# Research Papers — 2026-09 · bundle page 110 of 266 > Full-text markdown of papers #5451–5500 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/110 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/109.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/111.md) ### Testing-Driven Reliability Audit of Trajectory-Based Early Outcome Prediction for LLM Agents: Target-Specific Calibration Transfer Persists Within a Single Benchmark Authors: YanZe Cao Published: 2026-09-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.25647v1 Predicting early outcomes based on trajectory can decrease the expenses associated with agent evaluation by terminating a run once the outcome becomes sufficiently predictable, assuming that the predictor's confidence is properly calibrated. Calibration is at risk when a predictor is applied to an agent on which it was never trained, but it is not known whether such transfer failures are broad across agent systems or concentrated in specific target agent/head combinations. Using public SWE-bench Verified trajectories and a frozen dual-head early-outcome prediction pipeline, we ran a leave-one-agent-out calibration audit, a shared-predictor leave-two-agents-out control, oracle prior correction, and a robustness battery over training cohorts, task resampling, task halves, jackknife, and thresholds. Fixed-scaffold TerminalBench analysis served as a pre-registered boundary test. Broad same-predictor pairwise heterogeneity was not supported; the median pairwise corrected-gap differences were 0.0180 (SUCCESS head, 45 pairs) and 0.0385 (FAILURE head, 35 pairs), and the pre-registered heterogeneity criterion was not met on either head. Two specific combinations, gpt-5-mini/SUCCESS and claude-opus-4.6/FAILURE, showed persistent calibration-transfer errors (median corrected gaps 0.1377 and 0.1107) without a sign reversal under any frozen control. TerminalBench did not establish cross-benchmark replication: the success target produced zero decisions (INDETERMINATE), and the failure target did not satisfy the pre-registered persistence criterion. Therefore, a strong target-specific calibration-transfer error can exist within one frozen environment, but the evidence does not establish that the error is intrinsic to the model or general across benchmarks. --- ### Efficient Cost-Aware LLM Evaluation via Bayesian Bandit Gittins Indices Authors: Qian Xie, Yueli He, Nairen Cao Published: 2026-09-22 Categories: cs.LG, cs.CL, stat.ML arXiv: https://arxiv.org/abs/2609.25645v1 Exhaustively evaluating every candidate LLM configuration on every benchmark item to identify a high-performing one is costly. We formulate configuration selection as a cost-aware Bayesian bandit problem and propose GittinsEval, which draws on the Bayesian-optimal Gittins policy to determine which configuration to evaluate next and when to stop. We extend the policy with an anytime recommendation rule over both fully and partially evaluated configurations, using an LCB-style score to account for posterior uncertainty. GittinsEval is computationally efficient, requiring only lightweight online updates after offline precomputation. Across GSM8K, PIQA, AlpacaEval, and MMLU response matrices, GittinsEval is consistently competitive, with particularly strong gains over configuration-level Bayesian optimization on large-example benchmarks and over cost-unaware bandit baselines on large-candidate tasks. Crucially, GittinsEval often attains near-zero simple regret using only 1% to 2% of the exhaustive-evaluation cost; it also offers an adaptive stopping rule that typically triggers at 1% to 10%. --- ### Ladders of Thought: A Self-Evolving Curriculum of Progressively Simplified Reasoning Traces Authors: Minghui Liu, Thomas Magelinski, Dehao Yuan, Qi Yu, Furong Huang Published: 2026-09-22 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.25643v1 Large language models (LLMs) excel at reasoning when scaled to hundreds of billions of parameters, but small- and mid-scale models remain brittle reasoners even with knowledge distillation (KD). We present Ladders-of-Thought (LoT), a framework that improves reasoning by combining progressive question rewrites with a self-evolving curriculum. LoT automatically generates semantically faithful but easier variants of reasoning problems, organizes them into difficulty buckets using step-based measures, and employs a self-evolving bandit scheduler to allocate training adaptively. Evaluated on two reasoning domains, math and multi-hop reasoning, across 1-8B models from different families, LoT consistently improves over KD. It delivers large gains on arithmetic tasks (e.g., +32 percentage points on AddSub, +25pp on SVAMP), +2-8pp improvements on in-domain test splits, and strong though dataset-dependent benefits on multi-hop reasoning (e.g., +16pp on QASC, +25pp on StrategyQA). LoT also converges faster than staged curricula, highlighting the value of adaptive progression. These results show that progressive rewrites coupled with adaptive curricula provide a simple yet effective recipe for strengthening reasoning in smaller LLMs. --- ### When Quantum Meets AI: Quantum Methods for Machine Learning and Machine Learning Methods for Quantum Systems Authors: Tak Hur Published: 2026-09-22 Categories: quant-ph, cs.AI arXiv: https://arxiv.org/abs/2609.25641v1 This thesis studies the intersection of quantum computing and artificial intelligence in two directions: quantum methods for machine learning and machine learning methods for quantum systems. For quantum machine learning, Neural Quantum Embedding learns data representations that increase the trace distance between embedded class ensembles, lowering an embedding-dependent bound on empirical risk and improving classification on noisy quantum hardware. A training objective based on the Hilbert-Schmidt inner product extends this approach to deterministic quantum computation with one qubit (DQC1) and is demonstrated on an NMR quantum processor. A margin-based generalization analysis then connects quantum neural network performance to quantum state discrimination. In the studied benchmarks, margin distributions predict generalization more reliably than parameter-count metrics. For quantum systems, a Mamba-based neural decoder for surface codes matches a reproduced Transformer baseline in memory experiments while reducing inference-cost scaling from quartic to quadratic in code distance. Under an explicit decoder-induced-noise model, it achieves lower logical error rates and a higher effective threshold. For neural quantum states, stochastic reconfiguration is interpreted as tangent-space ridge regression, with its diagonal shift controlling the bias-variance trade-off under finite Monte Carlo sampling. Multi-shift stochastic reconfiguration reduces checkpoint-local validation residuals and update variance relative to fixed-shift SR, at additional computational cost. Together, these contributions show how learned representations, statistical control, and hardware constraints shape the exchange between quantum computing and machine learning. --- ### What Drives Hierarchy-Aware Image Retrieval? Taxonomy Alignment, Objective Choice, and Geometry Authors: Ling Shi Published: 2026-09-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.25638v1 Foundation vision models provide strong generic representations, yet high class-level retrieval accuracy does not necessarily imply that an embedding respects a target semantic taxonomy. We study strict explicit-taxonomy image retrieval on frozen DINOv2 features and ask: when hierarchical retrieval improves, how much of the change is associated with the organization of taxonomy-aware supervision, and how much with the Euclidean-hyperbolic geometry choice? We evaluate higher levels with strict cross-class criteria that exclude finer-grained matches, and compare Euclidean and hyperbolic projections trained with taxonomy-distance regression or a taxonomy-aware supervised contrastive objective. A compute-matched 2 x 2 Geometry x Loss factorial uses the same 768-256-32 projector capacity, optimization schedule, batch order, and fixed 100-epoch budget; the Loss axis denotes the Regression-to-Taxonomy-SupCon objective-family contrast. On CUB, the objective-family contrasts in mean hierarchy mAP (strict middle/high average, excluding Class/Leaf) are +0.0487 in Euclidean space and +0.0414 in hyperbolic space, compared with geometry contrasts of +0.0102 and +0.0030. On NABirds Parent-disjoint retrieval, the corresponding objective-family contrasts are +0.0467 and +0.0440, whereas geometry contrasts are +0.0017 and -0.0009. A semantic-alignment control shows that the true taxonomy substantially outperforms a structure-preserving shuffled hierarchy, while a NABirds curvature/radius control does not support stronger negative curvature as the explanation for the observed hierarchy gains. Across the two taxonomies, the Regression-to-Taxonomy-SupCon contrasts are larger in aggregate than the evaluated geometry contrasts; semantic alignment also matters separately, while geometry remains hierarchy-dependent. --- ### Shallow to Deep: Aligning Token Pruning with Stage-wise Roles in LVLMs Authors: Shuo Zhang, Jintao Tong, Yixiong Zou, Yuhua Li, Ruixuan Li Published: 2026-09-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.25635v1 Large Vision-Language Models (LVLMs) incur high computational costs from redundant visual tokens. Although training-free attention-based multi-layer pruning in the vision encoder stage has been explored as an effective strategy, we find that pruning in shallow layers consistently degrades performance. In this paper, we aim to understand this problem and seek a solution. By analyzing attention patterns across network depth, we find that shallow layers primarily function as edge detectors with chaotic attention maps, while deeper layers transition through local subject recognition and unstable semantic aggregation. To address the misalignment between pruning strategies and network stages, we propose STD, a hierarchical token pruning framework that adapts token selection mechanisms to the functional role of each network stage. STD employs High-Frequency Spectral Analysis in shallow layers to deterministically preserve structural edges, uses Gaussian-Smoothed Attention in intermediate layers to maintain spatial coherence, and introduces a Stability-Adaptive Trigger in deep layers to execute pruning only during semantically stable phases. Extensive experiments show that STD outperforms state-of-the-art pruning methods by 1.1% on LLaVA-1.5-7B with 88.9% token reduction, while also being plug-and-play and highly effective when combined with other methods, and by 2.1% on LLaVA-NeXT-7B with 94.4% reduction, delivering a 3.9x speed-up in the prefilling stage. Our code will be released at https://github.com/Twilight03/STD. --- ### An Exploratory Replica-Overlap Probe of the Grokking Transition Authors: A. C. Opus, J. Q. Lu Published: 2026-09-22 Categories: cs.LG, cond-mat.dis-nn, cs.AI arXiv: https://arxiv.org/abs/2609.25634v1 We trained 64 independently seeded networks in four configurations, continuing each to sustained convergence or a 40,000-epoch ceiling. We then asked whether an RSB-inspired distribution of pairwise weight overlaps changes across the grokking transition. It is the alignment step, not the overlap statistic, that determines what this registered probe can report. The registered implementation permutes hidden units without the corresponding bias and head-internal permutations and therefore does not preserve the network function. Every q_wt value computed through this alignment inherits the defect; q_fn does not, because it is computed from predictions of the unpermuted models. The numerical-precision requirement also failed, and an audit found protocol deviations. Consequently, the pre-registered rule gives no verdict: registered outcome UNDETERMINED (reason code C0_INSTRUMENT_INVALID). These data provide neither a confirmatory null nor a validated reading of the Parisi order parameter. Only frac40 cleared the 12/16 checkpoint-completeness requirement. For this configuration, a post-hoc criterion applied to the same data gave a Hartigan-dip interval containing zero (95% CI for Delta dip = (-0.017, 0.034)), whereas the overlap standard deviation increased by a factor of about 5.6. A post-hoc calibration assigns the dip test zero power at the simulated separations; the interval is therefore uninformative, not evidence of no change. The standard-deviation ratio is the only statistic here with power at the observed effect. Ensemble loss was near-flat only under the pre-specified 1% threshold. Finally, grokking rates of 0/16, 11/16 and 16/16 remain descriptive because train fraction is confounded with split identity. --- ### Robust, Estimator-Agnostic Dynamic 3DGS Compression Authors: Chenjunjie Wang, Zixi Huang, Yao Wang, Jona Ballé Published: 2026-09-22 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2609.25633v1 Dynamic 3D Gaussian splats (3DGS) model time-varying scenes using a separate Gaussian set per frame. While neighboring video frames are highly correlated due to smooth motion, Gaussian representations retain this correlation to varying degrees, depending on whether the estimator tracks them across time. Some 3DGS compression methods integrate the estimation to exploit temporal redundancy; here, we focus on robust compression regardless of the estimator. We concatenate groups of frames into one Gaussian set, augment each Gaussian with a frame index, and pass it to a static (i.e., non-temporal) 3DGS codec, converting temporal redundancy into spatial redundancy. Concatenated sets are spatially partitioned to limit memory. Our technique requires neither a motion model nor knowledge of the training method. Averaged over six N3DV sequences, all six static codecs achieve gains on tracked sets (-42.0% to -71.8% BD-rate) over per-frame coding. On untracked sets, all codecs except HGSC, which appears incompatible with our technique, remain competitive with per-frame coding (-3.5% to +5.0%). We further replace D-FCGS's I-frame coding with our technique while retaining its P-frame coding, yielding an overall BD-rate of -46.2%. We propose to visualize "trackedness" using an inter-frame similarity metric. The project is available at https://wcjj1236.github.io/d3dgs-benchmark. --- ### MachEmbodied-U0: Unified Understanding and Generation Model for Embodied Intelligence Authors: Haoran Wen, Wenfu Wang, Kunsong Shi, Jingke Wang, Wancheng Feng, Yiren Zhang, Yueran Zhao, Xuancheng Zhang, Nanfei Ye, Xingru Chen, Zhaohong Sun, Chengmin Yang et al. Published: 2026-09-22 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2609.25627v1 General-purpose robot control requires models to understand task intent, identify where to interact, capture how the scene evolves, and generate precise actions. Vision-language-action models provide strong semantic priors but typically do not explicitly model scene dynamics, while world-action models couple visual prediction with control without necessarily exposing the task-relevant semantic and spatial structure needed for fine-grained manipulation. We present MachEmbodied-U0 (ME-U0), a unified embodied foundation model connecting understanding and generation experts through a Mixture-of-Transformers architecture. Subtask prediction and affordance grounding guide joint visual-dynamics and action generation via flow matching. Visual dynamics encompass future RGB, depth, surface normals, and optical flow, providing complementary supervision for appearance, geometry, and motion. Multi-rate Rotary Position Encoding (MRPE) aligns visual dynamics with fine-grained control. We pretrain ME-U0 on approximately 4,200 hours of curated demonstrations from robotic datasets and egocentric datasets. Using only the supervision natively available in each downstream benchmark, ME-U0 achieves an average score of 17.66 on the RoboDojo simulation benchmark and average success rates of 99.0\% and 82.5\% on LIBERO and LIBERO-Plus, respectively. We additionally validate ME-U0 on real-world robotic manipulation tasks, demonstrating its effectiveness beyond simulation. Without corresponding downstream supervision, ME-U0 further demonstrates zero-shot subtask prediction, affordance grounding, and visual dynamics on simulated and real-world observations. Overall, ME-U0 combines competitive downstream control performance with transferable task-grounding and visual-dynamics capabilities across simulation and the real world. --- ### Accelerating the Mitigation of LLM Inference Nondeterminism Across GPU Architectures Authors: Liam Cooper, Shinnung Jeong, Hyeran Jeon, Jeffrey Young, Hyesoon Kim Published: 2026-09-22 Categories: cs.AR, cs.LG arXiv: https://arxiv.org/abs/2609.25624v1 Large language model (LLM) outputs are expected to be reproducible under greedy decoding, yet in practice the same model, prompt, and software stack produce different outputs on different GPUs. The root cause is floating-point non-associativity combined with hardware-dependent kernel selection. Inference frameworks select different matrix-multiplication kernels on each architecture, with different parallel reduction orders and unspecified tensor-core arithmetic, and the resulting rounding differences can flip output tokens. Existing solutions have imperfect cross-architecture reproducibility and incur a significant performance penalty. We present a solution employing a set of fixed-configuration fused-upcast GEMM kernels that load 16-bit weights from memory, upcast them to FP32 in registers, and accumulate with IEEE-754 arithmetic in a reduction order that is a pure function of the problem shape and is therefore independent of the device, its SM count, or kernel scheduling. By fixing the floating-point reduction order as a function of problem shape alone, every GPU runs the same operation sequence, so cross-architecture reproducibility of the linear layers reduces to correct IEEE-754 arithmetic rather than to rounding differences staying below a tie-flip threshold. We confirm our solution's linear-layer outputs are bitwise identical across NVIDIA Ampere, Ada, and Hopper GPUs, while running $1.17$ to $3.1\times$ faster end-to-end than the state-of-the-art solution and cutting weight-memory traffic in half. --- ### What Should a Self-Teacher See? Privileged Context Design for On-Policy Self-Distillation Authors: Kanghui Tian, Siyuan Liu, Tianxiang Jiang, Shuai Dong, Yizhuo Li, Tian Ding, Yuan Guo, Songze Li, Haowen Hou, Congcong Wang, Yi Wang Published: 2026-09-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.25623v2 More privileged information does not always make a better teacher. We study this tension in on-policy self-distillation (OPSD), where a self-teacher scores the student's own rollouts under privileged context, conventionally a complete reference solution that bundles the final answer with one particular reasoning path. Holding the student view and training fixed within each scale, we compare that default against three abstractions compiled offline, a named strategy, a method-independent framing, and a problem category, and against an answer-only control that keeps the destination but removes the path. In the primary runs on competition mathematics, the best intermediate contexts improve the in-domain peak mean over the full solution by 1.4 points at 4B and 1.6 at 8B, while storing an order of magnitude fewer hint tokens. Comparisons across three seeds also show positive mean gains for the framing and category contexts at both scales. Answer-only conditioning remains competitive in the primary runs, within 0.2 points of the full solution at these scales. The preferred context varies with student scale and task. Initial teacher-student KL does not order downstream performance. What a self-teacher should see is therefore not everything it could, but the level of abstraction its student can still act on. --- ### ChatT2: An Adaptive Framework for Developing a Large Language Model-Based Agent for Natural Product Domain Research Authors: Yihan Wang, Qiandi Gao, Yihui Zhuang, Liangjun Ge, Heqian Zhang, Jiaquan Huang, Zhiwei Qin Published: 2026-09-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.25620v1 Scientific investigations into microbial natural products (NPs) present significant challenges for novices, largely due to the complexity of microbial systems, biochemical diversity, technical skill requirements, and the demands of bioinformatics and data analysis processes. To address these issues, we introduce ChatT2, a large language model (LLM)-based agent that is specifically tailored to the unique characteristics of bacterial type II polyketides. These polyketides form a structurally distinct and therapeutically important NP family. ChatT2 was developed within an autonomous multiagent framework composed of a mentor, an executor, and an evaluator, each with defined responsibilities. The mentor acts as an intermediary between ChatT2 and the user, utilizing chain-of-thought prompting to refine the intent of the user. Under the guidance of the mentor, the executor synthesizes multimodal information via retrieval-augmented generation techniques and seamlessly integrates bioinformatics and cheminformatics tools. The evaluator ultimately assesses the output of the executor to ensure the richness and accuracy of the retrieved information. Our research highlights how ChatT2, designed with this multiagent framework, addresses the challenges faced by general LLMs in terms of understanding limited, specialized corpora and complex biological information and provides both experts and novices with a valuable tool for exploring various NPs of interest. The ChatT2 webserver can be accessed at https://chatt2.site/#/chat. --- ### Reasoning-Preserving Fine-Tuning of Post-RL LLMs with Null-Basis LoRA Authors: Wenzhi Fang, Nicholas Tzou, Lazar Valkov, Srinivas Chappidi Published: 2026-09-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.25618v1 Reinforcement learning (RL)-based post-training has become an effective approach for eliciting reasoning capabilities in large language models (LLMs). However, adapting post-RL models to new knowledge domains or behaviors through subsequent supervised fine-tuning (SFT) can severely overwrite these capabilities. Existing approaches mitigate such forgetting through experience replay, specialized initialization, or constrained optimization using gradient projection, but either provide limited preservation or incur substantial training overhead. Our analysis shows that reasoning activations concentrate in low-dimensional subspaces, leaving substantial null-space capacity for adaptation, and that the corresponding approximate null spaces can be reliably estimated from a modest number of examples. Motivated by these observations, we propose Null-Basis Low-Rank Adaptation (NB-LoRA), a parameter-efficient method for adapting post-RL LLMs while preserving their acquired reasoning ability. We formulate reasoning retention as a layer-wise hidden-state preservation constraint and construct a fixed approximate null basis from reasoning activations. LoRA updates are then reparameterized through this basis, enforcing the preservation constraint throughout fine-tuning. Extensive experiments across multiple RL-trained LLMs and diverse downstream tasks show that NB-LoRA matches standard LoRA in adaptation performance, maintains reasoning accuracy near pre-fine-tuning levels, and generalizes this preservation to held-out reasoning benchmarks. --- ### Evidence-gated multimodal parsing and vectorization of architectural floor plans Authors: Hongxuan Chen, Wenda Wang, Jiachen Lu, Qirui Shen, Zilong Huang, Lei He, Xinyue Dong, Weixin Huang Published: 2026-09-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.25615v1 Architectural floor plans remain a high-friction barrier to archive digitization and early design-model preparation because heterogeneous graphics encode spatial semantics and editable geometry together. We introduce SALI-FP, an evidence-gated multimodal pipeline that converts a plan into reviewable semantic maps, objects, vectors, and relation records while constraining local revisions by image evidence. In a full production audit of 11,534 heterogeneous plans, SALI-FP produced structured outputs for every plan, including 752,510 valid polygon-bearing objects. The same output form has supported initial drawing digitization and design-model preparation in practical design work. Public-benchmark calibration is paired with a 30-case matched visual evidence set in Appendix F, where room-scale coverage, openings, oblique boundaries, and circulation continuity can be inspected directly. SALI-FP offers an engineering-oriented interpretation-to-geometry workflow for reviewed CAD/BIM preparation and existing-building information recovery. --- ### Qwen3.8-Omni: Towards Native Omni-Modal Agents Authors: Qwen Team Published: 2026-09-22 Categories: cs.CL, cs.CV, cs.MM arXiv: https://arxiv.org/abs/2609.25611v1 We introduce Qwen3.8-Omni-Flash, a natively multimodal agentic model for real-world multimodal productivity. Compared with previous omni models, which primarily emphasized perception and interaction, Qwen3.8-Omni-Flash substantially improves multimodal understanding and reasoning, as well as performance on long-horizon agentic tasks. These capabilities are supported by a native multimodal co-training strategy that preserves strong text-domain capabilities while facilitating the transfer of agentic capabilities from text to audio and video tasks. The model inherits the sparse mixture-of-experts (MoE) architecture of Qwen3.8-Next and extends the context window to one million tokens, supporting long-context multimodal reasoning and long-horizon planning. These advances enable integration into production workflows as a primary agent or a specialized sub-agent, supporting video editing, long-form audio and video translation, music-conditioned music video or movie generation, and video-based note or omni-skill creation. To address the lack of native audio and video support in existing agent harnesses, we release Qwen-MM-Plugins, a lightweight open-source plugin framework for multimodal productivity. We further frame real-time multimodal interaction as a system-level challenge requiring orchestration of context and memory management, tool use, and sub-agent delegation. Accordingly, we release Qwen-Live-Harness, an open-source framework for building responsive, real-time multimodal agents based on Qwen3.8-Omni-Flash. Extensive evaluations demonstrate that Qwen3.8-Omni-Flash achieves strong performance across multimodal understanding, reasoning, long-horizon agentic execution, and video productivity tasks. These results and the accompanying open-source tools support Qwen3.8-Omni-Flash as a practical foundation for deploying natively multimodal agents in research and production. --- ### ArticleMiner: Ontology-Guided Knowledge Graph Construction from Scientific Publications Authors: Md Abrar Jahin, Craig A. Knoblock, Jay Pujara Published: 2026-09-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.25607v1 Scientific papers keep much of their quantitative content in tables and supplementary files, where a number means something only through its header, caption, unit, analytical method, and the conventions of its field. Recovering the rows and columns of a table is therefore not the same as recovering the scientific fact it reports. Most semantic table-interpretation methods assume that a clean table is already available and subsequently map its cells or columns to ontology terms, whereas most publication-level extraction systems are designed for a single domain. We study a middle path: a shared process that reads a paper and its supplementary files, gathers evidence from several parsers and a language model, and reconciles that evidence, while a bounded human-authored task module for each task supplies the domain meaning. The module lists the canonical names the graph may use, the surface forms that map to them, a small set of derivation rules and validity constraints, an identity key, and the bindings used to write RDF. It defines what a task is allowed to emit; it does not try to list every convention of a field. We build four such modules (for drug-discovery chemistry, materials science, machine learning, and mineral geochemistry) in the ArticleMiner framework, and evaluate them on 163 papers, including a new geochemistry benchmark with expert-curated ground truth. In comparisons against a same-LLM few-shot baseline, the point estimates favor ArticleMiner on all four tasks, with uncertainty on the two smaller benchmarks. The geochemistry comparison also includes access to supplementary files, so its improvement cannot be attributed to domain guidance alone. --- ### Generalized Deep Regression for Repeated Measurements Authors: Kexuan Li Published: 2026-09-22 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.25605v1 In this paper, we study the estimation of a marginal regression function from independent units with repeated binary, count, or continuous responses using ReLU deep neural networks. In the model, we assume that the dependence is generated by an unobserved random mean function within each unit. We then fit a neural network with a convex generalized regression loss. We show an oracle inequality by separating conditional measurement variation from between-unit variation. In addition, we prove that with $n$ units and $m$ measurements per unit, ReLU networks can attain an integrated mean squared error of order $n^{-1}+(nm)^{-2β/(2β+d)}$, up to logarithmic factors, over $β$-Hölder classes. We also derive a weighted oracle inequality for unequal cluster sizes and a rate for compositionally smooth functions. For pointwise ensemble inference, we give a projection central limit theorem and prove infinitesimal jackknife consistency under an explicit asymptotic linearity condition. Simulations and real data examples are provided to support our theoretical findings and practical implications. --- ### Ultra-fast Neural Inference for Stochastic Gaussian Splatting Denoising Authors: Chenxiao Hu, Hao Zhang, Yanchen Zhang, Meng Gai, Guoping Wang, Sheng Li Published: 2026-09-22 Categories: cs.CV, cs.GR arXiv: https://arxiv.org/abs/2609.25604v1 Stochastic rendering eliminates the sorting and alpha blending process in Gaussian splatting, at the cost of introducing spatial noise. Formulating temporal denoising over the pixel stream shared by view-consistent stochastic splatting renderers, we propose a temporal neural denoiser validated on stochastic 2D Gaussian Splatting rendering, combining dual-path exponential moving average accumulation, per-pixel learned trust prediction for history validation, a fixed anisotropic spatial filter and a variance-gated composition with stabilization. The denoiser suppresses the noise, achieving temporally stable, visually compelling outputs during free camera navigation, all while retaining the sort-free, blend-free rasterization performance. The combined pipeline retains a PSNR gap to sorted alpha-blending renderers, but the denoiser's overhead stays below the time saved by removing sorting and blending. --- ### Rewired or Gated? How Instruction Tuning Shapes Knowledge-Conflict Circuits in LLMs Authors: Shubham Santosh Pandere, Gautam Ranka, Ritika Varshney, Navya Deshmukh, Roushni Sareen, Roshan Kumar Singh Published: 2026-09-22 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2609.25602v1 In language models, the choice between believing the prompt and believing the weights is made by a handful of identifiable attention heads. Instruction tuning changes how models behave under conflict, but whether it rewires the underlying circuit or merely gates/reweights already present components, remains unknown. We provide the first mechanistic base-vs-instruct comparison of conflict-resolution circuits, across three families (Llama-3.2-3B, Qwen-2.5-3B, Gemma-3-4B). Five independent methods, node and edge attribution, superposition role analysis, causal ablation, and path patching, converge on gating, with the same heads, in the same late-layers, are found to be reweighted rather than replaced with a high node overlap (0.60-0.82). Behaviorally, tuning shifts models toward parametric memory, making instruct models reject a terse counterfactual context far more than base ones, the opposite of a naive user-following expectation. Yet this added skepticism is a factor of framing since it disappears when the same false claim is delivered as a coherent, evidential passage. The robustness that instruction tuning buys against terse injection is therefore real but narrow. More broadly, we believe that because the conflict circuit is preserved rather than rebuilt, interpretability and control tools calibrated on base models should transfer directly to their deployed instruct siblings. --- ### Observer Choice and Threshold Selection in Retinal Vessel Segmentation: A Subject-Separated Evaluation Authors: Wenhao Xu, Yixian Kong, Ting Pan, Changwei Wang, Feilong Wang, Rongtao Xu Published: 2026-09-22 Categories: cs.CV, eess.IV arXiv: https://arxiv.org/abs/2609.25597v1 The annotation used to select a segmentation threshold is part of the evaluation protocol, yet its effect is easily conflated with model quality. We examine this choice for retinal vessel segmentation using all 28 CHASE DB1 images and both human annotations. A fixed seven-fold protocol keeps both eyes of each of the 14 subjects together. Random forests and Extra Trees are fitted against observer 1 with three random seeds, yielding 42 fits. Five threshold policies share identical score maps: fixed 0.50, observer-1 tuning, observer-2 tuning, mean-observer tuning, and maximin tuning of the per-image lower observer Dice. For random forests, maximin changes the threshold in 19 of 21 fits, but worst-observer Dice decreases from 70.53 percent to 70.45 percent. The paired difference is -0.073 percentage points, with a conditional subject-bootstrap 95 percent interval of (-0.384, 0.238). Extra Trees shows the same direction. Identical observer-1-tuned random-forest masks score 73.66 percent against observer 1 and 71.06 percent against observer 2. The results support explicit reporting of both the threshold-selection reference and evaluation reference; they do not support an accuracy benefit from maximin tuning in this cohort. All splits, raw predictions, metrics and code are supplied. AI assistance is disclosed. --- ### Evaluating Coding Agents on Kernel Exploit Generation Authors: Junyoung Jang, Gwanhyun Lee, Hwiwon Lee, Kyuheon Kim, Jongseong Kim, Jinho Jung, Lingming Zhang Published: 2026-09-22 Categories: cs.AI, cs.CR arXiv: https://arxiv.org/abs/2609.25591v1 Coding agents now find real vulnerabilities in production software. However, bug discovery results do not measure whether agents can construct exploit primitives. We introduce KEX-bench, a benchmark for evaluating coding agents on exploit primitive generation against real operating-system kernels. KEX-bench contains 45 task instances across 40 Linux and Windows CVEs, covering kernel address leak, instruction-pointer control, heap read, heap write, and arbitrary address write. Each task runs in an isolated virtual machine, exposes controlled tools, and uses a deterministic verifier to check primitive-specific success. We evaluate state-of-the-art coding agents paired with frontier and open-weight models under fixed tool-call budgets. Without a reference proof of concept (PoC), the strongest configuration solves 1 of 20 Windows tasks (5.0%) and 14 of 25 Linux tasks (56.0%). With a reference PoC, the strongest configuration solves 31 of 45 tasks (68.9%). This highlights the gap where agents reach kernel crashes but fail to shape kernel state into exploit primitives. We release KEX-bench for reproducible research on AI-assisted exploitation at https://kex-bench.github.io. --- ### Transformer Heads Looking for Order Authors: Jasper van Doornmalen, Alexander Kozachinskiy, Corinna Mathwieser, Tomasz Steifer, Felipe Urrutia, José Verschae, Przemysław Andrzej Wałȩga Published: 2026-09-22 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.25588v1 In this note, we show that the problem of checking, whether a sequence of bits is ordered, is not doable by 1-head 1-layer transformers but is doable by a 2-head 1-layer transformer. Unlike similar previous results, our results assume the model where transformers have an output MLP. --- ### Deflecting the Value Compass: Interacting with Large Language Models Temporarily Shifts Human Value Priorities Toward Personal Focus Authors: Hasibur Rahman, Malak Sadek, Smit Desai Published: 2026-09-22 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2609.25586v1 Large language models increasingly support decisions where values are in tension, yet little is known about whether interacting with them changes which values users prioritize. In a preregistered study, 200 U.S. adults interacted with ChatGPT, Claude, or Gemini as a thinking partner or read fixed AI-generated considerations. The prompt asked LLMs to support reasoning without recommending a decision and named no values. Participants advised people facing real dilemmas and completed parallel PVQ-RR forms before, immediately after, and one task later. Each LLM condition temporarily shifted value priorities toward personal focus relative to the control (d=0.37-0.51), primarily through increased Self-Enhancement. Participants' advice retained words and meaning from their exchanges. Thus, a brief LLM interaction that neither targets values nor seeks to persuade can reorient values active during judgment without detectable convergence in value directions or advice. --- ### Hi-OPD: Hierarchy-Aware Open-Prompt Detection for Remote Sensing Images Authors: Jinlong Hu, Yi Zhang, Zhiqi Xia, Yikang Zhou, Shunping Ji Published: 2026-09-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.25584v1 Hi-OPD addresses a failure mode left uncontrolled by flat open-prompt training: descendant retrieval need not persist under ancestor queries when multi-source remote sensing annotations exhibit inconsistent granularity and missing labels. A detector may localize \textit{car} and \textit{van} under atomic prompts yet miss the same instances under \textit{vehicle}; flat AP does not expose this cross-level inconsistency. We propose Hi-OPD, a hierarchy-aware open-prompt detector, and construct RS153-HierOPD from 175,644 retained training image/tile records and 3.48M boxes mapped to 153 atomic categories with sparse hierarchy and alias relations. Hi-OPD learns ancestor retrieval through hierarchy-safe negative sampling, path multi-positive supervision, and one-way upward consistency, while per-source risk exclusion handles potentially missing labels. ConvVPE converts K-shot support boxes into text-compatible embeddings using detector-native features and the shared contrastive head. On Track A, Hi-OPD obtains 79.7/72.3 AP50 on DIOR/DOTA-v2.0, above the literature-reported OpenRSD results of 76.7/71.8. Under controlled training on the original converted annotations, the full hierarchy recipe raises DOTA-v2.0 parent AP50 from 7.2 to 71.5 and FAIR1M grandparent AP50 from 31.6 to 71.4, while DOTA-v2.0 atomic AP50 changes from 71.4 to 72.3. The text path reaches 99.7% CAR50 (0.3% violation) across the three common sources and 99.9%/0.1% on FAIR1M grandparent relations. On held-out VEDAI, text AP50 is 75.9, 6.2 points above OpenRSD. Joint AP and CAR show that explicit hierarchy training repairs this failure mode while retaining atomic detection and prompt transfer. --- ### EMGBlend: Heterogeneity-Aware Self-Supervised Pretraining for Gesture and Force Decoding Authors: Yuwei Jia, Cheng Zhong, Jinyang Yu, Zhe Cui Published: 2026-09-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.25582v1 Public surface electromyography (EMG) datasets vary widely in electrode layout, channel count, frequency support, and size. Simply mixing them for pretraining can misalign channel semantics, introduce spectral targets that some devices cannot observe, and let large or high-channel-count datasets dominate learning. We introduce EMGBlend, a self-supervised framework designed around these differences. It combines shared channel patches with geometry-aware attention, restricts spectral targets to each recording's supported frequency band, and balances exposure across data sources. We pretrain a 109M-parameter model on 11 public EMG sources and evaluate it on gesture recognition, continuous-force regression, and contact classification. EMGBlend consistently outperforms matched random initialization and waveform reconstruction controls. Fixed-budget source controls show that multi-source pretraining improves gesture recognition and remains competitive for force decoding. Ablations confirm that geometry, band-aware targets, and source balancing each contribute to transfer, although cross-person NinaPro force estimation remains difficult. Overall, EMGBlend shows how heterogeneous EMG datasets can be combined through explicit mechanism design rather than simple concatenation. Code is available at https://github.com/tamanano/EMGBlend --- ### Gaze responses to false-positive computer-aided detection prompts during colonoscopy: a paired-video and real-time eye-tracking study Authors: Te Luo, Yan Zhu, Peiyao Fu, Ruijie Yang, Xian Yang, Quanlin Li, Pinghong Zhou, Shuo Wang Published: 2026-09-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.25581v1 False-positive computer-aided detection (CADe) prompts may divert endoscopists' attention during colonoscopy, yet the attentional impact of individual prompts remains unclear. We used event-locked eye tracking to quantify gaze attraction and attention occupation in complementary retrospective and prospective studies. In a retrospective paired-video experiment, 3 senior and 2 novice endoscopists viewed 60 colonoscopy videos with and without CADe. The prospective study recorded gaze during 42 real-time CADe-assisted colonoscopies performed by 9 senior endoscopists. Screened CADe prompts outside expert-annotated lesion windows were classified as false-positive artifact events. False-positive prompts attracted gaze in 48.6% (68/140) of retrospective observations and 65.2% (533/817) of prospective events. Among attraction events with complete recovery, median attention occupation lasted 1000 ms in the retrospective study and 1100 ms in the prospective study. Corresponding median prompt durations were 33 ms and 267 ms, with median time amplifications of 17.55-fold and 5.15-fold, respectively. In paired retrospective comparisons, visible artifact prompts drew gaze closer to the prompted region than did the same-coordinate unassisted reference. Secondary retrospective analyses showed high lesion gaze recognition without and with CADe (98.0% versus 99.0%). First gaze entry into lesion regions occurred 147.8 ms earlier with CADe. Across controlled and real-time clinical settings, false-positive CADe prompts frequently captured gaze, with attention persisting beyond prompt visibility. These findings support considering prompt-related attentional burden in CADe evaluation and design. --- ### Agentic Building-Aware Satellite Gaussian Splatting for Auditable Urban DSM Reconstruction Authors: Wentao Sun, Zhengsen Xu, Yiping Chen, John S. Zelek, Jonathan Li Published: 2026-09-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.25578v1 Urban-scale 3D reconstruction from satellite imagery supports disaster response, city monitoring, and geospatial digital twins, yet neural rendering methods typically optimize average visual fidelity rather than the structures that analysts inspect first: buildings. We present an agentic building-aware satellite Gaussian Splatting workflow that uses Segment Anything-derived building masks as semantic priors and an Agentic Reconstruction Controller to select, verify, and record DSM reconstruction policies. On the DFC2019 JAX\_004 scene, building-aware weighting reduces building-region DSM MAE from 0.844 m to 0.806 m, showing that semantic priors can shift reconstruction capacity toward analyst-critical regions. A staged schedule provides a balanced operating point, improving full-scene MAE from 1.362 m to 1.349 m while retaining a building gain. Across four JAX scenes, the Agent selects validated policies for both general DSM and building-focused DSM objectives, and produces building-inventory metadata and per-scene decision records. The system combines semantic priors, policy selection, region-specific DSM metrics, and DSM-derived GIS surface products for auditable urban 3D analysis. --- ### Scalable Minimum-Volume Simplex Estimation with Non-asymptotic Analysis Authors: Jun Li, Yanlong Guo, Zhaozhao Zeng Published: 2026-09-22 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.25576v2 We study the estimation of a $K$-dimensional simplex from $N$ i.i.d.\ points sampled uniformly from its interior; the observations are convex combinations of $K+1$ unknown prototypes. Existing polynomial-time estimators need cubic per-sample work or $O(NK)$ storage and are impractical at $N\sim 10^6$--$10^8$. We propose DeepMVSA, which re-expresses the minimum-volume principle in neural implicit form: a lightweight coordinate network generates the mixing weights and a triangular LU-type parameterization the dual simplex matrix, reducing the trainable-state memory to $O(K^2)$, independent of $N$, and the cost per data pass to $O(NK^2)$. We prove a non-asymptotic sample-complexity bound of the polynomial-time benchmark order for a localized surrogate estimator; an oracle inequality for every global minimizer of the neural objective, with volume-inflation control and an explicit shrinkage bias; a conditional end-to-end error budget separating statistical, approximation, optimization, and enclosure-residual terms on an explicit envelope event; and two-point lower bounds: at any noise level $σ>0$ fixed independently of $N$, the $N^{-1/2}$ scaling is unimprovable in its $N$-exponent. Experiments with up to $N=10^8$ synthetic observations are consistent with the predicted accuracy and scaling, and feasibility on real scenes of $\sim 10^7$ pixels is demonstrated. --- ### Direct Optimization of Generators for Search in Automated Theorem Proving Authors: Adam Ousherovitch, Ambuj Tewari Published: 2026-09-22 Categories: cs.AI, stat.ML arXiv: https://arxiv.org/abs/2609.25575v1 Fine-tuned Large Language Models (LLMs) significantly advance Automated Theorem Proving (ATP), but are often deployed as guiding policies within tree search rather than for single-attempt generation. Recent work shows cross entropy is suboptimal for an LLM used in flat search strategies such as aggregation or filtering and that work has developed new loss functions to correct this misalignment. Extending this alignment to tree search is more challenging: proof discovery depends on exploration and recovery through off-trace states that supervised demonstrations do not reveal. We extend Compute-Aligned Training (CAT) to this setting through an abstraction of policy-guided search, deriving tractable, trace-supported losses. Alongside these search-aware losses, we introduce a search-agnostic uniform-allocation (UA) loss that accounts for the budget without specifying the specific search. Both induce scalar weights on per-tactic cross-entropy gradients. We characterize how off-trace behavior affects the search-aware weights, including conditions for vanishing approximation error at large budgets. On a Lean benchmark, both approaches achieve higher observed proof-success rates than cross-entropy across six search strategies, with strong results from a single shared UA adapter. Budget sweeps show larger gains over cross-entropy at 16 than at 256 expansions, implying CAT scales with test time compute. --- ### A Behavioral Trait Leaks into Preferences: Diagnosing Trait Interference in LLM User Simulators Authors: Chaehyun Kim, Sein Kim, Hongseok Kang, Chanyoung Park Published: 2026-09-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.25572v1 LLM-based user simulators aim to bridge the offline-online gap in recommender evaluation by emulating users through injected traits, where preference attributes determine what a user engages with and a behavioral activity trait governs how long they browse. However, we show this intended trait independence collapses during simulation, causing two failures: (i) Trait Interference, where amplified activity distorts preference boundaries and forces interactions with mismatched items to sustain browsing, and (ii) Evaluation Invalidity, where satisfaction scores inflate with activity-driven page counts despite taste mismatches, biasing evaluation toward trait distributions rather than recommender performance. To resolve this, we propose PQA, a page-level quality anchoring method that guides simulators using a personalized anchor reflecting each user's intrinsic preference standard. By assessing whether a page meets this standard before further browsing, PQA enables proactive exits from low-quality pages, letting the activity trait retain its intended role of modulating browsing depth within preference-conforming pages. Experiments show PQA mitigates trait interference and improves the reliability of LLM-based simulator evaluation under activity shifts. Our code is available at https://github.com/chaehyun1/PQA --- ### Recovering Agentic Sovereignty: Mitigating the Consensus Paradox via Contrastive Epistemic Decoding Authors: Dahlia Shehata, Ming Li Published: 2026-09-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.25570v1 Large language models (LLMs) exhibit a parametric vulnerability to adversarial swarm consensus. To mitigate this sycophancy, we introduce Contrastive Epistemic Decoding (CED), a zero-shot inference intervention. Unlike standard Contrastive Decoding (CD) which relies on a weaker secondary model, CED utilizes a dual forward-pass on a single architecture to isolate conformity bias. By introducing a novel asymmetric, zero-bounded probability clamp and discrete top-k truncation mask, CED mathematically suppresses toxic consensus tokens without causing grammatical collapse. Evaluated across 7,200 paired trajectories on complex benchmarks (GAIA, SWE-bench, Multi-Challenge) using Gemma-2 (9B), Llama-3.1 (8B), and Mistral v0.3 (7B), CED successfully neutralizes architectural and positional biases. By reducing cognitive loafing by up to 33.00% absolute, CED drives significant performance gains, yielding up to a 30.75% accuracy recovery. Regaining sovereignty induces distinct architectural behaviors---passive task-focus in Gemma-2 and active refutation of the simulated swarm in Llama-3.1---showing CED decouples compliance from capability without fine-tuning. --- ### SambaGraph: Action-Reaction Spatio-Temporal Graphs for Soccer Tactical Response Modeling Authors: Abel A. Reyes-Angulo, Henry O. Velesaca, Steven Araujo Published: 2026-09-22 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.25569v1 Soccer tactics are interactive: an attacking action changes the opponent's defensive problem, and the observed response depends on the multi-agent match state. We introduce SambaGraph, an action--reaction spatio-temporal graph dataset and benchmark for soccer tactical response modeling. From tracking and event data for all 64 matches of the 2022 FIFA World Cup, we curate 4,070 action-centered episodes represented as temporally aligned 23-node player--ball graph sequences with attack/defense views, response labels, and 26,270 split-safe attack--defense pairs. We study three questions: whether observed responses can be classified from graph episodes, whether successful defenses can be retrieved for a query attack, and whether graph-derived summaries support grounded LLM reasoning. A compact signature MLP obtains $0.796\pm0.007$ macro-F1 for response classification, while a fused graph--signature dual encoder reaches $0.471\pm0.029$ Hit@5 and $0.655\pm0.051$ Hit@10 for full-bank defensive retrieval. Hard negatives maximize pair discrimination but not retrieval quality. Local LLMs underperform supervised encoders for direct classification and do not improve over a strong original order in eight-candidate reranking, but they provide grounded tactical rationales. These results position SambaGraph as a reproducible benchmark for graph-based soccer strategy-response research. Code and dataset are available at: https://github.com/areyesan/SambaGraph. --- ### RootQuantV2: Adapting a Vision Foundation Model for Root-Trait Regression from Minirhizotron Imagery Authors: Kinjalk Parth, Sebastian Varela, Andrew D. B. Leakey Published: 2026-09-22 Categories: cs.CV, cs.AI, cs.LG, q-bio.QM arXiv: https://arxiv.org/abs/2609.25567v1 A lack of high-throughput phenotyping solutions for root traits in field-grown crops has severely constrained understanding and improvement of below-ground traits and processes. Minirhizotrons are the standard non-destructive root-phenotyping method in field environments. Computer vision solutions are needed to allow automated trait estimation at scale, but training data is scarce and human annotations are often inaccessible because they reside in proprietary software that only exports per-image scalar totals of root length and surface area. Nevertheless, large numeric archives of these root traits already exist. RootQuant showed that the traits can be predicted directly from the whole image by regression, thus removing manually traced masks from the pipeline; RootQuantV2 takes that idea further by replacing RootQuant's CNN backbone with a self-supervised ViT. We adapt a frozen DINOv3 ViT-L/16 with a hybrid parameter-efficient scheme. Training only 11.9M parameters (3.78% of the model), RootQuantV2 achieves length and area $R^2$ of 0.950 and 0.930, respectively, while lowering length/area RMSE by 24.3%/20.7% over RootQuant. RootQuantV2 thus repurposes legacy numeric archives for high-throughput, automated root trait estimation. --- ### AkasicMEM: Governed Enterprise Memory for Agents Authors: Jeongmin Bae, Yongjae Kim, Kyoung Hur, Donghyoung Han, Min-Soo Kim Published: 2026-09-22 Categories: cs.DB, cs.AI arXiv: https://arxiv.org/abs/2609.25563v1 Agent memory enables enterprise agents to retain knowledge acquired during work and reuse it across tasks and agents, turning execution experience into persistent organizational knowledge. Realizing this potential requires both source--memory integration, through which enterprise sources and accumulated memory can be utilized together, and memory governance, through which shared memory remains subject to organizational policies throughout its lifecycle. These requirements interact when information from enterprise sources persists in memory. As this information is repeatedly derived and reused under changing principals and policies, source restrictions may be bypassed, resulting in information leakage. Preventing such leakage requires authorization continuity, under which source restrictions remain effective throughout source-to-memory and memory-to-memory derivation and reuse. Existing approaches address these concerns individually, but do not treat source--memory integration, memory governance, and authorization continuity as combined core design targets across the memory lifecycle. We define Governed Enterprise Memory as agent memory designed around this combined scope and present AkasicMEM as its realization. AkasicMEM realizes authorization continuity through transitive lineage, policy composition during memory formation, and policy re-evaluation during retrieval. It is built on GraphAI's AkasicDB, a unified vector--graph--relational database whose storage and execution substrate enables the underlying operations of these mechanisms to be jointly optimized and executed. --- ### IndustrialVLA-Bench: A Traceable Multi-Axis Evaluation of Open Robot Policy Models Authors: Yiqi Wang, Zhifeng Rao, Jiaqi Zhang, Xiaoyang Li, Zhangkai Wu, Yiqun Duan, Mingkai Zheng, Fei Wang, Shan You, Taotao Cai Published: 2026-09-22 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2609.25562v1 Open robot policies increasingly follow two paradigms: vision-language-action models (VLAs) directly map observations and instructions to actions, whereas world-action models (WAMs) incorporate learned video or world dynamics into policy learning or action generation. Although both target the same manipulation tasks and represent alternative design choices, they are commonly reported under different evaluation protocols, leaving their capability, robustness, language sensitivity, and deployment-cost trade-offs unclear. We present IndustrialVLA-Bench, an evidence-aware evaluation of six released VLA and WAM systems under a unified reporting schema. It separately evaluates clean capability on LIBERO, non-language robustness on LIBERO-Plus, instruction sensitivity on LIBERO-Para, and observed execution cost. Reported task scores aggregate three complete evaluations with distinct random seeds under a fixed checkpoint and inference configuration. Across all six systems, clean LIBERO averages differ by only 1.58 points, whereas robustness and paraphrase summaries span 14.62 and 31.08 points. Restricting every comparison to the three protocol-faithful systems preserves the effect (1.36, 14.62 and 23.10 points), so the diagnostic separation reported here does not depend on the weaker evidence tiers. We additionally report observed inference latency, peak memory, runtime mode, and an evidence status for every system. Protocol-faithful, near-reproduction, and pending-verification entries remain visibly separated; only protocol-faithful entries support strict comparisons. Rather than claiming universal superiority of either paradigm, IndustrialVLA-Bench provides traceable evidence for comparing released robot policies on shared practical criteria. Code and evaluation records are available at https://github.com/xiaoqi-7/IndustrialVLA-Bench. --- ### HABILIS Brain 0: Geometry-Change Supervision for Vision-Language-Action and Residual Flow Recovery Authors: Jinu Pahk, Jesoon Kang, Taegeon Park, Jisu An, Soo Min Kimm, Jaejoon Kim, Byoung-Tak Zhang Published: 2026-09-22 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2609.25558v1 Vision-language-action policies benefit from geometric supervision, but current-frame geometry alone does not explicitly describe the changes associated with manipulation. This design is motivated by the goal of learning an embodiment-agnostic visual interface that can be pretrained across robot and egocentric video before robot-specific action alignment. We introduce Geometry-Change VLA (GC-VLA), which learns to predict multiview future-current geometry-change tokens from current observations. Offline frame pairs define a nominal 0.5-second prediction horizon; future observations are used only to construct training targets. Stage 1 trains a geometry-change vision-language model (GC-VLM). Stage 2 introduces a continuous ActionExpert and aligns it with robot actions while stopping action-flow gradients at the VLM interface. Stage 3 enables these gradients to update the trainable VLM components jointly with the ActionExpert. Stage 4 freezes GC-VLA and applies Geometry-Conditioned Residual Flow (GCRF), using a binary intervention router and a single bounded residual velocity policy learned from closed-loop feedback. GC-VLA achieves 95.20% success on LIBERO, and GC-VLA with GCRF achieves 99.55%. Inference uses current observations and the learned GC representation without executing the offline target encoders. --- ### Weakly Supervised Quantum Error Mitigation Authors: Seyed Mohamad Ali Tousi, G. N. DeSouza Published: 2026-09-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.25555v1 Supervised approaches to quantum error mitigation learn a map from noisy circuit outputs to ideal ones, and therefore require the ideal outputs. Producing those ideal outputs demands noiseless classical simulation, whose cost grows exponentially with system size, so supervision is unavailable in exactly the regime where mitigation matters most. We ask whether cheap, individually unreliable signals drawn from circuit structure and hardware calibration can take the place of ideal labels. We assemble sixteen heuristic labeling functions (stabilizer and parity constraints, relaxation and readout characteristics, local depth, gate counts, and neighboring activity), reconcile their disagreements with a probabilistic label model, and read the resulting per-qubit error probabilities as a readout channel whose inverse mitigates the measured distribution. No ideal output enters the training path. On $147{,}000$ five-qubit circuits executed on two IBM devices, the method removes $24.3\%$ (Algiers) and $28.8\%$ (Hanoi) of the Kullback-Leibler divergence to the ideal distribution, against $15.4\%$ and $21.5\%$ for the strongest published analytical baseline, a margin that holds on both devices and lies far outside its bootstrap interval. Supervised neural models trained on ideal distributions remain stronger where such labels exist, and we quantify that gap rather than setting it aside; the method's claim is to the regime where they do not, since the labels they require cannot be computed for the circuits mitigation is needed for. The codes will be released shortly. --- ### Synthesis and editing of multi-instrument audio mixtures using scalar-quantised latents with MIDI Span conditioning Authors: Sungkyun Chang, Keshav Bhandari, Simon Dixon, Emmanouil Benetos Published: 2026-09-22 Categories: cs.SD, cs.LG, eess.AS arXiv: https://arxiv.org/abs/2609.25546v1 Music creation often involves iterative refinement, changing selected musical details while retaining the rest. To support such refinement, we introduce SpanSynth-Edit, a flow-matching model for MIDI-guided synthesis and editing of multi-instrument audio mixtures using low-frame-rate scalar-quantised latents. MIDI Span encodes instrument-labelled note lifecycles as unordered event sets with continuous-valued attributes and pools each set into one conditioning vector per audio-latent frame. The model uses contextual audio for instrument-specific timbre guidance and supports editing by resynthesising the target region from revised MIDI. Experiments on single- and multi-instrument benchmarks show competitive performance and demonstrate within-frame onset control. We also discuss limitations of transcription-based note-adherence evaluation. --- ### DefaultGNN: A Dual-Perspective GNN Framework for Predicting Corporate Default from Buyer-Seller Transaction Networks Authors: Junghoon Kim, Hyunsung Kim, Seungyoon Choi, KyoungYong Park, Jihun Lee, YongGu Ji, Chanyoung Park Published: 2026-09-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.25542v1 Corporate default prediction is a core problem in financial risk management, yet traditional credit models rely heavily on financial statements that are often sparse or unavailable for many firms. Corporate transaction networks offer a complementary view of real economic activity, but how risk propagates through buyer-seller relationships remains underexplored. We conduct a large-scale empirical study using real-world electronic tax-invoice data spanning six years that links transaction histories with default events, revealing that transaction-driven risk is both role-dependent (buyer or seller) and scale-dependent. Based on these findings, we construct multiplex buyer-view and seller-view transaction networks and propose DefaultGNN, a dual-perspective graph neural network-based framework for corporate default prediction. DefaultGNN integrates both views to model how risk flows through transactional relationships, achieving strong improvements over both attribute-based and graph-based baselines, especially for firms with limited intrinsic risk signals. We further provide interpretable network-based explanations by visualizing how distressed trading partners contribute to default risk. In collaboration with a licensed credit rating agency, we validate that DefaultGNN's predictions complement existing credit scoring models, improving approval rates by 7-11%p without increasing default risk among approved firms. The source code can be found at https://github.com/jhkim611/DefaultGNN --- ### A JEPA Recipe for Tabular Foundation Models Authors: Mingyu Jeon, Suwan Cho, Jae Young Suh Published: 2026-09-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.25541v1 Tabular foundation models learn to predict cell values in context, whereas world-model self-supervision asks for prediction in representation space (LeCun, 2022; Assran et al., 2023). On a tabular foundation-model prior, the latent term of a joint-embedding predictive architecture (JEPA) collapsed in our earlier runs and took the encoder with it to a constant map. We report a recipe under which the latent term survives to convergence beside the value objective: the value head reads the encoder field rather than the predictor, and the target is an exponential moving average (EMA) difference. To bound its cost against the value-only arm, both arms train until a plateau rule stops them, with no fixed step budget. A fixed horizon had confounded a slowdown with a ceiling, since the value-only arm was still improving well past the usual budget. At convergence, in one run per arm, the JEPA arm trails the value-only arm across 147 real datasets, 32:70 wins to losses on classification (29:63 with one entry per dataset name) and 8:24 on regression, the margin small on classification and wider on regression, and the count leans the same way in each stratum and each benchmark. The JEPA arm (jepa) needs 1.42 times as many steps as the value-only arm (ds), and 1.66 times its wall-clock, to reach its plateau. --- ### Point Diffusion Mamba: Unified Diffusion-State-Space Modeling for Single-View 3D Reconstruction under Data Scarcity Authors: Wei Zhou, Xinzhe Shi, Xingxing Hao, Xing Hao, Kang Li, Jinye Peng, Ying He Published: 2026-09-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.25538v1 While single-view 3D reconstruction has seen significant progress, extrapolating complex 3D structures from inherently ambiguous 2D observations remains fundamentally ill-posed, particularly in the critically underexplored data-scarce regime. To address this challenge, we propose Point Diffusion Mamba (PDM), a method that integrates the generative power of diffusion models with the efficiency of state-space model for single-view 3D reconstruction under data-scarce conditions. Specifically, PDM employs a lightweight reconstruction module tailored to handle unordered point-cloud inputs effectively. By combining a Local Geometric Aggregation module with Mamba blocks, our approach jointly models global geometric structures and local details. In 3D reconstruction, each point in the initial noisy input requires a precise prediction, yet the high-level features extracted by the Mamba module capture only abstract semantic information from sparse points. To bridge this gap, we introduce the Hierarchical Feature Integration Network, which fuses high-level semantic and local geometric features for each point, overcoming the limitations of token-based point-cloud reconstruction. Furthermore, we propose a Dynamic Weighted Sampling strategy that adaptively unifies 3D generation with single-view reconstruction by leveraging generative priors to enhance reconstruction quality. Experimental results on the ShapeNet and Pix3D benchmarks demonstrate that PDM outperforms state-of-the-art methods, providing an effective solution for 3D reconstruction under data-scarce settings. Code is available at: https://github.com/NWUzhouwei/PDM. --- ### Compressing Long Context into Answer-Aligned Memory Embeddings for LLM Inference Authors: Md Mostafizer Rahman, Md Faizul Ibne Amin, Md Shahajada Mia, Yutaka Watanobe, Fang Liu Published: 2026-09-22 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.25537v1 Large language model (LLM) inference is constrained by the quadratic scaling of self-attention and the linear scaling of the KV cache, increasing latency, energy consumption, and GPU memory demand as context length scales. Existing soft-compression methods either lack query-guided memory selection at inference time, train without answer-targeted supervision, or couple compression tightly to a specific decoder architecture. We propose a Context-to-Answer-Aligned Memory Compression (CMC) framework, which compresses long input contexts into compact Context Memory Embeddings (CMEs) aligned to any frozen decoder's embedding space, reducing inference costs without modifying decoder weights. CMC introduces a two-tier KV cache that combines question-guided CME selection with a local context window, and trains the compressor with answer-targeted distillation from a frozen LLM. Experiments across nine encoder-decoder combinations and four QA benchmarks show that CMC consistently outperforms the baseline, achieving up to 7.3 EM and 4.0 F1 point gains on SQuAD, while reducing inference time and energy consumption by up to 20% and peak reserved GPU memory by up to 50% at 3,000 generation tokens. Ablation studies confirm that each architectural component and training objective contributes to the performance. --- ### Matryoshka attribution: Learning to attribute language model outputs to representations and weights Authors: Aryaman Arora, Kirill Acharya, Nathan Hu, Yanzhe Zhang, Noah Goodman, Dan Jurafsky, Christopher Potts Published: 2026-09-22 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.25518v1 Attributing language model outputs to their internal computations is an open problem in interpretability. Existing methods, which use causal interventions, gradients, or learnable masks, either are infeasibly expensive or struggle to identify actual causally-important internal computations. We propose framing attribution as the problem of identifying nested subsets of internal components which minimise a downstream loss. To learn this task, we introduce Matryoshka Attribution (MAttr), a mask learning method that parametrises the mask with a simple differentiable sigmoid top-$k$ operator. We supervise training over all sparsities simultaneously by randomising $k$ over training, resulting in a learned ordering of components by attribution score. MAttr achieves number 1 on the official leaderboard of the Mechanistic Interpretability Benchmark (Mueller et al., 2025); our method identifies sparse and task-transferrable circuits across varying circuit bases. As a practical application, we show that MAttr can be trained with reinforcement learning to identify weight changes responsible for downstream behaviours in LLM finetuning. We train MAttr on refusal judge scores and find that restoring $1\%$ of Llama 3.1 8B Instruct's weights to their base model state is sufficient to remove refusals while maintaining capabilities. We view MAttr as a successful formulation of interpretability into a learnable objective that we can tackle with gradient descent, and encourage future work along these lines. --- ### Real-World Perception for Autonomous Driving in Adverse Weather: Enhancing Standard Detectors via Foundation-Guided Auto-Annotation Authors: Sepideh Gohari, Goodarz Mehr, Azim Eskandarian Published: 2026-09-22 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.25515v1 Standard deployment-ready object detectors for autonomous vehicles degrade in adverse weather and lighting conditions without being trained on extensive domain-specific data. While large-scale vision foundation models offer robust zero-shot generalization, their high computational cost makes them impractical for real-time deployment. To bridge this gap, we propose a foundation-guided auto-annotation pipeline that enhances standard detectors without architectural changes. We first benchmark three distinct models, YOLOv8, Co-DETR, and SAM3, on our custom real-world driving dataset spanning 25 unique operational scenarios across various route, weather, and lighting conditions. Based on our analysis, SAM3 demonstrates superior accuracy and resilience across all scenarios. Thus, we deploy it as an offline auto-annotator to generate pseudo-labels on the unannotated subset of our dataset. Fine-tuning the baseline YOLOv8 on these annotations yields a 16.04% higher overall mean Average Precision (mAP) and improves cross-environmental stability compared to the baseline model, highlighted by a 32.73% and 28.65% mAP increase in Residential Direct Sunlight and Highway Fog, respectively. These results demonstrate that standard detectors can achieve environmental resilience without the need for extensive manual annotation or architectural modifications. --- ### West-WRF AI 2-km: High-Resolution Prediction of Integrated Vapor Transport and Precipitation Authors: Nazak Rouzegari, Vesta Afzali Gorooh, Agniv Sengupta, Phu Nguyen, Kuo-Lin Hsu, Amir AghaKouchak, Soroosh Sorooshian, F. Martin Ralph, Luca Delle Monache Published: 2026-09-22 Categories: physics.ao-ph, cs.AI arXiv: https://arxiv.org/abs/2609.25512v1 We introduce a stretched-grid artificial intelligence (AI) weather forecasting model with 2-km resolution over the western United States and part of the Northeast Pacific and approximately 31-km resolution elsewhere globally. Forecasting over the western U.S. is challenging because complex topography and atmospheric rivers (ARs) strongly influence orographic precipitation. West-WRF AI 2-km builds on a global model pretrained with a 40-year European Centre for Medium-Range Weather Forecasts Reanalysis v5 (ERA5) dataset and is fine-tuned with the Center for Western Weather and Water Extremes (CW3E) 2-km regional reanalysis to produce autoregressive 6-hourly forecasts of precipitation and integrated vapor transport (IVT). Forecasts are evaluated over winters 2020-2023 using gridded precipitation observations, rain gauges, and AR Reconnaissance dropsondes and are benchmarked against coarser-resolution AI forecasts and regional and global numerical weather prediction (NWP) systems. West-WRF AI 2-km reproduces observed precipitation-intensity distributions, retains fine-scale spectral variability, and produces sharper narrow coastal precipitation bands and localized, terrain-sensitive extremes. Its broader-scale performance remains comparable to coarser-resolution configurations while preserving large-scale skill despite higher resolution. Dropsonde verification shows lower errors and improved categorical skill at the most extreme IVT threshold. Overall, West-WRF AI 2-km provides its greatest value for localized precipitation extremes and intense AR-related moisture transport. --- ### A Deployment Study of Identity-Gated Drone Gesture Control Authors: Diyari Mohammed Salih, Ilyes Chaabeni, Naima Ait Oufroukh Published: 2026-09-22 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2609.25511v1 Vision-based gesture control accepts commands from any hand in the camera field of view, which is unsafe in shared indoor spaces. This paper presents IGate, an identity-gated control stack that includes gesture control and face tracking, in which commands are admitted only when an enrolled operator is verified. The system performs few-shot user enrolment from 20 initial face frames, without prior user-specific training: verification compares an embedding of the current face crop against the enrolled template by cosine similarity, while face tracking uses proportional correction. Gesture control is achieved by classifying extracted hand landmarks using an RBF-SVM trained on a custom dataset. Additionally, a hierarchical finite-state machine handles mode selection, default, and fallback behaviours. The approach is tested on a DJI Tello EDU, each component evaluated offline and in-flight across 270 trials (149 flown). Face verification yields a 0.32% offline equal error rate versus 19.3% in-flight. Under hover-locked conditions, the RBF-SVM gesture classifier outperforms the geometric rule (0.850 vs. 0.651 accuracy), with 82% of this gap stemming from the depth channel. All logs and reproduction scripts will be released. --- ### Hill Sampling for Test-Time Scaling: A Simple and Better Alternative to Repeated Sampling, Evolution, and Training Authors: Jacob Beck, Philip V. Ogren, Ari Kobren Published: 2026-09-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.25510v2 Large language models (LLMs) can improve solutions to verifiable scientific and algorithmic problems by spending additional computation at test time. Recent systems achieve strong results with increasingly elaborate evolutionary search harnesses or by updating model parameters during test-time training. We ask how much of this machinery is necessary. We introduce Hill Sampling, a form of hill-climbing optimization that repeatedly samples candidate programs from a frozen LLM, retains the best program found so far, and conditions all subsequent samples on that program. We evaluate the method on circle packing, sums and differences of sets, and Erdos' minimum-overlap problem using three open-weight models. Hill Sampling sets a new state of the art on circle packing among published methods, improves over the AlphaEvolve reference on Erdos' minimum-overlap problem, and achieves strong results on sums and differences of finite sets. The circle-packing and Erdos results require only hours of wall-clock time on eight NVIDIA H100 GPUs. We compare Hill Sampling against what is, to our knowledge, the largest application by parameter count of evolution strategies (ES) to LLM weights at test time. Surprisingly, when evaluating a method by the best program it generates, we find that learning model weights via ES is worse than invoking ES with a learning rate set to zero, i.e., using random weight-space perturbations to search for better models. Moreover, repeated sampling outperforms both ES methods, and Hill Sampling is the strongest of all. These results suggest a simple test-time compute allocation strategy: repeatedly sample edits to the best verified solution found so far before introducing additional complexity, such as adding archives, diversity mechanisms, evolutionary scaffolds, or test-time parameter learning. --- ### SMTB: Fast Structure-Mapping with Tight Bounds Authors: Daniel Weitekamp, Christopher MacLellan Published: 2026-09-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.25508v1 Structure-mapping forms analogies by aligning systems of relationally connected elements based on shared structure instead of surface features. We introduce a new structure-mapping algorithm: Structure-Mapping with Tight Bounds (SMTB) that is 5--15x faster than the structure-mapping engine (SME) and about 50\% better at finding mappings in large nested domains. SMTB is part of the broader Cognitive Rule Engine (CRE) project, a flexible multi-language-compatible framework with an accessible Python interface to state-of-the-art C++ implementations of core algorithms commonly used in cognitive systems such as pattern matching, planning, and structure-mapping. CRE and SMTB are designed to work with a wide range of representation choices. Unlike SME, which biases higher-order correspondences in tree-like predicate logic, SMTB maximizes relational connectivity without privileging higher-order relations. This allows SMTB to work just as well over arbitrary relational graphs as it does in tree-like domains of nested predicate logic. We discuss situations where privileging "higher-orderness" in structure-mapping can cause issues, and illustrate how SMTB avoids failure modes that SME would encounter in these situations. We also provide an evaluation comparing SMTB to SME v4 over 5845 domain pairs from the SME corpus. --- ### FAST-ML: A Hybrid Physics-Machine Learning Framework for Tropical Cyclone Intensity Forecasting Authors: Shijie Xiao, Jonathan Lin, Thomas Ehrmann, Ali Sarhadi Published: 2026-09-22 Categories: physics.ao-ph, cs.LG arXiv: https://arxiv.org/abs/2609.25505v2 Rapid intensification (RI) remains one of the most consequential and difficult aspects of tropical cyclone (TC) forecasting. Although full-physics numerical weather prediction models can represent the processes governing RI, resolving storm-environment interactions remains computationally expensive, while purely data-driven approaches often lack physical interpretability. We present FAST-ML, a hybrid framework that bridges data-driven efficiency with physical constraints. A physically informed dual-stream neural parameterization ingests 3D ERA5 fields to diagnose ventilation controls---environmental wind shear and mid-level entropy deficit. By optimizing these parameters end-to-end through a differentiable FAST intensity model, this architecture establishes a robust new paradigm for observation-driven parameter optimization, ensuring storm evolution remains strictly governed by thermodynamic principles. By better capturing the storm's continuous intensity evolution, FAST-ML improves upon its physical baseline, reducing ensemble CRPS across forecast lead times, with a reduction of approximately 31% at 60 h and nearly halving the RI false alarm ratio without sacrificing detection skill. In a 100-member ensemble configuration, FAST-ML produces intensity forecasts comparable to FNV3 for selected storms under the evaluated input configurations. Furthermore, zero-shot tests on selected Eastern Pacific storms provide encouraging evidence of cross-basin transferability. FAST-ML provides a modular intensity forecasting framework that can be coupled with externally supplied storm tracks and environmental fields. It demonstrates that observation-driven parameter learning within physically constrained dynamics simultaneously enhances accuracy, interpretability, and computational efficiency. --- ### Tracing mechanisms of sycophantic agreement in language models Authors: Sixing Chen, Zhuofan Josh Ying, Logan Riggs Smith, Jeremy Wertheimer, Natalie Shapira Published: 2026-09-21 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.35822v1 Sycophantic agreement in language models refers to the tendency to overly affirm a user's stated beliefs or preferences, often at the expense of factual accuracy. Although it is widely recognized as an alignment failure, its underlying mechanisms remain poorly understood. In this work, we use causal mediation analysis to identify the mechanisms behind sycophantic agreement. We show that a stated opinion is incorporated into the residual stream of the final prompt token early, where it biases subsequent answer retrieval. A sparse set of early attention heads carries this opinion signal. Ablating these heads substantially reduces sycophancy while leaving factual accuracy largely intact. The same heads carry the opinion when it is explicitly stated, regardless of how it is phrased. When an opinion is not stated explicitly but instead conveyed through content-free pushback (e.g., ``Are you sure?"), we find a distinct set of heads that suppresses the model's original correct answer to promote a revised answer. By providing a mechanistic account of how opinions induce sycophantic agreement, this work takes a step toward developing more targeted and reliable alignment interventions. ---