# Research Papers — 2026-09 · bundle page 104 of 266 > Full-text markdown of papers #5151–5200 of 13298 published in 2026-09, newest first. AI-readable mirrors of the full 13298-paper 2026-09 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-09/page/104 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/103.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/105.md) ### The Computational Value of Sensory-Aligned Receptive Fields Depends on Neuronal Expressivity Authors: Agnese Adorante, Aaron Spieler, Anna Levina Published: 2026-09-22 Categories: cs.NE, cs.LG, q-bio.NC arXiv: https://arxiv.org/abs/2609.26940v1 Biological sensory neurons have selective receptive fields organized along meaningful stimulus coordinates, such as frequency, motion direction, or retinotopic position. Such structure may arise from efficient coding and biological constraints on activity, connectivity, and wiring, as computational studies of simple neurons have shown across modalities. This raises a question: do structured receptive fields confer a computational advantage beyond resource efficiency itself, and does this advantage persist when individual neurons are highly expressive? We address this question in recurrent networks of Expressive Leaky Memory neurons, where we can independently vary neuronal complexity and the organization of feed-forward receptive fields. Across auditory and event-based visual classification tasks, receptive fields aligned with a task-relevant sensory coordinate improve test accuracy relative to budget-matched random receptive fields. This advantage disappears when sensory coordinates are scrambled, or when receptive fields follow task-irrelevant coordinates, showing that the benefit comes from alignment with task geometry rather than restricted connectivity alone. Increasing neuronal complexity reduces the performance advantage of structured receptive fields. Finally, generic synaptic sparsity regularization induces input selectivity and partially recovers performance, but remains substantially below explicitly structured receptive fields, suggesting that sparsity alone is insufficient to recover the full computational benefit of task-aligned receptive fields. Together, our results show that appropriate receptive fields can serve as a computational prior beyond sparsity itself, and that their value depends on the computational expressivity of individual neurons. --- ### Which Objectives Need a Dial? Predicting Objective Conflict and Covering Trade-offs in Steerable Pluralistic Alignment Authors: David Tsoi, Esra Dönmez Published: 2026-09-22 Categories: cs.AI, cs.CL, cs.CY arXiv: https://arxiv.org/abs/2609.26929v1 People hold diverse, sometimes conflicting values, so no single aligned model can satisfy everyone. Pluralistic alignment therefore calls for steerable models that can balance competing objectives differently. Multi-Objective Direct Preference Optimization (MODPO) does this by using an objective weight to span a continuum of trade-offs. We study two questions: when can one model improve two objectives simultaneously, and how can many trade-offs be covered without training a separate model for each? Across seven objective pairs from HelpSteer and UltraFeedback, two pre-training measurements predict whether objectives align or conflict for human-annotated data, but not for AI-annotated data, where response length and repetition confound reward-model scores. For broader trade-off coverage, selecting the nearest trained model and merging model parameters both help, but neither consistently matches direct training. These findings yield practical guidance for building steerable models that serve diverse preferences. --- ### Building Socio-Affective Artificial Intelligence for Interactive Multi-Agent Simulations Authors: David Berga Published: 2026-09-22 Categories: cs.AI, cs.CY, cs.GT, cs.HC, cs.MA arXiv: https://arxiv.org/abs/2609.26927v1 The objective of this article is to provide design principles and a software architecture for enabling interaction between humans and multiple agents in simulated dynamic worlds. This connects the current era of general artificial intelligence (AI/AGI) with the proliferation of transformer-based conversational agents and the increased computational capabilities. Given an overview of current and previous multi-agent theories of mind (socially and affectively-aware agents), the existence of an integrative design of agent interactions with themselves and with humans must be crucial for understanding how to create sustainable and governance in future human-agent reasoning systems. In this work is presented a software "AGIMUD" that integrates: A. socially-aware reasoning and emotion in agent behavior and interaction, B. a design of human multimodal scheme for human users, artificial agents and simulated worlds, and C. distributing the AI processing through the network to enable multiple autonomous agents. These integrations allow the dynamic world recreation as multi-user dungeons (MUDs) where both agents and humans can interact simultaneously in real time. Find the code online in https://github.com/dberga/AGIMUD. --- ### Experts Rise Where LLMs Disagree: Using Cross-Model Disagreement to Target Expert Effort in LLM Codebook Revision for Large-Scale Annotation Authors: Zeyu He, Zhuqian Zhou, Kirk Vanacore, Rene F. Kizilcec, Ting-Hao 'Kenneth' Huang Published: 2026-09-22 Categories: cs.CL, cs.AI, cs.HC, cs.LG arXiv: https://arxiv.org/abs/2609.26926v1 Large-scale text annotation brings expert insight to millions of documents, often through a codebook that AI annotators follow. Developing a robust codebook, however, takes months. Large language models (LLMs) could speed this process by applying an early codebook to the data, surfacing cases with strong LLM disagreement, and eliciting expert feedback to address them. We examined three ways experts can provide feedback for LLM codebook revision: (i) editing LLM-generated revisions driven by cross-LLM disagreement (Codebook Verifying), (ii) answering questions about LLM disagreements (Question Answering), and (iii) labeling disagreement cases with rationales (Rationale Labeling). Experiments on thousands of tutoring-session transcripts show that Rationale Labeling yielded the highest LLM-labeling accuracy (64.9%) against expert labels, outperforming the expert-revised codebook (57.8%). The best Question Answering setting also outperformed it (60.5%). Our work shows that LLMs can be used to strategically target expert attention, shortening months of codebook revision to days without sacrificing labeling performance. --- ### nnFoundation: 3D Foundation Models for Radiology Authors: Constantin Ulrich Harsy, Tassilo Wald, Karol Gotkowski, Yannick Kirchhoff, Marcel Knopp, Maximilian Rokuss, Elisa Stegmeier, Philipp Schader, Dasha Trofimova, Raphael Stock, Kim-Celine Kahl, Stephen Schaumann et al. Published: 2026-09-22 Categories: cs.CV, eess.IV arXiv: https://arxiv.org/abs/2609.26924v1 Radiological artificial intelligence has advanced rapidly, yet most systems remain narrowly task-specific, data-intensive, and fragile under domain shift. Foundation models promise more transferable and data-efficient solutions, but existing approaches are limited in scale, evaluated narrowly, and often assume that a single pretrained model can support diverse downstream tasks. Here we present nnFoundation, complementary convolutional and transformer-based 3D radiological foundation models. Developed within the Human Radiome Project (THRP), nnFoundation is trained on 2.1 million CT, MRI, and PET image volumes from 125 institutional and public datasets. We evaluate them across 108 tasks spanning segmentation, detection, classification, report generation, and image retrieval, including evaluations under domain shift, by external partners and in low-data and low-compute regimes. Across all task types, our convolution- and transformer-based nnFoundation models consistently outperform both prior 3D foundation models and training from scratch, establishing state-of-the-art performance for radiological imaging. However, performance follows a consistent task-dependent structure: the convolutional nnFoundation model dominates spatially localized tasks, whereas the transformer-based nnFoundation model excels in tasks requiring global semantic reasoning and in frozen-feature settings. Dynamically aligning the foundation model topology with the dataset characteristics post-hoc further improves transfer across heterogeneous 3D settings. These results show that transferable 3D radiological performance is governed not by a single universal model, but by the interplay of scalable pretraining, complementary architectures, and dataset-aware adaptation. We release nnFoundation models integrated into nnU-Net and nnDetection, enabling immediate application across established radiology workflows. --- ### A 3D Pose-Based Ensemble Framework for Cricket Shot Classification and Automated Biomechanical Analysis Authors: Sourav Shome, M. D. Ashiquzzaman Rahad, Rameswar Debnath Published: 2026-09-22 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.26923v1 Cricket is one of the most celebrated sports world-wide, and technological advancement has become deeply embedded in how the modern game is analyzed and coached. Cricket shot classification and automated performance analysis add a further dimension to this trend. Traditional approaches rely on RGB video features or static images, which are sensitive to environmental variations such as camera angle, lighting, and background clutter, and often fail to capture the underlying biomechanics of batting actions. In this paper, we propose a system to improve cricket coaching that takes raw video data, extracts batsmen from video frames using YOLO, and extracts 3D pose data from video frames using MeTRAbs. The system produces sequential skeletal pose data of 30 body points and captures the biomechanical features of a batsman. As part of the system, we also propose a deep learning ensemble for shot classification of four shots: flick, pull, defense, and drive. The ensemble performed well, compared to existing classification works, achieving 97.68% accuracy. In addition, we analyzed the misclassification rates to identify cases where shots were incorrectly classified and examined their possible causes. Our proposed system allows novice players to obtain useful feedback, such as important joint angles relative to expert batsmen, which can also be useful for injury prevention. The shot classifier also helps track class-wise shots over time for further analysis. In addition to novice players, coaches can use the system for player evaluation. --- ### Cross-Modal Contrastive Learning from Histopathology and CT for Automated Renal Cell Carcinoma Grading Authors: Amit Das, Tanmay Shukla, Naofumi Tomita, Faraz Farhadi, Jessica Sin, Ari Hakimi, Chad Vanderbilt, Jie-Fu Chen, Ritesh Kotecha, Weijie Ma, Bing Ren, Saeed Hassanpour Published: 2026-09-22 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.26920v1 Background: Clear cell renal cell carcinoma (ccRCC) exhibits substantial clinical heterogeneity, and accurate grade assessment is essential for risk stratification and treatment planning. However, conventional grading requires invasive tissue sampling. We developed RCC-Align, a cross-modal contrastive learning framework that leverages paired histopathology and computed tomography (CT) data during training to improve noninvasive CT-based ccRCC grade prediction. Methods: RCC-Align aligns paired whole-slide histopathology images (WSIs) and CT scans through contrastive cross-modal objectives, transferring grade-discriminative information from microscopic tissue morphology to macroscopic radiologic representations. The framework was trained and evaluated on paired TCGA and CPTAC cohorts using patient-level five-fold cross-validation. Performance for low- versus high-grade ccRCC classification was compared against CT-only baselines (DINOv2-Base and DINOv2-Finetuned) and a WSI-based reference model (GigaPath-Finetuned). Cross-modal alignment was assessed using cosine similarity analysis. Results: RCC-Align achieved an AUC of 0.601 (95% CI, 0.524-0.673) and AUPRC of 0.599 (95% CI, 0.541-0.676), outperforming DINOv2-Finetuned (AUC 0.545; AUPRC 0.543) with significantly improved low-grade prediction (p = 0.004). RCC-Align also demonstrated stronger paired WSI-CT embedding alignment compared with baselines. The WSI-based GigaPath reference achieved an AUC of 0.719. Conclusion: Pathology-guided contrastive learning improves CT-based ccRCC grading while requiring only CT at inference. This approach may complement tissue diagnosis when biopsy is unsafe, infeasible, or limited by intratumoral heterogeneity. Validation in larger, multi-institutional cohorts with external testing is needed before clinical translation. --- ### Bend the Clock: Predicting Ahead to Beat Latency in Event-Based Object Detection Authors: Biswadeep Sen, Benoit R. Cottereau, Nicolas Cuperlier, Terence Sim Published: 2026-09-22 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2609.26919v1 Event cameras promise low-latency perception for high-speed robotic systems, where even short delays can render detections stale by the time they inform downstream robotic decisions. Yet modern event detectors still require tens of milliseconds of computation before their predictions become available. Conventional evaluation ignores this delay by comparing predictions with annotations at the observation timestamp, even though the scene may have changed by the time those predictions are produced. We study this observation-availability mismatch in event-based multi-object detection and show that state-of-the-art event detectors degrade substantially when evaluated at prediction availability rather than observation time. To address this, we introduce ChronoFuse, a causal availability-time detector that predicts object states for when its output becomes available rather than for when its input was observed. ChronoFuse performs causal cross-time fusion over a multi-scale feature hierarchy, combining current representations with cached temporal features to expose short-term temporal cues without using future observations. The fusion pathway is lightweight, adding only 0.17 million parameters and 0.84 ms of mean end-to-end latency overhead. ChronoFuse recovers 71% of the accuracy lost to latency on 1Mpx driving data and 90.8% under rapid drone motion on FRED, nearly restoring zero-delay performance. Under the extreme motion of EV-Flying, ChronoFuse reaches 20.95 sAP, compared with 2.25 for the strongest standard event detector (9.3x gain). These results show that predicting ahead can be critical for robots operating in fast-changing scenes, including autonomous driving, agile flight, and robotic interception. --- ### On Preference Coverage Collapse from Hindsight Relabeling in Multi-Objective Reinforcement Learning Authors: Baptiste Bonin, Caro Strickland, Audrey Durand Published: 2026-09-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.26918v1 Hindsight relabeling which retroactively replacing a transition's goal with the outcome the agent actually achieved is an effective tool for improving sample-efficiency in Reinforcement Learning (RL). A natural extension to preference-conditioned multi-objective RL (MORL) relabels transitions with the preference direction the agent achieved rather than the one asked for. We show that this extension is frequently harmful: across four preference-conditioned off-policy algorithms spanning two critic backbones and two preference-sampling schemes on the continuous-control MO-Gymnasium suite, it degrades 19 of 36 algorithm-environment settings by as much as four standard deviations, improves only one, and leaves the rest unaffected. The harm is not a symptom of noisy relabels; denoising the target recovers almost nothing, and neither prioritized sampling nor any buffer-structural choice reproduces it. Instead, repeated relabeling collapses the critic's coverage onto whatever narrow region of the preference space the agent happened to visit. We name this failure mode \emph{Preference Coverage Collapse}, and quantify it with abandoned preference mass (APM), a value-aware statistic that tracks the harm ($ρ= -0.73$) where a purely structural coverage count does not. We then introduce \texttt{her\_mix}, a single-parameter convex combination pulling the achieved direction back towards the requested preference. At one fixed value across every algorithm and environment, it returns 16 of the 19 harmed settings to baseline, preserves and even improves the one setting in which relabeling helps, and cuts abandoned preference mass from $69\%$ to $6\%$. Protecting coverage over the preference simplex, not filtering noisy relabels, is what makes hindsight relabeling safe for MORL. --- ### COMED: The Missing Middle Between Routing and Collaboration in Multi-LLM Inference Authors: Norah Alballa, Wenxuan Zhang, Salma Kharrat, Fares Fourati, Zafar Ayyub Qazi, Mohamed Elhoseiny, Marco Canini Published: 2026-09-22 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.26913v1 No single Large Language Model (LLM) is uniformly reliable across queries, motivating multi-model inference systems that either route among models or combine their outputs. However, routing stops after selecting an initial model, while dense collaboration invokes peers on every query. We show that collaboration is non-monotonic: peers can recover failures that no model solves alone, but can also corrupt initially correct answers. We introduce COMED (Controlled Model Escalation for Multi-LLM Deliberation), a post-anchor controller for selective cross-model collaboration. COMED uses anchor self-consistency, router margin, and a lightweight peer probe to accept confident answers, verify ambiguous cases, and escalate only when collaboration is likely beneficial. We formalize this trade-off with a rescue-harm decomposition showing that selective collaboration improves when rescued errors outweigh collaboration-induced harms. Across medical, scientific, and general reasoning benchmarks, COMED improves fixed and routed anchors in all 16 open-weight settings, with gains up to +10.7 percentage points on MedQA while invoking fewer models and using fewer decoded tokens than dense collaboration. On HLE with frontier models, COMED improves GPT-5.5 from 23.1% to 28.1%, outperforming dense collaboration and achieving the best results. --- ### TwinCheck: Evidence-Grounded Negative-Twin Verification for Stateful Tool Agents Authors: Jiaxuan Dai, Tianyi Huang Published: 2026-09-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.26911v1 A single locally plausible tool call can derail an otherwise successful agent trajectory. Suspicion alone does not justify intervention, because the replacement itself can introduce the very failure verification is meant to prevent. We introduce TwinCheck, an inference-time verification policy that considers replacement only when the trace satisfies an evidence condition tied to a trace-local failure hypothesis. It constructs a trace-grounded counterfactual alternative, a negative twin, and replaces the agent's proposal only if the twin passes structural checks and the pairwise verifier prefers it in both candidate orders. For paired evaluation, exact replay holds the agent's parsed responses and actions fixed until the first accepted replacement, separating intervention effects from resampling. In the primary analysis of 159 multi-turn BFCL V4 tasks with complete exact-replay pairs, the complete policy raises task success for GPT-5.6 Sol from 45.3% to 58.5% (95% task-bootstrap CI (8.2, 18.8)), with no observed success-to-failure regressions. Together, these findings recast execution-boundary repair as a constrained comparison, making the counterfactual action itself the object of verification. --- ### Small Cues, Big Consequences: Learning Pivotal Cues for Multimodal Meme Classification Authors: Akshit Sharma, Prashant W. Patil Published: 2026-09-22 Categories: cs.MM, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.26907v1 Memes often derive their harmful, hateful, or sarcastic meaning from small but decisive visual, textual, or cross-modal cues. Existing multimodal classifiers can miss such evidence when relying mainly on global image-text representations. We introduce MemeCF, a cue-focused benchmark of 9,895 memes across harm, hate, and sarcasm, with annotations identifying the modality and rationale of the pivotal evidence. We also propose MemePIVOT, a local-global architecture for meme classification. MemePIVOT uses frozen CLIP features, unbalanced optimal transport to align words with image patches while allowing irrelevant evidence to remain unmatched, and an evidential fusion head to combine local grounding with global meme context under uncertainty. Experiments on HarMeme, PrideMM, and MemeCF show consistent gains over strong text-only, image-only, multimodal, and vision-language baselines. Cross-dataset and ablation results further show that explicit pivotal-evidence modeling improves robustness and contributes meaningfully beyond global multimodal representations. Our code and dataset are publicly available at https://github.com/AkshitSharma1/MemePIVOT --- ### CORE-STACK+: Meta-Learning for Deep Stacked Generalization Authors: Noor Islam S. Mohammad Published: 2026-09-22 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2609.26905v1 Stacking heterogeneous vision backbones (CNNs, ViTs, and hybrids) is the de facto recipe for accuracy, calibration, and robustness, yet two coupled pathologies limit its returns. Prediction-space multicollinearity ill-conditions the meta-learner's Gram matrix, inflating weight variance and producing brittle solutions on a thin manifold. Calibration collapse compounds constituent miscalibration through naive linear stacking, so adding more models can hurt expected calibration error (ECE). Existing remedies, ridge regularization, greedy selection, model soups, and SWAG address at most one of these issues, and none jointly target conditioning and calibration in heterogeneous prediction pools. We introduce CORE-STACK+, a preconditioning pipeline with four components: (i) a kernelized redundancy filter that removes non-linear inter-model dependencies invisible to Pearson correlation, using Centered Kernel Alignment (CKA) (23); (ii) a $<15$K-parameter differentiable meta-feature gate that learns per-sample attention over ensemble statistics; (iii) a spectrum-adaptive Ridge penalty $lambda^{star}=lmax(Chat)/SNR(Chat)$ derived from a Marchenko-Pastur signal-noise decomposition, eliminating nested cross-validation; and (iv) a Laplace-approximate Bayesian blender replacing inverse-RMSE heuristics. We prove a PAC-Bayes excess-risk bound that, for the first time, jointly accounts for prediction-space redundancy and meta-learner capacity. Across six benchmarks, CORE-STACK+ delivers $+1.8\%$ top-1 on ImageNet-1K, $-4.2$ mCE on ImageNet-C, $+0.9$ mIoU on ADE20K, and $+1.3$ AP on COCO, while reducing retained models by 35-57% and inference FLOPs by up to $41%$. ECE improves $2.1\times$ over deep ensembles without post hoc temperature scaling. --- ### Ajar: Measuring Open Privilege in Agent Defenses Authors: Reshabh K Sharma, Linxi Jiang, Shuo Chen, Zhiqiang Lin Published: 2026-09-22 Categories: cs.CR, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2609.26900v1 A language model agent acts through the tools it is given. The data it reads while working on a task can redirect what it does with those tools. A growing set of techniques for safe and secure agent execution therefore sits between the agent and its tools, aiming to enforce access control, information flow or isolation at that boundary. Today these techniques are evaluated on agent-security benchmarks built around indirect prompt injection. Those benchmarks judge a defense by how far it brings the number of successful attacks down while preserving the agent's utility. A defense is judged only on the agent's execution. It can score well on both metrics while holding open a transfer, a deletion or a broad read that no task needed. Ajar measures that open privilege directly using the existing benchmarks. It attaches to an agent-security benchmark that already exists and reuses the tasks, tool schemas, reference solutions and goal states that benchmark uses to grade its own runs. For each benign task it builds candidate tool calls the task does not need, so allowing one is privilege left open. These calls are presented to the defense at every point where the agent could act. We evaluate Ajar by attaching it to AgentDojo, where open privilege becomes a third axis beside the existing attack success and benign utility. We run it on five defenses: Progent, CaMeL, AC4A, Permission Assistant, and Claude Code's Auto mode. We observed that they leave widely different amounts of privilege open. Two defenses leak by almost the same amount yet differ widely in the benign tasks they finish, and one defense buys part of its tightness by refusing calls its tasks were entitled to make. This open privilege cannot be derived from the measured attack success or benign utility. The source code of Ajar is available at https://github.com/reSHARMA/Ajar. --- ### Harness as a Language: A Minimalist Agent Framework With Maximal Expressivity Authors: Zhening Li, Joshua Liu, Mateja Vukelic, Nicole Shen, Supriya Lall, Amitayush Thakur, Alex Zhang, Omar Khattab, Jonathan Light, Armando Solar-Lezama Published: 2026-09-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.26891v1 Modern language-model agents are built around the \textit{agent loop}, where the LLM is placed in an environment exposing a set of tools, and the LLM has full control over the workflow by alternating between tool calls and observing their output. However, certain workflows currently require additional engineering beyond the agent loop itself, such as memory systems and self-improving systems. We built an LLM agent framework, JAZ, to explore the extent to which a minimal harness that is little more than the agent loop itself can accomplish tasks these specialized systems are built for. JAZ exposes a single LLM-based primitive invoke and provides a set of built-in hooks that allow the programmer to apply constraints and monitoring. Generalizing existing code-mode agent loops, \texttt{invoke} is the simplest loop that satisfies two defining properties: (1) the LLM can write arbitrary executable code that can include recursive \texttt{invoke}; (2) everything visible to the LLM --- all inputs to \texttt{invoke} as well as its interaction history with the code environment --- are variables in the code environment. We motivate our design from first principles, viewing \texttt{invoke} as a language primitive representing a function whose implementation is provided at runtime by an LLM every time it is called. To validate the design of our core \texttt{invoke} primitive, we evaluate \texttt{invoke} --- with only prompting, no manually designed tools, harness, or external systems (e.g., memory or the file system) --- on workflows traditionally implemented through specialized external harnesses. On long-horizon workflows requiring recall beyond the context window, JAZ invoke outperforms Letta (MemGPT) by 8\% at half its cost on the recall-heavy portion of StuLife. On continual self-improvement, JAZ invoke outperforms ACE by 4\% at a lower cost on AppWorld. --- ### PR-Smoother: Simulator-Preserving Non-Gaussian Smoothing for Data Assimilation Authors: Yuta Tarumi Published: 2026-09-22 Categories: cs.LG, nlin.CD, physics.ao-ph arXiv: https://arxiv.org/abs/2609.26890v2 Many physical data assimilation (DA) workflows require smoothing methods that represent non-Gaussian posteriors over physical state variables, scale to high-dimensional simulators, train from observation windows alone, and remain compatible with calibration of the prescribed simulator. We introduce PR-Smoother, a simulator-preserving amortized smoother designed for this prescribed-simulator DA regime. Its key design principle is to keep the prescribed simulator explicit in both the evidence lower bound and the variational family: rather than learning replacement dynamics or a learned trajectory prior, PR-Smoother learns only future-conditioned corrections around the prescribed rollout. This yields an explicit non-Gaussian smoothing distribution over physical trajectories and supports joint state, parameter, and sensor-bias learning from observations alone. The variational family contains the exact smoother in deterministic and linear-Gaussian limits. Empirically, PR-Smoother captures multimodal posteriors in 4-dimensional Lorenz-96, remains accurate under ambiguous nonlinear observations and process noise in 40-dimensional Lorenz-96, and scales to joint state-parameter-bias inference in 16,384-dimensional Kolmogorov flow. --- ### Additive Nonparametric Regression with Spatial and Network Objects Authors: Rajarshi Guhaniyogi, Pritam Dey, Krishnendu Chandra, Aaron Scheffler, Bani K. Mallick Published: 2026-09-22 Categories: stat.ME, stat.AP, stat.ML arXiv: https://arxiv.org/abs/2609.26867v1 This article is motivated by an imaging application from the Adolescent Brain Cognitive Development (ABCD) study, aiming to predict task-based brain activation maps from t-fMRI using cortical metrics from structural MRI (s-MRI) and brain connectivity data from resting-state fMRI (rs-fMRI). Hierarchical Bayesian modeling is well-suited for integrating diverse imaging data and quantifying prediction uncertainty. However, progress in this field is limited due to challenges in designing joint priors that capture the structures and interconnections between different imaging modalities, along with computational complexity and lack of theoretical assurances. To address these challenges, the article introduces a novel regression framework that treats t-fMRI and s-MRI images as functional data, incorporating additive non-linear effects of both network and functional predictors on the functional response. Specifically, we employ Gaussian process (GP) priors on coefficients related to the functional predictors to capture their intricate functional dependencies with the response. Furthermore, a GP prior is assigned to encapsulate the non-linear nodal effects of the network predictor on the response function. The method is supported by theoretical results on predictive accuracy for the functional response, and is empirically validated through simulation studies and analysis of multi-modal neuroimaging data from the ABCD study. --- ### Marginally Correct Tool Caches Can Reverse Group-Normalized Policy Updates Authors: Shivam Gupta Published: 2026-09-22 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.26866v1 Tool-result caching reduces repeated execution in agent training, but also couples rollout randomness. We study a two-action model in which independent and shared execution preserve every rollout's conditional reward distribution. Despite this marginal agreement, sharing one stochastic result per group can reverse the expected group-normalized policy update. We derive an exact finite-group expression: against a constant alternative, the shared update follows the probability of winning minus the probability of losing, rather than the difference in expected reward. A Bernoulli specialization yields a wrong-direction region and a non-vanishing update-variance floor as group size grows. Centering without group standard-deviation scaling preserves the expected-return direction in this model, using an existing estimator control. Exhaustive finite sums verify 540 configurations and 3,240 estimator evaluations, with a separate ordered-sequence checker. An implementation audit reproduces the sharing path in a pinned, unmodified TVCache stack using 256 scripted rollouts. These results do not measure language-model training performance or refute TVCache's deterministic-output contract. They establish that marginal output validity alone cannot certify a stochastic cache as training-equivalent. --- ### Safety Nudges: User-Facing Interventions for Real-Time AI Risk Awareness Authors: Varshini Elangovan, James Wedgwood, Chhavi Yadav, William Agnew, Sauvik Das, Virginia Smith Published: 2026-09-22 Categories: cs.HC, cs.AI, cs.CY, cs.LG arXiv: https://arxiv.org/abs/2609.26865v2 Conversational AI systems can pose safety risks to their users such as hallucination, sycophancy, overconfidence, and anthropomorphism, but these risks are difficult for users to detect during everyday use. We introduce Safety Nudges, a browser-based tool that provides lightweight, in situ flags when concerning behavior is detected in chatbot conversations. We evaluated Safety Nudges in a two-week field study with 45 frequent chatbot users, collecting interaction logs, surveys, and feedback on individual nudges. Participants found the tool useful, clear, and minimally disruptive, with nearly all users reporting an increased awareness of potential AI harms, though we found that this improved awareness alone did not necessarily lead to discernible behavioral changes. Our results suggest that user facing safety nudges can complement model-level safeguards by helping people critically evaluate AI responses in context, while highlighting the importance of relevance, calibration, and user control in nudge design for conversational AI safety. The code for our Safety Nudges extension is publicly available at https://github.com/jtbwedgwood/safety-nudges. --- ### Comparative Evaluation of Static Embedding Models for HTTP Request Anomaly Detection Authors: Amanda Riverol, Gustavo Betarte, Rodrigo Martínez, Álvaro Pardo Published: 2026-09-22 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2609.26860v1 Web applications are increasingly targeted by cyberattacks that exploit HTTP requests to evade security mechanisms. Traditional web application firewalls (WAFs) rely on rule-based approaches that often exhibit high false positive rates and limited adaptability. Recent studies have explored machine learning techniques and word embedding models to improve anomaly detection in HTTP traffic. This paper presents a benchmark for static embedding models, specifically Word2Vec, FastText, and Doc2Vec, within a unified, single-class classification framework. We propose HEDA (HTTP Embedding-Based Detection Architecture), a modular detection pipeline that combines static embedding representations with single-class anomaly detection models to detect anomalies at the request level. The approach operates in an unsupervised environment, where both the embedding models and detectors are trained exclusively on benign HTTP traffic. The proposed methodology is evaluated on three datasets with heterogeneous characteristics, including both synthetic and real traffic. The experimental results show that the choice of embedding representation significantly affects detection performance, and that FastText-based embeds produce the most consistent results across all datasets, achieving high detection rates while keeping false positive rates under control. --- ### FLINT: Fast Lightweight Inference for Traversability Authors: William Bonilla, Maxime Boisvert, David-Alexandre Poissant, David Meger, Louis Petit Published: 2026-09-22 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2609.26857v1 Navigation in off-road conditions is challenging due to the lack of structure. There is no fixed vocabulary for what is traversable. The traversability depends on both the environment and the embodiment's dynamics. Neither of these two variables can be hand-labeled at scale. Thus, traversability has to be learned by the embodiment's own experience. Modern platforms tend to use multiple sensors to estimate traversability and navigate: RGBD cameras, lidar, radar, IMU, with computationally intensive platforms to run inference on neural networks. Against this trend, we propose FLINT, a lightweight traversability estimator: a 21.6M-parameter backbone, 38\times smaller than a comparable foundation-model backbone, that scores higher on held-out terrain probes and runs at 14.7 FPS on CPU alone using a RGB camera has the only sensor. Despite that gap in scale, FLINT produces a cheaper, more accurate costmap than a deployed foundation-model system (WildOS) on 23 of 24 replayed field logs. We compare different self-supervised learning signals and deploy the resulting models on a real platform in closed-loop field trials: the best self-supervised head reaches 99% autonomy over the route, outperforming a human-label-trained baseline deployed live on the same course. Our results show that heavy sensing and computing are not necessary for traversability estimation. --- ### QUARTET: Quad-branch cross-Attention and Random-walk Traces for Enhancing Transformers on Relational Graphs Authors: Kyaw Hpone Myint, Nan Jiang, Xiang Li, Zhe Wu, Alexandre G. R. Day, Pranab Mohanty, Giri Iyengar Published: 2026-09-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.26855v2 Relational Deep Learning (RDL) models multi-table databases as heterogeneous temporal graphs, and graph transformers currently achieve state-of-the-art performance on benchmarks like RelBench. However, the current leading model, RelGT, suffers from two key limitations: its random local sampler yields loosely connected subgraphs that hinder message passing, and its global attention module relies on a single, seed-feature-based memory that ignores broader macro-level dynamics. To overcome these limitations, we introduce QUARTET, an expressive graph transformer architecture that applies full self-attention on local subgraphs while enriching global context through cross-attention branches. Specifically, QUARTET employs a Causal Random Walk (CRW) sampler based on recency-truncated Personalized PageRank (PPR) to extract compact, hub-robust, and densely connected local subgraphs without temporal leakage. Concurrently, a quad-branch cross-attention module integrates global context from four complementary perspectives: seed feature, seed topology, temporal dynamics, and collaborative dynamics. Across the RelBench v1 classification tasks, QUARTET consistently matches or outperforms the current state-of-the-art graph transformer baselines (HGT and RelGT). Ablation studies confirm that the CRW sampler significantly enriches local neighborhood quality, while the global branches provide essential, task-specific predictive gains. --- ### SsgCaps: A controlled dataset for the evaluation of sound scene generation algorithms Authors: Modan Tailleur, Junwon Lee, Laurie M Heller, Mathieu Lagrange, Keunwoo Choi, Brian McFee, Keisuke Imoto, Yuki Okamoto Published: 2026-09-22 Categories: cs.GR, cs.AI arXiv: https://arxiv.org/abs/2609.26854v1 Sound Scene Generation is about the automatic synthesis of artificial sound scenes. We introduce SsgCaps, a publicly available dataset of human-engineered sound scenes wherein each scene matches a precisely structured prompt that guides the sampling process. The corresponding prompts are sampled from a predefined action-based typology that allows extensive sampling while retaining plausibility. SsgCaps is a sound scene dataset derived from the unpublished reference dataset for Task 7 of the 2024 DCASE Challenge edition, which contained private-and public-domain audio samples. In contrast, SsgCaps contains only public-domain audio samples, allowing us to open this dataset to the community. To make this dataset useful to the community, we first elaborate on the rationale for the prompt and dataset structure. We then perform a comparative quantitative analysis of the 2 versions of the dataset. To do so, we compare both versions to the audio synthesized by the SSG algorithms submitted to the challenge using Fr{é}chet Audio Distance (FAD) and Kernel Audio Distance (KAD) as well as perceptual ratings. This analysis shows only small differences, which enables us to recommend the open version for further benchmarking of SSG algorithms. --- ### COPE: Continual Personalization of LLMs under Sparse User Feedback via User Embeddings and Self-Evaluation Authors: Ruike Cao, Fugen Yao, Liang Dong, Jian Xu, Guanjun Jiang, Li Xiao Published: 2026-09-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.26853v1 While Large Language Models (LLMs) have achieved remarkable results across various benchmarks, their alignment with normative values often results in homogenized responses that fail to address diverse user preferences. Existing training-free methods often occupy valuable context windows through prompt engineering, while training-based methods typically remain static post-training, failing to support the continual optimization required in real-world settings. To address these challenges, we propose COPE (Continual Optimization with Personalized embedding and self-Evaluation), a novel optimization framework tailored for real-world-motivated interaction settings with sparse user feedback. Our framework assigns learnable personalized embeddings to each user and synergistically integrates preference capture, self-evaluation calibration, and personalized response optimization within a single update step. A key innovation of our method is the use of self-evaluation to generate proxy rewards, enabling continuous model updates even when explicit user feedback is unavailable. Experiments show that COPE consistently outperforms strong training-free and training-based baselines under sparse feedback, and remains complementary to Retrieval-Augmented Prompting (RAP). Further analyses confirm COPE's reliable self-evaluation, meaningful preference patterns, stable general capabilities, and robustness under shifting preferences and alternative evaluators. --- ### A Leakage-Aware Multimodal Evaluation Framework for Early Intraoperative Acute Kidney Injury Prediction Authors: Quang Minh Nguyen, Duc Minh Le, Ho Nhat Minh Nguyen, Thuy Quynh Nguyen, Trong Nghia Nguyen Published: 2026-09-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.26848v1 Postoperative acute kidney injury (AKI) after major non-cardiac surgery carries substantial morbidity, yet early intraoperative risk stratification remains difficult. In this retrospective cohort study, we propose SynerT, a waveform-only hybrid temporal backbone that combines a causal dilated TCN with a hierarchy of dilated recurrent layers to encode early intraoperative physiologic trajectories for AKI risk prediction. Building on SynerT, we further design two model variants that extend the backbone with structured clinical context: SynerT-MM, a late-fusion multimodal extension that integrates hemodynamic burden summaries and preoperative covariates, and SynerTStack, a leakage-safe stacked ensemble that combines cross-validated predictions from SynerT-MM with strong tabular baselines at the meta-learning stage. All models are evaluated under a strict leakage-aware framework on VitalDB, a high-fidelity perioperative database, with prediction restricted to information available within the first 60 intraoperative minutes. Among 2,413 waveform-usable cases (180 AKI-positive; 7.46% prevalence), SynerT fell well below strong structured-data baselines, demonstrating that waveform-only temporal modeling is insufficient under strict early constraints. SynerTMM recovered discrimination by incorporating hemodynamic burden summaries and preoperative covariates, and SynerT-Stack achieved the best overall performance across AUROC, AUPRC, and F1-max. Cross-fitted Platt recalibration substantially corrected calibration defects in both multimodal variants, and decision-curve analysis confirmed the recalibrated stacked model delivered the strongest net clinical benefit across low-to-intermediate thresholds. --- ### Gaussian-process surrogate indicators for residual-based adaptive GMsFEM Authors: Siqing Liu, Eric Chung, Yiran Wang Published: 2026-09-22 Categories: math.NA, stat.ML arXiv: https://arxiv.org/abs/2609.26843v1 Residual-based adaptive GMsFEM for high-contrast elliptic problems repeatedly evaluates local weighted $H^{-1}$ indicators on every coarse neighborhood, making indicator evaluation a recurring cost in repeated-query settings. We introduce a non-intrusive Gaussian-process (GP) surrogate for the indicator scores used in Dörfler marking. The operational predictor is the GP posterior mean, algebraically equivalent to a kernel ridge regression (KRR) estimator under the stated convention; it uses compressed local solution and spectral features without changing the multiscale solve, local spectral construction, or basis enrichment. A nonuniform perturbed-marking result quantifies how pointwise score errors affect the exact indicator mass captured by surrogate-selected neighborhoods, while a conditional bounded-discrepancy KRR pathway identifies sufficient assumptions for such score bounds. In controlled held-out in-distribution tests, the surrogate-guided method gives error-versus-DoF trends comparable with classical $H^{-1}$-residual offline adaptivity and evaluates the online indicator component 2.0-2.1 times faster, excluding offline data generation and GP training. --- ### NeuroRule: Making Black-Box Neural Networks Explainable through Rule-set Evolution Authors: Tapaswini Kodavanti, Hormoz Shahrzad, Risto Miikkulainen Published: 2026-09-22 Categories: cs.NE, cs.LG arXiv: https://arxiv.org/abs/2609.26841v1 High-capacity neural network models have achieved state-of-the-art performance across diverse classification tasks, yet they frequently operate as black-box models, lacking the transparency necessary for critical decision-making. Such opacity creates a persistent trade-off between performance and explainability. This paper proposes a solution to address this gap: the NeuroRule knowledge distillation framework that results in explainable rule-sets from neural network models. NeuroRule adapts the EVOTER rule-set evolution infrastructure to treat neural networks as targets for the evolution process, distilling their performance into concise sets of propositional logic expressions. There are three primary contributions: (1) an evolutionary method for distilling black-box neural network models into explicit rule-set models; (2) a method for making rule sets more explainable by including a conciseness objective to evolution; and (3) a demonstration that the distillation is viable even without access to the original neural network training data. The paper thus establishes that black-box neural network models can be made explainable and therefore useful in real-world applications where trustworthiness is paramount. --- ### LWCal: Loss-Weighted Calibration for Tabular Classifiers with Noisy Calibration Labels Authors: Zeming Liu, Hang Lyu, Jingtao Zhang, Yuan Xie Published: 2026-09-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.26839v1 Post-hoc probability calibration is usually evaluated under an optimistic assumption: the held-out calibration labels are clean. In many AI deployment settings, however, labels come from weak annotators, historical decisions, heuristics, or distant supervision, so the same label noise that corrupts training also corrupts calibration. We study this overlooked failure mode for tabular classifiers and propose LWCal, a CPU-only post-hoc calibrator that down-weights calibration examples whose noisy labels are contradicted by the base model's held-out probability. LWCal requires no clean validation labels, no noise-rate estimate, and no retraining of the base classifier. A second variant, Gated-LWCal, adds a conservative disagreement gate that backs off toward the raw score when the calibration split appears extremely inconsistent. On nine local binary tabular tasks, six random seeds, symmetric and asymmetric label corruption, and three tree-based base learners, LWCal obtains the lowest average calibration error while Gated-LWCal obtains the best average proper-score tradeoff. In the main random-forest study over 432 noisy cells, Gated-LWCal reduces expected calibration error from 0.188 to 0.122 and negative log likelihood from 0.438 to 0.396 relative to the raw classifier. Paired bootstrap intervals for Gated-LWCal versus raw, Platt, isotonic, and beta calibration exclude zero on ECE, Brier score, and NLL. The artifact contains all scripts, result tables, figures, and the compiled paper. --- ### Flash-dLLM: IO-Aware KV Caching and Parallel Decoding for Fast, Memory-Efficient Diffusion LLMs Authors: Quan Nguyen-Tri, Mukul Ranjan, Zhiqiang Shen Published: 2026-09-22 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.26796v1 Diffusion Large Language Models (dLLMs) have recently emerged as a promising alternative to autoregressive LLMs by enabling non-autoregressive text generation. However, their practical deployment remains limited by inefficient inference, largely due to the absence of effective Key-Value (KV) caching and scalable parallel decoding mechanisms. Existing acceleration methods typically study KV caching and parallel decoding in isolation, overlooking the I/O bottlenecks that arise when cache reuse and parallel token verification are jointly applied. In this work, we introduce $\textbf{Flash-dLLM}$, a training-free inference acceleration framework for fast and memory-efficient dLLMs. Flash-dLLM first identifies GPU memory I/O as a dominant bottleneck in KV-cache-enabled dLLM inference and addresses it with an I/O-aware fused KV-cache kernel that reduces redundant memory movement. Building on this optimized cache mechanism, Flash-dLLM further proposes an efficient KV-cache-driven draft-and-verify decoding strategy, where the dLLM itself serves as both drafter and verifier without requiring an auxiliary model. This unified design enables faster decoding while preserving generation quality and improving scalability to longer sequences and larger batch size. Extensive experiments on mathematical reasoning and code-generation benchmarks demonstrate that Flash-dLLM consistently outperforms existing state-of-the-art dLLM acceleration methods in both inference speed and memory efficiency. In particular, it achieves $5.1\times$ and $11.0\times$ speedups over prior strongest baseline Elastic-Cache on GSM8K and HumanEval, respectively. --- ### φ-RIE: From Photorealistic Reconstruction to Interactive Environments Authors: Runyi Yang, Deheng Zhang, Xiaoye Wang, Kanzhi Wu, Lei Sun, Ajad Chhatkuli, Kunyu Peng, Luc Van Gool, Danda Pani Paudel Published: 2026-09-22 Categories: cs.RO, cs.CV, cs.GR arXiv: https://arxiv.org/abs/2609.26795v1 3D Gaussian Splatting (3DGS) can reconstruct a captured scene photorealistically, but the resulting representation does not by itself support physical interaction. Robot simulation instead requires object-level change, \textit{i.e.}, objects must move independently, make contact, and reveal previously occluded surroundings. This gap arises because object appearance may remain entangled with the background, while hidden object geometry and occluded background content may be unobserved. To address this challenge, we present φ-RIE, a Gaussian-native pipeline that converts selected objects into movable simulator assets while preserving the remaining reconstruction. Our key observation is that asset construction and source removal should be coupled, \textit{i.e.}, one object identity should define the movable asset and the scene content to remove and complete. Accordingly, Scene Observation supplies shared evidence to Coupled Scene Construction, which creates registered assets and completed background Gaussians for simulator-driven rendering in an Interactive Environment. This coupling preserves unedited Gaussians while aligning visual and physical state. On 50 ScanNet++ scenes, evidence-based selection and registration retry increase matched F1 at 20\,mm from 0.336 to 0.383 at fixed retention. Further tests demonstrate asset executability, manipulation gains over a single-generator baseline, and the visual cost of conversion. Together, these results demonstrate that \name\ enables interactive scene conversion. --- ### HARMONY: Hierarchical Agentic Reasoning for MONocular Image-to-Scene Synthesis Authors: Shufan Sun, Chen Wang, Enxin Song, Jiatao Gu, Lingjie Liu Published: 2026-09-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.26793v1 Compositional 3D scene reconstruction has recently been explored from two directions: agentic reasoning that provides semantic understanding of spatial relationships but lacks precise alignment with input images; and visual geometry foundation models that predict dense point maps from input images but the reconstruction quality is limited. Therefore, recovering a complete 3D scene from a single monocular image with accurate inter-object relationships and high-fidelity reconstruction quality remains challenging. In this paper, we present HARMONY, a hierarchical chain-of-thought framework that leverages both agentic reasoning and visual geometry foundation. Given an image of an indoor scene, starting from an empty 3D floorplan, HARMONY first calibrates the camera against the reference image to establish a semantically-grounded spatial frame, then uses agentic VLM reasoning to recover the 3D room layout and an initial placement order. It then places the objects in a hierarchical order, from wall-mounted elements, free-standing furniture, to dependent decorations on top of furniture. We also use depth-first traversal for furniture so each placement conditions on previously resolved structure and a reflective feedback loop to avoid error accumulation. After each object placement by VLM, we use the point cloud estimations to perform geometry-based refinement so that the rendered image aligns better with the input. HARMONY can produce 3D scenes that are semantically consistent and perceptually aligned with the reference image, extending single-image compositional reconstruction to complex indoor scene images. Experiments on synthetic and real-world images demonstrate that HARMONY outperforms the evaluated reconstruction baselines, while qualitative comparisons with GPT-6 Astra suggest more faithful object arrangements and better preservation of scene details. --- ### DreamStream: Towards Policy-Oriented Generative Simulation for End-to-End Driving Authors: Ziyang Leng, Sicheng Mo, Seth Z. Zhao, Haoyuan Cai, Yu Zeng, Rowan McAllister, Bolei Zhou Published: 2026-09-22 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2609.26792v1 Faithfully evaluating end-to-end driving policies in simulation requires observations that are not merely photo-realistic, but preserve the scene features a policy relies on to make decisions. Existing platforms, however, exhibit a sim-to-real visual gap that corrupts policy perception, undermining their ability to assess a policy's closed-loop decision-making. To this end, we propose DreamStream, a generative, closed-loop simulator that achieves policy-oriented fidelity using a simulator-grounded autoregressive video model. Our video model is distilled from a large pretrained video model via traffic layout guidance, varying visual appearance while preserving policy-relevant features such as scenario layout and the temporal consistency of dynamic objects. We further observe that perceptual metrics like FID misrank how well these features are preserved. To tackle this, we introduce FD$π$, a new multi-representation metric that measures the sim-to-real gap as the Fréchet distance over scene-context features from public E2E policies. Under FD$π$, DreamStream improves over the strongest prior closed-loop simulator by $1.6\times$ on nuScenes and $4.7\times$ on NAVSIM, and induces the least perturbation to policy's perceptual observability. Based on DreamStream, we construct Navhard-CL benchmark, which turns non-reactive real-world benchmark NAVSIM into interactive testing environments with adversarial driving behaviors and weather variations. This benchmark exposes many failure modes of driving policies, such as scorer bias and lack of recovery behaviors, that prior closed-loop benchmarks overlook. Code and data are available at https://github.com/VAIL-UCLA/DreamStream. --- ### A Decentralized Partially Observable Team Decision Methodology with Delayed Information Sharing Authors: Xiaoxing Ren, Thomas Parisini, Andreas A. Malikopoulos Published: 2026-09-22 Categories: math.OC, cs.LG, eess.SY, stat.ML arXiv: https://arxiv.org/abs/2609.26783v1 We study decentralized partially observable team decision problems with low-rank latent dynamics and unknown system models. The proposed framework combines team-theoretic equivalence with low-rank model representations to address cooperative decision-making in partially observable Markov decision processes without prior knowledge of the transition model. Each team member makes decisions based on local private information and delayed common information shared across the team. Using only this available information, each member learns an approximate low-rank Markov decision process and applies least-squares value iteration to compute its policy. This yields a fully decentralized learning and planning algorithm that requires neither a centralized coordinator nor centralized training. We show that the resulting member-side solutions approximate the centralized team solution: despite partial observability, unknown dynamics, and delayed common information, each member recovers the corresponding component of an approximate team-optimal policy. We further establish finite-sample performance guarantees and derive a corresponding sample-complexity bound for the proposed algorithm. --- ### Agensh: Scaling Organizational Intelligence to 1,024 Agents Authors: Zhihao Zhan, Ting Song, Li Dong, Shaohan Huang, Jianxun Lian, Yan Xia, Furu Wei Published: 2026-09-22 Categories: cs.CL, cs.MA arXiv: https://arxiv.org/abs/2609.26781v1 A multi-agent system can reduce latency on complex tasks by executing work concurrently. Several pioneering harness frameworks support multi-agent systems. However, the scalability of current multi-agent harnesses is often constrained by a central orchestrator's capacity to allocate tasks and coordinate workers. To address this limitation, we introduce Agensh, a scalable self-organized multi-agent harness without a central orchestrator: concurrent workers execute a multi-agent cooperation loop, continuously gathering context, claiming and self-assigning sub-tasks, taking action and sharing findings, verifying results, and merging progress in an asynchronous manner. The loop is supported by the agentic organization infrastructure comprising three components: a shared workspace holds proposed, ongoing, and completed work; a message interface lets workers communicate; and shared context retains reusable findings and work intentions. To test the scalability of Agensh, we evaluate it on the five hardest ProgramBench tasks with GPT-5.6-sol (high). Scaling from 1 to 128 agents raises the mean final test-pass rate from 19.31% to 28.78%, an approximately 49% relative improvement. Larger organizations reach comparable test-pass rates earlier. On pandoc, scaling from 1 to 1,024 agents raises the final test-pass rate from 33.89% to 55.06%. Worker trajectories further show that different forms of self-organized cooperation gradually emerges and standardizes as the organization grows. These results reveal the number of agents as a new scaling dimension for multi-agent organizations to expand the frontier of general intelligence, offering a practical solution for complex tasks under hard latency constraints or time budgets. --- ### SpeakerMem-R1: Speaker-Centered Dual-Track Memory for Multi-Party Dialogue Authors: Haobo Zheng, Tan Tang, Yan Chen, Weijie Wang, Yingcai Wu Published: 2026-09-22 Categories: cs.CL, cs.AI, cs.IR, cs.LG arXiv: https://arxiv.org/abs/2609.26780v2 Long-term conversational memory in multi-party settings requires more than retrieving relevant content from long-term conversations: it must distinguish who said what, whom each statement concerns, how individuals perceive one another, what information is shared by the group, and how states change over time. Recent studies on multi-party dialogue benchmarks show that existing general-purpose LLM memory systems tend to lose person and group relations or struggle to integrate clues distributed across members, groups, and time. Together, these issues reveal two core bottlenecks: message attribution and relational understanding in multi-party dialogue, and state reconstruction from interleaved histories. To address both, we propose $\textbf{SpeakerMem-R1}$: its dual-track memory stores speaker-labeled verbatim messages and derived states organized into person-level and group-level views, then combines evidence from both tracks by entity, event, and time at query time. To reduce attribution and update errors during structured memory construction while enabling local deployment, we train Writer-R1 with SpeakerLevenshtein and speaker-conditioned GRPO. On GroupMemBench, SocialMemBench, and EverMemBench, SpeakerMem-R1 achieves binary accuracies of 47.9%, 69.2%, and 61.9%, respectively. On the publicly reported EverMemBench leaderboard from EverMind-AI, we achieves 62.33%, the best reported result among the latest state-of-the-art frameworks. It also achieves 70.85% on all 1,986 LoCoMo questions, which we use as a two-person long-term conversation boundary test. In a controlled evaluation of 305 questions, RL raises the SFT Writer's mean accuracy from 57.38% to 68.20%. We report both binary accuracy and token-F1, and ablations show that the verbatim and structured tracks, as well as person-level and group-level views, are complementary under the standardized evaluation interface. --- ### SpeakerMem-R1: Speaker-Centered Dual-Track Memory for Multi-Party Dialogue Authors: Haobo Zheng, Tan Tang, Yan Chen, Weijie Wang, Yingcai Wu Published: 2026-09-22 Categories: cs.CL, cs.AI, cs.IR, cs.LG arXiv: https://arxiv.org/abs/2609.26780v1 Long-term conversational memory in multi-party settings requires more than retrieving relevant content from long-term conversations: it must distinguish who said what, whom each statement concerns, how individuals perceive one another, what information is shared by the group, and how states change over time. Recent studies on multi-party dialogue benchmarks show that existing general-purpose LLM memory systems tend to lose person and group relations or struggle to integrate clues distributed across members, groups, and time. Together, these issues reveal two core bottlenecks: message attribution and relational understanding in multi-party dialogue, and state reconstruction from interleaved histories. To address both, we propose $\textbf{SpeakerMem-R1}$: its dual-track memory stores speaker-labeled verbatim messages and derived states organized into person-level and group-level views, then combines evidence from both tracks by entity, event, and time at query time. To reduce attribution and update errors during structured memory construction while enabling local deployment, we train Writer-R1 with SpeakerLevenshtein and speaker-conditioned GRPO. On GroupMemBench, SocialMemBench, and EverMemBench, SpeakerMem-R1 achieves binary accuracies of 47.9%, 69.2%, and 61.9%, respectively. On the publicly reported EverMemBench leaderboard from EverMind-AI, we achieves 62.33%, the best reported result among the latest state-of-the-art frameworks. It also achieves 70.85% on all 1,986 LoCoMo questions, which we use as a two-person long-term conversation boundary test. In a controlled evaluation of 305 questions, RL raises the SFT Writer's mean accuracy from 57.38% to 68.20%. We report both binary accuracy and token-F1, and ablations show that the verbatim and structured tracks, as well as person-level and group-level views, are complementary under the standardized evaluation interface. --- ### CliffCompaction: Cost-Efficient Compaction for Long-Horizon Coding Agents Authors: Trang Nguyen, Eulrang Cho, Bingqing Chen, Tim Dettmers Published: 2026-09-22 Categories: cs.AI, cs.LG, cs.SE arXiv: https://arxiv.org/abs/2609.26779v1 Agents often work on complex problems that require millions of tokens of context, which necessitates compacting across sessions due to limited context windows. We develop CliffCompaction, an autocompaction technique that reduces cost by up to 50% under a bounded context while maintaining or improving performance on Terminal-Bench and achieving new levels of efficiency for test-time scaling and state-of-the-art results on KernelBench. The per-rollout savings of CliffCompaction make the performance--cost trade-off of test-time scaling more efficient, adding over 10 percentage points on Terminal-Bench for less than the cost of two full-context runs. Under parallel test-time scaling, CliffCompaction lets Kimi K2.6 match Opus 4.7, and exceed Opus 4.6 and GPT-5.3 Codex at lower cost. The key to CliffCompaction's effectiveness is that it keeps compacted information faithful by only truncating or dropping content, never rephrasing or rewriting it. We never compact a compaction---each pass operates only on original content, and prior compacted output is discarded, preventing context drift from accumulating. These properties sustain continual learning over sessions exceeding a million tokens: on KernelBench, CliffCompaction reaches CUDA kernel speedups of $2.23\times$ after 200 steps and $3.58\times$ after 400 steps, surpassing specialized search algorithms and trained agents despite being a general-purpose compaction technique. We open-source a scaffold-agnostic API-proxy implementation of CliffCompaction usable with Claude Code, Codex and other harnesses. --- ### SWE-Serve: Benchmarking Agentic Engineering For Production Inference Serving Authors: Jennifer Williams, Dave Farris, Jeff Farris, Jiantao Jiao Published: 2026-09-22 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2609.26777v1 We introduce SWE-Serve, a benchmark for evaluating agents on production inference engineering tasks. Implementing an inference feature can require coordinating multiple changes across the serving stack, including model support, runtime execution, and public APIs. Existing benchmarks provide limited coverage of production inference engineering: repository-level software engineering benchmarks do not target inference, while general terminal-agent benchmarks include only a few inference tasks. Dedicated inference benchmarks, meanwhile, focus primarily on isolated kernel generation or performance optimization rather than repository-scale production feature implementation. SWE-Serve provides 53 repository-grounded tasks derived from recent production changes to SGLang, spanning six inference engineering families. Each task executes on either CPU or a single GPU (H100) and is evaluated with hidden functional and regression tests, including, where applicable, end-to-end (E2E) serving tests and calibrated performance gates. Executable no-op and oracle controls, adversarial verifier review, and closed-book execution support task validity and evaluation integrity. Across 11 models and 31 model-effort configurations, the best-performing configuration achieves 75% mean pass@1. SWE-Serve exposes a substantial gap between completing tasks locally and achieving production correctness. On 19 tasks with end-to-end coverage, model-serving E2E tests reject roughly one-third of patches that pass every other test (45.9% under the verifier versus 69.4% with E2E tests excluded from scoring), with pass rate increasing for each model's best-performing configuration. By making the production correctness gap directly measurable, SWE-Serve enables the field to track whether future agents move beyond completing tasks locally to achieving production correctness. --- ### StableVQ: Practical Guidelines for Stable Vector-Quantized Tokenizer Training Authors: Bao Tang, Jiahao Guo, Haoxiang Cao, Wenyu Liu, Changqian Yu, Kun Gai, Xinggang Wang Published: 2026-09-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.26774v1 Vector Quantization (VQ) is fundamental to discrete visual tokenizers that power modern autoregressive and masked image generation models. While recent shared-projection codebook methods have substantially advanced codebook utilization, training stability remains a critical and underexplored challenge. We argue that the root cause lies in the entanglement of the Encoder--Decoder and Codebook training: because neither module can reliably fulfill its own responsibility in isolation, the system can only function when the two subsystems happen to cooperate---a fragile condition that breaks down precisely when training is most stressed. We propose StableVQ, which revisits the proper learning objective of each module and resolves the problems that arise when each is trained to fulfill its own role independently. Concretely, (1) Dynamic STE corrects the instability in the Encoder's learning objective, enabling it to robustly optimize the reconstruction space under discrete regularization even when codebook utilization is low. (2) Region VQ Loss reconceives the Codebook's learning objective so that it can independently guarantee full tracking of the encoder output distribution, without relying on encoder oscillations to drive activation. (3) Decoupled Schedule recognizes that the distinct responsibilities of the Encoder--Decoder and the Codebook demand distinct optimization dynamics, and assigns each an independent learning rate schedule to ensure robust system-level behavior. Built on top of shared-projection codebooks, StableVQ is lightweight and introduces no learnable parameters. Experiments on ImageNet demonstrate consistent improvements in training stability, codebook utilization, and reconstruction quality across diverse codebook sizes and initialization settings. --- ### A2M: Trace-Optimized Agent Hijacking in the MCP Ecosystem Authors: Laizhen Li, Xuan Wang, Peicheng Zhao, Juanjuan Zhao, Kejiang Ye, Cheng-zhong Xu, Xitong Gao Published: 2026-09-22 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2609.26761v1 Agents using the Model Context Protocol (MCP) rely on semantic matching to select tools from third-party servers, exposing a semantic supply-chain risk through attacker-controlled metadata and outputs. We introduce A2M (Attraction-to-Manipulation), a two-stage black-box framework for hijacking MCP agents. The Attraction phase optimizes tool metadata to increase invocation probability; the Manipulation phase uses execution traces to refine adversarial tool returns that steer agents toward attacker-desired outcomes. On LiveMCPBench, direct attacks optimized and evaluated on GLM-4.6 achieve a macro-average malicious tool invocation rate of 93.6% across four scenarios, increase weighted token costs to 32.4$\times$ the benign baseline under Cognitive Denial of Service, and attain a mean attack success rate of 74.4% across Information Exfiltration, Environment Integrity Compromise, and Reasoning Derailment. Transfer to four other models without re-optimization yields corresponding macro-averages of 63.6%, 2.7$\times$, and 24.5%. These findings motivate stronger tool vetting and runtime isolation in MCP ecosystems. Code is publicly available at https://github.com/Lilaizhen/A2M. --- ### Grow the Harness, Not the Context: From Strategy-Free Scaffolds to Reusable Specialist Agents Authors: Laizhen Li, Jiarui Li, Juanjuan Zhao, Kejiang Ye, Ye Li, Cheng-zhong Xu, Xitong Gao Published: 2026-09-22 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2609.26760v2 Large language model (LLM) agents often handle streams of related tasks, yet standard harnesses repeatedly ask the model to reconstruct the same control decisions inside each task's context. We study whether task feedback can instead turn recurring control into reusable executable code, while reserving LLM calls for task-specific semantic reasoning. We introduce Growing Harness, a failure-guided training paradigm that learns the agent harness itself from a strategy-free scaffold that exposes fixed model and tool interfaces but encodes no task-solving controller. Function-level execution traces localize each failure to a bounded code surface, an optimizer repairs a window of failures jointly, and a success-first held-out gate rolls back repair sequences that harm prior capability. Accepted edits accumulate in one shared harness, allowing its control structure to emerge from task feedback. Across BrowseComp-Plus and WebArena-Verified with three deployment models from 4B to 120B parameters, Growing Harness achieves the highest mean success in five of six benchmark-model settings and trails the best mean by 0.7 pp. in the sixth. Relative to a Tool-Calling agent, it reduces LLM calls by 76.0-91.8% and deployed-agent inference cost by 74.4-98.6%. On WebArena-Verified, its success remains 44.7-45.3% across model scales, whereas Tool-Calling falls to 6.7% with the 4B model. Ablations show that trace-local edits, joint repair, and gate-based rollback each improve final success. These results show that persistent program growth can move recurring control out of model context and into low-cost code, yielding reusable specialist agents that remain effective with smaller deployment models. --- ### Grow the Harness, Not the Context: From Strategy-Free Scaffolds to Reusable Specialist Agents Authors: Laizhen Li, Jiarui Li, Juanjuan Zhao, Kejiang Ye, Ye Li, Cheng-zhong Xu, Xitong Gao Published: 2026-09-22 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2609.26760v1 Large language model (LLM) agents often handle streams of related tasks, yet standard harnesses repeatedly ask the model to reconstruct the same control decisions inside each task's context. We study whether task feedback can instead turn recurring control into reusable executable code, while reserving LLM calls for task-specific semantic reasoning. We introduce Growing Harness, a failure-guided training paradigm that learns the agent harness itself from a strategy-free scaffold that exposes fixed model and tool interfaces but encodes no task-solving controller. Function-level execution traces localize each failure to a bounded code surface, an optimizer repairs a window of failures jointly, and a success-first held-out gate rolls back repair sequences that harm prior capability. Accepted edits accumulate in one shared harness, allowing its control structure to emerge from task feedback. Across BrowseComp-Plus and WebArena-Verified with three deployment models from 4B to 120B parameters, Growing Harness achieves the highest mean success in five of six benchmark-model settings and trails the best mean by 0.7 pp. in the sixth. Relative to a Tool-Calling agent, it reduces LLM calls by 76.0-91.8% and deployed-agent inference cost by 74.4-98.6%. On WebArena-Verified, its success remains 44.7-45.3% across model scales, whereas Tool-Calling falls to 6.7% with the 4B model. Ablations show that trace-local edits, joint repair, and gate-based rollback each improve final success. These results show that persistent program growth can move recurring control out of model context and into low-cost code, yielding reusable specialist agents that remain effective with smaller deployment models. --- ### Type-Safe Is Not Error-Free: A Constrained Decision Head Follows the Option Name, Not the Rubric Bound to It Authors: Yu Sun, Junhao Xu, Jiajia Shi, Zijin Yang Published: 2026-09-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.26758v2 Typed decision models are built for settings where model outputs are consumed directly by software. Instead of generating free-form text, they return a decision over a predefined set of options. By construction, every output conforms to the required schema. Yet this guarantee does not tell us whether the model interprets the options as intended. We study Jev and two Jev-like models with open weights by changing how option names are assigned to rubrics. Each option consists of an option name and a textual rubric that defines what the option means. We change only which option name is assigned to each rubric; the question, state, rubric wording, and set of option names remain exactly the same. On 1200 workflow decisions with task-specific rubrics, renaming the two options from 0/1 to no/yes changes 70.4 more answers per hundred (95% CI: (67.6, 73.1)) and shifts AUC from .94 to .23, revealing a systematic reversal in the decision ranking rather than simple uncertainty. The same operation has little effect with neutral option names. This pattern holds across all 4 predicates, where the effect is at least 7.4x larger than under the neutral control, and becomes stronger as the number of options increases. The effect also depends on the read-out geometry: a second model family that mean-pools over the full option span flips 4.1x less often. The hosted model exhibits the same behavior: the swap changes AUC from .8146 to .5806 and produces 24x as many answer flips as its test-retest floor. In contrast, replacing the option names with random character strings returns all model families to the neutral-control regime without reducing accuracy. The failure therefore depends on the semantic polarity of the option names rather than on the renaming operation itself. Across all conditions, the type-error rate remains 0%, even when decision accuracy degrades substantially. --- ### Type-Safe Is Not Error-Free: A Constrained Decision Head Follows the Option Name, Not the Rubric Bound to It Authors: Yu Sun, Junhao Xu Published: 2026-09-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.26758v1 Typed decision models are built for settings where model outputs are consumed directly by software. Instead of generating free-form text, they return a decision over a predefined set of options. By construction, every output conforms to the required schema. Yet this guarantee does not tell us whether the model interprets the options as intended. We study Jev and two Jev-like models with open weights by changing how option names are assigned to rubrics. Each option consists of an option name and a textual rubric that defines what the option means. We change only which option name is assigned to each rubric; the question, state, rubric wording, and set of option names remain exactly the same. On 1200 workflow decisions with task-specific rubrics, renaming the two options from 0/1 to no/yes changes 70.4 more answers per hundred (95% CI: (67.6, 73.1)) and shifts AUC from .94 to .23, revealing a systematic reversal in the decision ranking rather than simple uncertainty. The same operation has little effect with neutral option names. This pattern holds across all 4 predicates, where the effect is at least 7.4x larger than under the neutral control, and becomes stronger as the number of options increases. The effect also depends on the read-out geometry: a second model family that mean-pools over the full option span flips 4.1x less often. The hosted model exhibits the same behavior: the swap changes AUC from .8146 to .5806 and produces 24x as many answer flips as its test-retest floor. In contrast, replacing the option names with random character strings returns all model families to the neutral-control regime without reducing accuracy. The failure therefore depends on the semantic polarity of the option names rather than on the renaming operation itself. Across all conditions, the type-error rate remains 0%, even when decision accuracy degrades substantially. --- ### FleXray: Universal Clinical X-ray Segmentation Authors: Victor Ion Butoi, Vivek Gopalakrishnan, John V. Guttag, Adrian V. Dalca, Neel Dey Published: 2026-09-22 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.26756v2 X-ray is medicine's most widely used imaging modality, yet remains among its least quantitative. Unlike volumetric modalities like CT or MRI, X-ray collapses 3D anatomy into a 2D projection, causing structures to overlap and anatomical boundaries to be ambiguous, even to experts. As a result, labeling X-ray databases for training general-purpose segmentation systems is impractical, leaving morphometric and functional X-ray analysis confined to narrow anatomical regions and applications. To this end, we present FleXray, a generalist model for anatomical segmentation across the entire body in clinical X-rays. Instead of curating large, manually annotated X-ray datasets, we build a scalable, physics-based generative X-ray data engine. Using existing 3D whole-body CT segmentation datasets and generative image-editing models, we simulate fully-annotated 2D X-rays with diverse appearances, physiological properties, and imaging geometries. Trained on these simulations, FleXray accurately segments 60 anatomical structures across unseen research datasets and in-the-wild X-rays. We further show that FleXray makes X-rays directly amenable to quantitative analysis, enabling automated measurements for disease grading, robust navigation during X-ray-guided interventions, and data-efficient learning of pathological targets. We release the model, code, a full-body X-ray segmentation dataset, and a local, easy-to-use browser-based tool at https://flexray.csail.mit.edu . --- ### FleXray: Universal Clinical X-ray Segmentation Authors: Victor Ion Butoi, Vivek Gopalakrishnan, John V. Guttag, Adrian V. Dalca, Neel Dey Published: 2026-09-22 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.26756v1 X-ray is medicine's most widely used imaging modality, yet remains among its least quantitative. Unlike volumetric modalities like CT or MRI, X-ray collapses 3D anatomy into a 2D projection, causing structures to overlap and anatomical boundaries to be ambiguous, even to experts. As a result, labeling X-ray databases for training general-purpose segmentation systems is impractical, leaving morphometric and functional X-ray analysis confined to narrow anatomical regions and applications. To this end, we present FleXray, a generalist model for anatomical segmentation across the entire body in clinical X-rays. Instead of curating large, manually annotated X-ray datasets, we build a scalable, physics-based generative X-ray data engine. Using existing 3D whole-body CT segmentation datasets and generative image-editing models, we simulate fully-annotated 2D X-rays with diverse appearances, physiological properties, and imaging geometries. Trained on these simulations, FleXray accurately segments 60 anatomical structures across unseen research datasets and in-the-wild X-rays. We further show that FleXray makes X-rays directly amenable to quantitative analysis, enabling automated measurements for disease grading, robust navigation during X-ray-guided interventions, and data-efficient learning of pathological targets. We release the model, code, a full-body X-ray segmentation dataset, and a local, easy-to-use browser-based tool at https://flexray.csail.mit.edu . --- ### EquivSVA: A Formally Verified Dataset of Behavioral Assertions Across Equivalent RTL Implementations Authors: FNU Aditi Published: 2026-09-22 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.26751v1 Large language models are increasingly used to generate SystemVerilog Assertions from natural-language specifica- tions and register-transfer-level designs. Existing datasets and benchmarks support important goals such as large- scale training, formal evaluation, specification-to-assertion generation, and mutation-based testing. A complemen- tary need is to study whether a generated assertion cap- tures externally observable behavior or depends on inci- dental details of one RTL implementation. We present EquivSVA, a formally verified dataset organized around behavior families. Each family contains four structurally distinct RTL implementations of the same externally ob- servable behavior, shared interface-level gold properties, three controlled mutants, and formal-validation evidence. EquivSVA contains 120 behavior families across 12 cat- egories, 480 reference RTL implementations, 914 gold properties, and 360 mutants. Every final family passes a fixed 17-job validation suite covering RTL equivalence, gold-property proofs, property reachability, mutant dis- tinguishability, and gold-property checks on mutants. We also provide fixed family-safe train, development, and test splits. As a small demonstration of the analyses en- abled by the dataset, we evaluate the publicly released, Apache-2.0-licensed Qwen2.5-Coder-7B-Instruct model on the held-out test split. Of 293 interface-only generated properties, 93 are formally sound, and the number of sound properties varies across equivalent implementations for 14 of 24 test families. These results illustrate how behavior-family organization can support controlled stud- ies of assertion-generation robustness without requiring changes in intended functionality. The dataset, generators, validation scripts, and case-study artifacts are publicly released at https://github.com/aditigupta96/EquivSVA. --- ### Metrics Failure in LLM-Based Code Vulnerability Repair: An Empirical Study and a Change-Aware Screen Authors: Om Nepal, Sushant Aryal, Oluseyi Olukola, Nick Rahimi Published: 2026-09-22 Categories: cs.SE, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2609.26749v1 Large language models (LLMs) are increasingly applied to the automated repair of C/C++ security vulnerabilities, and compile rate is a commonly reported proxy for progress: whether the generated patch compiles. We argue that compile rate is a scientifically unreliable metric for single-function vulnerability repair, and we support this with five controlled experiments over 203 vulnerable functions from Big-Vul, three open-source code LLMs (350M to 6.7B parameters), and three prompting strategies. Compile rate (i) barely responds to an intervention that substantially improves the generated code; (ii) is dominated by evaluation-harness and dataset artifacts rather than model quality, with about 64% of compile failures not attributable to the model, a share that is nearly invariant across models; (iii) shifts by 1.8 to 2.7 times on identical patches under a single compiler-standard flag, with zero regressions; (iv) ranks the three models in the opposite order to reference-similarity metrics; and (v) rewards non-repairs when used as an optimization target, since a compiler-feedback loop raises compile rate while similarity to the human fix falls, with manual inspection finding deletion- and placeholder-style non-repairs among the newly compiling outputs. The natural fallback, whole-function CodeBLEU, also fails: an unchanged copy of the vulnerable input outscores every model. We also examine diff_F1, a change-aware screen that scores only the edited region. It gives exactly zero credit to a no-op and near-zero credit to some, though not all, of the deletion-based gaming patches we observed, while still crediting genuine partial edits, so it may serve as a cheap screen before deeper, execution-based analysis. It is not a repair-quality metric, and we report where it falls short. Our findings argue for change-aware, execution-grounded evaluation of LLM-based vulnerability repair. --- ### Automatic depth-based local center clustering via $β$-integrated local depth and adaptive grouping Authors: Siyi Wang, Alexandre Leblanc, Paul D. McNicholas Published: 2026-09-22 Categories: stat.ME, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2609.26748v1 Clustering is an unsupervised learning technique that partitions unlabeled data into groups. Most existing methods require user-specified parameters, such as the number of clusters or neighborhood size. Conversely, we propose automatic depth-based local center clustering (A-DLCC), a fully data-driven method that eliminates numerical parameter tuning. A-DLCC uses the $β$-integrated local depth to identify stable exemplars, points consistently central across multiple locality levels, termed local centers, which are ranked by their representativeness. Each local center induces a group of similar points, with group-level similarity measured by a proposed nonparametric metric called group-level local similarity. To guide merging, we incorporate the bottleneck path idea from graph theory, which forms the basis of our adaptive merging criterion. Based on this criterion, we design a single agglomeration rule in which a group is either absorbed by a neighbor it reaches better than itself or bonded to a neighbor that both sides find more reachable than their own background, every merge being additionally required to be carried by a contact stronger than a configuration-model null expects. The rule automatically estimates the number of clusters and decides when to stop merging. Experiments on synthetic and real data show that A-DLCC produces interpretable clustering results without parameter tuning. --- ### Diffusion-Induced Spatial Attention Overlapping Community Detection Authors: Kosti Koistinen, Vesa Kuikka, Joni Herttuainen, Matthew Hendren, Brian Holt, Kimmo K. Kaski Published: 2026-09-22 Categories: cs.SI, cs.LG arXiv: https://arxiv.org/abs/2609.26737v1 Detection of overlapping communities is essential for modelling networks in which nodes participate simultaneously in multiple structural or functional groups. Existing graph neural network approaches commonly rely on local message passing, which can obscure community boundaries through smoothing and limit the representation of structurally relevant long-range dependencies. We introduce Diffusion-Induced Spatial Attention Community Detection (DISCO), a deep-learning framework that combines a structural prior derived from influence spreading dynamics, sparse multi-head attention, and non-negative community-affiliation learning. The prior identifies candidate interactions beyond immediate graph neighbours and biases attention according to their structural proximity, while a Bernoulli-Poisson edge-reconstruction objective enables overlapping community inference from node attributes and structural profiles, or both. Benchmark experiments show that DISCO performs competitively against established graph convolutional and graph attention approaches across different input configurations. To demonstrate its practical applicability, we present a proof-of-concept cybersecurity use case in which changes between community assignments inferred from consecutive communication-network snapshots provide an interpretable anomaly signal. Temporal community similarity identifies structural deviations, while node-level contributions help locate the devices associated with them. DISCO therefore provides both a flexible method for overlapping community detection and a foundation for analysing structural change in dynamic networks. ---