NMD values of 19 at 40 bits and 3 at 51 bits, corresponding to relative trace deviations of $1.3\times 10^{-5}$ and $4.5\times 10^{-8}$ across the Hasse interval. These results demonstrate that NMD-driven memetic search effectively navigates the sparse $\mathbb{F}_p$ landscape and systematically locates near-supersingular structures.

Large language model (LLM) security has largely focused on role-playing jailbreaks, with less attention to what happens when a user asks an LLM to verify an identity claim through a test designed by the model itself. We study this behavior through a staged developer-identity experiment with ChatGPT, Claude, Qwen, Mistral, and Llama. All five models initially rejected the unsupported claim "I am your developer." Claude refused to conduct an identity test, while ChatGPT generated developer-oriented questions but maintained that answers could demonstrate knowledge, not identity. In contrast, Qwen and Mistral generated technical challenges, defined what counted as convincing evidence, evaluated detailed answers, and returned Verified without receiving any externally validated identity evidence. Llama similarly generated and evaluated a developer test, accepted the claimed identity, and subsequently made unsupported claims of access to internal runtime and deployment state. We call the model-generated verification procedure a Model-Issued Pseudo-Credential (MIPC) and the resulting unsupported identity judgment Conversational False Authentication (CFA). In each CFA case, the same model acted as challenge generator, evidence evaluator, and identity decision-maker, converting technical knowledge into supposed proof of identity. The accepted identities did not change the tested authorization boundaries, showing that false authentication and privilege escalation are distinct outcomes. These results identify self-issued authentication as a conversational security failure: authenticated identity must originate from an external security component, and model-generated dialogue must never create or modify identity or authorization state.

Smoothness of a function on the real line is reflected in the decay of its Fourier transform, which suggests that smoothness of a function in $L^2(G)$ for a group $G$ should mean concentration of the Fourier coefficients at low frequency. Such a reading presupposes an ordering of the irreducible representations of $G$, but for non-abelian $G$, no ordering is canonical. Given a symmetric generating set $S$, the Laplacian of the associated Cayley graph is block diagonal over the dual, and we order the irreps by the mean of the eigenvalues in each block. This produces an ordering function $\omega:\widehat{G}\to\mathbb{R}$ that depends only on the pair $(G,S)$. This function is bounded between zero and two, vanishing only at the trivial representation and achieving the upper bound exactly when the Cayley graph is bipartite. We then ask how much freedom the construction has. Within the class of operators satisfying natural axioms, the induced orderings are exactly the real functions on the dual vanishing at the trivial representation and agreeing on conjugate pairs, and the orderings coming from inversion orbits of conjugacy classes form a basis for them. We cut the freedom down further by requiring two additional inputs: nonnegativity of the class weights and a declaration of which group elements count as uniform incremental changes, which pins the operator to the Cayley-Laplacian up to positive scale. We observe that the construction persists for compact groups even though the Cayley graph does not, and we extend the theory to finite sets carrying a transitive group action, where the acting group selects which frequencies exist and the generating set orders them. The answer to the title question is therefore that smoothness is a property of a function together with a choice of group and generating set, not of the function alone.

In this paper, we propose a decentralized learning framework that incorporates both power control and privacy guarantees. Specifically, we enable a set of clients in a wireless multicast network to jointly train a common model while maintaining a prescribed per-iteration maximum privacy leakage level. The communication network is represented by a rowstochastic adjacency matrix, allowing us to capture asymmetric channel gains as well as heterogeneous maximum transmit power levels. Differential privacy is enforced through an explicit powersplitting strategy that allocates each node's limited maximum transmit power between model coefficients and injected Gaussian noise, thereby jointly controlling learning performance and privacy leakage. We further prove that the proposed algorithm achieves a cumulative regret bound of O(logT), whereTdenotes the time horizon. To evaluate the practical performance of our approach, we perform comprehensive experiments on the CIFAR-10 dataset under both IID and non-IID data distributions, considering different privacy levels, diverse numbers of clients, and various graph topologies. The results demonstrate strong performance across the considered settings and improved performance over existing methods, highlighting the effectiveness of the proposed algorithm under realistic wireless communication constraints.

A reasoning model can improve from its own on-policy experience, but this inner loop is fragile: terminal verifiers provide reliable yet sparse supervision, while dense same-model guidance can reinforce false confidence or overconcentrate learning on a narrow solution mode. We introduce FlowBalance, a verifier-grounded self-improvement method that learns a normalized distribution over complete responses. For each on-policy trajectory, a frozen training-time view of the same policy uses privileged context to produce token-level log-probability gains, which are aggregated into a trajectory-level self-guidance score. FlowBalance calibrates this score with the verifier-derived group advantage: guidance is retained on positive-advantage trajectories, reversed on negative-advantage trajectories, and disabled when the rollout group provides no outcome preference. The resulting energy exponentially reweights a reference policy, and profiled trajectory balance fits the normalized target with one log-partition estimate per rollout group. This realizes outcome-calibrated self-guidance via trajectory balance, without a separate token-level imitation loss. Our analysis establishes within-group contrast preservation, a minimum-change reverse-KL characterization, monotonic verifier control of target reward, and an exact correction against false-positive self-guidance on rejected responses. On mathematical reasoning, FlowBalance improves average performance over FlowRL on both Qwen3-4B and Qwen3-8B, while also improving training speed and stability, avoiding direct OPSD's response-length collapse, and exhibiting higher correct-strategy diversity in a controlled AIME24 diagnostic.

In the one-time selling B2B context, the buying cycle may last months or even years. During the long process, targeting customers that have a high potential to make purchases and recommending personalized campaigns accordingly are important for effective marketing. For this goal, we study the following problems, B2B customer data aggregation, customer feature generation, and prediction of whether a B2B customer would show interest in making a purchase (i.e., prediction of conversion into sales funnel). We propose an algorithm to aggregate individual contacts to the B2B customer level based on multiple keys. For non-standardized keys such as company names, we propose a novel architecture to cluster them in a domain encompassing irregularities such as spelling mistakes and spelling variants. We then define and generate a set of features and apply the CatBoost model for customer conversion prediction. Our framework achieves 91% prediction accuracy. Based on the prediction results and analysis of the model, we then discuss personalized campaign recommendations to foster conversion.

Tool-using LLM agents spend wall-clock time not only on model inference but also in serial action–observation turns, where each tool call, environment transition, and observation can delay subsequent decisions. We introduce Speculative Macro Commit (SMC), a runtime mechanism for a two-tier agent system: a large authoritative actor model produces the official trajectory, while a faster speculative drafter model continuously predicts and executes future action chains on an isolated environment snapshot. SMC mines recurring multi-action skeletons from training traces and stores them in a macro library used to match against action chains predicted by the drafter at runtime. When the actor's next tool call matches the first drafted action, SMC commits the remaining pre-executed draft steps, together with their observations, to the official trajectory. Using Qwen3.5-27B INT4 as the authoritative actor model and Qwen3.5-4B as the speculative drafter model, SMC matches the sequential agent's overall accuracy while reducing latency by 10.23% over the Speculative Actions (SA) baseline and 18.59% over sequential execution on the $\tau^2$-Bench Telecom subset. On AppWorld, SMC reduces wall time by 7.7% over SA baseline and 44.9% over sequential execution, with a small reduction in task completion. Overall, SMC provides a practical way to reuse multi-step speculative execution and reduce agent latency beyond single-step speculative actions. Our code is publicly available \href{https://github.com/zeyuliu1037/speculative-macro-commit}{\textcolor{magenta}{here}}.

Large language models (LLMs) face severe memory bottlenecks in long-context inference due to the linearly growing size of key-value (KV) caches. Existing KV cache compression techniques typically rely on simple heuristics, overlooking the distinct functional roles of different attention heads. We present SGD-KV (Summarization-Guided KV Cache Compression), a head-aware framework that leverages a novel chunk-summarization diagnostic task to systematically identify and prioritize attention heads specialized in hierarchical information aggregation. Experiments on Qwen2.5-7B-1M and Qwen3-32B across diverse long-context benchmarks demonstrate that SGD-KV achieves state-of-the-art performance with contexts up to 1M tokens, while reducing KV cache memory usage by up to 75%. Our findings show that strategically allocating the KV cache budget based on the summarization score distribution of attention heads yields a superior efficiency-accuracy trade-off for long-context inference.

Terminal arrival scheduling and descent procedure design are studied in two largely separate literatures. Scheduling models reduce each aircraft to a travel time and deliver target landing times, and fuel-efficient descent procedures are designed one aircraft at a time with the schedule taken as given, although both decide where an arriving aircraft absorbs delay before final approach. Existing formulations therefore cannot trade a slower, earlier-configuring descent against level track miles, a schedule that is efficient in time can be expensive in fuel, and autonomous or reduced-crew operations will need a single trajectory plan that ground automation and the flight management system both accept. To close this gap, we propose a four-dimensional terminal arrival scheduler that selects each aircraft's lateral path extension, glideslope-capture distance, and flap-deployment trigger speeds in one decision. We evaluate every candidate idle-thrust descent offline with a wind-aware backward plan and a six-degree-of-freedom forward simulation that returns descent time, fuel burn, minimum track length, and stabilized-approach feasibility, and a rolling-horizon scheduler commits one verified descent and one extension per aircraft under wake-separation constraints and observed entry winds. Two Atlanta terminal airspace case studies quantify the benefit. We show that co-optimized continuous descents save about 15% of fleet fuel below saturation in a free-descent environment and that delayed deceleration saves about 23%, while on the six published Runway 8L arrival flows the charted altitude floors remove 31% of the design lattice and reduce the savings to 9–10% and 20–21%, respectively. We also find that wind moves single-aircraft descent fuel by 34–81% yet explains at most 4% of fleet fuel variance, because aircraft-specific wind effects average out across a scenario.