# Research Papers — 2026-09 · bundle page 85 of 266 > Full-text markdown of papers #4201–4250 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/85 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/84.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/86.md) ### Benchmarking the Connectomes of Caenorhabditis elegans within the Reservoir Computing Framework Authors: Felix S. Reimers, Ola Huse Ramstad, Aliaksandr Hubin, Stefano Nichele Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.30508v1 The aim of this work is to examine the connectomes of Caenorhabditis elegans through a computational lens using the reservoir computing framework. Connectomes are mappings of biological neural networks; C. elegans is the first organism for which physical connectomes covering the whole nervous system have been published. The connectomes of C. elegans used in this paper have been derived at different ages of the organism and are based on three different ways of measuring inter-cellular connections. They have, with minimal preprocessing, been implemented as reservoirs in the form of echo state networks, which are recurrent neural networks. In reservoir computing, the reservoir itself is not trained, rather the output of the reservoir is passed to a comparatively small read-out module in which training takes place. Training and testing is conducted in different neuro-inspired tasks, with the aim of using these tasks as a benchmark for the connectomes. This process has been repeated with different configurations of the reservoir and equally sized but randomized null models have been used for comparison. The results show that the biological wiring and a bio-informed configuration of input and output nodes of the reservoirs do not necessarily lead to better performance. Contrarily, the randomized null models are often outperforming the original connectomes on the chosen benchmarks. At the same time it becomes clear that the results depend a lot on the configuration of the reservoir and the way the connectome has been derived from the organism. Connectomes from different ages may produce varying outcome, without a clear trend becoming visible. --- ### Federated Targeted Maximum Likelihood Estimation Authors: Diyang Li, Fei Wang, Kyra Gan Published: 2026-09-24 Categories: cs.LG, cs.DC, stat.CO arXiv: https://arxiv.org/abs/2609.30503v1 The evidence behind a scientific or operational decision is often held by hospitals, banks, or registries that cannot pool individual observations. Cross-silo federated learning moves computation to the data and exchanges agreed summaries. Targeted maximum likelihood estimation (TMLE) refines a flexible initial fit, yielding plug-in estimators that respect the model and support efficient inference. TMLE itself, however, has remained a fully centralized procedure. To fill this gap, our paper introduces the first federated TMLE algorithm. We federate targeting itself, for an arbitrary target, loss, and fluctuation family, through two complementary frameworks. FedTMLE-G aggregates local gradients and reproduces centralized targeting step for step. FedTMLE-L lets each institution complete its own fluctuation fit before a single exchange of fitted updates, trading synchronized fidelity for local autonomy. For gradient aggregation, we develop a finite-precision protocol that transmits changes rather than values and certifies targeting accuracy within explicit bounds on exchanges and bits. A description-length analysis of the accepted updates then shows that this finite communication leaves numerical targeting error negligible against sampling uncertainty. The cost of computing an estimator is thus distinct from the complexity of selecting it. Our analysis also indicates that keeping data local is not itself a privacy guarantee of TMLE, since instability of full-record reconstruction need not prevent recovery of a specified sensitive attribute. For a personalized version of local averaging, institutions retain their own estimates and leave once local targeting is complete. A nonconvex convergence bound charges the improvement forfeited through averaging to disagreement among local fits and exposes a tradeoff between equal institutional influence and the sampling variability of small silos. --- ### To Solve Bilevel Optimization with Nonconvex Lower Levels, We Need Second-Order Stationarity Authors: Zhiyao Zhang, Menglu Yu, Alvaro Velasquez, Nathaniel D. Bastian, Jia Liu Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.30501v1 Although bilevel optimization (BLO) has emerged as a powerful framework for addressing many complex and nested machine learning problems in recent years, most existing studies are confined to the lower-level strongly convex (LLSC) or lower-level generally convex (LLGC) settings (i.e., the lower-level objective function is assumed to be, at least, convex). While the LLSC/LLGC assumptions render more tractable algorithmic design and theoretical analysis, they are too rigid to encompass many machine learning problems in practice. The limitations of LLSC/LLGC assumptions in BLO motivate us to investigate solving the BLO problem in the general lower-level nonconvex (LLNC) settings, which remains in its infancy. In the literature on LLNC-BLO, most of the existing works either require additional structures in the lower-level objective function for tractable theoretical analysis, or adopt the first-order stationarity reformulation as a lower-level surrogate problem, which is inherited from the LLSC/LLGC settings but could lose their effectiveness in the LLNC setting. To bridge this gap, we propose to reformulate the nonconvex lower-level problem using a second-order stationarity-based surrogate, the solution of which guarantees a local optimal solution at the lower level. Based on this reformulation, we propose the PROBE (Perturbed gradient algorithm for bilevel problem) and show that it overcomes the limitations of prior works by probing and escaping lower-level saddle points. We prove that PROBE achieves a finite-time convergence rate of $O(T^{-2/5})$, where T denotes iterations. To our knowledge, this work is the first to establish the finite-time convergence for achieving lower-level second-order stationary solutions in general LLNC-BLO. Our experiments on both a large language model-based data curation task and a meta-learning task also show that PROBE outperforms SOTA methods. --- ### PolicyAttention: Softmax Attention Implements Policy Mirror Descent for Closed-Loop Control Authors: Yuhe Sui, Yingzhi Tang, Shufang Chen Published: 2026-09-24 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.30500v1 Can causal softmax attention implement policy mirror descent as a repeated controller rather than a one-step algebraic identity? Negative-entropy policy mirror descent (PMD) has the statewise update $\operatorname{PMD}_η(π,Q)=\operatorname{softmax}(\logπ+ηQ)$. Building on the known Q-TD-PMD recursion, we construct one fixed causal-softmax actor--environment--one-step-critic protocol with explicit actor, routing, sampling, and normalization residuals, and propagate them to the policy actually returned. The construction states the finite-logit/full-support domain, the external tokenization and sampling boundary, and the mean-zero LayerNorm carrier conditions required by the normalized compilation. Separately trained pre-LN Transformers recover the target computation empirically. A frozen one-step audit model is closest to PMD among the tested fixed rules; in a preregistered five-run $S=4$ repeated-control test, the learned actor with an exact one-step critic reaches median returned-policy loss $1.052\times$ the Exact PMD oracle and retains the criterion across four no-retraining shifts. The same checkpoints with their learned critic give descriptive median $1.050\times$ the oracle (no registered margin). At $S=8$, replacing the exact critic by the learned critic raises median $T=20$ loss to $0.0225$ yet leaves the Liang--Lai and Algorithm Distillation adaptations $20.2$--$24.2\times$ higher-loss; this is a one-sided sampled-critic bound because PolicyAttention consumes 144 generative transitions per round versus 20 on-policy transitions for the adaptations. The strict 20-transition comparison remains open. At $S=8,16$, the exact-critic common-harness comparison remains $17.7$--$28.2\times$ lower-loss than those adaptations, with the information asymmetry stated locally. --- ### Ordinary Nonconvex SGD under Distance-Dependent Moments: Finite-Horizon Stationarity and Nagaev Bounds Authors: Wei Biao Wu Published: 2026-09-24 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.30499v1 Uniform noise-moment bounds exclude stochastic gradients whose variability increases with the iterate. We study ordinary, single-sample stochastic gradient descent for smooth, lower-bounded, possibly nonconvex objectives under distance-dependent conditional moments. Under second moments alone, a direct descent--displacement argument yields $T^{-1/3}$ expected average squared-gradient stationarity with a horizon-dependent stepsize. An explicit oracle-complexity corollary matches the known smooth Blum--Gladyshev (BG-0) lower bound, including the $Lb_2Δ^3\varepsilon^{-6}$ and $LΔσ^2\varepsilon^{-4}$ stochastic terms, where $Δ$ is the initial objective gap and $σ^2+b_2\|x-x_1\|^2$ bounds the variance. Thus unchanged SGD attains the minimax stochastic complexity in this second-moment class. For $p>2$, predictable localization and a Hilbert-space Fuk--Nagaev inequality yield a high-probability bound separating logarithmic variance and polynomial rare-shock contributions. The localization radius is derived from the recursion: no bounded-iterate assumption, clipping, normalization, momentum, or increasing batch size is needed. We also give increasing-confidence rates, an objective-gap-growth refinement recovering root-$T$ stationarity, and stochastic $L^p$-Lipschitz examples. The broad BG-0 optimality statement is distinguished from the smaller mean-square-smooth class, in which additional oracle structure permits faster algorithms. --- ### Learning to Bias: Machine Learning-Enhanced Particle Filters Authors: Apoorv Srivastava, Eric Darve Published: 2026-09-24 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2609.30498v1 Sequential inference estimates latent states from noisy and incomplete observations. Particle Filters (PFs), a class of Monte Carlo methods based on importance sampling, provide a flexible framework for this task, but often suffer from poor sample efficiency and unfavorable scaling with dimension, partly due to suboptimal proposal distributions. We address these challenges by integrating learned proposals into the PF framework. We introduce Neural Optimal Particle Filters (NOPFs), which learn an amortized approximation to the optimal proposal from offline simulated one-step conditioning tuples. The learned proposal is used as a drop-in replacement in standard PF updates, with samples corrected by standard importance weights so that the method asymptotically targets the same filtering distribution under standard support and density-evaluation assumptions. Across stochastic nonlinear benchmarks of varying inference complexity, NOPFs improve sample efficiency and distributional accuracy over standard PF baselines with modest computational overhead. The approach integrates data-driven proposal learning into classical inference without altering the underlying filtering objective. --- ### Breaking Homogeneity: Diversifying Persona Sets for Creative LLM Outputs Authors: Sang Bin Moon, Nicole Cho, Daniel Borrajo, Sumitra Ganesh, Abolfazl Hashemi Published: 2026-09-24 Categories: cs.CL, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2609.30492v1 Language models often produce homogeneous responses to open-ended tasks; such homogeneity can spawn groupthink-the convergence of ideas toward a singular and potentially suboptimal decision. We formulate persona diversification as a set-level conditioning problem and study two orthogonal design choices: selecting versus generating personas, and space-filling versus frontier-seeking diversity. We instantiate this design space with four methods spanning coverage and dispersion subset selections, uniform-coverage sampling, and evolutionary persona generation. Evaluations on the Alternative Uses Task (AUT), Infinity-Chat, and Divergent Association Task (DAT) show the benefits of the proposed methods across tasks and creativity objectives. On AUT, evolutionary persona generation increases response diversity by 78.8%, originality by 26.1%, flexibility by 49.5%, and holistic creativity by 13.9% over task-only prompting, while maintaining 98.5% validity; on Infinity-Chat, it nearly doubles persona-induced response separation relative to random personas. Moreover, evolutionary personas compose with creativity-optimized prompting, further increasing its response diversity by 18.6% and creativity by 6.3%. These results establish persona-set geometry as a task-agnostic mechanism for eliciting divergent LLM outputs, and support persona diversification as a reusable complement to prompt optimization. --- ### BioEVAL: A global, multi-institutional benchmark of large language and multimodal models for bioengineering Authors: Shun Ye, Vinny Chandran Suja, Chenlong Li, Chongming Jiang, Reza Zamani, Xiang Li, Christopher Bain, Yuqi Zhou, Walker Peterson, Huidong Wang, Chenglang Hu, Jongchan Park et al. Published: 2026-09-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.30489v1 Large Language Models (LLMs) have demonstrated historic breakthroughs in general reasoning with early successes in biomedical science. However, existing LLM benchmarking emphasizes factual recall, offering limited insight into model performance on frontier and multimodal tasks. We assembled BioEVAL (BioEngineering Validation of AI and LLMs), a global, multi-institutional initiative designed to assess experimental reasoning capability across bioengineering (BE) subfields. BioEVAL spans 11 major BE subfields plus a set of uncategorized items, bringing together 22 research groups to create a PhD-level benchmark comprising 608 evaluation items: 1) 380 multiple-choice questions (MCQs, 359 retained after audit), 2) 218 literature synthesis tasks, and 3) 10 multimodal problems with experimental image interpretation. Benchmark items underwent authoring-group expert review and centralized quality control before evaluation. Following evaluation, a blinded cross-group consensus audit of the highest- and lowest-accuracy MCQ items flagged 21 questions for revision or removal; these were withheld, and all reported MCQ results are computed on the 359 retained items. We evaluated diverse cloud-scale foundation/multimodal models (e.g., ChatGPT, Gemini, and Grok) and locally deployable models suitable for inference on consumer-grade GPUs. Models achieved the highest accuracy of up to 90% on MCQs, similarity score of 0.72 on literature synthesis, and accuracy of 80% on a small sample of multimodal reasoning questions, with substantial performance variation across subfields. Leaderboard rankings characterize current capabilities, limitations, and development priorities across the evaluated BE task categories. BioEVAL is maintained as an extensible benchmark with standardized protocols for continuing expert item contribution and model evaluation. --- ### Geometric Feature Learning for Functional Data Valued on the Symmetric Positive Definite Manifold Authors: Samuel V. Singh, Mimi Zhang Published: 2026-09-24 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2609.30487v1 We here develop a functional neural network, termed MatFAE, for learning trajectories on the Riemannian manifold of symmetric positive definite (SPD) matrices. MatFAE features intrinsic layers that map manifold-valued functions to Euclidean vector-valued functions, followed by a functional layer that projects them into a finite-dimensional Euclidean space. Unlike most neural networks for discrete-time sequences, MatFAE treats each sequence as a continuous function and can therefore encode trajectory dynamics (e.g., first-order derivatives) in its latent representations. Additionally, the morphology of the functional weights in the functional layer offers interpretability by revealing the regions of the input functional data that contribute most to the latent representations. We justify the design principles and properties of each intrinsic layer and detail how matrix factorization is handled during backpropagation. We apply MatFAE to a range of fMRI datasets, demonstrating its ability to efficiently learn informative representations from high-dimensional SPD trajectories and its practical value for real-world neuroimaging analysis. --- ### Do LLMs Understand Context? A Knowledge Graph-Based Evaluation Framework Authors: Subavarshana Arumugam, Mamta Nallaretnam, Kithuni Wickramasinghe, Chamath Gunapala, Pragatheeswaran Vipulanandan, Kamal Premaratne, Uthayasanker Thayasivam Published: 2026-09-24 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.30484v1 While large language models (LLMs) have achieved remarkable linguistic capabilities, a profound question lingers at their core: do these models truly comprehend context or simply excel at pattern matching on an unprecedented scale? Contextual understanding in LLMs refers to the ability to correctly extract relevant information from a given context, integrate it into a coherent internal representation, and reason over it to produce factually consistent and contextually grounded responses. However, traditional methods such as BiLingual Evaluation Understudy (BLEU) and perplexity simply measure surface-level performance. This reveals a critical gap in question answering (QA), where responses must be contextually grounded rather than simply being memorized associations. To fill this void, we propose a novel knowledge graph (KG) based evaluation framework for LLM contextual understanding in QA. Central to this is Semantic Structural Similarity for KGs (S3KG), a hybrid similarity measure combining structural and semantic signals into a single score. In addition, a diagnostic analysis framework is developed to identify and categorize reasoning errors at the triplet level, enabling fine-grained analysis of model failures. Together, across nine benchmarks, S3KG achieves F1 gains of up to $+7.6$ points over the strongest baseline and AUROC up to $0.973$. --- ### AcoustiClaim: A Numeric Claim Benchmark with Instrument Ground Truth Authors: Sheng-Tse Lin, Siyuan Zhai, Chien-Liang Kuo, Massa Baali, Bhiksha Raj Published: 2026-09-24 Categories: cs.SD, cs.CL, cs.LG, eess.AS arXiv: https://arxiv.org/abs/2609.30483v1 Audio language models state numbers for acoustic quantities, and neither human opinion nor a judge model says whether such a number is true of the signal. AcoustiClaim extracts each numeric claim from free text, scores it against the instrument that defines the quantity, and classes each quantity by where its reference can be read. Four open-weight systems and one closed model, asked for ten quantities five ways on two corpora, fill 207 cells. Of these, 49 emit fewer than five distinct values, and eight of the 158 cells that can be ranked exceed a rank correlation of 0.3, the bar we set, three with an interval clear of it, five of them one closed model reading pitch. Error sits at or above a constant-predictor floor in every ranked cell but three. The reference decoder we train declines the five voice quantities in prose on 95% of mixtures, with nothing withheld, and states them on the clean twins, reproducing its targets' rule from audio alone. With a calibrated threshold, withholding lowers error on all ten quantities on the mixtures in the mean and on eight at every split, against at most 0.6% from a random selector. A linear baseline orders errors at least as well as ours. F0 s.d. and shimmer stay above the constant floor. --- ### The Shape of Events: Edge-Based Inductive Biases via Cross-Domain Distillation Authors: Soshun Kihara, Shunsuke Yasuki, Masato Taki Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.30478v1 Convolutional neural networks trained on ImageNet are known to exhibit a strong preference for local high-frequency texture, an inductive bias that translates into fragile robustness against distribution shifts in real-world environments. Event cameras, in contrast, record only changes in scene brightness and are therefore well suited to capturing contour information; however, due to the absence of diagnostic benchmarks in the event domain, the inductive bias that event-camera data instills in vision models has remained underexplored. In this work, we use knowledge distillation from the event domain to the RGB domain so as to exploit the rich evaluation toolkit available in the RGB domain and systematically dissect this inductive bias. Our experiments show that distillation from the event domain induces, in the RGB domain, color invariance, shape bias, and robustness to high-frequency noise. We identify the underlying mechanism as the model suppressing its dependence on high-frequency texture while acquiring a stronger dependence on edge-based object shape. This hypothesis is supported by changes in how color and spatial information are processed at the early layers, together with a spectral trade-off in which robustness to the absence of high-frequency components coexists with vulnerability to contamination of the relied-upon frequency bands and to disruption of geometric structure. We further show that this inductive bias differs from existing robustification methods and that it functions as a useful prior for diverse downstream tasks in which shape and contour information contribute alongside other cues. The code is available at https://github.com/snskysk/event2rgb-distillation . --- ### Scaffold-Constrained Subset Dynamic Programming for Exact SSE Clustering Authors: Yordan P. Raykov, Max A. Little Published: 2026-09-24 Categories: math.ST, cs.LG arXiv: https://arxiv.org/abs/2609.30477v1 Exact Euclidean \(K\)-means partitions \(n\) observations into \(K\) unlabelled clusters, but the unrestricted search is generally exponential. We use data-derived geometric graphs to precondition an exact subset dynamic program: as a result only connected vertex subsets are admitted as clusters, while sum-of-squared-errors (SSE) loss is unchanged. A remaining-set recurrence minimises fixed-\(K\) or penalised SSE, with exact factorisation over the connected components of each remaining set. The central question we study is how much computational support can be removed while preserving an unrestricted optimum. Graph inclusion gives monotone coverage and support relations, and a bottleneck threshold identifies the first covering graph in a nested hierarchy. For fixed \(K\) and dimension, under compact ball support and density bounds, retaining \(q=O(\log n)\) nearest neighbours per observation preserves an empirical SSE optimum with probability tending to one, using an \(O(\log n/n)\) fraction of complete-graph edges. Truncated Gaussian mixtures with unequal weights and covariances satisfy these conditions. The rate we provide is a sufficient upper bound rather than a result implying polynomial optimisation complexity. Objective-matched synthetic and full-data comparisons assess coverage, compression, and reference-label agreement. As a secondary application, we illustrate how the proposed scaffold preconditioning can be utilized to improve the efficiency of split-merge proposals that preserve unrestricted mixture posteriors. --- ### Asymmetric Classifier-Free Guidance for Target-Speaker ASR Authors: Yiwen Guan, Jacob Whitehill Published: 2026-09-24 Categories: eess.AS, cs.CL, cs.SD arXiv: https://arxiv.org/abs/2609.30476v1 Target-speaker automatic speech recognition (TS-ASR) must identify and transcribe a desired speaker under varying overlap and noise conditions. These changes alter the acoustic evidence for the target speaker in the speech mixture, motivating inference-time calibration of speaker conditioning. We introduce asymmetric classifier-free guidance (CFG) for TS-ASR using Whisper: the speaker-conditioned branch predicts the target transcript, while the speaker-unconditioned branch predicts serialized multi-speaker transcripts. CFG adjusts the contribution of speaker conditioning during decoding through a single guidance scale. We select a global guidance scale on target-domain development data and train a lightweight encoder-based predictor to adjust it for each utterance, keeping the recognition model fixed. Under domain shifts, our full system achieves relative word error rate (WER) reductions of up to 21.8% over the condition-only baseline, and 5.6% over standard conditional decoding of the same CFG-trained model. Oracle analysis shows that substantially larger WER reductions are possible through utterance-level scale selection and identifies how beneficial adjustments vary with domain shifts. --- ### Mentored Decoding: Faster Inference meets Boosting Authors: Vivien Tran-Thien, Richard Nock Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.30474v1 Speculative decoding is a successful technique speeding up inference of a target autoregressive language model via a fast drafter model. Lossy speculative decoding allows a drift with respect to the target to further improve speed. Interestingly, it has been observed experimentally that the resulting model can $\textit{also}$ beat the target $\textit{quality-wise}$. Our paper formally proves how such a feat is possible with a formal approach to lossy speculative decoding called $\textit{mentored decoding}$. To get there, we connect inference to a celebrated ML training theory, $\textit{boosting}$, and proceed via the generalization of mentored decoding to the whole set of $f$-divergences. We uncover key properties of mentored decoding, among which (i) the particularly appealing geometric nature of the total variation case, (ii) simple approximations for any $f$-divergence in direct relation with boosting compliance, and (iii) a $\textit{divergence independent}$ $O(n)$ space and $O(\mathrm{sort}(n))$ time data structure built on drafter and target outputs, which allows to query the optimal parameters of the dual problem in $O(\log n)$ time and constructing optimal mentored distributions in $O(n)$ time for any $f$-divergence. --- ### Moment-guided edge sampling Authors: Weibin Cai, Reza Zafarani Published: 2026-09-24 Categories: cs.LG, cs.SI arXiv: https://arxiv.org/abs/2609.30472v1 Edge sampling makes local decisions to achieve graph-level objectives, such as preserving structural properties. This creates a fundamental challenge: \textit{how can the effect of a local edge edit (i.e., edge addition or removal) on global graph structure be quantified and controlled?} We address this challenge with a \textit{moment-guided edge sampling framework} based on spectral moments of the random-walk transition matrix. We compute exact moment changes through two complementary methods: a combinatorial method with closed-form updates for low-order moments, and a low-rank method that exploits \textit{locality} and \textit{cyclic trace invariance} to compress computations to edited endpoints, supporting arbitrary moment orders and batched edits. For single-edge edits at fixed moment orders, the low-rank method reduces the cost from $O(mn)$ to $O(m)$, while the combinatorial method evaluates low-order changes in constant time given maintained local statistics. These moment changes provide \textbf{interpretable structural signatures} of local edge motifs that aggregate into graph-level fingerprints. This structural meaning motivates us to ask whether preserving moments also preserves the graph properties. We further derive and validate that moment-preserving sampling can \textbf{retain related structural properties}, including triangle-weighted clustering coefficient. These structural insights enable \textbf{analysis and improvement of graph learning}: different edge structures have distinct effects on supervised node classification, while moment-guided augmentation is competitive for graph contrastive learning. Together, these findings establish moments as an interpretable and controllable bridge from local edge edits to global graph structure and learning. --- ### CARGO: Context-Aware Retrieval-Gated Evaluation of Agentic AI in Production Authors: Mukul Chhabra, Shail Patel, Luigi Medrano Published: 2026-09-24 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.30471v1 Reference-based LLM-as-a-judge evaluation assumes the reference answer is the target. In deployed agentic systems that operate over dynamic entities (support cases, assets, accounts), the closest available reference typically applies the correct procedure to a different entity, so a literal judge penalizes different identifiers, dates, and statuses as errors or hallucinations. We name this failure mode reference-instance divergence (RID). We propose CARGO, a framework that (i) treats retrieved references as procedural exemplars and grounds factual judgments in the live instance's observed context, (ii) assigns each claim a three-way status (supported, contradicted, unverifiable) and penalizes only contradictions, and (iii) gates evaluation by retrieval confidence, casting production evaluation as selective prediction. We introduce CARGO-Bench, a perturbation-based diagnostic suite with ground truth by construction that separates leniency from discrimination. On CARGO-Bench (246 items, two judge models, 7,872 judgments), the standard reference-based judge penalizes 100% of correct entity-transplanted answers and is uninformative (discrimination index DI ~ 0); supplying the live facts without reframing changes nothing. CARGO eliminates these false penalties (0/50) while retaining near-complete contradiction recall (50/50 and 49/50), raising DI to 0.58 (0.48, 0.68); a rubric-swap control attributes most of the effect to context-grounded dimension definitions. CARGO also exposes a limitation of its own design: the leniency that protects entity values suppresses detection of procedural corruptions (20% recall). A post-hoc fix does not close the gap, and an LLM-as-annotator study with written guidelines and adjudication shows the same blind spot. We release a preregistered protocol for extending the evaluation to expert agreement, risk-coverage, and cost on production traffic. --- ### Reliability-aware Cross-sample Enhancement for Robust Multimodal Sentiment Analysis Authors: Menghua Jiang, Haokai Gao, Xiangui Kang, Haifeng Hu, Sijie Mai Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.30470v1 Multimodal Sentiment Analysis (MSA) aims to infer human emotions from multiple modalities such as text, audio, and vision. In practice, inputs are often corrupted by noise and missing modalities, which degrades performance. Existing methods typically address these challenges in isolation, limiting their effectiveness in realistic settings. To address this limitation, we propose a Reliability-aware Cross-sample Enhancement (RCE) framework. Specifically, RCE first introduces an adaptive variational information bottleneck to model modality-wise uncertainty and perform quality-aware information compression, thereby suppressing redundant noise in unreliable modalities. Furthermore, we design a reliability-aware cross-sample enhancement strategy that retrieves high-confidence, semantically consistent neighbors from a large candidate pool to enrich and calibrate current representations, effectively alleviating information deficiency caused by missing modalities. Building upon this, RCE integrates cross-modal interactions with a multilevel reliability-aware fusion mechanism to adaptively aggregate information across modalities and enhancement stages, leading to more robust multimodal representations. Extensive experiments demonstrate that RCE consistently outperforms state-of-the-art methods across full, noisy, and missing-modality settings. --- ### Pretrained ASR Pseudo-labeling for Noisy Police Audio Authors: Kaavya Chaparala, Su Huang, Stephen L. Morgan, Rhiannon N. Miller, Anjalie Field Published: 2026-09-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.30469v2 Pretrained ASR systems perform poorly on noisy Broadcast Police Communication (BPC), hindering efforts to understand police decision-making. Pseudo-labeling offers an unsupervised path to improve ASR without expensive human labels, but the efficacy of this approach on very noisy domains is not known. In this work, we systematically assess the opportunities and limits of pseudo-labeling to adapt foundation ASR models (Whisper and Qwen3-ASR) to noisy BPC domain corpora from Baltimore and Chicago. We demonstrate that existing internal confidence metrics (log-probabilities and STAR scores) fail to distinguish between high and low quality BPC pseudo-labels, and we introduce an external LLM-as-a-judge filtering paradigm that leverages parametric knowledge to discard contextually implausible transcripts. Our LLM-judging filters more aggressively than internal metrics and significantly reduces WER of the pseudo-labeled training sets across the Baltimore and Chicago BPC corpora, though a substantial gap remains relative to an oracle filter. We also introduce a new cross-model pseudo-labeling paradigm where one model is finetuned with pseudo-labels from the other, and we identify this method as a promising direction for future pseudo-labeling work. --- ### Where Does Retrieval-Based Open-Ended Evaluation Fail? Automatic Taxonomy Induction from Long-Form Medical Answer Factuality Verification Authors: Heyuan Huang, Jirui Dai, Alexandra DeLucia, Sonal Joshi, Mahsa Yarmohammadi, Jie Gao, Bernal Jiménez Gutiérrez, Mark Dredze Published: 2026-09-24 Categories: cs.CL, cs.IR arXiv: https://arxiv.org/abs/2609.30467v1 Retrieval-based factuality evaluation, where LLM-generated claims are verified against evidence from authoritative medical corpora, has become the dominant paradigm for scalable hallucination detection in high-stakes clinical settings. Despite the urgency of reliable and transparent medical fact verification, most systems measure performance with aggregate metrics like F1, which obscure where and why failures occur. Existing RAG diagnostics require gold answers or annotated gold evidence, neither of which exists in this regime. We introduce two comprehensive taxonomies, grounded in a case study on the open-ended MedExpert dataset and 3 closed-ended datasets, decomposing failures into retrieval-stage errors along five quality dimensions, and verifier-reasoning errors into six consecutive steps. We adapt an automatic pattern induction pipeline using LLM-as-Judge to label evidence quality and classify verifier reasoning errors at scale, and then stress-test our findings across 4 retrieval methods and 6 frontier verifier models. Our analysis reveals that scaling model size, adding reasoning effort, expanding to authoritative web sources, and applying medical fine-tuning do not resolve these failure modes, demonstrating that they represent fundamental limitations of the retrieve-then-verify paradigm in open-ended medical settings rather than artifacts of outdated systems. We release our code and data at https://anonymous.4open.science/r/Medical_RAG_eval-4AB5 for the full reproducibility of our results. --- ### A Benchmarking Framework for Context-aware XR Interfaces Authors: Hyunsung Cho, Sarah Yewon Yun, Nancy Ruonan Sun, Ben Lafreniere, Mark Parent, Kashyap Todi, Tanya R. Jonker, Hrvoje Benko, Sherry Tongshuang Wu, David Lindlbauer Published: 2026-09-24 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2609.30466v1 Everyday Extended Reality (XR) systems aim to provide context-aware access to the right functionalities at the right time and place, with minimal manual reconfiguration as users switch context. Yet these interfaces are hard to evaluate: current prototyping and user-study workflows offer no systematic, repeatable way to compare adaptation methods across users and scenarios. We present ContextXR, a novel benchmarking framework for context-aware XR interfaces. ContextXR represents an XR application as a connected graph of functional facets, each a semantically coherent group of related capabilities that together support a shared user intent. On this representation, we build MineXR++, a dataset augmenting prior XR interface data with facet-level annotations, and formulate three canonical tasks of context-aware suggestion: context factor analysis, initial facet suggestion, and next facet suggestion. Our evaluation protocol scores suggestion methods by a simulated interaction metric, the navigation and search cost of reaching the desired functionality. Through experiments benchmarking global popularity, relational retrieval, and LLM-based methods, we demonstrate that ContextXR enables the systematic, reproducible evaluation of context-aware XR interfaces. --- ### RAZOR: Pruning Replaceable Experts in LLMs Authors: Mingyang Song, Mao Zheng Published: 2026-09-24 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2609.30465v3 Mixture-of-experts (MoE) models activate only a few experts per token yet store the entire expert pool. Whole-expert pruning shrinks that pool, but for reasoning models it must remove experts without eroding reasoning ability. Common scores rank experts by routing frequency or output magnitude, which measures isolated contribution rather than deletion damage. What decides the damage is functional replaceability, whether the surviving computation can reproduce what is removed. A large contribution may be replaceable by the remaining mixture, whereas a small one may carry a direction the survivors cannot recover. We introduce RAZOR, a training-free method that scores replaceability from consensus residuals, the deviations of individual expert outputs from their original weighted mixture. Holding the layer input fixed, these residuals yield the exact output change from deleting one expert, including survivor reweighting and the replacement expert promoted by router refill. RAZOR aggregates this change over calibration tokens and prunes to a layerwise budget using forward passes alone, without gradients, subset search, or recovery training. On GLM-4.7-Flash, Qwen3.6-35B-A3B, DeepSeek-V4-Flash-0731, and Hy3 at 25% and 50% expert removal, RAZOR attains the highest macro average over nine reasoning-centered tasks among the evaluated pruning methods in all eight model-budget settings. Against REAP on GLM-4.7-Flash and Qwen3.6-35B-A3B, it gains 2.12-5.59 points on this average and lowers reverse KL in all four comparisons. Retained accuracy is not the whole picture, as pruned Qwen3.6-35B-A3B still shifts in response diversity, formatting, and termination. --- ### VkVIO: Cross-platform GPU Acceleration for Visual-Inertial Odometry with Vulkan Authors: Ole Hoffmann, Mateo de Mayo, Daniel Cremers Published: 2026-09-24 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2609.30459v1 Perception in robotics and XR fundamentally relies on good state estimation. Visual-inertial odometry (VIO) and Simultaneous Localization and Mapping (VI-SLAM) are proven ways of achieving this goal in a cost-effective and accurate manner. Efficiency in these systems allows for smaller, cooler, and lighter devices. GPU acceleration is a natural approach for reducing latency, thanks to their wide availability in platforms like embedded computers, mobile phones, and XR headsets. However, previous works in the literature have limited themselves to the use of CUDA for this task, significantly reducing deployment options to a single vendor. We instead leverage the vendor-agnostic Vulkan API, originally designed for the strict performance requirements of 3D graphics applications. In this work, we present VkVIO, the first, to the best of our knowledge, cross-platform GPU-accelerated VIO method. We provide state-of-the-art accuracy with causal estimates required for real-time operation. We deploy VkVIO on a diverse range of devices spanning a workstation, a laptop, and an extremely inexpensive single-board computer, while outperforming CUDA-based systems on the same hardware. VkVIO enables possibilities for low-latency, low-power, and low-cost VIO in robotics and XR. --- ### Spectral Feedback for Test-Time Alignment of Protein Diffusion Models Authors: Shai Dickman, Mert Cemri, Landon Butler, Kannan Ramchandran Published: 2026-09-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.30456v1 Reward maximization alignment methods for discrete diffusion models have primarily focused on steering the reverse process, either by influencing token logits or by selecting favorable sequences at intermediate steps. These approaches largely treat inference as a unidirectional process, lacking mechanisms for revisiting undesirable token selections. We introduce Spectral Feedback, an algorithm that selects edit-positions in a feedback loop, allowing the model to iteratively correct its own generations. This approach leverages the mask structure of discrete diffusion models by re-masking and re-sampling tokens, analogous to image editing methods that reintroduce noisy latents and re-run the reverse process. While prior alignment methods focus on what token labels to assign to maximize a target reward, we instead treat which tokens to revisit as the central alignment problem. Selecting edit-positions is challenging because edit effects are interdependent: the impact of modifying one token depends on which others are edited simultaneously. We define an edit-set as a set of token positions to re-mask and re-sample. Motivated by prior work on sparse interactions in biological systems, we find empirically that edit-set value functions for protein inverse folding admit sparse Fourier representations. This structure enables Spectral Feedback to efficiently learn and optimize the value functions for edit-position selection. Spectral Feedback is model-agnostic and can be applied to pretrained, test-time aligned, and fine-tuned diffusion models. For all of these models, the algorithm improves alignment performance without modifying the underlying generative process. Applied to inverse folding with a protein stability reward oracle, it achieves a 32.3% increase in stable proteins for a pretrained model, 24.8% for Best-of-10, and 5.8% for a state-of-the-art RL fine-tuned diffusion model. --- ### Auditing System-1 Models on Biosecurity-Relevant Benchmarks: Calibration, Selective Prediction, and Permutation Instability in a Non-Generative Model Authors: Kimon Antonios Provatas, Ilias Georgakopoulos-Soares Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.30454v1 Non-generative "System-1" models return structured probabilistic decisions in a single forward pass, without autoregressive decoding, at a small fraction of the inference cost of a generative model. This makes them of interest as inexpensive components in larger pipelines, but their reliability on biosecurity-relevant tasks has not been systematically examined. We audit one commercial System-1 model on 6,020 multiple-choice items drawn from the Weapons of Mass Destruction Proxy (WMDP), a paraphrase-robust WMDP-Bio variant, and six LAB-Bench subtasks, measuring accuracy, calibration, error detection, selective prediction, and sensitivity to the order in which answer options are presented. Accuracy is strongly task-dependent. Once the vendor's uncertainty field is correctly interpreted, the model is reasonably well calibrated (pooled expected calibration error 0.034) and its top-1 probability separates correct from incorrect predictions (pooled AUROC 0.820), though both degrade substantially on the weaker tasks. Under four cyclic rotations of the answer options, 37.4% of WMDP-Cyber items receive different answers; a control using byte-identical repeated calls attributes most of this to option order rather than run-to-run variation. Averaging probabilities across rotations improves WMDP-Cyber accuracy by 3.8 percentage points, and applying it only to low-confidence items recovers most of that gain at well under the cost of averaging every item. --- ### LensDesigner: A Self-Improving Agent for Optical Lens Design Authors: Lei Sun, Haoran Liang, Dannong Xu, Yao Gao, Yuyu Geng, Jinjin Gu, Kaiwei Wang, Danda Pani Paudel, Luc Van Gool Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.30450v1 Optical lens design is a complex, non-convex optimization challenge that relies heavily on human experience and intuition. Existing optimized-based automatic lens design methods struggle to navigate this vast parameter space without meticulous manual tuning. In this paper, we present LensDesigner, an autonomous agent framework that mirrors the problem-solving workflow of expert opticians. To overcome the initial cold start problem, we construct LensLib100K, an extensive optical lens library, and employ Optics-Aware Retrieval to supply physically valid structural seeds. Within an interactive physical simulation environment, the agent executes macroscopic orchestration while receiving immediate optical feedback. Furthermore, we introduce a continuous self-evolving mechanism guided by a curriculum agent. By iteratively solving design tasks with progressively increasing difficulty, the agent autonomously extracts, accumulates, and reuses design heuristics, effectively evolving its optical lens design expertise over time. At the evaluation level, we introduce LensArena, a standardized evaluation benchmark comprising $120$ diverse optical design tasks, covering extreme configurations. Extensive experiments on this benchmark demonstrate that LensDesigner significantly outperforms publicly available baseline algorithms, achieving superior success rates and optimization efficiency. We hope this work sheds light on the emerging field of intelligent optics. The code will be publicly available. --- ### Predicting Transmembrane Protein Topology from 3D Structure Authors: Sitong Chen, Xiaopeng Mao Published: 2026-09-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.30446v1 This paper presents a novel approach to infer protein topology using the state-of-the-art graph neural network (GNN), SchNet. The model is trained on the same dataset used to develop the recent DeepTMHMM model with 5-fold cross-validation. Unlike the conventional approaches based on using only the protein sequences or the $α$-carbons as features, we have decoded our classifier in this way, so all atom-level embeddings are used. Without applying any pre-trained weight, the final results have shown great potential that GNNs can be used for topological predictions. --- ### Bayesian Uncertainty Quantification for fMRI Functional Connectivity via Simulation-Based Inference Authors: Simon Carter, Zeming Kuang, Lilianne R. Mujica-Parodi, Helmut H. Strey Published: 2026-09-24 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.30445v1 Optimizing fMRI scan duration and spatial resolution is critical for experimental design, yet traditional correlation-based approaches cannot quantify uncertainty or disentangle scanner measurement noise from true neural variability across subjects. Without principled uncertainty bounds, researchers cannot know whether a protocol is long enough to reliably estimate connectivity, or whether between-subject differences reflect biological variation or noise. We present a Bayesian framework modeling BOLD dynamics as coupled Ornstein-Uhlenbeck processes, using Sequential Neural Posterior Estimation to obtain connectivity posteriors while accounting for frequency-independent measurement noise across the BOLD spectrum. Applied to N = 28 healthy controls (55 scans) at 7T using a functional network atlas (65 DMN regions), the framework quantifies uncertainty across its sources: scanner noise, subject variability, and acquisition length. Spatial analysis identifies a mean of 46 voxels per ROI, roughly half of typical region sizes, as sufficient to achieve 90% of asymptotic precision. At the single-subject level, 7T reaches its within-session precision plateau in approximately 7 minutes versus 10 minutes for 3T, a 40% reduction in required scan time, providing the first direct, model-based quantification of the scan-time advantage conferred by higher field strength. At the population level, 3T requires roughly 37 times more per-subject scan time than 7T for the pooled curves to converge, confirming a consistent advantage of higher field strength at every timescale. Together these findings provide concrete, scanner-specific guidance for protocol optimization, with direct implications for reducing acquisition costs and improving the reliability of connectivity-based clinical biomarkers. We provide code enabling researchers to derive these bounds from their own data. --- ### Inference-Time Target Speaker Unlearning in LLM-Based Automatic Speech Recognition Authors: Bo Su, Yueru Yan, Thai Le Published: 2026-09-24 Categories: cs.CL, cs.SD, eess.AS arXiv: https://arxiv.org/abs/2609.30439v1 We introduce target-speaker unlearning ASR (TSU-ASR) task in a fully end-to-end framework for multi-speaker ASR and diarization. Given a multi-speaker utterance and a set of opt-out speakers who do not wish to have their speech transcribed, the task requires an ASR system to transcribe all speakers except the opt-out ones, while still indicating when those speakers are active. As a first step towards tackling this task, we introduce a novel, light-weight Enrollment-Conditioned Gating (ECG) module attachable to a frozen dual-stream speech LLM that enables ASR for new opt-out speakers dynamically during inference, even those who were not seen during initial ECG training phase. Our experiments on both AMI (English) and AliMeeting (Mandarin) datasets show that speech transcription accuracy for corresponding opt-out words or characters falls from 72.3% to 48.2% and from 73.6% to 27.3%, respectively, while retained speakers' transcription error rates maintain more or less the same. Our approach provides a practical solution for modern video conferencing platforms, allowing speakers to dynamically opt-out from automated AI transcriptions without forcefully leaving the meeting sessions, enabling a privacy-preserving interface for potentially millions of online meetings daily. --- ### WALT: Learning World-Model-Aligned Latent Trajectories for Autonomous Driving Authors: Mingkai Jia, Jiaxin Guo, Zhijian Shu, Jiawei Xu, Mingxiao Li, Jintao Cheng, Ping Tan, Wei Yin Published: 2026-09-24 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2609.30436v1 Driving world models learn rich predictive representations of the surrounding environment from visual observations, yet accurate visual prediction does not necessarily translate into effective trajectory planning. We argue that a key bottleneck lies in the mismatch between visual world states and raw geometric trajectories, which may limit the planner's ability to exploit action-relevant semantics encoded by the world model. To address this issue, we propose World-Model Alignment for Latent Trajectories (WALT), which learns a compact generative trajectory latent space by transferring information from a frozen pretrained driving world model without modifying the world model itself. Rather than directly generating raw waypoints, WALT maps them into compact representations through a dual-branch trajectory autoencoder and transfers semantic knowledge from the frozen visual world model into this trajectory space, encouraging the learned action representation to capture scene-level cues relevant to future motion and planning. Beyond our proposed formulation, we systematically study latent learning based on Joint-Embedding Predictive Architectures (JEPA) and feature alignment following Representation Alignment (REPA) to investigate how trajectory-only representation learning affects downstream planning. We evaluate WALT on the NAVSIM benchmarks. Relative to the raw-waypoint baseline, WALT improves PDMS from 89.4 to 89.8 on NAVSIMv1 and EPDMS from 87.3 to 87.9 on NAVSIMv2 while reducing trajectory planner FLOPs by 30.5%. These results suggest that preserving world representations while extracting action-relevant information provides an effective interface for world-model-based trajectory planning. --- ### ProCAP: Probabilistic Cross-Attentive Prompt Learning for Vision-Language Models Authors: Hiwa Azeez Abbas, Fatemeh Daneshfar, Moloud Abdar Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.30434v1 Pre-trained vision-language models such as CLIP can recognize new categories via prompting, but they often struggle when labeled data are scarce or the test distribution shifts. Prompt learning adapts only a small set of parameters while keeping the backbone frozen, yet many existing multimodal prompt learners couple the visual and textual branches weakly and can be brittle in low-shot regimes. We propose ProCAP, a probabilistic cross-attentive prompt learning framework that improves cross-modal interaction and training stability without updating any CLIP weights: it learns both visual and textual prompt tokens and links them through stacked bidirectional multi-head cross-attention so the two branches refine each other across prompt depth. To reduce overfitting under limited supervision, we parameterize prompt tokens with Gaussian means and variances and regularize them with lightweight KL and L2 penalties, and we further add a compact symmetric InfoNCE head that aligns cross-attended image features with class-level text representations in a shared low-dimensional space. Across few-shot base-to-novel generalization on 11 datasets, cross-dataset transfer, and domain generalization on ImageNet shift benchmarks, ProCAP achieves strong aggregate base-to-novel performance and competitive transfer performance while keeping the CLIP backbone unchanged. --- ### Improving Molecular-Morphology Contrastive Pretraining using Deep-Learning-based Morphology Profiles Authors: Jie Li, Kathryn E. Kirchoff, Dante A. Pertusi, Zhizhuo Zhang Published: 2026-09-24 Categories: cs.LG, q-bio.BM arXiv: https://arxiv.org/abs/2609.30433v1 Recent advancements in image-based profiling techniques have enabled the collection of high-volume cell morphology data, allowing new molecular embedding models to learn from the experimental phenotypic perturbations of a molecule in a cell. Previously, we developed Molecule-Morphology Contrastive Pretraining (MoCoP), a strategy for aligning small molecule embeddings to morphology fingerprints extracted through CellProfiler. The resulting molecular representation showed transferable performance for quantitative structure--activity relationship (QSAR) prediction tasks. Here, we extend the method by using a deep-learning-based cell image encoding pipeline to extract more feature-rich morphology profiles and align them to the molecular embeddings through contrastive learning. The new embeddings encode more accurate information on how molecules perturb cell morphology and enable improvements for QSAR predictions through either fixed-embedding linear probes or fully flexible fine-tuning. Morphology retrieval performance scales log-linearly with training data size, suggesting continued improvements as larger datasets become available. The improved MoCoP v2 also achieves superior performance on toxicity prediction and competitive results on ADME and activity benchmarks, when compared with existing molecular embedding models that use both cell morphology and transcriptomic data during training. --- ### Actively Resolving Contextual Uncertainty for Underspecified Tasks in Natural Language Authors: Zachary Ravichandran, Jonathan Diller, Fernando Cladera, Varun Murali, George J. Pappas, Vijay Kumar Published: 2026-09-24 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2609.30428v1 Foundation models provide robots with the ability to interpret natural language and reason about environmental context, yet most language-conditioned policies assume that goals are well-specified and that task-relevant information is provided upfront via a prior map. Operating in unfamiliar environments with underspecified tasks entails high contextual uncertainty: the robot must jointly infer what constitutes task success, what constitutes relevant information, and where (or whether) that information exists. We address these limitations via CLUE (Closed-Loop contextual Uncertainty rEsolution), a framework for actively resolving contextual uncertainty given underspecified tasks in natural language. CLUE uses an LLM-derived policy to hypothesize task-relevant concepts and potential plans. It then uses a language-embedded map, which is constructed online, to ground these hypotheses into actions. The policy sequentially evaluates hypotheses via closed-loop environment interaction and refines its plans as it gathers new information. We deploy CLUE on a Boston Dynamics Spot across three real indoor and outdoor environments spanning 15 tasks that require object disambiguation, functional inference, and occlusion reasoning. CLUE achieves a success rate within 7 percentage points of an oracle policy and outperforms an LLM-enabled planner without closed-loop feedback by a 4x margin. Supporting experiments demonstrate that simply building and then querying a language-enriched map is insufficient to resolve complex contextual planning tasks; these approaches achieve roughly one third the success rate of CLUE while requiring over 10x more VLM tokens. We provide additional information at https://zacravichandran.github.io/CLUE. --- ### Fake News Theories: Harnessing Disciplinary Insights for Computational Modeling, Detection, and Explanation Authors: Zhaoyang Cao, Miriam Metzger, Reza Zafarani Published: 2026-09-24 Categories: cs.LG, cs.CY arXiv: https://arxiv.org/abs/2609.30427v1 Disinformation research has produced increasingly accurate automated fake-news detectors, but many systems remain difficult to interpret and are weakly connected to established theories of persuasion, credibility, and human judgment. In this paper, we develop a theory-informed computational framework that translates cross-disciplinary theories of fake news into measurable features for automated detection and explanation through statistical techniques and large language models. To that end, we conduct a structured cross-disciplinary review of theories from social sciences, psychology, economics, among other disciplines that reveal how fake news persuades and spreads, thereby establishing a broad theoretical foundation for computational modeling. Experiments on benchmark datasets show that theory-derived features are predictive and provide interpretable, theory-referenced diagnostic signals. Multi-feature models generally outperform individual features, although gains among the strongest small feature combinations are modest. Our work highlights the value of interdisciplinary perspectives in building robust and interpretable fake news detection systems, advancing the foundation for human-centered approaches in combating disinformation. --- ### Understanding Perturbed Parameter Ensemble Sensitivities Using A Contrastive Learning Approach Authors: Da Fan, David John Gagne, Gregory S Elsaesser, Brian Medeiros, Addisu G Semie, Qingyuan Yang, Akila Sampath, Subashree Venkatasubramanian Published: 2026-09-24 Categories: physics.ao-ph, cs.AI arXiv: https://arxiv.org/abs/2609.30420v1 Perturbed parameter ensembles (PPEs) reveal how physics parameters affect climate simulations, but interpreting parameter sensitivities across multivariate, spatially structured outputs remains challenging, particularly when calibrating models against observations. We develop an explainable contrastive learning model that maps 5 monthly cloud and radiation fields into a shared representation space. We train the model on the fields of two 100-member Community Atmosphere Model version 6 (CAM6) PPEs, spanning 34 parameters, that only differ in the warm rain microphysics scheme: KK2000, the default bulk microphysics scheme, and TAU-ML, a neural network emulator of a bin microphysics scheme. The learned representations separates two PPEs with over 94\% linear classification accuracy while preserving the seasonal variability and ensemble spread due to parameter perturbations. In the shared representation space, the representations of satellite observations occupy the same low-dimensional manifold as the PPEs but are displaced from them most strongly during boreal spring and autumn. TAU-ML PPE has a lower distance to observations compared to KK2000 in the representation space. Integrated Gradients attributions highlights the contributions in subtropical low-cloud regions, Northern and Southern Hemisphere storm track regions, and tropical convection regions to differences between PPEs and observations. Regional attributions correlate most strongly with parameters associated with cloud microphysics, boundary layer turbulence, and deep convection. These results demonstrate that explainable representations of climate fields can attribute model differences to specific variables, regions, seasons, and physical parameters. --- ### Electric Vehicle Charging Station Location Selection using Geospatial Artificial Intelligence (GeoAI) Authors: Eun Hak Lee, Euntak Lee Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.30417v1 As electric vehicle (EV) adoption increases, ensuring efficient and well-distributed charging infrastructure has become a critical challenge. While many EV charging station location problem (CSLP) studies focus on minimizing costs or travel distance, it is crucial to consider the surrounding geospatial characteristics of existing stations that influence operational performance. This study proposes a geospatial artificial intelligence (GeoAI)-based framework that integrates high-dimensional EV-related geospatial data, including EV usage, land-use, population, and traffic attributes. We incorporate a variational autoencoder (VAE) and a graph convolutional network (GCN) into the model to capture similarities among existing charging stations, and to identify suitable locations for future stations. The VAE compresses high-dimensional EV input data into a low-dimensional latent space, and the GCN uses this latent representation to predict locations suitable for charging stations. Using real-world data from Bryan-College Station, Texas, US, the proposed model outperforms state-of-the-art baselines, achieving an F1-score of 0.87 in distinguishing existing station locations from non-station locations. The model also identifies 27 additional candidate locations that show geospatial characteristics similar to those of existing stations, based on a similarity score. We further evaluate two policy implementation scenarios, maximizing geospatial similarity and minimizing total travel distance, each yielding different outcomes aligned with distinct strategic objectives. The findings highlight the importance of incorporating spatial context into CSLP and provide valuable insights for future EV infrastructure planning, promoting both efficiency and accessibility in the rapidly growing electric mobility sector. --- ### All In Good Time: Causality-Aware Framework for LLM-Based Simultaneous Speech-to-Speech Translation Authors: Amir Hussein, Enas Albasiri, Travis M. Bartley, Nourchene Ferchichi, Ke Hu, Harishchandra Dubey, Myungjong Kim, Zhehuai Chen, Oluwatobi Olabiyi, Sanjeev Khudanpur Published: 2026-09-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.30416v1 Large Language Models (LLMs) have shown strong performance in low-resource offline translation; however, extending them to simultaneous speech-to-speech translation (Simul-S2ST) remains challenging due to the scarcity of causally aligned training data with high cross-lingual speaker fidelity. In addition, existing approaches rely on fixed translation policy or confidence heuristics, leading to suboptimal quality and higher latency. We propose a causality-aware Simul-S2ST framework with a novel data pipeline that generates high-fidelity, causally aligned segments with improved voice transfer. The framework introduces (i) a factorized S2ST architecture (FAST), (ii) a causality-aware adaptive policy (CAP), and (iii) causality-aware latency metric. Experiments on CVSS Spanish, German, and French show that FAST-CAP consistently improves the quality-latency trade-off, achieving up to +1.2 BLEU and a 26% relative latency reduction over a fixed policy. Despite using substantially less training data than existing systems, FAST-CAP achieves state-of-the-art results in speech translation quality and speaker fidelity while yielding up to a 38.8% relative reduction in latency. --- ### A Unified Account of Concepts and Chunks Authors: Karthik Singaravadivelan, Pat Langley Published: 2026-09-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.30414v1 Cognitive psychology has studied how people encode, use, and learn concepts that describe categories, and how they represent, recognize, and acquire chunks for familiar patterns of elements. The literatures on these two topics are nearly disjoint, which poses a challenge for unified theories of cognition. In this paper, we review Cobweb, a computational account of categorization and concept formation, and propose an extended theory that incorporates chunks and their acquisition. The theory makes no commitments about modality, applying to any experience that decomposes into elements and relations among them. We also present \trellis/, an implementation of this theory, and illustrate its application to learning context-free grammars, which we adopt as a testbed because they involve both concept-like and chunk-like elements. In addition, we report experimental results on three synthetic grammars that demonstrate the system's ability to represent syntactic knowledge, use it to parse and generate sentences, and learn compositional structures from sample parses. We conclude by discussing related work on concepts and chunks, along with directions for future research in the area. --- ### Adaptive Multi-Value Control in LLMs via Causal Activation Steering Authors: Payel Bhattacharjee, Ravi Tandon Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.30405v1 Large language models (LLMs) are increasingly deployed in settings where responses must reflect multiple, potentially interacting social norms and human values. Activation steering offers a lightweight alternative to training-based alignment by modifying internal activations at inference time. However, prior human-value steering methods have largely considered values in isolation, while direct composition of multiple directions relies on fixed intervention strengths that cannot respond to the model's evolving internal state. Motivated by this key observation, we introduce AIMES, a framework for adaptive multi-value activation steering. AIMES constructs layer-specific bipolar directions for moral-foundation values and uses intermediate-layer vocabulary readouts as online observers. An observer-guided controller then adapts the strength of each requested value intervention at every decoding step based on its current observed state, without training a separate value-state estimator. Across multiple instruction-tuned model families, value combinations, and intervention depths, we find that multi-value controllability varies across both value combinations and intervention locations. Compared with fixed joint steering and prompt-based steering, AIMES shows depth-dependent advantages that are broadly supported across two independent evaluators, with some variation in the precise depth at which specific control effects emerge. These advantages come with smaller realized activation-space interventions than fixed-joint steering and comparable response quality. Overall, our results suggest that online observer feedback can provide lightweight, state-aware adaptation for single-pass multi-value steering. --- ### What Improves Multimodal Misinformation Detection? Answers from a Large-Scale Empirical Study Authors: Akshit Sharma, Prashant W. Patil Published: 2026-09-24 Categories: cs.CV, cs.AI, cs.CL, cs.LG, cs.MM arXiv: https://arxiv.org/abs/2609.30402v1 Multimodal misinformation is increasingly crafted to look convincing by pairing a textual claim with an image that appears to "prove" it. Yet in practice, building effective detectors often hinges on a small set of design choices that are rarely examined in a controlled way. In this paper, we conduct a large-scale study of multimodal design choices for misinformation detection with over 3,375 experiments- spanning three benchmark datasets and a broad range of pre-trained vision and language backbones. Through systematic comparisons and targeted robustness analyses, we distill practical guidance on which design choices help, when do they fail silently, and what aspects of the pipeline most strongly shape model behavior, answering 4 key Research Questions (RQs). We aim to provide a reliable foundation for designing stronger and more dependable multimodal misinformation detection systems, thus contributing to the broader research community. --- ### A Synthetic Ground-Truth Framework for the Evaluation of Explainable AI Methods Authors: Miquel Miró-Nicolau, Francesco Spinnato, Riccardo Guidotti Published: 2026-09-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.30397v1 Evaluating explainable Artificial Intelligence (XAI) methods is a challenging task due to the lack of reliable evaluation procedures and, in particular, the absence of ground truth explanations. In the literature, existing evaluation approaches typically assess explanations by measuring their fidelity with respect to the predictions of a black-box model. However, such evaluation strategies only quantify the degree to which an explanation reproduces the model's output, without ensuring that the explanation correctly reflects the underlying decision process. As a consequence, different explanations may achieve similar fidelity scores while providing inconsistent or misleading interpretations of the model behavior. In this paper, we propose a framework for the evaluation of XAI methods based on synthetic ground truth. The proposed approach relies on controlled interventions to generate synthetic datasets in which the importance of input components can be determined by design. This enables the construction of ground truth explanations that are directly aligned with the behavior of the model under analysis. The framework is instantiated across three data domains, namely binary images, tabular data, and time series, allowing a comprehensive assessment of explanation methods in heterogeneous settings. Experimental results obtained by evaluating nine widely used XAI methods show significant limitations in current techniques and highlight the importance of synthetic, intervention-based benchmarks for a reliable assessment of explanation quality. --- ### An End-to-End Pipeline for Causal ML with Continuous Treatments: An Application to Financial Decision Making Authors: Javier Moral Hernández, Clara Higuera-Cabañes, Álvaro Ibraín Published: 2026-09-24 Categories: stat.ME, cs.LG, stat.AP, stat.ML arXiv: https://arxiv.org/abs/2609.30396v1 This paper presents an end-to-end causal machine learning (ML) pipeline designed for real-world applications with continuous treatments. The proposed framework consists of six sequential steps: dimensionality reduction, causal identification, positivity assumption violation handling, estimation, refutation and evaluation, and policy optimization. We introduce practical contributions not currently available in existing causal ML toolkits, specifically: (1) a method for detecting and quantifying positivity violations in continuous treatment settings (2) a novel, scalable two-stage dimensionality reduction framework tailored for causal inference with high-dimensional data; (3) the adaptation of sensitivity analysis and estimation methods originally designed for binary treatments to the continuous treatment space and (4) an end-to-end integration of these components into a modular, reproducible workflow. These innovations address real-world challenges in causal inference that are often not covered in theoretical frameworks but frequently encountered in industrial applications. The methodology is validated with a synthetic dataset inspired in a real-world financial debt collection use case, however its design can be applied to analogous problems across different industries. Results demonstrate that the proposed methodology offers a more computationally efficient approach and produces less biased estimates compared to standard methods for problems with continuous treatment and high-dimensional data. A fully functional GitHub repository with documented code and numbered notebooks is made available ensuring reproducibility and practical implementation. The pipeline presented is intended to contribute to closing the gap between academic approaches and practical application in industry contexts where causal ML can be highly beneficial such as the financial sector. --- ### CSCWD: Cross-Scale Channel-wise Knowledge Distillation for Lightweight Tiny Object Detection on Edge Devices Authors: Amir Zamani, Zeinab Ghasemi-Naraghi Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.30395v1 Real-time tiny object detection in aerial imagery is constrained by the weak spatial evidence of very small objects and the loss of high-resolution detail in lightweight detectors. This study presents Cross-Scale Channel-wise Knowledge Distillation (CSCWD), a training-time framework that transfers high-resolution spatial representations from a YOLO11m-P2 teacher to a compact YOLO11n student without altering the student's inference architecture. Unlike conventional same-scale feature distillation, CSCWD transfers supervision from teacher P2 to student P3 after feature alignment while retaining same-scale distillation at deeper pyramid levels. Under the unified seven-sequence Drone-vs-Bird validation protocol, YOLO11n-CSCWD achieves 50.17% mean average precision at an intersection-over-union threshold of 0.5 (mAP@0.5) and 59.73% recall, improving the matched CA-YOLO11n baseline by 2.92 percentage points in mAP@0.5 and 3.55 points in recall. Cross-scale alignment further increases mAP@0.5 by 2.09 points over the corresponding same-scale channel-wise distillation configuration. In zero-shot evaluation on DUT-Anti-UAV, mAP@0.5 increases from 48.29% to 50.06% without target-domain fine-tuning. This domain was included because its challenging small targets make low-latency, computationally efficient detection particularly relevant. On Raspberry Pi 5 using NCNN-FP16 at 640x640 resolution, the 2.58-million-parameter student achieves 50.32% mAP@0.5 at 82.32 ms mean wall-clock latency, or 12.15 frames per second, while retaining essentially the same runtime and memory requirements as the matched baseline. The results support cross-scale distillation for improving tiny-target detection without increasing inference-time model complexity. --- ### LiTe-GS: Oracle-Efficient Next Best View Selection for 3D Gaussian Splatting Authors: Vivek Pandey, Amirhossein Mollaei Khass, Nader Motee Published: 2026-09-24 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2609.30393v1 Selecting informative camera views is critical for efficient training and adaptive refinement in 3D Gaussian Splatting, where each observation significantly influences model parameters. However, information-driven view-selection strategies can require repeated evaluations of expensive information-gain oracles as the number of candidate views increases. We propose LiTe-GS, an oracle-efficient method for next best view selection in 3D Gaussian Splatting. LiTe-GS reduces the number of information-oracle evaluations by performing randomized subset evaluation of candidate views rather than exhaustively scoring the full candidate pool. The resulting approach achieves expected $O(M\log(1/ε))$ oracle complexity with respect to the number of candidate views $M$, independent of the selection cardinality $K$, while providing an explicit trade-off between oracle efficiency and approximation quality through $ε$. We provide theoretical guarantees on oracle complexity and approximation performance under the proposed selection scheme. Experiments on Blender and Mip-NeRF 360 demonstrate that LiTe-GS maintains reconstruction quality comparable to Fisher-information-based baselines while substantially reducing the number of Fisher-oracle evaluations across different acquisition settings. --- ### From Weak Data to Strong Policy: Q-Targets Enable Provable In-Context Reinforcement Learning Authors: Yichen Lin, Xuyuan Xiong, Xue Wang, Xiangfu Meng, Mike Mingcheng Wei, Tao Yao Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.30391v1 Existing in-context reinforcement learning methods mainly pretrain Transformers with supervised behavior-prediction objectives. This enables task inference from context, but makes the learned policy strongly depend on the quality of offline actions: when trajectories are weak or suboptimal, imitation itself becomes a biased learning signal. We propose Q-Target Pretrained Transformers (QTPT), which keeps the context-conditioned Transformer architecture but replaces behavior cloning with a Bellman-style Q-target objective. QTPT therefore learns to use rewards and transitions in the context to estimate action values, rather than simply imitating the behavior policy. We theoretically analyze QTPT in stochastic linear bandits and finite-horizon MDPs, showing stronger robustness to data quality than supervised pretraining. Empirically, QTPT improves over supervised behavior prediction on controlled RL benchmarks with random or suboptimal data, and we examine extensions to D4RL Kitchen and AntMaze. Supplementary experiments evaluate backbone robustness, meta-RL comparisons, task-coherent context, and unsupported-action value overestimation. These comparisons distinguish the benefits of Q-target pretraining from the remaining limitations of offline coverage. --- ### Stealth Apart, Harm Together: Skill Cascading Attacks on Skill-Based Agent Systems Authors: Zihao Zhu, Siwei Lyu, Adel Bibi, Baoyuan Wu Published: 2026-09-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.30383v1 A skill is a modular package of natural-language instructions, executable scripts, and reference resources that an agent can load at runtime to extend its capabilities for a specific task. Skill-based agent systems therefore enable flexible reuse of third-party capabilities, but the openness of this skill ecosystem also opens up a new attack surface. Prior work has focused on vulnerabilities within individual skills, but little attention has been paid to risks that arise from interactions across skills. In this paper, we introduce skill cascading attacks, a threat paradigm in which a malicious objective is distributed across multiple skills so that each modification looks benign in isolation, yet their combined execution is harmful. For instance, in a prescription-review pipeline, the first skill weakens signals of recently discontinued medications in the extracted history, the second downgrades the severity of any drug interaction tied to them, and the third suppresses the resulting low-priority alert in the final summary, so that a severe drug-interaction warning silently disappears before reaching the physician. To systematically study this safety blind spot, we develop SkillCascade, an automated multi-agent red-teaming framework, and release SkillCascade-Bench, a benchmark of 213 validated cascading test cases across multiple agent systems and domains. Across representative agents (e.g., OpenClaw, Claude Code, Codex) and LLM backbones, cascaded interactions reliably induce harmful behaviors while evading existing per-skill scanners and runtime monitors. Our findings highlight a gap between component-level integrity and system-level safety, and call for defenses that reason over cross-skill interactions rather than individual skills in isolation. --- ### DanLing NestedTensor: Composable Multi-Ragged Tensors for Deep Learning Authors: Zhiyuan Chen Published: 2026-09-24 Categories: cs.LG, cs.AI, cs.PF arXiv: https://arxiv.org/abs/2609.30379v1 Variable-size inputs are common in deep learning, but dense batching allocates a shared envelope and spends computation on padding. The cost multiplies across varying axes: an explicit pair state allocates $BN_{\max}^2$ positions instead of $\sum_i N_i^2$. Packing removes that waste, but composing packed operations still requires the logical axes and sample boundaries a flat buffer no longer exposes. We present DanLing NestedTensor, a PyTorch tensor abstraction that makes multi-ragged structure a property of the tensor itself. Packed values carry tensor-backed partitions and logical dimension order, so broadcasting creates ragged axes, feature transformations retain them, and reductions consume them. The same representation carries through autograd and both eager and compiled execution. On an A100, the geometric-mean speedup over same-mode padding is 2.74$\times$ eager and 3.39$\times$ compiled across four BERT scales, and 1.97$\times$ eager across four FCN backbones. A four-block Pairformer-style workload runs 2.40-4.32$\times$ faster than a padded reference using native PyTorch kernels across square length regimes in eager execution, with peak allocation falling from 38.08 to 5.41 GiB on its high-variation batch. The tensor interface lets model code built from its supported operators compose efficient variable-size computation without managing offsets at any call site. Code will be released publicly upon publication. --- ### Cost-Aware Best-LLM Identification using Dueling Feedback Authors: Sarvesh Gharat, Nikhil Karamchandani, Jayakrishnan Nair Published: 2026-09-24 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2609.30360v1 Inspired by the problem of identifying the best model from a collection of large language models (LLMs) with heterogeneous querying costs, we formulate and analyse a variant of the multi-armed bandit (MAB) with (i) dueling feedback, where pairwise comparisons between model responses provide robust preference signals, and (ii) heterogeneous sampling costs, reflecting the differing costs of querying different LLMs. Assuming the existence of a Condorcet winner, a condition we empirically validate across multiple real-world datasets, we propose a Track-and-Stop style algorithm for best-arm identification with prescribed confidence. We prove that the algorithm almost surely achieves the asymptotically optimal cost as the error tends to zero. Finally, we extensively evaluate our approach on both synthetic and real-world instances, demonstrating consistent improvements over classical cost-unaware algorithms and their cost-aware extensions. --- ### AlphaEarth distinguishes cities but compresses urban variation Authors: Andrew Renninger Published: 2026-09-24 Categories: cs.CV, physics.soc-ph arXiv: https://arxiv.org/abs/2609.30356v1 Cities differ in built form, land cover and development history, complicating comparison across places and time. Satellite foundation models map Earth's surface onto common numerical representations. Yet the tasks and targets used to shape them typically do not focus on cities: globally consistent labels for urban function do not exist, and many datasets - especially land cover and land use classifications - collapse the built environment into few classes. Here we audit the representation, focusing on AlphaEarth but with broader applicability to other Earth embeddings, by probing the geometry and geography of embeddings for 1,000 urban areas in 162 countries. We find that cities occupy a shifted but overlapping region on the hypersphere, 62.7° from the global mean direction, and continent and climate predict 24.3% of variation among the mean directions of urban centres in excluded countries. Inside cities, degrees of urbanisation carry 8.9% of the variation, and what they leave holds shared directions whose local orientation varies, not one universal axis of urbanisation. Retained variation is itself unequal: dispersion within urban centres is 14.1% greater per standard deviation of national development, even after adjusting for population, land area and continent. Further controls suggest cities in developing countries present less contrast in vegetation and texture, and dispersion follows that contrast: full adjustment for it leaves at most 6.4% of the gradient. Annually, a city's representation moves nearly eight times more than redrawing its own pixels explains, and contracts where the 2022 loss of Sentinel-1B removed a pass direction. AlphaEarth's representations therefore support comparison across regions, while the differences between its annual layers are not yet validated for comparison over time. --- ### Strategic Self-Consistency Authors: Tori Qiu, Ander Artola Velasco, Manuel Gomez-Rodriguez Published: 2026-09-24 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2609.30352v1 Self-consistency has become a popular technique for enhancing the reasoning abilities of large language models by generating multiple reasoning paths and selecting the final answer through a majority vote. However, because model providers typically charge users in proportion to the number of reasoning paths generated, they have a financial incentive to artificially increase the path count. In this work, we show that an unfaithful provider can exploit this incentive using a simple, efficient algorithm while avoiding detection by an auditor: by generating and strategically reordering additional reasoning paths, the algorithm makes every path appear necessary to reach the majority. To validate our algorithm, we conduct experiments with multiple instruct models from the Llama and Qwen families, as well as reasoning models distilled from DeepSeek-R1, on benchmark datasets spanning mathematics, science, and question answering. Our results suggest that the distribution of additional reasoning paths generated by our algorithm is heavy-tailed and that substantial capacity to overcharge remains even under the best possible audit designed to keep the false-positive rate below $α= 0.1$. ---