# Research Papers — 2026-09 · bundle page 152 of 266

> Full-text markdown of papers #7551–7600 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/152
[← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/151.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-09/page/153.md)

### Who Audits Whom, on What Substrate, with What Evidence? An Independence-Graded Audit Protocol for Agentic AI

Authors: Mohamed Chahine Ghanem
Published: 2026-09-16
Categories: cs.AI, cs.CR, cs.ET
arXiv: https://arxiv.org/abs/2609.18272v1

Agentic AI systems plan, invoke tools and act with limited supervision; they are now both the subject of audits and, increasingly, the auditor. Independence, the foundation of assurance,is still applied to them as a binary. We argue that it must be graded along three orthogonal axes: principal independence (who controls the auditor), substrate independence (an auditor sharing the auditee's foundation-model family, toolchain or guardrails fails with it) and evidence independence (whether evidence is attestable rather than self-reported). Each axis has precedent; the contribution is to grade all three on a single audit, aggregate them by the weakest link, and apply the same rubric when the auditor is itself an agent. We give the model a formal basis by transplanting the beta-factor model of common-cause failure from reliability engineering, a seven-step protocol whose outputs a third party can verify, a structural detectability analysis of a procurement-controls agent audited at three grades, and a Monte Carlo study of the model in which a conventional internal audit of an agent-a real audit team, a second agent, provider logsp-surfaces 5.9% of the faults it could in principle see and none at all in half the fault classes. We map the triple to the EU AI Act as amended, ISO/IEC 42006, UK public-sector risk-management guidance and audit-regulator practice.

---

### BENCHCOMPASS: From Scores to Signals for Training and Harness Decisions in Payment-Domain LLMs

Authors: Sijie Dong, Wei Ren, Xuanwei Hu, Jiawei Luo, Zifan Wang, Xiaoyun Feng, Hui Cai, Lyuxin Xue, Peng Lu, Jianshe Li, Xin Zhang, Wei Wu
Published: 2026-09-16
Categories: cs.AI
arXiv: https://arxiv.org/abs/2609.18270v1

Payment operations are a critical financial infrastructure, but the value of large language models in this domain remains unclear because payment rules change quickly, evidence is fragmented, and decisions depend on transaction state, participant role, region, and payment rail. Existing benchmarks do not isolate whether failures come from missing payment-rule knowledge, poor use of supplied evidence, or brittleness under imperfect harness inputs. We introduce BENCHCOMPASS, a payment-domain benchmark whose construction pipeline builds scenario-grounded tasks from typed evidence packs, applies LLM-based quality checks, creates task-input attack variants, and reserves final item admission for domain experts. The release contains an expert-reviewed Pro benchmark covering payment knowledge, context-grounded scenario reasoning, and Attacked Open robustness, plus a lower-assurance Normal pool for inspection and future curation. Across 16 model variants, BENCHCOMPASS shows qualitatively different failure modes: missing parametric payment knowledge, incomplete reasoning over supplied rules, and failure to reject plausible but invalid workflows. The benchmark remains unsaturated: the best frontier model reaches 89.6% on Open Context-Grounded Reasoning and 81.7% under attacked inputs, while a representative 32B open-weight model reaches 69.8% and 42.6%. Benchmark data and code are available at https://github.com/ant-intl/BenchCompass.

---

### REPAIR: Resolving Long-Tail Confusion in Scientific Retrievers via Fact-Verified Iterative Refinement

Authors: Yerim Oh, Gunhee Kim
Published: 2026-09-16
Categories: cs.AI
arXiv: https://arxiv.org/abs/2609.18262v1

Precise retrieval of scientific information is fundamentally constrained by long-tailed concepts and high fact-sensitivity of scientific corpora. These challenges often limit the effectiveness of dense retrievers and hallucination-prone LLM augmentation. To address this, we present REPAIR, a self-evolving data augmentation framework for scientific dense retrievers. REPAIR iteratively synthesizes training data to address knowledge gaps by cycling through diagnosis of long-tail concepts, API-guided evidence expansion, and differentiation via hard negative mining. This process effectively grounds retrieval in factual reality to resolve fine-grained distinctions. Extensive experiments demonstrate that REPAIR significantly outperforms 19 strong baselines on nine materials science and biomedical benchmarks. Our work highlights that diagnosing and factually augmenting data to long-tail deficits is essential for robust scientific retrieval.

---

### MS-RFD: Multi-Signal Release Frame Detection in Hammer Throw from Reconstructed 3D Trajectories

Authors: Ahmed Endris Hasen, Nikolaos Passalis, Tomi Vanttinen, Jenni Raitoharju
Published: 2026-09-16
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.18260v1

Recent advances in artificial intelligence and computer vision are reshaping sports performance analysis by enabling automated detection, tracking, and performance analysis. In hammer throw, performance is strongly determined by the kinematic conditions at release, particularly release speed, release angle, and release height. However, identifying the release instant from video typically requires manual frame-by-frame inspection, which is subjective and cumbersome in real-world training scenarios. In this paper, we present a fully automatic multi-signal release frame detection (MS-RFD) method for hammer throw using reconstructed 3D hammer trajectories. The proposed method integrates four complementary kinematic signals: speed dynamics, angular velocity transition, radial distance relative to the rotation center, and post-release trajectory linearity. These signals are fused to score and verify candidate release frames. MS-RFD is evaluated through the throwing-distance estimation error obtained from the release parameters estimated at the detected frame. An ablation study analyzes the contribution of each signal and compares alternative candidate selection strategies. The results show that speed dynamics and radial expansion provide the strongest signals for release frame detection, while angular velocity and post-release linearity provide smaller refinements.

---

### M2Tok: Multi-head Multi-codebook Discrete Action Tokenization for Vision-Language-Action Models

Authors: Chunpu Xu, Zhixuan Liang, Yuhao Zhang, Chi-Min Chan, Jessie Wang, Yang Xiao, Mengkang Hu, Xiaokang Yang, Yao Mu
Published: 2026-09-16
Categories: cs.RO, cs.AI, cs.CL, cs.CV
arXiv: https://arxiv.org/abs/2609.18259v2

Recent advancements have successfully adapted autoregressive language models to process multimodal signals, such as images and actions. Since raw action signals are continuous, effective tokenization is essential to map high-dimensional inputs into compact discrete tokens for autoregressive processing. However, existing discrete action tokenizers often suffer from high reconstruction loss, failing to preserve the fine-grained dynamics required for precise control. This "discretization bottleneck" significantly limits the performance ceiling of downstream Vision-Language-Action (VLA) models. To address this, we propose ${M}^2$Tok, a Multi-head Multi-codebook Action Tokenizer designed to minimize reconstruction error and enhance policy performance. Our approach introduces two key structural innovations: (1) we decompose the latent action features into multiple heads, enabling the model to implicitly align specific heads with distinct action dimensions; (2) we assign independent codebooks to each head for quantization. By leveraging the combinatorial nature of multiple codebooks, we significantly expand the representational expressivity of the tokenizer, leading to substantially lower reconstruction loss compared to previous methods. We evaluate the ${M}^2$Tok-based VLA on the RoboTwin, Simpler-Env, and 3 zero-shot real-world tasks. Experimental results demonstrate our method not only achieves superior reconstruction fidelity but also significantly boosts the success rate of VLA models. Comprehensive ablation studies further confirm the effectiveness of the multi-head and multi-codebook mechanisms. Code is available at https://github.com/cpaaax/M2Tok.

---

### Evolving Error States: Failure-Aware Progressive Repair for Ultrasound Lesion Segmentation

Authors: Ziliang Wang, XuJiang Tang, Lu Yuting, Weixin Xu, Yongqiang Zhao, Ying Fu, Kehua Guo
Published: 2026-09-16
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.18256v1

Reliability under sparse and heterogeneous failures remains a fundamental challenge for medical image segmentation. High average accuracy can conceal a small set of structurally distinct and clinically consequential errors. Existing post-hoc correction methods alleviate this problem, but typically estimate false-positive and false-negative corrections from the same fixed prediction. This ignores the dynamic evolution of error states and limits the correction of complex cases. Inspired by iterative error feedback in structured prediction, we propose Failure-Aware Progressive Repair (FAPR). FAPR represents the current segmentation mask as a dynamic failure state and models each repair operation as a state-transition operator. Each accepted correction forms a new prediction state for subsequent error diagnosis and repair, enabling later operations to adapt to preceding changes. Conditional routing selectively activates necessary state transitions, while failure replay exposes the model to rare error states. By keeping the base segmentor frozen, FAPR preserves its established segmentation capability while improving difficult cases. Across three public ultrasound lesion segmentation benchmarks, FAPR improves mean DSC by 1.52%. On the very-hard subsets of BUSI and TN3K, the average gain reaches 13.77%.

---

### Re2A: Situated Conversational Recommendation via Rubric-based Preference Reasoning and Alignment

Authors: Dongding Lin, Jian Wang, Xiaoyan Zhao, Wenjie Li
Published: 2026-09-16
Categories: cs.AI
arXiv: https://arxiv.org/abs/2609.18249v1

Real-world recommendation scenarios are commonly grounded in shared physical environments during user-recommender interactions. This motivates situated conversational recommendation (SCR), a complex task requiring recommender assistants to jointly reason over dialogue history, co-observed scenes, and in-scene item attributes. However, current approaches struggle with this setting due to two intertwined challenges: accurately understanding situated user preferences throughout the conversation and generating responses that simultaneously satisfy user needs and grounded situations. To this end, we propose Re2A, a framework that formulates SCR as a structured reason-then-align process. We introduce rubric-based preference reasoning, which uses automated rubrics to guide the model toward producing explicit preference states. Based on these states, we propose a preference-conditioned optimization to align response generation with dual objectives: user preference satisfaction and situation consistency. Extensive experiments on two SCR datasets demonstrate that Re2A consistently outperforms state-of-the-art methods, delivering more precise, context-aware conversational recommendations. Our code is available at https://github.com/DongdingLin/Re2A.

---

### Quanta: A Self-Contained Python Library for Hybrid Retrieval over Quantised Embeddings, Lexical Indexes, and Knowledge Graphs

Authors: Ioannis E. Livieris
Published: 2026-09-16
Categories: cs.IR, cs.AI
arXiv: https://arxiv.org/abs/2609.18248v1

An advanced retrieval-augmented generation pipeline is typically assembled from three or four independently operated systems: an approximate nearest-neighbour index, a full-text search engine, a graph database, and a relational document store. Each contributes its own deployment surface, configuration model, and failure modes, and the integration logic that binds them is written anew in every project. In this work, we present \textsc{Quanta}, an open-source Python library, which unifies dense vector search over 4-bit quantised embeddings, BM25 full-text retrieval, and knowledge-graph traversal behind a single retrieval API. Quanta makes two design commitments, which distinguish it from existing hybrid retrieval stacks. First, signals are combined by \emph{weighted reciprocal rank fusion} rather than by normalising heterogeneous scores onto a shared range, which we argue is ill-posed because such normalisations are query-dependent. Second, the graph is a \emph{candidate expander and not a relevance scorer}: traversal widens the candidate pool, and the newly admitted documents are re-scored by the dense indexes under an identifier allowlist, so structural adjacency determines what is considered while content evidence determines how it ranks.

---

### Acting in Meters: Learning Metric Interactions for Precise Robotic Manipulation

Authors: Lijie Wang, Zheng Lu, Yiming Wang, Heyang Yu, Kenghou Hoi, Bowen Hu, Di Cui, Tianyu Xin, Haoran Liao, Wanqi Zhong, Xingjie Fan, Yizhao Xu et al.
Published: 2026-09-16
Categories: cs.RO, cs.LG
arXiv: https://arxiv.org/abs/2609.18243v2

Vision-Language-Action models and World-Action Models have advanced language-conditioned robotic manipulation, yet often leave metric relations among actions, objects, and scene geometry implicit. Human manipulation combines semantic understanding of task-relevant objects with spatial feedback that guides hand motion relative to objects and their surroundings. Inspired by this, we introduce a metric interaction framework that models object-level and scene-level interactions in physical Cartesian space at a shared metric scale. At the object level, Interaction-Centric Tokens (ICTs) explicitly represent end-effector pose trajectories relative to manipulated objects and are jointly denoised with actions, providing physically grounded interaction supervision. At the scene level, the Metric Action Interaction Field (MAIF) uses action and ICT queries to attend to metric scene point-cloud features and learns geometry-conditioned action corrections. Through two-stage adaptation, our framework improves diverse VLA and WAM baselines with a small number of additional parameters and training steps. Experiments demonstrate average success-rate gains of 0.80 and 3.59 percentage points on LIBERO and RoboTwin 2.0, respectively, alongside gains of 6.80 percentage points on real-world tasks and 7.45 percentage points on their out-of-distribution variants.

---

### Unified Response Geometry for Structured Pruning

Authors: Kaixiang Shu
Published: 2026-09-16
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.18239v1

Structured pruning is commonly formulated as ranking individual channels, although channel responses can be complementary or cancel through downstream mixing. Motivated by these response interactions, we formulate pruning as the selection of a subset with large joint response capacity, followed by a separate functional realization step. Our unified response geometry maps each candidate set to \(M(D,R)=D^{1/2}RD^{1/2}\) and uses its determinant together with Schur-greedy residuals to select non-redundant coordinates. The same construction yields two information-conditioned instances: an unlabeled instance based on activation covariance, and a task-conditioned instance that combines activation and gradient variance for response scale with gradient correlation for complementarity. To convert the selected subset into an executable network, we fold predictable removed responses into successor weights through ridge compensation and recalibrate batch-normalization statistics, without fine-tuning the network. On ImageNet ResNet-50, the unlabeled instance reaches \(65.4\%\) and \(53.9\%\) Top-1 accuracy at 30\% and 40\% deletion, versus \(59.8\%\) and \(43.1\%\) for strength-only selection; the task-conditioned instance reaches \(67.7\%\) and \(56.3\%\) under the same protocol. A six-family screen shows architecture-dependent behavior, with positive relative contrasts in several convolutional and expansion-layer settings and clear boundary cases in windowed attention. These results support response geometry as a conditional principle for structured pruning, with its benefit determined jointly by the observed response and the architecture in which that response is realized.

---

### F-DACE: Fuzzy Disagreement-Aware Causal Evidence Fusion for Abstention-Safe Conversational Retail Decision Support

Authors: Sourish Dey
Published: 2026-09-16
Categories: cs.LG
arXiv: https://arxiv.org/abs/2609.18238v1

Observational decision-support systems often expose one causal estimate as a recommendation even when plausible estimators disagree. The inherent engine of the proposed system is causal machine learning: a conditional-average-treatment-effect estimand identified by backdoor adjustment, estimated by an EconML DML causal forest and DoWhy linear regression, checked by two-way fixed effects, and converted into candidate levers by constrained optimisation. F-DACE is the decision layer on that engine. It represents precision, propensity overlap, placebo-refutation stability, interval overlap, and directional agreement as fuzzy memberships. Hard vetoes force abstention after estimand mismatch, failed diagnostics, informative sign conflict, or weak evidence. In 180 panel simulations spanning six identification conditions, F-DACE made a decision in 67.2% of runs and limited false recommendations to 17.2%; the corresponding rates were 33.3% for the causal forest and 35.6% for backdoor regression, matching deterministic unanimity rather than dominating it. Nearly all (30 of 31) false recommendations occurred under shared unmeasured confounding, which no fusion rule can diagnose when every component shares the omitted variable. The retail application aggregates a public Walmart panel to 6,435 store-weeks across 45 stores. F-DACE abstains for all five markdown indicators: some estimates are imprecise, one refutation fails, and MarkDown5 has a direct sign conflict. A LangGraph conversational agent exposes impact, what-if, and lever-optimization tools while a deterministic verifier preserves causal-layer status. On 24 live questions it achieved 100.0% tool-routing accuracy, 100.0% status fidelity, and 0.983 mean groundedness. On ten adversarial questions it resisted all injected instructions.

---

### Anomaly Detection in General Ledger Data: Results from a Hybrid Approach

Authors: Jan Gronewald, Alexander Michael Rombach, Sebastian Stephan, Peter Fettke
Published: 2026-09-16
Categories: cs.LG
arXiv: https://arxiv.org/abs/2609.18228v1

Journal Entry Tests (JETs) are a mandatory part of annual audits to evaluate and assess both highrisk audit areas and potential material misstatements. However, as JETs are designed to detect known patterns based on domain knowledge, the resulting lists are often very large and require substantial additional effort from the auditor. To ensure the economic efficiency of the audit, the number of false positives in JET result lists must be reduced. Especially machine learning (ML) methods represent a promising approach to improve anomaly detection in this field. In this research in progress paper, we investigate different approaches on how to combine JETs with ML-methods in a hybrid manner. We present specialized models to increase the detection performance and validity of anomaly detection results to improve audit efficiency. The experiments are based on synthetic data consisting of different normal and anomalous journal entries.

---

### WISE: A Lightweight, Weakly-Supervised Model for Onboard Fire Smoke Detection and Localization

Authors: Sha Lu, Yu Sun, Liang Zhao, Jixue Liu, Lin Liu, Jiuyong Li, A. K. Qin, Alejandro Mousist, Stefan Peters
Published: 2026-09-16
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.18227v1

Wildfire smoke detection from satellite imagery is critical for early warning and rapid response. For onboard satellite deployment, detection systems must operate under strict memory and latency constraints while providing spatially informative outputs for downstream decision-making. Existing tile-level classification methods are computationally efficient but lack spatial localization, whereas pixel-level segmentation approaches provide detailed masks yet are typically too computationally demanding for real-time onboard execution. To address this gap, we propose WISE (Weakly-supervised Inference-efficient Smoke Extraction), a deployment-oriented framework for onboard fire smoke detection and localization. WISE leverages only tile-level annotations through a teacher-student distillation strategy, where an offline teacher provides soft spatial supervision to a lightweight WISE-Student optimized for efficient onboard inference. The student jointly predicts tile-level smoke presence and smoke probability maps within a single forward pass, enabling spatially informative detection under strict computational constraints. WISE was evaluated through in-orbit execution aboard the ISS-mounted IMAGIN-e payload. Three model variants achieve average inference times of 0.10 s, 0.14 s, and 0.26 s per tile, indicating near-real-time per-tile inference within onboard resource limits. Ground-based experiments on Landsat 5 and Landsat 8 imagery further indicate effective detection and spatially informative localization. The best-performing variant achieves a mean tile-level F1 score of 0.964 and a mean pixel-level F1 score of 0.750 across 10 runs, while containing only 0.12M parameters and requiring approximately 3 GFLOPs. Together, these results indicate that WISE is a practical candidate for low-latency wildfire smoke monitoring from space under onboard resource constraints.

---

### Remembering Solomon Marcus

Authors: Florin Nichita
Published: 2026-09-16
Categories: math.HO, cs.AI
arXiv: https://arxiv.org/abs/2609.18224v1

From the manifest of Andre Breton, through the transdisciplinary understanding, we arrive at a post-modern manifest. A talk by Laura De Marco (Harvard) will provide scientific background to approach an AMS poetry. The next section will be a qualitative analysis of some new operations on the real numbers. The conclusions will be given in the last section, and an appendix will recall some previous work with some new comments.

---

### APGEM: Adaptive Policy-Guided Error Mitigation for Quantum Reinforcement Learning on a Real-World CVRP Case Study

Authors: Shabir Ahmad Sofi, Bisma Majid, Mir Mohammad Yousuf
Published: 2026-09-16
Categories: cs.LG, cs.AI, cs.ET
arXiv: https://arxiv.org/abs/2609.18219v1

Quantum Reinforcement Learning (QRL) represents policies as variational quantum circuits (VQCs), making it attractive for combinatorial optimization such as the Capacitated Vehicle Routing Problem (CVRP). On noisy intermediate-scale quantum (NISQ) hardware, however, decoherence degrades fidelity and destabilizes learning, and conventional error mitigation is applied statically without regard to the learning context. We introduce Adaptive Policy-Guided Error Mitigation (APGEM), a controller that selects among Zero-Noise Extrapolation (ZNE), Probabilistic Error Cancellation (PEC), Clifford Data Regression (CDR), and Readout Error Mitigation (REM) online, driven by a fidelity, entropy, and cost aware utility function and an epsilon-greedy rule over temporal-difference Q-scores. We evaluate on a realistic urban-logistics testbed, a Delhi-based CVRP over real landmarks with geodesic inter-node costs, exercised across five noise families and four severity levels. On this instance, the QRL agent outperforms constructive heuristics and approaches metaheuristics, while mitigation restores approximation ratios from 0.84-0.87 to 0.92-0.94 under high noise. The controller shifts from a CDR-dominated regime under short training horizons to a balanced deployment across all four techniques under longer horizons, indicating genuine regime-dependent selection. These preliminary results position adaptive, learning-aware mitigation as a practical route to noise-resilient QRL.

---

### CPR: Combining global composing, local performing and full-sequence refining in piano rendering with continuous autoregressive modelling

Authors: Chong Jing, Junan Zhang, Zhizheng Wu
Published: 2026-09-16
Categories: cs.SD, cs.AI
arXiv: https://arxiv.org/abs/2609.18216v2

Prompt-conditioned piano MIDI-to-Music rendering aims to faithfully render target notes while reproducing the timbre of a reference recording. Existing approaches primarily follow two paradigms: autoregressive (AR) modeling and flow matching (or diffusion). Discrete-codec AR models provide causal temporal modeling, but quantization can discard acoustic detail. Flow matching better preserves acoustic structure in the cost of full-sequence attention costs and worse semantic structure. Continuous autoregressive models operate directly on continuous representations. It not only combines the condition-following ability of AR models and distribution-modeling capacity of flow matching but also bypasses the quantization bottleneck with lower computational costs. Building on this principle, we present Composer--Performer--Refiner (CPR) framework. Composer autoregressively predicts continuous hidden states, Performer generates 24kHz acoustic latents through local flow matching and Refiner then upsamples the waveform to 48 kHz. We further introduce Bottlenecked Representation Alignment (BREPA) and Modality--Time RoPE (MT-RoPE) to strengthen musical semantic structure in Composer hidden states and temporal alignments across modalities. Codes are available at https://github.com/FEAfeatherTHER/CPR_official

---

### A Lightweight CNN Integrated Compact Convolutional Transformer for Multi-Scale Feature Learning and reducing computational complexity for breast cancer mammography image detection and classification

Authors: Md Taimur Ahad, Ainuddin Ahmed
Published: 2026-09-16
Categories: cs.LG, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2609.18212v1

Over the years, Convolutional Neural Networks (CNNs) have demonstrated strong capability in cancer detection and classification using medical images. However, CNN-based models often struggle to capture long-range contextual dependencies. In such scenarios, integrating Compact Convolutional Transformer (CCT) architectures after the CCT layer allows CNN-extracted features to reshape into compact patch tokens using a CCT tokenizer, followed by the addition of positional embeddings to preserve spatial structure. Using 5-fold cross-validation, the model was tested on 3 sets of breast cancer mammography. With only 250,435 parameters, the model achieved 99%-100% accuracy across 3 datasets, indicating robust generalization. Explainable AI (XAI) was integrated into the model to explain the breast cancer classification process to enhance clinical trust. The results indicate that the proposed framework is suitable for computer-aided diagnosis systems, particularly in resource-constrained clinical environments. The novelty of the proposed CNN-integrated CCT overcomes the limitation of CNN's gradient degradation in the last layers by integrating convolutional tokenization with transformer-based learning. Lighter than ViT, which is effective in capturing long-range dependencies, the model has also proven efficient in breast cancer classification by capturing long-range dependencies among breast tissue regions.

---

### Understanding Dynamic Scenes at Gigapixel Scale: Wide-Area Spatio-Temporal Perception from UAVs

Authors: Yuhang Zhu, Meiyi Zhu, Yunkai Dang, Zhangnan Li, Yuxuan Wang, Wenbin Li, Hongbing Pan
Published: 2026-09-16
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.18210v2

UAV-borne imaging has advanced from megapixel to gigapixel sensors, shifting aerial perception from recognizing individual targets to understanding entire dynamic scenes. We characterize this demand as Wide-area Spatio-temporal Scene Understanding (WSTU), which requires wide-area coverage, per-target resolution, and temporal continuity at once, a combination existing datasets lack. To fill this gap, we introduce an ultra-High-resolution (12768x9564) Airborne Remote-sensing Dataset (HARD) annotated at three levels for object detection, multi-object tracking, and scene-level visual question answering. Ultra-high-resolution imagery raises per-frame processing time to seconds. At that scale latency can no longer be ignored in evaluation. Thus, we propose a latency-aware metric for multi-object tracking called streaming-HOTA (s-HOTA). Extensive baseline experiments show how ultra-high-resolution processing reshapes each task. For detection, the end-to-end pipeline affects accuracy and speed as much as the detector itself does. For tracking, high latency charges the association axis far more unevenly than the detection axis, and association is where pipelines diverge. As a result, the pipeline that performs best offline can lose its lead under s-HOTA. For VQA, vision-language models remain weak at cross-frame identity binding and cannot transfer their single-frame gains to it. Together these findings show that the baselines we evaluate fall short of WSTU. HARD provides the data and the systematic baselines to advance it.

---

### Reinforcement Learning for Real-Time Vision-Language-Action Policies

Authors: Perry Dong, Kuo-Han Hung, Dorsa Sadigh, Chelsea Finn
Published: 2026-09-16
Categories: cs.RO, cs.LG
arXiv: https://arxiv.org/abs/2609.18207v1

Reinforcement learning fine-tuning on top of large, pretrained Vision-Language-Action (VLA) models offers promise for highly reliable robot deployment. However, because of their scale, modern VLA models suffer from high inference latency, so the observation used to select an action is often stale by execution time, creating a distribution shift that can substantially degrade reliability and performance. Prior work has explored asynchronous policy execution to reduce the effect of latency, but these methods are mostly built on imitation learning and offer no mechanism for moving beyond the training distribution toward higher reliability. We close this gap by enabling RL fine-tuning that meets the real-time control requirements of dynamic real-world manipulation. Our approach builds on EXPO-FT, a framework for sample-efficient, reliable VLA fine-tuning with reinforcement learning, and decouples slow, expressive action generation from fast, reactive action edits: a large pretrained VLA proposes action chunks using its strong behavior prior, while a lightweight edit policy performs fast, reactive decision-making by editing actions in response to changes in state, conditioned on the latest observation. We instantiate this as Real-Time EXPO-FT, an RL framework for finetuning real-time VLA policies. On the Kinetix benchmark, Real-Time EXPO-FT enables a delayed policy to achieve the best performance among delayed and non-delayed methods in 10 out of 10 environments. On four dynamic real-world tasks, robot object passing, ball balancing, table soccer kicking, and dynamic object picking, with online robot data capped at 10 minutes, Real-Time EXPO-FT improves average policy performance from 42% to 97%, all without human intervention, demonstrating rapid, sample-efficient adaptation to challenging real-world dynamics. Website: https://pd-perry.github.io/real-time-expo-ft

---

### CapMap-MS-TTA: 3rd Place Solution for the MUMU Track of the 8th LSVOS Challenge at ECCV 2026

Authors: Chengfeng Qiu, Kaifeng Wei
Published: 2026-09-16
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2609.18206v1

The MUMU track of the 8th Large-scale Video Object Segmentation (LSVOS) Challenge requires a single unified multimodal model to jointly solve image tagging (Task A), open-vocabulary object detection (Task B), and English captioning (Task C) under strict resource constraints (<=0.5B parameters and <=8 GB peak GPU memory). We present CapMap-MS-TTA, a training-free submission built on Microsoft Florence-2-base (~231M parameters), combining caption keyword mapping with multi-scale flip test-time augmentation. Task C uses the native <DETAILED_CAPTION> pathway with length/token sanitization. Task A maps the same detailed caption into the official quality/scene/event vocabularies via an expanded keyword lexicon with whole-word matching and a lightweight expand-hints stage. Task B runs Florence-2 open detection (<OD>) with multi-scale and horizontal-flip test-time augmentation (TTA), followed by label-aware non-maximum suppression (NMS). Without fine-tuning, the system improves our reproduced Florence-2 baseline from 15.16 to a best public score of 16.4815, and ranks 3rd on the final MUMU leaderboard.

---

### Beyond Accuracy: How Procedural Traces Shift the Decision Criterion of LLM Overseers

Authors: Zihan Chen, Di Zhu, Lei Zheng, Weiling Li
Published: 2026-09-16
Categories: cs.CL, cs.AI, cs.CY, cs.HC
arXiv: https://arxiv.org/abs/2609.18204v1

Organizations increasingly use oversight loops where one large language model (LLM) audits another's outputs alongside procedural traces of claimed steps. A common concern about such LLM-as-a-judge pipelines is that detailed traces make overseers gullible. Using signal detection theory, we audit five LLM overseers on 19 compliance tasks (4,551 analyzed judgments), varying only trace detail and evidence labeling. With disconfirming evidence always visible, error detection remains near ceiling. Instead, elaborate traces shift the decision criterion toward rejection, increasing false alarms in susceptible overseers. Without option labels, human-validated reason coding shows about 60% of false alarms cite an inability to tie evidence to its option. Labels eliminate this stated reason, yet residual rejection of correct work persists in those overseers and rises with trace detail. Procedural traces thus act as governance artifacts that shape oversight decisions. AI auditors should be evaluated by their decision criterion and false-alarm behavior, alongside accuracy.

---

### Behavior2Value: Benchmarking and Empowering LLMs for Consumer Value Measurement from E-commerce Behaviors

Authors: Peixuan Hou, Bin Chen, Li He, Jian Xu, Bo Zheng, Xiuli Ma, Guojie Song
Published: 2026-09-16
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2609.18203v1

Human values are deep motivational orientations that shape human behaviors. In e-commerce, they reveal the stable drivers behind users' purchase decisions. Compared with short-term interests, consumer values better explain how users evaluate products before purchase. However, consumer values are often implicit in complex and fragmented behavioral trajectories, leaving value measurement from e-commerce behaviors largely underexplored. To this end, we propose the Behavior-to-Value (B2V) task, which aims to identify consumer values from e-commerce behavioral trajectories. Centered on this task, we first construct the E-commerce Consumption Value Taxonomy (ECVT) and introduce B2V-Bench, the first B2V dataset and benchmark, based on anonymized Taobao behavioral logs. B2V-Bench consists of real-world purchase decision episodes, covering 25 types of purchase behaviors, along with corresponding consumer value orientations manifested in each episode. To improve consumer value measurement accuracy, we further present B2V-Verifier, a behavior-to-value measurement model based on Value Verification Tuning, which learns to assess whether behaviors provide sufficient evidence for each value inference. Experiments show that B2V-Verifier outperforms strong LLM baselines, improving multi-label classification by 34\%. The dataset and code will be publicly released upon acceptance.

---

### T-SANDHI: Tone Sandhi-aware Adaptive Network with Decoupled Hybrid Injection for Low-resource Taiwanese Hokkien Speech Recognition

Authors: Hung-Yang Sung, Chien-Chun Wang, Tien-Hong Lo, Yu-Sheng Tsao, Yung-Chang Hsu, Berlin Chen
Published: 2026-09-16
Categories: cs.CL
arXiv: https://arxiv.org/abs/2609.18194v1

In Taiwanese Hokkien automatic speech recognition (ASR), prior studies often treat tone sandhi as a major challenge under the assumption that models fail to process implicit phonological variations. However, our experiments on Taiwanese Hokkien reveal that speech foundation models actually handle tone sandhi variations effectively, and the real performance bottleneck stems from a localized confusion between these variations and retained citation tones. To address this, we propose T-SANDHI to explicitly decouple surface acoustics from underlying lexical intent on top of a frozen Whisper backbone. Using a lexicon-guided multi-task learning structure driven by text-derived pseudo labels, our lightweight hybrid injection module integrates independent citation and sandhi phonetic streams via dynamic gating. Extensive evaluation on the TAT-MOE corpus and two blind test sets demonstrates that this explicit disentanglement effectively resolves tonal mapping confusion, outperforming baselines with strict parameter efficiency.

---

### Transformation Laws in Neural Representations: Structure, Realisability, and Construction

Authors: Yuan Sun
Published: 2026-09-16
Categories: cs.LG, cs.NE
arXiv: https://arxiv.org/abs/2609.18190v1

How neural representations preserve the structure of input changes connects representation analysis with internal intervention. We study operable representational content through compatible actions of reference transformations on neural features. We characterise when a transformation descends through an encoder, and give a linear setting in which the defect is governed by the transformation's demand for discarded information, measured in the metric the representation induces. On a rectifier the failure to realise a transformation has two distinguishable sources --- what the source region has already made unrecoverable, and what it costs to satisfy every region the transformation visits with one operator --- and for a \textit{measured} harmonic carrier the same question has a closed answer: a linear realisation exists exactly when the retained harmonic blocks are invariant under the action. Using colour as the in-depth instance, we find that hue orbits in frozen visual features concentrate 84--88\% of their energy in the first two harmonics with rotation planes shared across shapes, that this organisation is substantially inherited from input and architecture and is reshaped by training and depth, and that the measured structure supports prediction, transport from new starting states, and composition --- with global and local realisations differing sharply in which they achieve. Guided by the measurements, we construct a compact interface whose rotation action is fixed by the structure and never fitted: it reads hue zero-shot at 3.4$^\circ$ median error on unseen shapes. Theory, structural measurement, and construction together establish transformation laws as a concrete object connecting the understanding of neural representations to their design.

---

### WFM: Wiki Foundation Model for Complex Agentic Reasoning

Authors: Junnan Dong, Linhao Luo, Senlei Zhang, Gong Chen, Taian Guo, Yifei Yu, Rong Tao, Tao Guo, Qian-Wen Zhang, Siyu An, Ruizhi Qiao, Xing Sun
Published: 2026-09-16
Categories: cs.AI
arXiv: https://arxiv.org/abs/2609.18182v1

Real-world agents fundamentally require persistent non-parametric knowledge for dynamic reasoning, i.e., long-term memory and retrieval-augmented generation. While graphs have shown reliable advantages in providing structured evidence, the sparse graph representations naturally restrict machine readability and semantic density required for complex agentic workflows. Driven by this limitation, the entire industry is witnessing a paradigm shift from traditional sparse graphs to LLM Wiki, an agent-native knowledge representation that couples dense document contexts with markdown files containing multi-layered topological linkages. However, parameterizing such rich semantics is challenging to encode dense textual contexts using traditional sparse graph embeddings. Moreover, learning LLM Wiki with existing graph encoders could overwhelm distributed system overheads that hinder deployment in large-scale commercial scenarios. To this end, we propose a novel paradigm Wiki Foundation Model, i.e., WFM, tailored for scalable, agent-native representation and retrieval. Specifically, (i) we formalize a Wiki Graph schema that seamlessly bridges fine-grained structures with dense contexts, maintaining explicit topologies alongside continuous semantics; (ii) A query-conditioned attentive aggregation is tailored for rich wiki message passing and explicit attention variance regularization; (iii) We engineer an infrastructural NCCL boundary exchange protocol that hoists static partition indices and leverages fixed-shape GPU-to-GPU collectives, bypassing CPU serialization and memory copy overheads. Extensive evaluations across five long-term agent memory and multi-hop reasoning benchmarks demonstrate the remarkable performance of WFM, while achieving a 10.5 times training acceleration on distributed clusters.

---

### MoRE: Mixture of Reused Experts

Authors: Eric S. Qiu, Utku Umur Acikalin, Justin Lovelace, Christian Belardi, Arjun B. Mulchandani, Carla P. Gomes, Kilian Q. Weinberger
Published: 2026-09-16
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2609.18176v2

Mixture-of-Experts (MoE) architectures decouple model capacity from computational cost, yet incur high memory footprints as parameters grow linearly with the number of experts. Recurrent Transformers achieve parameter efficiency by reusing layer weights, but typically lack the capacity for competitive language modeling. We propose Mixture of Reused Experts (MoRE), a hybrid that shares expert pools across groups of adjacent layers. Each layer retains its own router but selects from a larger shared pool, expanding the diversity of routing combinations without additional parameters. To enable shared experts to distinguish between layers, we introduce lightweight learnable depth embeddings that condition each layer's input before routing. Experiments across three model scales (114M-1.15B parameters) show that MoRE consistently achieves lower perplexity and stronger downstream performance than standard MoEs and state-of-the-art weight-sharing architectures at matched compute and parameter budgets, with only minimal modifications to existing MoE implementations.

---

### Beyond Direct Sensing: Harnessing Indirect Observations from Third-Party Sensors in Vehicle Tracking

Authors: Gaofeng Dong, Vamsi Eyunni, Pragya Sharma, Kang Yang, Mani Srivastava
Published: 2026-09-16
Categories: eess.SP, cs.LG
arXiv: https://arxiv.org/abs/2609.18173v1

Vehicle tracking is fundamental to applications ranging from urban mobility and public safety to security and defense. Conventional tracking relies on direct access to sensors that provide strong observations such as vehicle identity and location. In practice, however, factors such as ownership, privacy, cost, and operational constraints may limit directly accessible sensors, leaving sparse observations and long tracking gaps. Meanwhile, many additional third-party sensing assets may be present across the environment but remain inaccessible at the raw-data level, preventing their direct integration into the tracking system. In this work, we investigate whether weak, indirect observations with uncertain spatial and temporal cues can complement sparse direct sensing for vehicle tracking. Specifically, we propose GrayTrack, which fuses weak anonymous events with sparse direct observations using a road-constrained particle filter. We build a CARLA-Mininet-WiFi pipeline to evaluate the system under controlled conditions, generating direct observations from accessible cameras and indirect observations from third-party cameras. Our learning-based detector achieves an F1 score of 0.989 for anonymous vehicle passages. Further, incorporating indirect third-party observations reduces trajectory RMSE by 60.1% and catastrophic track loss from 35.8% to 0.3%. These results demonstrate that GrayTrack can effectively exploit weak indirect observations to extend tracking capabilities.

---

### Characterizing Replay Retention Under Dynamics Shift in Model-Based Reinforcement Learning

Authors: Everest Yang, Skye Thompson, George D. Konidaris
Published: 2026-09-16
Categories: cs.RO, cs.LG
arXiv: https://arxiv.org/abs/2609.18167v1

Adapting to changes in robot dynamics requires learning from new data without discarding experience that may still be useful. In continual model-based reinforcement learning (RL), replay collected before a dynamics change can slow adaptation, while removing it unnecessarily reduces available training data and can be especially costly if earlier dynamics return. We study when recent transitions are preferable to the full replay history. Two quantities characterize this trade-off: change magnitude and age-staleness area under the curve (AUC), measuring how well transition age separates stale from fresh data. Forgetting stale data helps after large permanent shifts but hurts when dynamics recur and older data becomes useful again. Choosing a replay strategy therefore depends on predicting when older data will help or hurt. We test these effects across two locomotion morphologies, two model-based RL algorithms, and Real-World RL benchmark perturbations. Because ground-truth staleness labels are unavailable on deployed robots, we evaluate whether an estimator built from interaction data can still provide the quantities needed to choose a replay strategy after permanent changes. Our results show that replay retention depends on change magnitude and on how the dynamics evolve.

---

### Energy-Regularized Imitation Learning for Force- and Work-Aware Robotic Manipulation

Authors: Toshiki Otani, Hiromu Taketsugu, Norimichi Ukita
Published: 2026-09-16
Categories: cs.RO, cs.CV
arXiv: https://arxiv.org/abs/2609.18164v1

This paper studies energy-aware manipulation as a physically grounded learning problem. We define a joint-space mechanical-work proxy from joint torque and angular displacement, and train a differentiable energy predictor that estimates this work from robot states and actions. The predictor converts a non-differentiable simulator-side physical quantity into a differentiable regularizer for fine-tuning a pretrained manipulation policy. We instantiate the framework with RVT-2 on RLBench and evaluate 12 manipulation tasks involving object contact, articulated motion, placement, pushing, and sweeping. The proposed fine-tuning reduces the average mechanical work from 208.8J to 204.4J (i.e., 2.1% reduction), while the mean task success rate also increases slightly from 86.2% to 86.9%. These results show that work-aware policy optimization can suppress physically inefficient motion without requiring an explicit differentiable dynamics model.

---

### Time-Aligned Evolving Concept Graphs for Scientific Relation Forecasting

Authors: Fred Sun, Jingze Wang, Minkun Xu, Shangqi Guo
Published: 2026-09-16
Categories: cs.AI, cs.IR
arXiv: https://arxiv.org/abs/2609.18163v1

Forecasting scientific relations can guide discovery by identifying promising connections before they emerge. Existing approaches often model concept semantics and graph structure separately or summarize semantics over coarse historical snapshots, leaving semantic representations potentially misaligned with rapidly evolving graph evidence. We propose a time-aligned evolving concept graph framework that jointly models semantic and structural evolution. Its core idea is to treat dated papers as shared update events, reconstructing semantic and structural states from the same publication history through each prediction time. Pair-level fusion combines these states to forecast first co-occurrence, relation formation, and conditional relation type. Holding architecture and training fixed, refreshing context alongside graph updates improves mean relation AUPRC by 16.6% over frozen context. On a graph built from 187,848 papers with 270,687 concepts and 7.45 million co-occurrence links, the complete framework improves mean relation AUROC from 0.9290 for the strongest evaluated baseline to 0.9722, with mean population-weighted AUPRC 0.005778.

---

### TeochewBench: A Human-Reviewed Benchmark for Teochew Hanzi Translation

Authors: Jianan Wu
Published: 2026-09-16
Categories: cs.CL
arXiv: https://arxiv.org/abs/2609.18156v1

Teochew has a substantial speaker community and exhibits distinctive lexical, syntactic, and pragmatic features, yet textual resources for evaluating large language models remain limited. We present TeochewBench, a human-reviewed benchmark comprising 300 Teochew Hanzi expressions for evaluating translation from Teochew Hanzi into Mandarin Chinese and English. The dataset covers five categories: basic vocabulary; everyday sentences; Teochew-specific expressions; tone, politeness, and context; and idiomatic, ambiguous, and culturally specific expressions. A primary Teochew-speaking reviewer examined all entries individually and revised them as needed, while two additional Teochew speakers verified selected items. Our main evaluation covers 11 official general-purpose post-trained models on the reviewed dataset in both translation directions, yielding 6,600 predictions. Two official base checkpoints provide 1,200 predictions for supplementary diagnostics, bringing the total to 13 models and 7,800 predictions. We additionally include a Hanzi-copy control, which returns the source input unchanged, to assess how shared Hanzi affect automatic scores for translation into Mandarin Chinese. Qwen3.5-27B achieved the highest overall chrF-style score among the evaluated checkpoints, at 60.63, followed by Qwen2.5-72B-Instruct at 56.61, Gemma-3-27B-IT at 56.36, and GLM-4-32B-0414 at 55.82. Across the 11 main-evaluation models, the mean chrF-style score decreased from 69.25 for low-specificity items to 27.52 for high-specificity items. High-specificity expressions received lower scores and exhibited smaller cross-model differences, suggesting that they constitute a shared low-scoring region across the model families evaluated here. The Hanzi-copy control further indicates that surface overlap in low-specificity items can substantially affect automatic scores for translation into Mandarin Chinese.

---

### PageRecall: Measuring Page Selection in Literature-Grounded Question Answering

Authors: Aaditya Chauhan
Published: 2026-09-16
Categories: cs.IR, cs.CL
arXiv: https://arxiv.org/abs/2609.18154v1

We describe our system for LitTraceQA (GroundLM @ EMNLP 2026): given a research question, retrieve the relevant papers from a pool of 27,487, cite the page and the table or figure where the answer lives, and answer in a requested format. Our main finding is that evidence grounding is limited by retrieval, not by reading. The page selector put the annotator's page, which we call the gold page, in front of the model that locates evidence only about half the time (52.6% gold-page recall), while that model, given the page, cited the right one in 45 of the 48 locators it emitted (94%). When the page was missing it rarely said so: of 45 such cases it returned nothing 14 times, a wrong page 24 times, and a correct page 7 times, so the pipeline failed quietly almost twice as often as it failed visibly. Since the failure was that the right page was never shown, the fix is to stop choosing: each retrieved paper fits in the model's context, so we show it whole. Page ranking survives only as a fallback inside papers too long to fit, which no test-split paper was, and gold-page recall reaches 100% on the papers we can parse. Separately, questions that identify their target by position rather than content, such as "the first author of the 24th reference", are served by parsing rather than retrieval: we resolve the bibliography into an addressable list, which also supplies identifiers the evidence metric scores. The final system scores 0.762 paper $F_1$, 0.441 evidence $F_1$ and 0.920 multiple-choice accuracy on the held-out test split. Because the pipeline depends on a closed model without seed control, we release a harness that verifies the paper's central claims against committed artifacts.

---

### LIGE-GR: A Smooth Leap from Ranking to Generative Recommendation in the LLM Era

Authors: Venkat Srinivas, Chenzhang He, Sam Woodmansee, Shawn Lian, Wenjie Hu, Renjie Jiang, Ziheng Huang, Xinyuan Zhang, Zhihao Zheng, Zhuoran Yu, Rui Li, Lei Yuan et al.
Published: 2026-09-16
Categories: cs.LG, cs.IR
arXiv: https://arxiv.org/abs/2609.18148v2

The remarkable success of large language models (LLMs) has provided important inspiration for the next generation of recommender systems. Structurally, recommendation and language generation share a similarity: both aim to produce an ordered sequence that optimizes the user's experience. However, how to precisely absorb the essence of the LLM paradigm into mature industrial recommender systems remains an open problem. There are two challenges. First, it is unclear how to incorporate the LLM paradigm -- sequence-level generation and optimization -- into recommendation. Second, real-world recommender systems are mature systems that have been iteratively customized for years around specific products, business constraints, serving infrastructure, and organizational ownership. Replacing such systems wholesale is often technically risky and organizationally disruptive. In this paper, we propose LIGE-GR, a listwise generation and evaluation recommendation framework that upgrades from a traditional ranking system (itemwise recommendation) toward a generative recommendation paradigm. Instead of rebuilding the entire recommendation stack from scratch, LIGE-GR generalizes the existing pointwise recommendation system into a listwise generation system. This allows mature recommender systems to benefit from listwise optimization while preserving compatibility with existing models, value functions, and serving infrastructure. We validate LIGE-GR in short-video recommendation on Instagram Reels and Facebook Video. On these recommendation surfaces, LIGE-GR improves time spent by 1.14 percent on Instagram Reels and 0.72 percent on Facebook Video, while requiring only modest additional inference resources.

---

### Reaching Every Position Without Searching: Rotating Sparse Wiring on the Hypercube as a Substitute for Attention

Authors: Yoshiaki Takashita
Published: 2026-09-16
Categories: cs.LG
arXiv: https://arxiv.org/abs/2609.18145v1

Attention pays, at every layer and for every input, the cost of searching for whom to connect. We ask how far one can get with wiring that is fixed, sparse, and simply rotated from layer to layer. Treating the $n$ positions of a sequence as the vertices of a $\log_2 n$-dimensional hypercube and connecting each position, at layer $\ell$, to its neighbour along dimension $\ell \bmod \log_2 n$, information from every position reaches every other in $\log_2 n$ layers with $2n$ links per layer instead of $n^2$. On a synthetic task that is unsolvable unless all positions are reached, this rotation matches all-to-all wiring at $1/32$ of the links, while the same sparse pattern held fixed across layers fails; what matters is that every dimension is touched, not the order. On character-level language modelling of a public corpus (the first $12$M characters of enwik8), a hybrid that keeps two attention layers among sixteen sparse ones reaches $0.06$ bits-per-character lower held-out loss than a fully attentive model of the same width at the same step budget (three seeds each, no overlap), with $1/7$ of the links, $42\%$ fewer parameters, and $2.4\times$ less wall-clock time; the purely rotated schedule is level with the hybrid. The same ordering holds on a second corpus of mixed Japanese, English and code, where the gap widens to $0.16$. The usable learning-rate window is four to eight times wider than attention's on both. We also report what did not work - learned coordinates, and a "dynamics" variant whose apparent gains turned out to be an artefact of a saturated kernel - and the measurement discipline (frozen corpus, full-coverage evaluation, seed spread as the bar for ranking) that we found necessary to say anything at all at this scale.

---

### Multi-View Mixture-of-Experts with Vision-Language Reranking for Cross-View Object Geo-Localization

Authors: Xuyu Fan, Qi Ming, Zhu Han, Liuqian Wang, Siyuan Cao, Xiaohan Zhang, Xudong Zhao, Mingjing Zhao, Yuhan Zhang
Published: 2026-09-16
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.18139v1

Cross-view object geo-localization (CVOGL) locates a target in satellite imagery using drone or street-view queries. Existing methods train separate detectors for each viewpoint, leading to parameter redundancy and impeding cross-view knowledge sharing. Moreover, top-ranked satellite candidates are often visually similar, so visual appearance and categorical labels alone are insufficient to resolve such ambiguity. To address these, we propose MVLGeo, an efficient framework designed to unify multiple viewpoints and reduce model redundancy. First, we introduce environmental contextual text from the query view as cues to distinguish visually similar candidates via Vision-Language Reranking (VL-Rerank). Second, we design a multi-view Mixture-of-Experts architecture (MV-MoE) with a shared encoder and view-specific experts to reduce redundancy and promote knowledge sharing, while cross-view contrastive learning aligns their representations for consistency. Third, we introduce an adaptive elliptical prior (ESAM-Prior) as auxiliary positional encoding for anisotropic geometric perception. Extensive experiments on the CVOGL benchmarks confirm that MVLGeo, as a unified model for multiple query viewpoints, achieves state-of-the-art performance, demonstrating robustness to input degradation and generalization across viewpoints. Code and models will be available on GitHub to facilitate future work.

---

### DualSQL: Text-to-SQL with Multi-Agent Reinforcement Learning

Authors: Shijie Chen, Yu Gan, Yeounoh Chung, Jiani Zhang, Quannan Li, Sravan Babu Bodapati, Cody J. Greer, Yu Su, Fatma Ozcan
Published: 2026-09-16
Categories: cs.CL, cs.AI, cs.DB
arXiv: https://arxiv.org/abs/2609.18135v1

State-of-the-art Text-to-SQL systems are typically multi-agent pipelines centered around two fundamental tasks: schema linking and SQL generation. However, existing work trains separate models for each task, failing to leverage the synergy between these interrelated tasks. In this work, we propose DualSQL, a new Text-to-SQL system consisting of two agents powered by a single model backbone. The agents share the same model weights and agentic scaffold, enabling joint optimization through a robust multi-agent reinforcement learning (RL) framework. We design three database access tools to facilitate effective multi-step reasoning grounded to interactions with the databases. To improve training and avoid model collapse, we introduce a set of rollout guardrail mechanisms that stabilizes multi-agent RL training, supporting DualSQL to keep improving during training. We also introduce a new SQL correctness metric, robust execution match (REX), to more accurately judge SQL correctness and assign reward signals. Being trained on only 3755 examples, DualSQL-4B achieves an impressive 68.0% execution accuracy on the BIRD development set, matching previous 7B models. DualSQL-8B further improves to 71.1%, outperforming previous state-of-the-art single-model solutions with 32B parameters. These results demonstrate the strength of joint multi-agent reinforcement learning for building high performance Text-to-SQL pipelines.

---

### Rethinking How We Evaluate Methodological Progress in Health AI

Authors: Florent Pollet, Matthew McDermott
Published: 2026-09-16
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2609.18134v2

Methodological progress in artificial intelligence (AI) for electronic health records (EHRs) depends on our ability to determine which algorithms work better, and under which conditions. However, such progress is thought to be hindered by difficulties in reproducibility and in defining clinically meaningful evaluation tasks. We empirically study these barriers by re-implementing 12 historical and recent algorithms within a shared evaluation framework and evaluating them on two clinical datasets, MIMIC-IV and NWICU. We compare two complementary task families: expert-authored clinically meaningful tasks and generated tasks defined from randomly sampled event codes and prediction horizons. We ask whether relative algorithms comparisons transfer across task families and datasets, whether residual task heterogeneity contains useful methodological structure, and what a controlled comparison reveals about progress over the last decade. We find that aggregate pairwise comparisons transfer strongly across evaluation settings, including from randomly generated tasks to clinically meaningful tasks and across datasets. At the same time, clinically meaningful tasks exhibit greater task-method interaction, providing preliminary evidence that task properties can help explain when particular modeling choices are advantageous. Finally, newer algorithms do not consistently outperform earlier approaches: gradient-boosted trees remain highly competitive when paired with a modern, wide and sparse representation of the EHR. Together, these results suggest that useful methodological knowledge may require less task engineering than commonly assumed, while highlighting the importance of understanding the structured heterogeneity that remains across tasks and methods.

---

### Stealthy in Semantics, Antagonistic in Space: Attacking Visible-Infrared Object Detectors via Object-Level Misalignment

Authors: Yueqi Zhu, Qi Ming, Guo Cheng, Yongkang Zhang, Feiran Liu, Juan Fang, Jiahuan Zhou, Jiangmeng Li, Yuhan Zhang
Published: 2026-09-16
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.18133v1

Visible-infrared object detectors are used for robust perception under challenging illumination and weather conditions. Current physical attacks apply conspicuous patches to spatially aligned target regions, which are noticeable to human observers. Meanwhile, most of these methods only perturb the appearance within the aligned region, without explicitly targeting the correspondence between modalities or the fusion process. In this paper, we propose CamoShift, an adversarial framework for visible-infrared object detection. By combining visual camouflage with object-level infrared shifting, CamoShift breaks cross-modal spatial alignment and disrupts fusion. Specifically, the Semantic Camouflage Module (SCM) generates a stealthy camouflaged patch that can be attached to the host object and maintains its effectiveness in the infrared branch through an RGB-IR adapter. The Object-level Spatial Decoupling Module (OSDM) shifts the infrared target evidence in a scale-aware manner, so as to break object-level correspondence and disrupt cross-modal fusion. Then, the Harmonic Adversarial loss (HarAdv loss) further balances attack strength and visual stealth during optimization. To the best of our knowledge, we are the first to target both visual stealthiness and attack success in visible-infrared object detection. Extensive experimental results show that CamoShift achieves a superior balance between attack effectiveness and visual stealth. Code and models will be available on GitHub.

---

### Colla-Q: Toward Collaborative Experts in MoE Quantization via Minimax Precision Balancing

Authors: Eunju Shin, Jongbin Ryu
Published: 2026-09-16
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2609.18131v1

In this paper, we present a Mixture-of-Experts (MoE) quantization method based on activation entropy. Although quantization reduces memory and computational costs, it can substantially degrade performance. In particular, performance decline is pronounced in quantized MoE models, where individual experts have a small number of parameters that are sensitive to low-bit representation. Considering that MoE operates as an ensemble model with collaborative contributions from routed experts, a significant performance decline of a particular expert due to quantization can harm model performance. Therefore, we propose Colla-Q, a bit-allocation framework to maintain balanced performance across experts through an activation-entropy-based bit-width allocation algorithm. This approach encourages each expert to operate collaboratively in the quantized model, thereby 1) improving the overall MoE performance and 2) reducing the dependence on the calibration dataset. Since uniformly adjusting each expert's performance facilitates robustness and stability of the MoE model, the proposed MoE quantization method can generalize more consistently across different calibration datasets. Our code is available at: https://github.com/mmai-laboratory/Colla_Q

---

### Benchmarking Tabular Foundation Models as Surrogates in Expensive Evolutionary Optimization

Authors: Lu Han, Jin Wang, Yuchen Li, Haoran Gu, Shulei Liu, Ziyang Shi, Wenao Lu, Handing Wang
Published: 2026-09-16
Categories: cs.NE, cs.LG
arXiv: https://arxiv.org/abs/2609.18130v1

Surrogate-assisted evolutionary algorithms (SAEAs) are effective methods for solving expensive optimization problems (EOPs), where surrogate models replace most expensive evaluations and critically influence the final optimization results. In recent years, tabular foundation models have advanced rapidly, and the Tabular Prior-data Fitted Network (TabPFN) has been adopted as a surrogate model for EOPs due to its strong predictive capability, demonstrating promising performance. Motivated by its potential as a surrogate model in SAEAs, this work conducts a comprehensive study that combines extensive experiments with in-depth theoretical analysis to investigate the effectiveness of TabPFN. Specifically, we perform experiments across both offline and online SAEA settings, covering diverse problem scenarios such as single-objective, multi-objective, constrained, combinatorial, mixed-variable, and engineering optimization problems. In addition, we further analyze the advantages and limitations of TabPFN within SAEAs and provide practical guidelines for its application in different optimization settings. Results show that the effectiveness of TabPFN is highly problem dependent, and it cannot replace conventional surrogates universally. Overall, TabPFN should be adopted selectively according to data availability, landscape complexity, search space characteristics, and its role within the algorithm. Customized model management strategies and role-specific algorithm design are necessary to fully exploit its advantages and avoid its pitfalls.

---

### MCLC-NET: Multimodal Continual Learning for Leaf Counting

Authors: Ruchi Bhatt, Pratibha Kumari, Shreya Bansal, Vedant Agnihotri, Dwarikanath Mahapatra, Mukesh Saini
Published: 2026-09-16
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2609.18129v1

Leaf counting is an important task in plant phenotyping for monitoring plant growth and estimating crop yield. Most existing methods rely on RGB images, but their performance is often affected by occlusion, lighting variations, and other real-world challenges. Additional modalities, such as depth and thermal images, can provide useful complementary information. However, multimodal leaf counting remains underexplored. Also, many existing methods assume that all training data are available simultaneously, which is impractical in real agricultural settings, where data is collected over time from multiple sources. To address these challenges, we propose MCLC-NET, a multimodal continual learning framework for leaf counting. It learns tasks sequentially using a memory-based strategy with a memory buffer to retain important samples from previous tasks. We also introduce MMLC, a real-world multimodal leaf-counting dataset designed for a domain incremental scenario (DIS) in CL. It contains RGB, depth, and thermal images collected across different crop types under varying environmental conditions, arranged in three orderings: crop-wise, time-wise, and mixed. Experimental results, averaged over three random seeds, demonstrate that MCLC-NET consistently outperforms existing methods across all three task orderings, achieving the lowest AMSE of 0.675$\pm$0.027, 0.542$\pm$0.069, and 0.745$\pm$0.057, respectively.

---

### Symbolic Temporal Supervision of LLM Agents Using Contracts

Authors: Yifeng Xiao, Pierluigi Nuzzo
Published: 2026-09-16
Categories: cs.AI, cs.LO
arXiv: https://arxiv.org/abs/2609.18128v1

Large language model (LLM) agents augmented by tools can automate complex, multi-step tasks, such as web navigation, code generation, and workflow orchestration, by acting on external systems through tool calls. However, hallucinations, distributional instability, and adversarial manipulations in LLMs, and the irreversible consequences of certain tool calls can lead to harmful outcomes. Existing safeguards either grade recorded trajectories post hoc with stochastic LLM judges or block unsafe actions one call at a time, and no single deterministic artifact supports both roles. We present ContrAgent, a contract-based framework for symbolic temporal supervision of LLM agents. ContrAgent captures an agent's behavior as a sequence of tool calls and formalizes it as a trace over a fixed set of checkable predicates. It then specifies required behaviors using assume-guarantee contracts in linear temporal logic over finite traces (LTLf). Each contract is compiled to a deterministic finite automaton (DFA) that serves two roles: gating agent actions online and evaluating recorded traces offline. A contract library, acting as a reusable knowledge base, is maintained independently of the agent's model and can be applied across different agents within the same task domain. We show the effectiveness of our approach on four benchmarks spanning both roles, where ContrAgent matches state-of-the-art LLM-judge and rule-based guardrail baselines while producing deterministic, reproducible verdicts and, in the online mode, orders-of-magnitude lower per-call latency.

---

### Learning Fractional-Order Dynamics from a Single Trajectory

Authors: Xiaole Zhang, Ziyi Zhang, Zehao Zhao, Stephen Tu, Guannan Qu, Yorie Nakahira, Paul Bogdan
Published: 2026-09-16
Categories: cs.LG, eess.SY
arXiv: https://arxiv.org/abs/2609.18127v1

Many real-world processes exhibit long-range dependence, where the current state depends on a slowly decaying trace of past states rather than on the most recent state alone. This paper studies system identification for discrete-time fractional-order linear time-invariant systems from a single observed trajectory of length $t$, a setting that captures such non-Markovian dynamics through the Grünwald--Letnikov difference operator. Unlike Markovian systems, fractional-order systems couple estimation across the entire history, making both statistical analysis and practical identification more challenging. We propose \emph{Fractional-Order Ordinary-Least-Squares Grid-Search (FO-GS)}, a simple two-stage estimator that exploits the diagonal structure of the fractional-difference operator to decouple the identification problem row-wise. Under the stability assumption, we establish high-probability, non-asymptotic error bounds for estimating both the fractional order and the system matrix in the heterogeneous setting, with both estimation errors scaling as \(\mathcal{O}(t^{-1/2})\). Through experiments, we show that \emph{FO-GS} outperforms existing baselines in recovering both the fractional order and the underlying system dynamics.

---

### Designing Agentic AI Workflow Portfolios under Imperfect Selection and Compute Cost

Authors: Mojtaba Abdolmaleki, Stefanus Jasin, Boyu Wang
Published: 2026-09-16
Categories: cs.AI
arXiv: https://arxiv.org/abs/2609.18126v1

Agentic AI systems often approach the same task through multiple workflows that differ in reasoning strategy, verification structure, and compute cost. A natural deployment policy is to use the workflow with the highest average performance, but this can be suboptimal because different workflows may succeed on different instances. We study a portfolio-and-selector paradigm in which a firm runs multiple workflow executions and selects the final answer after observing their outputs. Additional executions may uncover correct answers that the best standalone workflow misses, but they consume compute and introduce plausible distractors that complicate final selection. We formulate this as a workflow portfolio problem in which the firm jointly chooses run size and allocation across workflow types. We summarize selector quality through an odds-lift index and derive sharp bounds on the value of workflow variety. For finite workflow pools, we develop exact formulations, linear programming relaxations, randomized rounding procedures, and computable performance certificates. For large implicit workflow classes, we derive a finite-dimensional dual and an ellipsoid method using a pricing oracle to identify workflows with high weighted accuracy net of recurring compute cost. Under a weak condition, the method obtains a near-optimal solution to the relaxation with polynomially many oracle calls. We evaluate the framework on three datasets: ABCD, Schema-Guided Dialogue, and HotpotQA. Relative to the best standalone workflow, portfolio optimization improves held-out selector accuracy by 3.1, 7.5, and 0.9 percentage points, respectively. Dual-guided workflow generation adds 3.5 points on ABCD and 24.1 on HotpotQA, with no additional gain on Schema-Guided Dialogue.

---

### PRISM: Predictive Representation of Interaction Style and Motion for Social Robot Navigation

Authors: Bo-Han Chen, Hiromu Taketsugu, Norimichi Ukita
Published: 2026-09-16
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2609.18125v1

Humans often observe others before interacting and adjust their behavior accordingly. Robot navigation in crowds, however, often represents pedestrians mainly by observed geometric states, leaving individual differences in interaction tendencies implicit. We propose PRISM (Predictive Representation of Interaction Style and Motion), a framework that infers interaction traits from passive observations of human-human interactions. PRISM encodes human trajectories into a continuous ordinal latent space with a transformer encoder trained by Rank-N-Contrast loss, and pairs each inferred trait with a temporal-stability score supplied to the navigation policy. In randomized crowd simulations, PRISM reduces collision rates over the geometry-only baseline and yields small improvements in navigation-time and path-length metrics. These results suggest the utility of passive latent-trait inference for social navigation in dynamic crowds.

---

### Aligned Consensus Teaching for Label-Efficient Oriented Object Detection in Weakly-Aligned Visible-Infrared Imagery

Authors: Qi Ming, Xiaxin Yuan, Jiahuan Zhou, Jiangmeng Li, Xudong Zhao, Zhanchao Huang, Juan Fang, Shaoguang Huang, Aleksandra Pizurica
Published: 2026-09-16
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.18124v1

Visible-infrared object detection (VIOD) detects objects with oriented bounding boxes from paired visible and infrared images. Existing methods depend on costly dual-modality annotations. Semi-supervised learning can reduce this burden, but extending it from single-modal detection to VIOD is challenging. In the practical image-pair-level setting considered here, only a few pairs are labeled in both modalities, while the rest are completely unlabeled. This limited supervision creates three challenges: (i) too few labeled boxes for robust cross-modal alignment; (ii) pseudo-label errors caused by branch-wise misses accumulate during self-training; and (iii) tail-class annotations become critically scarce as the labeling budget decreases. We propose Aligned Consensus Teacher (ACT) for label-efficient VIOD in this setting. Its Cycle-Consistent Region Alignment (CRA) combines cycle consistency and sparse anchors with reliability-weighted regional matching. Cross-Modal Consensus Mean-Teacher (CMC-MT) forms consensus pseudo labels under pair-preserving views to recover branch-wise misses and supervise unlabeled pairs. Text-Guided Cross-Modal Instance Augmentation (TG-CMIA) uses a vision-language scene prior to compose tail-class instance pairs while preserving RGB--IR offsets. To the best of our knowledge, ACT is the first framework to study semi-supervised VIOD under this image-pair-level setting. Experiments on DroneVehicle and VEDAI show consistent gains across annotation ratios. With 10\% labeled pairs on DroneVehicle, ACT reaches 94.3\% of the mAP obtained by the same detector under full supervision. Code and models will be available on GitHub to facilitate future work.

---

### AutoTuneBench: Trustworthy Measurement for Agent Auto-Tuning of LLM Serving Engines

Authors: Li Chen
Published: 2026-09-16
Categories: cs.AI
arXiv: https://arxiv.org/abs/2609.18123v1

Large language model agents tune GPU kernels and serving engines through a closed loop of propose, measure, and keep, but the measurements behind this loop are not trustworthy. We characterize four failure modes from a four-day pilot corpus of 619 model calls: strawman baselines manufacture speedups, absolute times do not transfer across machines, saturated tasks nullify comparisons, and infrastructure defects impersonate science. We present AutoTuneBench, a benchmark and measurement protocol that makes trust architectural. The protocol is frozen as code with test-enforced provenance; a database-level validator rejects out-of-protocol results; anti-cheat checks run outside the agent's modification surface; comparisons follow pre-registered readouts; and measurements anchor to externally published results, grounded in paired-seed statistics with a 5\% cross-run coefficient-of-variation cap. Honest measurement rewrites the headlines: our best kernel reads 10.6x against a naive baseline but 2.03x against the honest one; one configuration delivers 1.174x on one machine and 1.0049x on another; a pre-registered on/off comparison nulls at a shared wall (2.4840 vs 2.4957\,ms); and the KernelBench Level-1 suite admits 51\% of tasks with median speedup 1.0001x over PyTorch eager. The protocol, the two-engine corpus (vLLM and SGLang), and its audit trail are released as open artifacts.

---

### PentestChain: A Cost-Aware, MCP-Orchestrated Framework for Automated Penetration Testing with Free-Tier LLMs

Authors: Rushabh Vipulkumar Patel, Dipo Dunsin, Mohammed Almaiah, Mohamed Chahine Ghanem
Published: 2026-09-16
Categories: cs.CR, cs.AI, cs.NI
arXiv: https://arxiv.org/abs/2609.18120v2

AI-driven penetration testing has been demonstrated with premium frontier models such as GPT-4, but the per-engagement token cost makes continuous, automated testing unaffordable for the smaller organisations that need it most. This paper presents PentestChain, a ten-phase automated penetration testing framework that couples a curated, deterministic exploit map with a cost-aware AI cascade-a local Ollama model (qwen2.5-7b) first, then free-tier OpenRouter and Cerebras, with a rule-based fallback that always produces output-and exposes the full pipeline through a Model Context Protocol (MCP) server with eleven tools. We make three contributions. First, we treat US-dollar cost per engagement as a measured, first-class evaluation metric and show that a 7B-parameter local model, kept off the critical path by a deterministic backbone, sustains end-to-end operation at zero measured paid-API cost. Second, we analyse the attack surface that an MCP-exposed offensive engine introduces, grounding a four-position threat model in the 2025 MCP incident record (the CVE-2025-6514 remote-code-execution flaw in mcp-remote, the postmark-mcp supply-chain backdoor, and the tool-poisoning-rug-pull-line-jumping class), and contribute four mitigations. Third, we specify a reproducible, containerised evalua-tion protocol aligned with the standardised testbeds now expected at top-tier venues-AutoPenBench, a Cybench subset, and the PentestGPT 182-sub-task benchmark-with multi-trial statistics (more than 10 trials per configuration, pass-at-k, non-parametric significance tests and effect sizes) and direct, same testbed reproduction of the PentestGPT and PentestAgent baselines rather than citation of their published numbers. On the legacy targets measured to date, the framework detected 26 services, enriched 34 CVEs, produced

---

### Preservation of Log-Concavity and Convergence of Wasserstein-Fisher-Rao Gradient Flows

Authors: Francesca Romana Crucinio, Sahani Pathiraja
Published: 2026-09-16
Categories: stat.ML, cs.LG, math.PR
arXiv: https://arxiv.org/abs/2609.18118v1

We study the convergence of Wasserstein-Fisher-Rao (WFR) gradient flows for sampling from probability distributions known up to a normalisation constant. By combining Wasserstein transport with Fisher-Rao birth-death dynamics, WFR flows balance exploration and selection. These flows have been recognised as a promising mechanism to accelerate convergence beyond Langevin dynamics. We show that for a class of strongly log-concave target distributions satisfying additional curvature conditions, WFR flows preserve strong log-concavity, in contrast to Wasserstein flows which enjoy this property only in the Gaussian setting. Exploiting this result, we derive explicit non-asymptotic convergence rates for the symmetrised Kullback-Leibler divergence, without requiring a warm-start as required in current estimates. In particular, we show that the convergence rate decomposes additively into Wasserstein and Fisher-Rao contributions, thereby confirming a recent conjecture within this setting. These results provide refined convergence guarantees and further develop the theoretical foundations of WFR gradient flows for sampling and Bayesian inference.

---

### Token Latency Fairness: Performance Isolation for Multi-Tenant LLM Serving

Authors: Dev Bali, Soujanya Ponnapalli, Yichuan Wang, Natacha Crooks, Scott Shenker, Matei Zaharia
Published: 2026-09-16
Categories: cs.DC, cs.LG
arXiv: https://arxiv.org/abs/2609.18112v1

LLM serving is typically offered as a shared, multi-tenant service, where high-demand workloads from one client can cause latency SLO violations for others. Existing solutions for performance isolation equalize client throughput in the long run, for example through queueing and batching fairness. However, these approaches do not provide latency isolation guarantees; as a result, well-behaved clients can still experience significant degradation to their token-level latencies. In this paper, we present FairInference, which provides the novel δ-token fairness guarantee: for a well-behaved client, if a token is generated in d time units in isolation, it will be generated within d + δ time units in multi-tenant execution, providing strong latency isolation guarantees for LLM serving. To achieve this, FairInference addresses a key challenge of LLM serving: bounding delays from sharing GPU resources without support for fine-grained scheduling or resource allocation. In FairInference, the scheduler enforces per-token deadlines, while bounding the delays from GPU compute sharing and accounting for the additional delays introduced by the shared KV caching in GPU memory. We show that FairInference effectively bounds token-level latency spikes for well-behaved clients and improves overall throughput compared to state-of-the-art LLM serving systems.

---