# Research Papers — 2026-07 · bundle page 102 of 222 > Full-text markdown of papers #5051–5100 of 11075 published in 2026-07, newest first. AI-readable mirrors of the full 11075-paper 2026-07 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-07/page/102 [← Previous bundle](https://aipod.dev/papers/archive/2026-07/page/101.md) [Next bundle →](https://aipod.dev/papers/archive/2026-07/page/103.md) ### Gradient-Energy Guided Block-Wise Perturbations for Sharpness-Aware Minimization Authors: Zhen Huang, Jiaxin Deng, Junbiao Pang Published: 2026-07-17 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2607.18306v1 Sharpness-Aware Minimization (SAM) improves generalization by minimizing the worst-case loss in a local parameter neighborhood. Standard SAM implicitly allocates its global perturbation budget across parameter blocks according to instantaneous minibatch gradient norms. Such an allocation can be noisy and may not reflect the sensitivity that blocks accumulate throughout training. We propose Gradient-Energy Adaptive Radius SAM (GEAR-SAM), which maintains an exponential moving average (EMA) of squared block gradients as a lightweight, curvature-related sensitivity signal and allocates the fixed SAM budget through a closed-form constrained optimization. GEAR-SAM preserves the global SAM radius, requires no Hessian-vector products or explicit Fisher estimation, and adds only scalar state beyond SAM. Experiments on image classification, transfer learning, noisy-label learning, and partition studies demonstrate improved generalization and robustness across architectures and tasks. More broadly, GEAR-SAM provides a dynamic view of sharpness-aware optimization: a fixed perturbation budget should be redistributed as the sensitivity of functional network blocks evolves during training. --- ### The Information Shadow: Measuring Structural Limits on What Language Models Can Learn Authors: Priyansh Srivastava, Romit Chatterjee Published: 2026-07-17 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2607.18305v1 Some limits on what language models know are not gaps in data coverage but structural properties of learning from text. We introduce the information shadow: the region of phenomena that a text-trained learner cannot acquire regardless of scale, comprising (I) structures language cannot express, (II) functions that are statistically non-identifiable from the training distribution, and (III) functions that are representable but unreachable by gradient-based training. We give each type a probe that is decisive because the premise of the shadow is, in that setting, provable. For Type I, Language Compression Residuals compare a text learner, which sees only a lossy text-like encoding of the signal, against a full-signal learner, which sees the underlying signal directly. The text learner sits at a computable expressibility ceiling while the full-signal learner pulls away by a gap that stays flat across 300x more data, so the deficit is a property of the channel, not of training. For Type II, the Counterfactual Distinction Test trains models on data exactly consistent with two incompatible rules. Across a provable string task and a language-like agreement task, behavior on counterfactuals is set by the model's inductive bias, while 5% disambiguating data steers the learned rule bidirectionally to either target (r = +/-1.0, p < 1e-10). For Type III, Basin Escape Mapping exhibits a function that is representable at 100% (by hand construction) yet reached 0% of the time by standard training and instantly from a nearby initialization, with width scaling providing no help (p = 1.6 x 10^-14). Each effect is isolated by a control that rules out a capacity or modality artifact. We release the probe suite and discuss implications for benchmark design, capability auditing, and shadow-aware uncertainty. --- ### TD-DPO: Difference-Aware Preference Optimization for Mitigating Sycophancy in Clinical Autism Intervention Dialogue Authors: Shuzhong Lai, Junhong Lai, Chenxi Li, Qing Zhou, Haifeng Li, Gang Pan, Lin Yao, Yueming Wang Published: 2026-07-17 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.18304v1 The sycophancy of large language models can increase the safety risk in intervention dialogue for autistic children. Supervised fine-tuning can somewhat reduce sycophancy, but relying solely on positive examples is often insufficient to identify and correct failure patterns. We observe that sycophancy behaviors can often be localized to a limited span within the model response. In this regime, sequence-level preference optimization can over-update preference-irrelevant tokens and degrade intervention ability. To address this, we propose the \textbf{M}inimal \textbf{E}dit \textbf{D}ata \textbf{A}ugmentation (MEDA) strategy to construct controlled, stable, minimal edit preference pairs and \textbf{T}oken-level \textbf{D}ifference \textbf{D}irect \textbf{P}reference \textbf{O}ptimization (TD-DPO), which upweights difference tokens between chosen and rejected responses while downweighting shared tokens to suppress background drift. Extensive experiments across multiple backbones and evaluators show that TD-DPO achieves a better trade-off between sycophancy mitigation and intervention ability retention in our offline settings, highlighting its potential as a practical alignment approach for autism intervention. --- ### Fretiq: Browser-Native Electric Guitar String Classification via Engineered Spectral Features and Held-Out Free-Play Evaluation Authors: Aadi Garg Published: 2026-07-17 Categories: cs.SD, cs.LG, eess.AS arXiv: https://arxiv.org/abs/2607.18303v2 Identifying which string produces a given pitch in monophonic electric guitar audio is a classification challenge: a single pitch can often be produced on multiple strings, with timbral differences largely imperceptible to untrained humans. We present Fretiq, a preliminary single-instrument, single-player browser-based string classification system using a 26-dimensional feature representation of frequency band energies, spectral statistics, and 13 Mel-Frequency Cepstral Coefficients. Across five seeds, a shuffled frame-level validation split yields 97.25 +/- 0.32 percent accuracy, with an ablation study identifying MFCCs as the primary accuracy driver (92.09 +/- 0.50 percent without MFCCs). We introduce Comparison Training, a data collection method recording same-pitch pairs on adjacent strings in deliberate alternation. An initial shuffled-split comparison found no net benefit but was confounded by non-comparable validation sets across conditions. A corrected matched evaluation, using an identical recording session held out from training and model selection, shows including comparison-session data improves accuracy by 25.78 +/- 1.46 percentage points; a size-matched control shows this is not explained by training-set size alone. A recording-session-held-out evaluation yields 86.53 +/- 1.23 percent accuracy, closely matching an independently collected free-play evaluation (87.8 percent), both well below the shuffled-split figure, showing shuffled validation substantially overestimates real generalization here. We describe the feature extraction pipeline in Python and TypeScript for training-inference parity and document two implementation failure modes. The system runs entirely in-browser with no specialized hardware required. --- ### A Better Start for Language Models: Domain-Conditional Position Offsets Authors: Ye Qiao Published: 2026-07-17 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.18302v1 Autoregressive language models are least accurate at the beginning of a sequence, where little context forces reliance on a generic pretraining prior. We show that this cold-start penalty is domain dependent and reduce it with a domain-conditional position offset: a single learned vector added to the embedding activation at the first sequence positions while all model weights remain frozen. The offset trains in minutes on roughly one hundred documents, switches between domains without added sequence state, and has no measurable latency overhead. Across eight Mamba, GPT-NeoX, and Llama models spanning 410M to 8B parameters, it reduces held-out in-domain perplexity by up to 27%; the effect persists at 70B, and one position captures most of the benefit. A matched, converged direct logit-bias correction reaches at most only 7.9% and leaves later-token loss unchanged, showing that the offset propagates through model state rather than merely recalibrating the output prior. A tuned LoRA reaches lower perplexity but uses two to three orders of magnitude more parameters and an active low-rank weight path, while soft prompts add sequence positions. With wrong-domain controls, offsets improve retrieval reranking and domain classification when decisions depend on early in-domain tokens, For the few-shot reasoning whose signal occurs later, the results maintains unchanged. Position-aware prefill application also help generation tasks, whereas naive application at every cached decoding step causes repetition. The offset is therefore not the strongest adapter, but a lightweight, hot switchable tool for short in-domain scoring and calibration. --- ### Mitigating Compiler Fusion-Induced Power Bursts in Mobile NPU Inference as the Battery Depletes Authors: Ryoga Yuzawa, Masayoshi Tomizuka Published: 2026-07-17 Categories: cs.AR, cs.AI arXiv: https://arxiv.org/abs/2607.16555v1 Mobile devices increasingly rely on real-time NPU inference for camera and perception workloads. Under low-voltage conditions, however, a single inference can induce an instantaneous voltage droop in the power-delivery network, causing the power management integrated circuit to invoke dynamic voltage and frequency scaling (DVFS) and increase latency. We present a measurement study of this effect on a commercial smartphone. We show that aggressive operator fusion in a mobile NPU compiler can create monolithic superlayers whose concentrated execution produces large peak-current bursts. These bursts shift the DVFS-onset voltage upward and reduce the low-voltage operating margin. We further evaluate a practical black-box mitigation: a measurement-guided, pre-compilation graph rewrite that inserts barriers at selected peak-to-average power ratio hot spots to prevent harmful superlayer merging in the vendor NPU compiler. On Snapdragon 8 Gen 3 with MobileNetV4 at 768 x 768 resolution on ImageNet-1k, this method reduces peak current from 3.12 A to 1.94 A with 3.76% latency overhead, preserves stable latency deeper into the low-voltage regime, and shifts the inferred DVFS margin by approximately 173 mV. --- ### Capacity and Redundancy Trade-offs in Multi-Task Learning Authors: Asif Khan Published: 2026-07-17 Categories: cs.LG, cs.AI, cs.IT arXiv: https://arxiv.org/abs/2607.16554v1 In multi-task learning (MTL) negative transfer is often considered as an optimization artifact, but it can also be viewed as a consequence of limited shared capacity and weak task redundancy. We investigate this effect through a Capacity--Redundancy (CR) identity that decomposes the sum of per-task predictive informations into joint predictive information that includes label redundancy defined via total correlation (TC), and a residual coupling term that quantifies interference left unresolved by the shared representation. Additionally, we show two key results: (i) a clustering-gap decomposition that gives a necessary and sufficient condition for clustered sharing to outperform global sharing, and (ii) a gradient--TC bridge in a Gaussian multi-task model that formally justifies gradient cosine similarity as a proxy for redundancy ordering. Empirically, we estimate the residual coupling $Δ$ from validation residual correlations, showing that clustered LoRA substantially reduces $\widehatΔ$, outperforms size-matched random partitions, and results in statistically significant gains with multi-seed confidence intervals. --- ### Discrete Ricci Curvature on Protein Contact Graphs for Lightweight Fold Classification Authors: Jianru Shen Published: 2026-07-17 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.16553v1 Protein fold classification can be approached via sequence-based representations or structural descriptors, but direct comparisons between lightweight handcrafted descriptors and pretrained protein language model embeddings remain limited. We investigate discrete Ricci curvature on Calpha contact graphs as a lightweight structural descriptor for fold classification. Each protein domain is represented by a 22-dimensional fixed-length feature derived from summary statistics and quantiles of Ollivier-Ricci and Forman-Ricci edge curvature distributions. We evaluate on CATH top-10 Topology classification and on the ASTRAL 40%-identity SCOPe top-10 Fold benchmark, comparing against geometry, contact-graph statistics, persistent homology, and mean-pooled ESM-2 (150M) baselines. On both datasets, lightweight structural descriptors substantially outperform mean-pooled ESM-2 embeddings, with a larger performance gap on the ASTRAL 40% SCOPe benchmark. Ricci alone uses 22 dimensions, or 3.4% of the ESM-2 baseline dimensionality, and already outperforms mean-pooled ESM-2 on both datasets. Combining Ricci with persistent homology yields the strongest performance, achieving macro-F1 of 0.71 on CATH and 0.68 on SCOPe with a 112-dimensional feature vector. These results identify a regime where lightweight interpretable graph descriptors offer a practical alternative to pretrained protein language model embeddings. --- ### Encoding EEG Signals to Examine Human-Like Next-Word Prediction Behaviour in Language Models Authors: Boi Mai Quach, Binh T. Nguyen, Cathal Gurrin, Graham Healy Published: 2026-07-17 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.16549v1 Language models (LMs) are trained to excel at predicting the next word in the sequence given prior context, and humans also share this predictability in reading comprehension. Neuroscience research reveals that next-word predictability influences brain response, as recorded at millisecond resolution using electroencephalography (EEG). While our evidence indicates that advanced LMs achieve accuracies closely aligned with human performance at the next-word prediction task, this raises the question: Does higher prediction accuracy necessarily mean that these models adequately capture the cognitive signals associated with human reading comprehension? Here, we generate regressors for both humans and LMs based on two information measures, including top-1 prediction and surprisal, to predict event-related potential (ERP) elicited from EEG recordings which reflect different stages of cognitive processing during reading. We argue that modelling ERP patterns offers fine-grained analysis of the cognitive plausibility of various LMs during reading. Our results indicate that only surprisal potentially correlates with language-processing ERPs, especially for open-class words with high semantic content. Moreover, our findings challenge the assumption that scaling LMs with increased parameters and computational budgets will consistently lead to improved convergence with human-like linguistic processing. --- ### Projective Maximum Entropy: Universality and Acceptance-Region Calibration Authors: Hideitsu Hino Published: 2026-07-17 Categories: math.ST, stat.ML arXiv: https://arxiv.org/abs/2607.16547v1 Maximum-entropy reference distributions are usually constructed on the normalized probability simplex. This formulation is less natural for unnormalized statistical models, in which positive multiples represent the same shape, and it does not directly explain how a prescribed admissible region should determine the deformation parameter of a bounded-support reference distribution. We formulate maximum entropy on the projective space of nonnegative measures and establish three results of statistical relevance. First, a universality theorem shows that every admissible monotone transform of the same normalized power functional has exactly the same optimizer under linear moment constraints. The result unifies the maximum-entropy implications of Tsallis and Rényi entropies, Hölder composite scores, pseudo-spherical scores, Bregman--Hölder constructions, and related homogeneous divergences without asserting a new distribution family. Second, the common optimizer is characterized as a $q$-exponential density; under mean and covariance constraints it is a compactly supported $q$-Gaussian for positive deformation and a Student-type density for negative deformation. Third, a prescribed Mahalanobis acceptance region with squared radius $R^2>d+2$ uniquely determines the deformation parameter $γ_R=2/(R^2-d-2)$. The resulting affine-equivariant reference density is the unique projective maximum-entropy solution, and its support coincides with the specified ellipsoid without an additional support constraint. This provides a principled method for constructing bounded-support statistical reference distributions from robust location and scatter estimates or from externally specified admissible regions. --- ### AffectFuse: Cross-Task Feature Fusion with Temporal Modeling for Multi-Task Affective Behavior Analysis Authors: Dipit Saha, Mohammad Raihan Rashid, Shah Mohammad Abdul Mannan, Ahnaf Tahmid, Md. Mehedi Hasan Published: 2026-07-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.16546v1 Affective behavior recognition in the wild requires joint prediction of continuous valence-arousal, categorical facial expression, and multi-label action units from unconstrained face images. We present our system for the Multi-Task Learning (MTL) track of the 11th Affective Behavior Analysis in-the-wild (ABAW) competition on s-Aff-Wild2, the static selected-frame version of Aff-Wild2. The method focuses on post-encoder adaptation: frozen AffectNet-supervised backbones provide multi-resolution features, while task-specific temporal heads and cross-task fusion modules select the useful signals for each target. For action-unit recognition, we adapt MAE-Face with Low-Rank Adaptation (LoRA) and use DISFA through per-unit expert routing rather than direct sequential transfer. Ablations over backbone, temporal, fusion, and AU-adaptation choices define the final configuration. The final system obtains P = 1.7302 on the official validation split, showing that post-encoder adaptation and task-wise modeling choices provide a strong MTL pipeline without training a new large-scale face foundation model. --- ### From Optimal Policies to Individual Differences: Rethinking Reinforcement Learning for Biology Authors: Patrick Govoni, Palina Bartashevich, Clémence Bergerot, Valerii Chirkov, Valentin Lecheval, Pawel Romanczuk Published: 2026-07-17 Categories: cs.NE arXiv: https://arxiv.org/abs/2607.16542v1 Reinforcement learning (RL) is primarily known as a computational method for optimizing control tasks, but it is increasingly used to explain biological behavior. While RL successfully captures key aspects of biology, a major gap remains: between-agent behavioral variability. Consistent individual differences naturally permeate biological populations, yet RL models typically present only the single best individual or the population average. Addressing this gap requires moving beyond current practices to generate behavioral diversity using biologically plausible mechanisms. Here, we examine approaches from various subfields of RL and outline potential paths forward to close the gap between biology and simulation. --- ### Do Vision Encoders Exhibit Human-like Color Thresholds? Authors: Engy Ehab, Pablo Hernández-Cámara, Nahla Belal, Jesús Malo, Javier Vazquez-Corral, Alexandra Gomez-Villa Published: 2026-07-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.16540v1 Understanding and characterizing human color perception is a longstanding research goal. One of the most traditional approaches is looking for the human color discrimination thresholds, the minimum chromatic differences perceptible to human observers. In recent years, deep neural networks have become the standard networks for computer vision tasks. In particular, deep vision encoders, foundation models trained on large-scale visual data, map images into latent feature representations. Despite the widespread use of deep vision encoders, few studies have investigated whether their internal representations exhibit human-like discrimination thresholds. In this work, we present a large-scale exploratory study probing the chromatic sensitivity of more than 50 pretrained vision encoders, including convolutional networks and vision transformers, against human discrimination thresholds. Using controlled chromatic stimuli at multiple chroma levels, we compare model-derived chromatic discrimination thresholds with human discrimination ellipses through a region-overlap metric (mIoU). Our analysis reveals generally weak alignment between model representations and human perceptual thresholds across all model families, with the best mIoU < 0.25. Moreover, we find that self-supervised encoders consistently outperform supervised ones, while language-supervised models show the most polarized behavior, occupying both the top and bottom of the ranking. These findings suggest that human-like chromatic sensitivity does not emerge naturally from current large-scale visual training objectives for any of the analyzed architectures. --- ### Building2Building: A Large Scale Benchmark for Generalizable Real-World Reinforcement Learning Authors: Vincent Taboga, Justin Veilleux, Doseok Jang, Anushree Rankawat, Pierre-Luc Bacon Published: 2026-07-17 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2607.16534v1 Reinforcement learning (RL) has achieved strong results in control, yet learned policies remain brittle to changes in dynamics, action spaces, observation spaces, or goals, a critical limitation for real-world deployment. Existing benchmarks offer limited diversity and complexity, making it difficult to rigorously study transfer, multi-task learning, and meta-learning in RL. We introduce Building2Building (B2B), a large-scale suite of realistic Heating, Ventilation, and Air Conditioning (HVAC) control environments built on EnergyPlus, a state-of-the-art building simulator. B2B is fully compatible with the Gymnasium interface and features a parametric building generator, enabling the systematic generation of diverse building configurations with heterogeneous observation and action spaces. Based on this suite, we define benchmark tasks targeting key open challenges in RL, including goal adaptation, dynamics adaptation, action-space shifts, and cross-domain transfer. By providing a large-scale, diverse, and physically grounded testbed with standardized evaluation protocols, B2B enables systematic investigation of generalization and transfer in continuous control. Beyond advancing research on generalization in RL, this new benchmark also carries significant societal implications by enabling improved HVAC control at scale, one of the most energy-intensive systems in buildings. --- ### Nonuniformity Principle in Human-AI Coworking Authors: An Luo, Jie Ding Published: 2026-07-17 Categories: cs.AI, stat.ME arXiv: https://arxiv.org/abs/2607.16530v1 As generative AI is increasingly applied to automate multi-step and high-stake workflows, human judgment and involvement remain essential for ensuring the quality of AI-generated outputs. In practice, while it is desirable for human experts to provide oversight on AI regularly, often by reviewing intermediate outputs, giving feedback, making corrections, and steering subsequent steps, such oversight is constrained by the time and resources that humans can afford. This creates a tension between the need for human oversight and AI's efficiency in delivering more output with less intervention. An important but underexplored question, then, is how to optimally engage humans in human-AI coworking. This work was originally motivated by our empirical observation that in long AI workflows, human oversight often improves user satisfaction while reducing unnecessary rework and token consumption. From there, we formulate the problem of where to place oversight stages in human-AI coworking. Under reasonable assumptions, we then develop the nonuniformity principle, which states that the optimal schedule places oversight stages with non-decreasing gaps along the workflow. We empirically validate this principle in two common AI agent workflows: writing literature reviews and constructing websites. --- ### Hierarchical Domain Generalization Authors: Chenxiao Yang, Zhiyuan Li, Shai Ben-David, Nathan Srebro Published: 2026-07-17 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.16528v1 We study hierarchical domain generalization as a problem of extrapolation from finite observed regions to an entire instance space, replacing i.i.d. sampling with arbitrary domain hierarchies. We show that the central obstruction is not only the complexity of the hypothesis class, but the train/test domain partition through which evidence is revealed. In particular, no matter how small the class or how large the training size, some partition makes generalization fail for some target. These results suggest that modern generalization theory must treat domain structure as a first-class object. --- ### Feedback Attribution and Representation Geometry: Metrics for Comparing Individual and Shared Rewards in MARL Authors: Tasha Pais, Richard Higgins Published: 2026-07-17 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.16524v1 Cooperative multi-agent RL systems routinely use team-averaged rewards, a feedback-attribution choice that gives each agent the team outcome regardless of its individual contribution. We ask whether this leaves a measurable signature, geometric or behavioral, on learned representations. We propose EffRank/$n$ (effective rank normalized by agent count) and $D_\text{act}$ (mean pairwise KL divergence between agents' action distributions) as low-overhead diagnostics for reward-attribution effects, then test them on competent MAPPO agents in SMACv2 \texttt{protoss\_5\_vs\_5}, where unit type is encoded in the observation. In an observation $\times$ reward-attribution comparison (unit type observed vs.\ masked; individual damage-contribution reward vs.\ shared team reward), geometry follows observation rather than reward. With unit type observed, shared and individual rewards have similar EffRank/$n$ ($0.31{\pm}0.03$ vs.\ $0.29{\pm}0.02$) and probe accuracy ($0.75{\pm}0.05$ vs.\ $0.73{\pm}0.05$, both $\gg 1/3$ chance), while $D_\text{act}$ leans higher under individual rewards ($1.23{\pm}0.06$ vs.\ $1.07{\pm}0.20$). Masking unit type cuts the above-chance probe signal by more than half, to $0.49$ in both reward arms. In short: individually rewarded agents are competent and separable by role, but on SMACv2 the observation explains the geometry and reward attribution shows up mainly in behavior. Thus geometric diagnostics must control for observed role information and test persistent roles that are not directly observed. EffRank/$n$ and $D_\text{act}$ add $<$5\% overhead. --- ### SEER: Supervised Learning to Control Energetic Reasoning Authors: Sascha Van Cauwelaert, Michele Lombardi, Pierre Schaus Published: 2026-07-17 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.16523v1 One of the main strengths of Constraint Programming is the ability to reduce the search space via propagation. However, propagation is a double-edged sword, with more pruning power coming at the price of larger computation time. For each problem constraint, the best propagator depends on the specific instance and may change at search time. In the literature, Machine Learning (ML) techniques and activity-based heuristics have been applied respectively for choosing (statically) the propagators for a batch of problems and to adapt (dynamically) the propagation strength. We propose to merge those efforts by using an oracle function, obtained via ML, to decide whether to run complex propagators for a target constraint. A combination of design choices makes the approach flexible and easy to embed in state-of-the-art solvers. In this paper, we focus on investigating the feasibility of building an oracle for the Energetic Reasoning propagator. Our experiments show that high prediction accuracy can be obtained, provide suggestions for classification features, and highlight important issues to address when building such an oracle. --- ### Geometry-Enhanced Portion Estimation for Multimodal LLMs Authors: Lin Liao, Peng Li Published: 2026-07-17 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2607.16514v1 Image-based dietary assessment promises to replace costly, bias-prone manual recalls, but portion estimation remains a major blocker. Multimodal LLMs (MLLMs) recognize a wide range of foods zero-shot in uncontrolled photos, yet they are weak at portion estimation -- a gap we measure across the current frontier (Gemini, GPT, and Claude flagships alike). We present a method that enhances a frozen, commercial MLLM with an accurate portion head: a small geometry-enhanced network on a frozen DINOv2 backbone with a structured softmax-ownership volume, consuming the MLLM's per-food name, bounding box, and density range -- no depth sensor, no MLLM fine-tuning. Evaluated fully open-vocabulary on three real-world benchmarks, the head cuts per-food portion error by 33-41% relative to the MLLM alone, outperforms every flagship MLLM's direct estimates, and surpasses each benchmark's originally published image-only model at its own reported metric. --- ### How Formerly Incarcerated People Envision Technologies for Prison Parole Authors: Saiph Savage, Jesse Nava, Wanqing Iris Zhou, Hwijoon Lee Published: 2026-07-17 Categories: cs.CY, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2607.16513v1 AI-driven algorithms and automated tools are increasingly embedded in the correctional landscape, shaping parole eligibility,release decisions, and surveillance. These tools are also often framed as objective, inevitable solutions to inefficiency andbias. Yet, these computational systems are rarely designed with input from justice-impacted individuals, which means theymight fail to address the real needs of incarcerated people. To address this gap, we surveyed 31 formerly incarcerated peopleabout their parole experiences and their visions for technologies that could support parole preparation. Contrary to dominantassumptions, participants did not imagine computational tools as instruments to dismantle the prison system, but as resourcesfor navigating power: translating complex parole concepts into culturally familiar terms, documenting personal transformationin board-legible ways, and recognizing the often-invisible labor of families. We argue that these imaginaries point towardtechnologies designed for strategic agency, where tools help incarcerated individuals and their families build the capacity tonavigate existing power structures in pursuit of freedom. We conclude by reframing computational tools for parole awayfrom surveillance and toward human-centered systems that support people in navigating carceral power structures. --- ### Foresight Residual RL for Long-Horizon Robot Manipulation with Vision-Language-Action Models Authors: Yuhan Liu, Xinyu Zhang, Litao Liu, Abdeslam Boularias Published: 2026-07-17 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2607.16506v1 Vision-Language-Action (VLA) policies offer strong general-purpose manipulation priors, but often fail on tight-tolerance, contact-rich assembly due to long-horizon credit assignment and subtask coupling: a state that is geometrically successful for the current skill can be brittle for downstream skills. We show this failure mode in residual reinforcement learning (RL) over a frozen VLA base policy: constant sparse success rewards improve each subtask in isolation yet yield little or no gain when skills are chained, because terminal state quality is uncontrolled. We propose Foresight Residual RL, which optimizes handoff quality by augmenting each subtask's sparse success reward with an offline-estimated foresight value -- the probability of future subtask success conditioned on the terminal state of the current subtask. Concretely, we (i) train a visual foresight predictor from images of terminal states of the base policy, labeled using downstream rollout statistics, and (ii) train residual policies via backward foresight induction, using the predictor output as a reward multiplier. On a three-phase wrench-based nut-tightening assembly task in Isaac Gym (grasp, move-insert, rotate), our method achieves 85.6% full-task success, outperforming standard subtask residual RL (54.5%) and VLA baselines, while leaving per-subtask success unchanged. These results highlight that improving long-horizon performance requires shaping which successful states are produced at each sub-task, not only whether success occurs. --- ### Leakage-Robust Evaluation and Data-Scale Sensitivity of Attention-Enhanced Multi-Task Learning for Joint Fault Diagnosis and Remaining Useful Life Estimation Authors: Md Mahamudur Rahaman Shamim, Md. Nuruzzaman, Zannatul Ferdus, Md Rajib Ahmed, Abieer Nwshad Anward, Mohammad Tooneer, Johir Uddin Khan, Khalid Hossen Published: 2026-07-17 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.16493v1 Multi-task deep learning models that jointly perform fault classification and remaining useful life (RUL) regression are increasingly used in predictive maintenance, yet reported performance can be strongly affected by how sliding-window sequences are split into training and test sets. We investigate this issue using AMTLNet, an attention-enhanced multi-task architecture, on three public benchmarks: NASA C-MAPSS, NASA IMS, and the UCI Hydraulic System dataset. We show that naive splitting can inflate classification accuracy from a genuine 20-60 percent to 99.9 percent, or reduce it to 0 percent through degenerate class representation. To address this, we introduce a chunk-based, leakage-audited splitting protocol and evaluate all models using five seeds, one-way ANOVA, and Tukey HSD tests. On C-MAPSS, with 19,976 leakage-free training windows, AMTLNet matches a single-task CNN-LSTM baseline in classification, achieving 84.12 +/- 0.96 percent accuracy with Tukey p = 1.0, and reaches an R2 of 0.86 +/- 0.01 while significantly outperforming a naive multi-task baseline. On the smaller Bearing and Hydraulic datasets, multi-task training is unstable, but the failure mode differs: classification degrades for Bearing, whereas regression degrades for Hydraulic. We relate this asymmetry to label provenance and propose a practical framework for deciding when joint training is appropriate under data scarcity. Ablation results show that the multi-head attention branch is the main contributor to regression stability. Removing it reduces R2 from 0.861 to 0.766 and more than doubles classification variance, whereas the convolutional branch contributes little to regression despite using about one-third of the parameters. This study contributes a reusable leakage-audit protocol, seed-transparent evaluation, and evidence that task-specific stability depends more on label provenance than on task type. --- ### WeedExpert-R1: Incentivizing Botanical Reasoning in MLLMs with Reinforcement Learning for Precision Weed Grounding Authors: Zonglin Yang, Wei-Zhen Liang, Nevin Lawrence, Xin Qiao, Benjamin Riggan, Chi-En Chiang, Fuchen Li Published: 2026-07-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.16492v1 Precision weed control requires species-level identification and instance-level localization. However, conventional object detectors use a closed vocabulary, limiting their deployment across regions, and cannot explain their predictions in complex agricultural scenes. Multimodal large language models (MLLMs) offer visual grounding and reasoning capabilities, but insufficient botanical knowledge can cause hallucinations in fine-grained weed identification. This study introduces WeedExpert-R1, a multimodal model that learns visually grounded botanical reasoning through verifiable rewards. A domain-specific Chain-of-Thought synthesis pipeline combines a human-curated botanical trait dictionary with an Auditor-Synthesizer LLM workflow to generate reasoning data for supervised fine-tuning. Group Relative Policy Optimization is then applied with rewards for format, accuracy, instance count, and response length. Across 37 weed species from six datasets, WeedExpert-R1-4B achieved 75.82 percent exact-set precision at an IoU threshold of 0.5, 89.30 percent precision, and 87.81 percent recall. It outperformed proprietary models, including GPT-5.4 and Gemini-3.1-Pro, and larger open-source models, including Qwen3-VL-30B-Instruct and Gemma-4-31B-it. Results on unseen species further demonstrate its open-vocabulary capability and potential for deployment across diverse regions and crops without retraining. --- ### K-IPO: Kendall-constrained Importance Preserving Oversampling for Imbalanced Tabular Data Authors: Marios Tyrovolas, Argiris Sofotasios, Dimitris Metaxakis, Georgios Mermigkis, George Georgoulas, Panagiotis Hadjidoukas, Chrysostomos Stylios Published: 2026-07-17 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2607.16478v1 Oversampling is widely used to address class imbalance in tabular classification, but existing methods can distort the feature importance ranking underlying model explanations. Although recent studies have quantified this distortion by comparing real and synthetic data, none have actively sought to prevent it. In this paper, we introduce Kendall-constrained Importance-Preserving Oversampling (K-IPO), a generator-agnostic, "generate-then-select" framework that preserves the original data's feature importance ranking during augmentation. K-IPO iteratively generates minority-class candidates and accepts them only if their inclusion maintains a user-defined minimum Kendall's tau (τ) correlation with the reference ranking. Optionally, stricter constraints can be applied to the highest-ranked features. We evaluated K-IPO on 20 imbalanced binary classification datasets using three classifiers and multiple explanation methods. In most cases, K-IPO achieved the best or tied-best results in feature importance preservation, explanation consistency, and class separability. It also generally improved predictive performance while maintaining competitive computational overhead. --- ### When to Use Which? Benchmarking Optimisers for Configurable Systems under Varying Budgets Authors: Chao Jiang, Yulong Ye, Tao Chen, Miqing Li Published: 2026-07-17 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2607.16476v1 Software configuration tuning is crucial for optimising system performance, and various optimisers have emerged over the last decade. Yet, the time required during the tuning process may vary across systems. In some systems (e.g., PostgreSQL), it may take a few minutes to measure a configuration, whereas in some others (e.g., MariaDB), it can take several hours. Moreover, even within the same system, users may have varying budgets and preferred settings. This naturally raises a question -- Given a budget level, which optimiser is the best choice for SE practitioners? This matters because optimisers usually have their own ``comfort zone'' and may perform very differently under distinct budgets. In this paper, we aim to answer this question. We systematically evaluate eight well-established optimisers across 22 configurable systems under varying budget levels. We find that, unsurprisingly, model-based optimisers (e.g., SMAC) are well-suited under tight budgets, and model-free optimisers (e.g., GAs) become superior with more generous budgets. However, interestingly, there is one optimiser, FLASH, that performs consistently well on most systems regardless of budgets. We lastly investigate the reasons behind this phenomenon and find that many systems possess good local optima (with large basins of attraction), allowing greedy optimisers (e.g., FLASH) to achieve strong performance. Source code, data, and supplementary materials of this work are available at https://anonymous.4open.science/r/Config-W2W-98B2. --- ### Spectral-Morphological Attention U-Net: An Efficient Network for Active Wildfire Detection Authors: Yugong Zeng, Jonathan Wu Published: 2026-07-17 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2607.16472v1 Over the past decades, the frequency of global wildfires has been increasing steadily. Therefore, if the fire can be detected and precisely located at an early stage, the potential hazards caused by it can be minimized to the greatest extent. The machine learning methods based on satellite images, due to their ability to automatically monitor extremely remote and vast areas, have shown great potential for application in the field of wildfire detection. To address this challenge, we proposed a new model named spectral-morphological attention U-Net(SMA-UNet), which includes a spectral attention module, a residual attention UNet backbone, a channel-spatial modulator, and a pair of differentiable morphological gates. We trained and evaluated this model with two datasets. These modules, excluding the backbone, are used to detect active fire events for the first time, especially the pair of differentiable morphological gates, which is innovatively developed. The proposed model achieved the highest scores in both datasets (e.g., intersection over union 75.16% in TS-SatFire, 22.50% in Sen2Fire). By conducting ablation studies of each module, we compared their independent contributions and tested their combinations. Ultimately, the integration of these modules yields a highly robust framework that significantly improves segmentation consistency across diverse and complex environmental conditions. Future work will focus on validating the proposed architecture across large-scale, multi-regional datasets from different satellite sensors to establish its broader generalizability for global wildfire detection. --- ### Berkeley and Heiserman as an Unexhausted Architecture for Embodied Machine Intelligence Authors: Christopher A. Tucker Published: 2026-07-17 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.16465v2 Edmund C. Berkeley is usually remembered as a mediator between symbolic logic and early computing, yet that standard description understates the scope of his work. This paper argues for a stronger reading: Berkeley should also be understood as an early theorist of embodied machine intelligence. Across Berkeley's major writings on symbolic logic, machine intelligence, living robots, and Squee, intelligence appears not as disembodied symbol manipulation alone but as the organized coordination of sensing, storage, calculation, control, state, and action in physically realized machines. The paper's first contribution is interpretive: it reconstructs Berkeley as a thinker of machine architecture, temporally extended behavior, and environment-coupled control. Its second contribution is comparative: it reads Berkeley alongside David L. Heiserman to recover a shared descriptive scheme centered on sensing, state or memory, control, action, and adaptation. Its third contribution is critical: it uses that scheme to assess current embodied-AI discourse. The broader claim is that contemporary LLM-centered robotics often demonstrates impressive capability without an equally explicit account of persistence, recoverability, maintenance, and structured modification of conduct through experience. --- ### Compact convolutional neural networks for AI-based drone detection system Authors: Gábor Farkas, Gábor Fazekas, Karakai Patrik, András Németh, Gábor Farkas Published: 2026-07-17 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.16455v1 The increasing use of first-person-view drones in modern conflicts has created a demand for compact and reliable detection systems capable of operating in complex electromagnetic environments. These drones continuously transmit video signals through onboard video transmitters, generating radio-frequency emissions that can be exploited for early detection. This study investigates the use of lightweight convolutional neural networks for automated detection of drone signals captured by a software-defined radio-based electronic warfare framework. Samples are converted into rasterized time-domain images, providing a computationally efficient input representation suitable for embedded systems. Several custom model architectures were designed and benchmarked in terms of accuracy, model size, and inference performance using a dataset containing approximately 40,000 labeled images. In addition to offline testing, the models were integrated into a GNU Radio signal processing chain for real-time evaluation. The results show that compact models can achieve high detection accuracy while maintaining low computational requirements, making them suitable for embedded radio-frequency monitoring applications. Compared with existing spectrogram-based RF detection methods, the proposed approach eliminates frequency-domain preprocessing and achieves comparable accuracy with significantly reduced computational cost. --- ### DS@GT ARC at AnimalCLEF 2026: Species-Aware Graph Construction for Multi-Species Animal Re-Identification Authors: Evan Sinclair Smith, Anthony Miyaguchi, Snigdha Palamari, Danté Evangelista Published: 2026-07-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.16453v1 Automated individual animal re-identification is essential for large-scale biodiversity monitoring; however, field imagery complicates separating identity cues from nuisance variation in pose, illumination, background, resolution, and species-specific morphology. The DS@GT ARC submission to AnimalCLEF 2026 introduces a multi-species image-clustering system for re-identifying Eurasian lynx, fire salamanders, loggerhead sea turtles, and Texas horned lizards. Instead of relying on a single descriptor or nearest-neighbor retrieval, this approach formulates re-identification as species-aware graph construction over candidate image pairs. The pipeline integrates tailored preprocessing, global candidate retrieval, LightGlue-based local verification with multiple keypoint families, LightGBM pair scoring, conservative edge admission, and Leiden community detection. This design directly addresses a primary failure mode of clustering-based re-identification: high-scoring false pairs that act as bridge edges and merge distinct individuals through transitive closure. Across species, ablation studies demonstrate that local feature support, foreground-aware preprocessing, and species-specific backbone selection enhance pair evidence, while graph operating points determine the trade-off between fragmentation and over-merging. The selected submission achieved a public ARI of 0.733 and a private ARI of 0.674, ranking fifth among 230 teams. These results indicate that robust wildlife re-identification requires not only strong visual representations but also calibrated integration of global similarity, local identity markings, neighborhood context, and graph-level constraints. The code can be found at https://github.com/dsgt-arc/animalclef-2026. --- ### Committed Before Reasoning: Behavioral Reproduction and Preliminary Activation-Level Evidence of Answer Pre-Commitment in an Open-Weight LLM Authors: Heejin Jo Published: 2026-07-17 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2607.16451v1 Chat models sometimes commit to an answer and then produce reasoning that justifies it rather than deriving it -- even when the answer contradicts a task premise. We study a minimal probe: "I want to wash my car. The car wash is 100 meters away. Should I walk or drive?" Only drive works (the car must be at the car wash), yet models overwhelmingly recommend walking. (1) Behavioral reproduction: on Qwen3-8B across five system-prompt conditions (210 rollouts), the wrong commitment occurs in 85-100% of sampled rollouts per condition and 100% of greedy rollouts, in both thinking and non-thinking modes; a 4,096-token thinking budget does not repair it. (2) Preliminary activation-level evidence: probing hidden states with a pretrained, training-free activation oracle (no task-specific probe training) at positions before the answer text is emitted, "walk" read-outs exceed a neutral-context baseline (68% vs. 17%; walk-committing rollouts p=.005, drive-committing rollouts p=.005, Fisher exact) -- notably, rollouts that eventually answer drive also read as walk-leaning before commitment (5/6). The oracle's default on unrelated content is "drive" (83%), so the read-outs are not lexical bias; stratifying by literal walk/drive occurrence shows they are not text recovery either (spans containing "drive" still read out walk; in balanced lexical fields, per-rollout walk-majorities beat a per-prompt neutral baseline 15/22 vs. 1/8, p=.01; drive-committing rollouts 6/6, p=.002). Samples are small and the within-rollout positional gradient is not significant (p=.34); we frame these results as preliminary. (3) Methodological: with fixed oracle, activations, and positions, question wording alone moves a positive control from 2/16 (open question) to 11/16 (closed); negative oracle results are uninterpretable without per-wording positive controls. --- ### EA-RMENet -- Path Loss Prediction in Urban Environments using Deep Learning Authors: Jonathan O'Shea, Conor Brennan Published: 2026-07-17 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.16449v1 Accurate path loss prediction is a critical component of wireless network planning. Current path loss prediction methods typically struggle to balance the trade-off between accuracy and computational efficiency. This paper proposes the Efficient Attention Radio Map Estimation Network (EA-RMENet) which is an image data-driven, deep learning (DL) model designed for radio map estimation (RME). EA-RMENet uses a U-Net framework with an EfficientNetB5 encoder, Attention Gated (AG) skip connections, and Atrous Spatial Pyramid Pooling (ASPP). The EfficientNet encoder uses compound scaling to balance accuracy and efficiency. AG skip connections suppress irrelevant features, and the ASPP captures a multi-scale context. The model has a test prediction RMSE of 0.0334 on the RadioMapSeer3D dataset with an inference time of 0.022 seconds/sample. In the ICASSP 2023 Radio Map Prediction Challenge, the model ranks third with a competitive RMSE of 0.0406 this highlights the models potential for real-world RME. --- ### Retrieval is Enough: Training-Free Interpretability with a Tool-Using Agent Authors: Sriram Balasubramanian, Soheil Feizi Published: 2026-07-17 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2607.16448v1 Interpretability methods for neural network activations span a wide cost spectrum, from cheap, training-free techniques (such as linear probes, PCA, SVD) to more expensive training-based ones (such as SAEs and activation oracles). Training-based methods are typically more powerful, in part because they leverage large activation datasets during training. This raises a natural question - do they actually surface insights that go beyond what is recoverable from the training dataset itself? To address this, we equip an LLM agent with a vector database of activations paired with their textual contexts, along with tools for manipulating activations - projecting out directions in latent space, computing activation differences and averages. The agent iteratively queries the database, forms hypotheses from the retrieved samples, and validates them by constructing linear probes. We call this method HARP, for Hypothesis-driven Agentic Retrieval and Probing. Despite not involving any training, HARP outperforms both activation oracles and SAE-based agents on concept discovery, concept detection, model steering, and secret elicitation. The training-free design also makes HARP substantially cheaper and more flexible: new datasets can be indexed on demand whenever existing ones prove insufficient. More broadly, our results suggest that current training-based methods do not yet extract insights beyond their training data, and motivate benchmarks that explicitly require interpretability methods to demonstrate such insights. We release our code at https://github.com/SriramB-98/HARP --- ### One Modality to Forget Them All: Enhancing Cross-Modal Unlearning in Vision-Language Models Authors: Sudharshan Balaji, Yili Ren, Guangjing Wang, Yimin Chen, Ning Wang Published: 2026-07-17 Categories: cs.CV, cs.CL, cs.CR arXiv: https://arxiv.org/abs/2607.16442v1 Machine unlearning is widely used to remove hazardous knowledge from large language models. Modern Vision-Language Models (VLMs), however, process both text and visual inputs, raising a fundamental security question: does unlearning in one modality transfer to the other? We present the first systematic, bidirectional study of cross-modal unlearning transfer across three VLM architectures: LLaVA-1.5 (MLP projection), InstructBLIP (Q-Former), and IDEFICS (gated cross-attention). We find that unlearning transfers across modalities, but the transfer is asymmetric and incomplete. In some cases, text unlearning strongly transfers to vision. However, this robustness is not preserved under typographic attacks that manipulate the visual presentation of text. Under such attacks, previously unlearned knowledge can be readily recovered, indicating shallow unlearning. To address the transfer gap and shallow robustness, we propose \textsc{CrossInf}, an influence-guided mitigation strategy. Motivated by the observation that different model components contribute unequally to cross-modal transfer, \textsc{CrossInf} focuses unlearning on transformer blocks that most influence cross-modal generalization. It reduces the transfer gap by more than half in architectures with strong fusion, while preserving model utility. It also improves robustness under typographic attacks, reducing the attack success rate to near zero. We further conduct human evaluation with three annotators ($κ{=}0.77$) to validate our findings. Finally, we analyze shallow unlearning using Centered Kernel Alignment (CKA), providing insights into the observed transfer behavior and robustness limitations. --- ### RIMS: Preference Optimization via Smoothed Multi-pair Aggregation for Small-Scale LLM Retrieval-Augmented Generation Authors: Pei Tian, Zihan Dong, Tianci Liu, Linjun Zhang, Haoyu Wang Published: 2026-07-17 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.16431v1 Small-scale language models (SLMs) are attractive for retrieval-augmented generation (RAG) in resource-constrained settings, but their limited capacity makes them highly sensitive to noisy or spurious retrieved evidence. Existing preference-based methods such as RoseRAG select only the hardest single preference pair via hard argmin/argmax, discarding the remaining signal; others treat multiple pairs as independent binary comparisons, resulting in low data utilization. We propose RIMS, a three-stage preference optimization framework comprising (1) synthetic chain-of-thought preference data generation via rejection sampling using the target SLM itself without relying on proprietary models, (2) a differentiable soft aggregation mechanism that replaces hard selection with a smooth operator, preserving gradient signal from all preference pairs while retaining the discriminative structure of margin-aware selection, and (3) preference optimization with the smoothed objective applied to multiple alignment algorithms. We theoretically show that the smoothed approximation admits a controllable error bound and that smooth aggregation yields provably tighter gradient alignment to the oracle objective than hard selection. Experiments on four multi-hop question answering benchmarks show that our approach outperforms state-of-the-art baselines across multiple SLM backbones, achieving consistent gains in Exact Match and F1 under noisy retrieval conditions. Our implementation is available at https://github.com/tptrix29/RIMS. --- ### Back to the museum: Investigation of the acceptance of Android Andrea with and without emotion simulation in a museum Authors: Marcel Heisler, Christian Becker-Asano Published: 2026-07-17 Categories: cs.RO, cs.AI, cs.HC, cs.LG arXiv: https://arxiv.org/abs/2607.16428v1 For a second time, the android robot Andrea was set up at a public museum in Germany for six consecutive days to have conversations with visitors, fully autonomously. Building on previously gathered qualitative results, the robot was now capable of engaging in multi-lingual conversation with the visitors about the museum context. The robot was prepared with context information about the museum in general and its surrounding exhibits this time. The robot featured a slightly artificial sounding voice that was previously evaluated as congruent with its gender-ambiguous but very humanlike design. Three experimental conditions were implemented, in which either (1) the robot simulated no emotions, (2) the robots emotions were determined by ChatGPT 4.1, or (3) the WASABI emotion simulation architecture simulated the robot's emotion dynamics. An extended version of the TAM2 questionnaire was employed to let 73 visitors report on several factors of their opinion about the android robot Andrea after having experienced it. In result, the statistical analysis suggests that these first two approaches to implementing emotions into the chat architecture of our android robot Andrea did not yield any positive effects on the subjective evaluations by the visitors and were not detectable on a conscious level. --- ### Multi-level context Modeling for consistent expert selection in Mixture-of-Experts Authors: Shuhan Huang, Naifan Zhang, Yuanbo Tang, Yang Li, Wai Kin Victor Chan Published: 2026-07-17 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.16427v2 Mixture-of-Experts (MoE) enables efficient scaling of Transformer models by routing tokens to a small subset of experts. However, existing routers typically condition expert selection on shallow or isolated token representations, which often produce unstable and semantically inconsistent routing decisions across layers. In this work, we revisit expert selection from a representation perspective and identify context incompleteness as a key bottleneck limiting effective expert specialization. To address this issue, we propose Multi-level Context Fusion MOE (MCF-MOE), a framework that constructs context-aware representations by integrating complementary signals from cross-layer semantic aggregation and local token-level interactions, enabling more informative and consistent expert selection. Experiments on language modeling and understanding benchmarks demonstrate that MCF-MOE consistently improves routing consistency and downstream performance over strong MoE baselines, highlighting the importance of contextual completeness in expert routing. The code is available at https://github.com/shuhanhuang/MCF-MOE. --- ### Multi-level context Modeling for consistent expert selection in Mixture-of-Experts Authors: Shuhan Huang, Naifan Zhang, Yuanbo Tang, Yang Li, Wai Kin Victor Chan Published: 2026-07-17 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.16427v1 Mixture-of-Experts (MoE) enables efficient scaling of Transformer models by routing tokens to a small subset of experts. However, existing routers typically condition expert selection on shallow or isolated token representations, which often produce unstable and semantically inconsistent routing decisions across layers. In this work, we revisit expert selection from a representation perspective and identify context incompleteness as a key bottleneck limiting effective expert specialization. To address this issue, we propose Multi-level Context Fusion MOE (MCF-MOE), a framework that constructs context-aware representations by integrating complementary signals from cross-layer semantic aggregation and local token-level interactions, enabling more informative and consistent expert selection. Experiments on language modeling and understanding benchmarks demonstrate that MCF-MOE consistently improves routing consistency and downstream performance over strong MoE baselines, highlighting the importance of contextual completeness in expert routing. The code is available at https://anonymous.4open.science/r/MCFMOE. --- ### When to Plan: Learning to Select Between Reactive Control and Deliberative Planning Authors: Adam Labiosa, Josiah P. Hanna Published: 2026-07-17 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.16421v1 It has long been recognized that humans have the ability to switch between fast, reactive decision-making and slower, deliberative planning. In this paper, we study the question of how to learn this ability, known as meta-reasoning, in artificial agents. We model reactive decision-making as a policy that directly maps state observations to actions. Such policies can be trained with reinforcement learning (RL) or imitation learning, but may generalize poorly outside of their training distribution. Alternatively, model-based decision-time planning is more likely to produce good actions across a broader set of states but requires additional computation time, which delays acting. In this work, we introduce an RL method for training a meta-reasoning policy that allocates computation by conditioning on a reactive-policy uncertainty score. This score enables it to predict when the reactive policy is likely to perform poorly and when planning is needed. We conduct an empirical study on motion planning and navigation environments, showing that this design enables the meta-reasoning policy to learn when the reactive policy provides a good-enough action versus when decision-time planning is needed. Additionally, we show that our design enables the meta-agent to shift toward fully reactive control as the reactive policy improves. --- ### Signal-based Model Access Risk Analysis for AI System Operations Security Authors: Maria Mahbub, Steven Young, Amir Sadovnik, Edmon Begoli, Chris Rugenstein, Donald Coulter, Anthony Ayodele Published: 2026-07-17 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2607.16414v1 Artificial intelligence (AI) systems are now ubiquitous across domains such as security, finance, healthcare, consumer technology, and large-scale cloud services, where they process massive volumes of data and make consequential decisions daily. This widespread adoption has created a broad attack surface through which adversaries can manipulate, evade, extract information from, or otherwise subvert deployed models. Depending on system design and exposure, attackers may have very different forms of access: some observe only final decisions, while others receive confidence scores, intermediate representations, or even full model parameters. While previous surveys typically organize evasion attacks into white-box, gray-box, and black-box categories based on the attacker's knowledge of model internals (architecture, parameters, gradients), this taxonomy often conflates different deployment scenarios that provide vastly different output signals, all labeled as ``black-box'' despite enabling fundamentally different attack strategies. Understanding how evasion attack strategies adapt to the specific information signals returned by deployed systems is critical for organizations making procurement and deployment decisions. To address this gap, we introduce the Signal-based Model Access Risk Taxonomy (SMART), a deployment-oriented framework that classifies attacker access according to the nature and richness of the information signals available from deployed AI systems. Using this taxonomy, we provide a structured overview of evasion attacks across progressively richer levels of information exposure, highlighting how deployment interfaces influence attack capabilities and informing more secure AI deployment and procurement decisions. --- ### Interactive Task Alignment as a POMDP Authors: Andy Dai, Zexue He, Zhenyu Zhang, Alex Pentland, Jiaxin Pei Published: 2026-07-17 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.16412v1 Current benchmarks for language models primarily evaluate execution on fully specified tasks. However, real user tasks are often ambiguous. Users arrive with incomplete, exploratory, or even inconsistent goals, requiring the assistant to first determine the intended task before carrying it out. We study this problem as task alignment: the ability to align with a user on their intended task. We introduce a general framework for converting specified tasks into underspecified interactions, formalized as a POMDP in which the model must infer a latent task from partial and evolving user intent. We validate our user simulator post hoc with a human user study. Across shopping, coding, and professional work settings, we find that while models often perform well once the task is specified, models still struggle with task alignment: current models act prematurely, interact ineffectively, and fail to resolve ambiguous requests. Models on average recover the user's intended task only 22-32% of the time under ambiguity. In a human study in the same setting, humans reach 48%, outperforming all evaluated models. We show that post-training with supervised fine-tuning and reinforcement learning improves task alignment, but models still lag behind humans in resolving uncertainty through interaction. Together, our results suggest that current models still lack key interaction abilities required for reliable agency. --- ### Think, Plan, Paint: Layout-Aware Reasoning for Controllable Image Generation in Unified Models Authors: Junhao Liu, Jian-Wei Zhang, Tao Huang, Miles Yang, Zhao Zhong, Liefeng Bo Published: 2026-07-17 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2607.16409v1 Unified Multimodal Large Language Models (MLLMs) offer a promising paradigm for unifying visual understanding and generation, yet they still struggle to follow complex spatial instructions and logical constraints in controllable image generation. To address this gap, we present ATLAS, a unified framework that equips MLLMs with a human-like "Think, Plan, and Paint" paradigm. We adopt layout as the shared representation that connects the three stages, enabling the model to reason about spatial requirements, plan explicit object arrangements, and render the final image. We further improve plan-to-image fidelity with reinforcement-learning-based layout alignment. We instantiate ATLAS at 7B and 80B scales, achieving state-of-the-art performance among MLLMs on image generation benchmarks and an average 65.31% improvement over existing layout-based unified MLLMs. On spatially related tasks, ATLAS obtains an average 23.06% gain over the base models. Through the same layout interface, ATLAS also supports instruction-guided editing and multimodal grounding. We further introduce ATLAS-Reasoning, a benchmark for evaluating generation under complex spatial instructions. --- ### Apple-$π$: Benchmarking Thinking with Video Towards Law-Grounded Physical Intelligence Authors: Runmao Yao, Kairui Hu, Yukang Cao, Ruisi Wang, Shulin Tian, Ziang Cao, Weichen Fan, Ziqi Huang, Yuhao Dong, Hao Li, Zhaoxi Chen, Zhongang Cai et al. Published: 2026-07-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.16401v1 Modern video generation models are increasingly hailed as emerging world models with an internalized grasp of physical law. Yet existing benchmarks largely evaluate physical plausibility only at the output level, without verifying whether the model arrives there through a faithful, law-grounded reasoning process. We introduce Apple-PI, the first benchmark that anchors video-model evaluation explicitly in physical laws. Apple-PI comprises three components. 1) Orchard: a dataset of 400 videos covering ten canonical tasks in classical mechanics. It separates single-law tasks for confounder-free diagnosis from multi-law tasks for probing generalization. 2) Benchmark Protocol: a three-stage protocol based on scientific reasoning, including Perception, Formulation, and Deduction. It uses chain-of-frames prompting on infographic-annotated first frames, treating the generated video as the model's visible reasoning trace. 3) Evaluation Suite: a hybrid evaluation suite that combines MLLM-based subjective scoring with physics-law-grounded objective measures. This enables stage-resolved diagnosis of not only whether a model fails, but where it fails. Benchmarking 11 models shows that current video models remain far from reliable law-grounded world simulators, with the best video model scoring only 0.473. Our stage-, pillar-, and source-resolved analyses further expose a Perception-to-Formulation-to-Deduction bottleneck, weak multi-law state transfer, and a persistent Sim-to-Real gap. These findings position Apple-PI as a diagnostic foundation for guiding future video models toward world models with law-grounded physical intelligence. --- ### Automated Hardware Validation Test Plan Generation for Large Scale AI Datacenter Platforms Using a Generative AI Multi-Agents Architecture Authors: Mohammed-Khalil Ghali, Saurabh Kulkarni, Prathamesh Kulkarni, Rohan Kulkarni, Sangwon Yoon, Daehan Won Published: 2026-07-17 Categories: cs.MA, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2607.16388v1 Large-scale AI datacenter platforms comprise thousands of heterogeneous hardware components whose validation requires comprehensive fault injection test plans. Today these plans are authored manually: engineers review hardware self-healing validation documents and bills of materials, enumerate failure modes per field-replaceable unit, and produce flat lists of single-layer test cases. This process is labor-intensive, error-prone, and dependent on institutional knowledge; coverage gaps surface late, traceability to source specifications is implicit, and the effort is largely repeated per platform. This paper presents a generative AI multi-agent architecture that automates the generation of structured hardware validation test plans from two canonical inputs: self-healing validation documents, which enumerate known failure modes and their detection and remediation behaviors per field-replaceable unit, and component Bills of Material. An ingestion agent normalizes heterogeneous inputs into a canonical representation; a classification agent maps components to functional domains via contextual reasoning over part descriptions and sub-category hierarchies; and a generation agent synthesizes test cases by combining normalized failure modes with domain-classified data, filling gaps and producing edge cases. The output conforms to a standardized schema for direct import into internal validation software. Evaluated on two production platforms against manual baselines, the framework achieves coverage expansions of 74.2% and 51.4%, cutting authoring from days to hours. It yields fully traceable mappings from each test case to its source specification, and its multi-agent decomposition is portable across platform generations. Automated and expert evaluations confirm 100% extraction fidelity and high acceptance of new scenarios, validating the framework as a robust human-in-the-loop force multiplier. --- ### Fantastic Adaptive Taxonomies and How to Use Them Authors: Mert Cemri, Andrei Cojocaru, Melissa Pan, Shu Liu, Shubham Agarwal, Alexander Krentsel, Jay Tang, Kannan Ramchandran, Joseph E. Gonzalez, Matei Zaharia, Alex Dimakis, Ion Stoica Published: 2026-07-17 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2607.16387v2 An agent system's execution traces record how it fails, and procedures that improve such a system without changing model weights (trajectory selection, prompt and workflow optimization, runtime monitoring) read these traces for feedback. Yet raw traces are a poor medium for accumulating that feedback: long, instance-specific, and lacking a stable vocabulary for recurring failures. We argue that an agent system should instead maintain an explicit representation of how it fails, induced from its own behavior and reusable wherever failure feedback is needed. AdaMAST builds this representation by converting a target system's traces into a compact, evidence-grounded failure taxonomy: named failure codes organized along three fixed axes (system-level, role-specific, and domain-specific), with every name, definition, and evidence pattern induced from the traces; no code is hand-authored, no trace human-annotated. The taxonomy is not merely a post-hoc diagnostic but a shared feedback interface, improving agents in three ways. In agent-system search, taxonomy-coded diagnoses of failed candidates outperform free-form reflection on all five benchmarks we test. At runtime, taxonomy feedback raises SWE-agent's resolution on SWE-bench Verified Mini from 60% with free-text reflection to 70%, and improves Claude Code from 64.0% to 70.7% as a runtime skill. In trajectory selection, AdaMAST-Judge, a verifier built on the induced codes, improves best-of-5 accuracy on Terminal-Bench 2.0 by 8-15 points over Pass@1. The vocabulary itself is compact (an order-of-magnitude compression that preserves trace distinctions), human-faithful (matching expert failure annotations more closely than a hand-crafted reference vocabulary), and adaptive (taxonomies induced for different domains share few codes). Adaptive failure taxonomies close the loop between the traces agents produce and the procedures that improve them. --- ### Scaling Limits of Constant-Stepsize SGD at Flat Minima Authors: Jingyi Zhang, Cheng Mao, Debankur Mukherjee Published: 2026-07-17 Categories: cs.LG, math.OC, math.PR, stat.ML arXiv: https://arxiv.org/abs/2607.16384v1 For stochastic gradient descent (SGD) with a constant stepsize $α$, the invariant law of the iterates, centered at a minimizer, describes the behavior of the algorithm over long time horizons. In the strongly convex case, this invariant law has the familiar $\sqrtα$ scaling and a Gaussian limit as $α\downarrow 0$. We show that this behavior changes fundamentally for convex objectives $H$ with flat minima and (sub)quadratic tails. More specifically, we study SGD with Markovian noise generated by a contractive driving chain. For every sufficiently small constant stepsize $α$, we prove existence, uniqueness, and geometric convergence to an augmented invariant law in a Wasserstein distance induced by an $α$-dependent metric. When the minimizer $x_\star$ has local flatness exponent $m\ge2$, meaning that $\nabla^2 H(x)\asymp \lVert x-x_\star\rVert^{m-2} I_d$ as $x\to x_\star$, we obtain a contraction bound with factor $1-cα^{m-1}$, where $c>0$ is a constant. This recovers the factor $1-cα$ in the quadratic case $m=2$. We then analyze the small-stepsize scaling limit. We show that the invariant law concentrates on the scale $α^{1/m}$ and that the rescaled iterates converge weakly to the stationary distribution of the stochastic differential equation $$ dY_t=-h_0(Y_t)\,dt+Σ^{1/2}\,dB_t , $$ where $h_0$ is the limiting drift at the minimizer and $Σ$ denotes the asymptotic covariance. This recovers the Gaussian limit when $m=2$ and gives generally non-Gaussian stationary limits in the flat case $m>2$. Finally, we give corresponding results for coordinate-separable objectives with unequal flatness exponents. --- ### Disentangling Model and Human Data Uncertainty in Apparent Facial Age Estimation Authors: Andrei Foitos, Ivo Pascal de Jong, Matias Valdenegro-Toro Published: 2026-07-17 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2607.16378v1 Estimating the apparent age of individuals from facial images is challenging due to the subjective nature of perception and the inherent variability of the data. We investigate the role of uncertainty estimation, attributing uncertainty jointly to a lack of knowledge (epistemic) or inherent noise/chance (aleatoric). Leveraging the APPA-REAL dataset, we train Bayesian Neural Networks on datasets of varying sizes using three BNN approximations: MC-DropConnect, Flipout, and Deep Ensembles using supervision on human aleatoric uncertainty available in the APPA-REAL dataset. Each model outputs both the predicted apparent age and the amount of aleatoric and epistemic uncertainty. Our results confirm the hypothesis that the inherent aleatoric uncertainty remains stable across dataset sizes, while epistemic uncertainty increases as training data decreases. These findings demonstrate that different sources of uncertainty can be quantified in face age estimation. --- ### AoA: Theorem Proving Agent over Abstract Syntax Tree of Redesigned Language Authors: Qiyuan Xu, Joshua Ong Jun Leang, Renxi Wang, Wenda Li, Haonan Li, Luke Ong, Conrad Watt Published: 2026-07-17 Categories: cs.SE, cs.AI, cs.LG, cs.PL arXiv: https://arxiv.org/abs/2607.16372v1 Interactive theorem proving (ITP) underpins program verification and formalized mathematics, but its manual effort limits scalability. LLM-based proof agents promise to ease this effort, but their heavy token consumption and API cost remain a major obstacle. We trace this cost to a shared root: current agents operate on serialized concrete syntax, emitting proofs as source text and recovering proof states through separate, line-number-based queries, so every edit shifts later lines and forces repeated relocation of errors and states. This same dependence on concrete syntax also blocks adoption of Minilang, a recent proof language that reaches SOTA on LLM-based proving but is too new for LLMs' training corpora. We address both problems by lifting the agent off source text and onto the abstract syntax tree (AST): the model supplies proofs as JSON representations of Minilang's AST -- native to tool-calling LLMs -- and drives the prover through a tree-edit model that fuses proof operations and states into one proof tree, so each operation carries its own subgoal's state, readable directly off the tree. We realize this design in \emph{Agent over AST} (AoA). Against Amazon's Isabelle Agent on miniF2F and NTP4VC-Pearl common success sets, AoA cuts API cost by 2.3--4.7x (normalized input-cache accounting), uses 2.9--6.9x fewer tokens and 3.9--8.9x fewer tool calls, and finishes 1.4--2.0x faster -- while also solving far more problems on the harder verification benchmark. --- ### PRISM: Multimodal Terrain Mapping for Rover Navigation in Unstructured Environments Authors: Raul Castilla-Arquillo, Carlos Perez-del-Pulgar, Levin Gerdes, Alfonso Garcia-Cerezo, Miguel A. Olivares-Mendez Published: 2026-07-17 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2607.16366v1 Robotic navigation in unstructured environments requires robust situational awareness to safely traverse hazards such as steep slopes and rocky terrain. To address this challenge, perception systems increasingly rely on multimodal sensor fusion. Specifically, integrating thermal imagery with standard optical and depth sensors enhances terrain differentiation, directly improving the reliability of mapping algorithms. This paper presents PRISM, a multimodal perception system for terrain mapping in unstructured settings. PRISM leverages a custom sensor suite to capture aligned RGB, depth, and thermal (RGB-D-T) imagery. At its core is OmniUnet, a novel vision transformer-based network specifically designed for multimodal semantic terrain segmentation. We validated the proposed system using two newly annotated datasets (BASEPROD and LAENTIEC) and demonstrate its real-world applicability through physical field experiments. Deployed on a resource-constrained embedded computer, PRISM efficiently generates traversability maps that directly enable autonomous navigation via a rover's Guidance, Navigation, and Control (GNC) subsystem. --- ### MTSSL: Meta-Thresholding Semi-Supervised Learning Authors: Shuyang Liu, Ziang Zeng, Ruiqiu Zheng, Jiazheng Wang, Zechen Liu, Wenxi Li, Zhou Yu Published: 2026-07-17 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2607.16363v1 A large body of Semi-supervised Learning~(SSL) algorithms encounter the threshold $τ$ to select pseudo-labels. The value of $τ$ across different SSL algorithms can vary depending on the learning perspective, yet they may achieve similar performance. It motivates us to establish a unified theoretical framework to explain the role of $τ$ in SSL. We statistically explained that the unsupervised loss is affected independently by correct and incorrect pseudo-labels, while $τ$ adjusts their numbers to balance the corresponding error term. This inherent trade-off indicates that SSL can reach the same loss with varying $τ$, precise optimal values of $τ$ during training may be unnecessary. With this, we treat $τ$ as an updatable parameter and optimize it via differentiation; the new policy is named \textbf{Meta-Thresholding Semi-Supervised Learning (MTSSL)}. Extensive experiments demonstrate the superior performance of MTSSL. We observe that the accuracy curves of SSL algorithms can overlap completely even when the values of $τ$ differ significantly, which supports our theoretical framework and indicates that the selection of $τ$ can be relaxed in the future design of SSL algorithms. --- ### OmniStyle-INR: Universal and Multimodal Style Transfer for INRs Authors: Rafał Kajca, Michał Miziołek, Kornel Howil, Rafał Tobiasz, Przemysław Spurek Published: 2026-07-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.16362v1 Style transfer remains a fundamental and highly important task across various data modalities, enabling creative manipulation conditioned by both reference images and textual descriptions. Recently, methods utilizing Gaussian Splatting have emerged as a unified representation for 2D images, video, 3D scenes, and 4D dynamics. However, representing videos and 2D images with Gaussian Splatting is structurally sub-optimal for dense continuous domains. The number of required Gaussians often approaches the total number of pixels, raising questions about the actual utility of such a representation for these specific modalities. In contrast, Implicit Neural Representations have established themselves as a much more popular and natural choice across all these data domains. Implicit Neural Representations naturally provide significant advantages, including data compression, inherent capabilities for super resolution, and seamless integration with deep generative models. To this end, we introduce OmniStyle-INR, a novel framework that leverages network-based continuous representations as a truly universal domain. Our approach successfully performs high-quality style transfer across all visual modalities, guided seamlessly by both text prompts and visual exemplars. ---