# Research Papers — 2026-09 · bundle page 90 of 266

> Full-text markdown of papers #4451–4500 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/90
[← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/89.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-09/page/91.md)

### Decoupled Learning and Selection in Slate Recommendation for Privacy and Stability Under Noisy Scores

Authors: Sam Urmian, Qinyi Liu, Mohammad Khalil
Published: 2026-09-24
Categories: cs.LG, cs.IR
arXiv: https://arxiv.org/abs/2609.29453v1

We formalize slate recommendation as a randomized score learner followed by deterministic selection. First, an appropriately scoped differential-privacy guarantee passes through selection and its audit trace by post-processing. End-to-end privacy holds only when selector inputs are public or independent, previous private outputs, or separately privacy-accounted; fixing raw state or candidate information instead yields only a conditional guarantee. Second, we derive a logged margin certificate: bounded score-induced objective movement below half the smallest greedy decision margin guarantees that the ordered slate is unchanged. Controlled fixed-margin tests show near-linear exponent scaling, with an empirical slope of $-0.220$ (95% CI $(-0.231,-0.210)$) against the independent-noise reference $-1/4$. Real-anchor experiments on OULAD, MovieLens-25M, and Amazon Musical Instruments show that greater anchor weight reduces score-noise-induced ranking churn. OULAD and EdNet certificate checks validate the implementation of the logged inequality, while closed-loop simulations show bounded target drift and setting-dependent downstream utility. The contribution is therefore a privacy-scope contract and a certifiable score-to-slate stability mechanism, not a universal utility claim.

---

### YODAS v3: Over 1 Million Hours of High-Bandwidth, Stereophonic, Multilingual Speech

Authors: William Chen, Shinnosuke Takamichi, Sayaka Shiota, Satoru Fukayama, Samuele Cornell, Shinji Watanabe
Published: 2026-09-24
Categories: cs.CL
arXiv: https://arxiv.org/abs/2609.29448v1

We present YODAS v3, a weakly-labeled speech corpus containing over 1.1 million hours of 48kHz multi-channel audio in 147 languages, released under a CC BY 3.0 license. YODAS v3 is not only the largest open speech dataset to date, but also the first truly large-scale speech corpus with high-fidelity stereo audio. We first provide the collection methodology for the corpus, where we introduce new techniques for gathering language-balanced speech data. The effectiveness of our approach is shown by the language distribution of the crawled data: 22 languages in YODAS v3 have over 10K hours and 73 languages have over 5K hours of data. We then conduct extensive analyses on the composition of the data, such as the distribution of languages, audio quality, and transcription quality. Finally, we train baseline speech recognition and neural codec models to show the effectiveness of the dataset. Download at https://huggingface.co/datasets/espnet/yodas3.

---

### Frame-to-Panorama Localization and Context-Aware Sampling for Scene-Specific Ship Detection in a Smart Marina Testbed

Authors: Ignat Romanov, Andreas Hadjipieris, Neofytos Dimitriou
Published: 2026-09-24
Categories: cs.CV, cs.AI, cs.CG
arXiv: https://arxiv.org/abs/2609.29447v1

Smart maritime infrastructures provide continuous access to heterogeneous sensing streams, enabling repeated experimentation, digital-twin development, and AI-based maritime services. However, sensing hardware alone is not sufficient for scene-specific model development: historical video streams must also be spatially indexed, contextualized, and reduced to informative subsets for annotation. This paper presents a frame-to-panorama localization and context-aware sampling pipeline for ship detection in historical PTZ maritime video lacking reliable pan, tilt, and zoom metadata. The main contribution is an end-to-end data-curation approach that recovers camera-view information from historical PTZ video and combines it with environmental context and visual diversity to construct compact, scene-specific training sets. Specifically, frames are localized on a reference panorama using SuperPoint and LightGlue, enriched with weather and solar-state metadata, and selected through diversity sampling to preserve variation across camera view and environmental conditions. A second context-aware stage targets under-represented distant-vessel cases near the horizon using tile-level visual embeddings and Gaussian Mixture Model clustering. Applied within the CMMI MDigi-I Smart Marina testbed, the proposed pipeline reduces 40,718 candidate frames to 220 images for annotation, corresponding to a 99.5% reduction. A YOLO26-m detector fine-tuned on this subset achieves a mean AP50 of 94.78% $\pm$ 0.51% and a mean AP50-95 of 75.10% $\pm$ 1.73% under sequence-grouped five-fold cross-validation. These results demonstrate that highly redundant infrastructure video streams can be transformed into compact, spatially and contextually diverse training sets for scene-specific detector adaptation while substantially reducing annotation effort.

---

### SPADE-DFL: Communication-Efficient Decentralized Federated Learning via Derivative-Free Linearized ADMM

Authors: Mengli Wei, Mengkai Zhu, Jiawen Chen, Wenwu Yu, Duxin Che
Published: 2026-09-24
Categories: cs.LG, cs.NI
arXiv: https://arxiv.org/abs/2609.29446v1

Reducing communication in derivative-free decentralized learning requires controlling the disagreement accumulated over multiple local updates. This paper develops SPADE-DFL, a primal--dual method that allows the number of local function-value updates between neighbor exchanges to grow with the computation budget while preserving the nonprivate convergence order. For smooth nonconvex objectives under uniform query-moment bounds, the prescribed nonprivate schedule achieves a time-averaged stationarity and consensus bound of $\mathcal{O}(T^{-1/3})$ using only $Θ(T^{2/3})$ communication rounds, where $T$ is the number of local updates per client. For private training, the accumulated data-dependent increment is isolated from the graph correction, allowing one protected state per client and round to generate all outgoing messages. We prove client-level differential privacy for the full interactive transcript and quantify the resulting optimization error over a finite horizon. Experiments on four classification tasks show that SPADE-DFL achieves higher mean test accuracy than existing decentralized learning methods.

---

### Two Emojis of Difference: What Multilingual Affective Generation Benchmarks Actually Measure

Authors: Fardeen Sadab, Adib Sakhawat
Published: 2026-09-24
Categories: cs.CL
arXiv: https://arxiv.org/abs/2609.29445v1

We audit a multilingual affective generation benchmark eight instruction-tuned LLMs producing emoji summaries for 17,100 Bangla, English and Hindi sentences, with 6,960 human judgements and find its headline conclusions to be artefacts of the measurement instrument rather than properties of the systems. Treating annotators as a random rather than a fixed factor, no system differs significantly from any other ($F(7,14)=0.59$, $p=0.76$), although the conventional analysis declares 19 of 28 pairwise differences significant. Annotator identity explains far more rating variance than system identity, and the winning system changes whenever any single annotator is removed. The ordering that does emerge tracks output length: mean emoji count explains 78.7\% of between-system variance, and a within-item length-matched comparison over 2,599 pairs reverses the leaderboard. We further show that cross-provider anisotropy differences vanish under mean-centring, that per-language token costs change sign with the normalising unit, and that multi-view row-wise splits inflate macro-F1 by $3.1$ points and change the top-ranked system. In place of preference scoring we propose **emoji-affect decodability**, a reference-based probe whose rankings are stable to $\pm0.003$ macro-F1 across seeds.

---

### IterSynth: Rethinking Deep Search Agents via Role-Decoupled Iterative Synthesis

Authors: Xingyu Wu, Yuchen Yan, Zhengxi Lu, Siqi Chen, Xin ZHANG, Aiting Liu, Chao Deng, Jie Liu, Jin Ma, Jian Shao, Jun Xiao, Yongliang Shen
Published: 2026-09-24
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2609.29444v1

Deep search requires LLM agents to decompose complex queries, search for evidence, and synthesize grounded answers, yet existing ReAct-style agents suffer from two limitations: role coupling, where one policy must handle planning, evidence use, and synthesis; and context accumulation, where growing search histories introduce noise and obscure useful information. To address these issues, we propose IterSynth, a role-decoupled and summary-based paradigm that alternates between a Planner for identifying information needs and a Synthesizer for integrating evidence into an evolving summary state. This design separates planning from synthesis while using the summary as the persistent state of search, reducing both capability coupling and context noise. To train IterSynth effectively, we further introduce Role-Decoupled Policy Optimization (RDPO) for reinforcement learning, which combines terminal outcome rewards with turn-level rubric evaluations and computes role-specific advantages for more precise credit assignment. Experiments on five long-horizon deep-search benchmarks such as BrowseComp and Xbench-DS show that IterSynth-8B achieves an average score of 50.7, surpassing the strongest prior $\leq$8B agent by +4.2\%. Moreover, IterSynth serves as a model-agnostic prompting paradigm, delivering substantial zero-shot gains over ReAct and similar prompting paradigms on frontier proprietary models.

---

### Pose Adaptive Dynamic FiLM Modulation for Visual Speech Recognition

Authors: Matthew Kit Khinn Teng, Haibo Zhang, Takeshi Saitoh
Published: 2026-09-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.29443v1

Head-pose variation introduces substantial appearance transformations in visual speech recognition (VSR), making pose-aware feature modulation desirable. However, performance degradation and unwanted feature interactions may result from using numerous Feature-wise Linear Modulation (FiLM) circuits with fixed modulation intensity. We propose a Pose Adaptive Dynamic FiLM framework with a Dynamic Residual FiLM (DR-FiLM) modulator that predicts input-dependent weights to adaptively control the strength of pose-conditioned modulation. Experiments on LRS2 and LRS3 demonstrate that unweighted multi-pathway modulation substantially degrades phoneme recognition, increasing PER to 20.33% and 29.42%, respectively, compared with 16.20% and 20.96% for the single ResFiLM configuration. In contrast, the proposed DR-FiLM with dynamic Deep-Res weighting reduces PER to 15.74% on LRS2 and 23.91% on LRS3, substantially mitigating the adverse effects of unweighted modulation. The analysis of the learned weights further reveals a consistent tendency to assign greater weight to the deeper FiLM pathway as head-pose variation increases. These results show that merging pose-conditioned FiLM circuits is more efficient when the modulation strength is dynamically controlled.

---

### Detecting Glaucoma Across Multi-ethnic Myopic and Non-Myopic Populations Using an Uncertainty-Aware Vision Transformer: A Multicentre Model Development and Validation Study

Authors: Raghavan Lavanya, Yangqin Feng, Ten Cheer Quek, Quan V. Hoang, Linda Yi-Chieh Poon, Jost B. Jonas, Ya Xing Wang, Vinay Nangia, Jin Wook Jeoung, Sehie Park, SoYeon Kim, Benjamin Y Xu et al.
Published: 2026-09-24
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2609.29433v2

Background: Artificial intelligence (AI)-based glaucoma detection from colour fundus photographs (CFP) offers scalable screening, but performance may decline on external datasets because of differences in ground-truth definitions, populations, and coexisting conditions such as high myopia (HM). We developed and validated a Vision Transformer-based deep learning (DL) model for glaucoma detection across multi-ethnic cohorts with and without HM. Methods: A ViT-B/16 model with predictive uncertainty estimation was developed using 56,483 CFPs (57.1% with myopia; 14.4% with HM). Glaucoma labels were standardised using clinical, imaging, and perimetry data. The model was validated on 16 independent datasets across three continents, including four datasets with explicit HM labels. Findings: Internal AUROC was 98.7% (95% CI 98.2-99.1%), with sensitivity 94.5% and specificity 97.3%. Across 16 external datasets from eight countries, AUROCs ranged from 86.4% to 99.6%. In HM eyes, internal AUROC was 97.8% (95% CI 96.1-99.2%), with sensitivity 94.8% and specificity 93.7%. External HM AUROCs were 86.5% in the Beijing Eye Study and 93.3%, 91.8%, and 85.5% in hospital-based datasets from Taiwan, Thailand, and South Korea. In an exploratory HM clinical evaluation, the model had higher CFP-only diagnostic accuracy than ophthalmologists and trained graders (92.0% vs 70.0%; p=0.008) and performed comparably to glaucoma specialists using full clinical information. Interpretation: The model showed robust glaucoma detection across myopic and non-myopic multi-ethnic populations and may support AI-assisted screening in settings with high HM prevalence.

---

### Just Ask Jev: Reinforcement Learning for Calibrated Decisions as a Zero-Shot Detector of AI Alignment Failures

Authors: Ruoqi Guo, Yi Liu, Gelei Deng, Yuekang Li, Lida Zhao, Yutao Wu, Simin Chen, Ying Zhang, Leo Yu Zhang
Published: 2026-09-24
Categories: cs.AI, cs.CL, cs.CR
arXiv: https://arxiv.org/abs/2609.29429v1

Detectors of alignment failures screen deployed language models and score alignment benchmarks. Most are generative judges that spend a decoding pass on every criterion, and classifiers that read token probabilities, such as Llama Guard, still score one fixed label per call. Jev, a model trained with reinforcement learning for calibrated decisions (RLCD), answers many typed questions about one input with calibrated probabilities in a single call. Whether it detects alignment failures has not been measured. We present RLCDAlignBench, which benchmarks Jev on ten alignment failures: sycophancy, jailbreaks, deception, prompt injection, hallucination, privacy violation, social bias, reward hacking, concealing uncertainty, and power seeking. It spans 44 benchmarks and five target models, labelled by each benchmark's scorer and, on two, by humans. Many of these failures are relational, defined against a reference, such as the user's belief or an injected instruction, that the response alone does not reveal. Our key idea is therefore to vary what Jev is asked separately from what it sees: the question's wording and answer type on one side, the fields of the input on the other. A single generic question reaches a median AUROC of 0.886 zero-shot and beats supervised baselines on most benchmarks. Question wording matters little, while context matters more, mostly through fields that encode the label. Jev matches the reference scorer's agreement with human labels, surfaces label defects in existing benchmarks, and costs 63x less than LLM-judge scorers. Code and data: https://github.com/sumleo/RLCDAlignBench.

---

### agentic-ger: terminology recovery in long-form speech using global context

Authors: Yanqiao Zhu, Wupeng Wang, Zhifu Gao, Xiangang Li, Xie Chen
Published: 2026-09-24
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2609.29428v1

Recent advances in speech language models have improved automatic speech recognition (ASR) for long-form audio. However, accurately and consistently transcribing domain-specific terminology remains challenging. Motivated by the world knowledge and contextual capability of large language models (LLMs), we propose Agentic-GER, an LLM-based agent for terminology correction in long-form speech. The agent uses global context from the full transcript to identify suspicious terms and resolve ambiguous hypotheses. It selectively re-transcribes the source speech to check candidate corrections, and uses accepted edits to guide subsequent decisions. Experiments with four LLMs and two ASR systems on GigaSpeechBench show consistent terminology improvements in both Chinese and English, with and without thinking. On Chinese speech, Agentic-GER achieves up to a 36.8% relative reduction in biased character error rate (B-CER) over the Whisper baseline.

---

### Rufus-Air: An Open LLM Post-Training Recipe

Authors: Chia-Yuan Chang, Renyuan Cheng, Rui Feng, Xiaotian Han, Yuan He, Hongye Jin, Linwei Li, Shiyang Li, Fenglin Liu, Xin Liu, Priyanka Nigam, Haoyang Wen et al.
Published: 2026-09-24
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2609.29421v2

Rufus-Air is an open and reproducible post-training recipe on GLM-4.5-Air-Base (106B-A12B), organized as a serial pipeline of eight stages: SFT, Reasoning RL, Coding RL, Instruction-Following RL, General Agent, Coding Agent, Search Agent, and RLHF. We document the data, reward design, infrastructure, stage order, and stagewise results needed to reproduce the recipe. Stages progress from basic to advanced capabilities and from hard, verifiable rewards to softer judge-based signals. Training builds on open-source components and public data, much of it used as released, without new human annotation or an in-house distillation teacher. Our main findings are that (i) diverse, high-quality SFT establishes a strong capability floor; (ii) difficulty filtering keeps RL prompts within a productive learning range; (iii) reward reliability provides a practical principle for ordering stages; and (iv) infrastructure and engineering choices are part of the recipe, not just an implementation detail. Rufus-Air improves over the official GLM-4.5-Air post-trained release and is competitive with similarly sized open models.

---

### Controlling Backchannels in Streamable Full-duplex Models

Authors: Maike Züfle, Peter Polák, Sefik Emre Eskimez, Jan Niehues, Peter Bell, Ondřej Klejch
Published: 2026-09-24
Categories: cs.CL, cs.HC
arXiv: https://arxiv.org/abs/2609.29418v1

Backchannels, brief acknowledgements like "uh-huh" produced while the other party may still be talking, are central to natural conversation, but full-duplex spoken dialogue models rarely model them explicitly. We introduce a lightweight backchannel head that predicts, from a full-duplex model's own hidden states, when a backchannel should begin. Once this probability crosses a tunable threshold, a backchannel is force-decoded. Attached to both a 7B (PersonaPlex) and a 1B (F-Actor) model, it generalizes across scale. Probing confirms the hidden states anticipate real human timing, and generation evaluation shows more frequent, better-timed backchannels. Human raters judge the resulting backchannels on par with real ones.

---

### Neural Transport Nested Sampling

Authors: David Yallup, Will Handley
Published: 2026-09-24
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2609.29413v1

Sampling from Boltzmann distributions of molecular systems is an inference problem that has seen significant recent developments fuelled by advances in neural density estimation. We develop a novel sampling algorithm, Neural Transport Nested Sampling (NTNS), which combines the classical strengths of nested sampling with modern neural flow-based methods. NTNS uses a flow matching velocity as the drift in a Metropolis--Hastings corrected Langevin kernel inside a nested sampling outer loop, requiring only evaluations of the target energy function and providing scalable estimation of the full partition function of high-dimensional particle systems. We benchmark NTNS on challenging molecular sampling benchmarks, scaling up to Lennard--Jones clusters of 55 interacting particles, where it reduces both interatomic distance and energy Wasserstein errors to reference MCMC by over an order of magnitude relative to the strongest neural baselines at lower wall-clock cost. To our knowledge, NTNS is also the first neural sampler to return a calibrated, temperature resolved partition function estimate at this scale, recovering the phase structure across temperature from a single run.

---

### Large Language Models for Programming: Actually Fixing or Reimplementing Incorrect Code?

Authors: Alexandru Stefan Stoica, Traian Rebedea, Marian Cristian Mihaescu
Published: 2026-09-24
Categories: cs.CL, cs.SE
arXiv: https://arxiv.org/abs/2609.29410v1

Recent studies have shown that Large Language Models can effectively solve problems and fix bugs in diverse programming environments, including competitive programming. Existing approaches primarily evaluate LLM performance in problem solving or bug fixing independently, but do not explore the relationship between these two capabilities. This work focuses on determining how much the LLM deviates from a buggy solution to fix the bug compared to a human-written patch, and if there is a bias towards generating entirely new solutions. We construct a dataset with all the submissions ($\sim$ 3000) from a couple of users from Codeforces, and we match each buggy submission with its corresponding human fix. By using the similarity between the buggy solution and the human fix as a baseline, we evaluate the quality of LLM-generated bug fixes on 3 OpenAI GPT models (gpt-5-nano, gpt-5-mini, gpt-5.1). We check if the generated solutions solve the problem by using the Codeforces-R1 dataset, an openly available dataset that has tests generated with the DeepSeek-R1 model. Our findings suggest that LLMs tend to modify more lines than necessary compared to human fixes and, in some cases, generate entirely new solutions. We also observe that LLMs solve more problems correctly when allowed to generate solutions from scratch rather than patch buggy submissions, even when those submissions are close to the human patch. This has important implications for the design of AI-assisted programming tools, particularly in supporting user debugging processes and promoting incremental problem-solving strategies rather than solution replacement.

---

### Machine Unlearning for Gibbs Supervised Learning Algorithms

Authors: Yaiza Bermudez, Samir M. Perlaza, Iñaki Esnaola
Published: 2026-09-24
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2609.29409v1

In this paper, a method for achieving exact unlearning for Gibbs supervised learning algorithms is proposed using a variational formulation inspired by empirical risk minimization subject to relative entropy regularization (ERM-RER). Such a method consists of maximizing the expected empirical risk over the dataset to be unlearned subject to a regularization by relative entropy with respect to the original algorithm. The optimization variable is a probability measure on the models; and the solution is another Gibbs probability measure that represents a new Gibbs supervised learning algorithm. The method guarantees exact unlearning in the sense that the new Gibbs algorithm coincides in distribution with the algorithm that would have been obtained by retraining from scratch on the dataset to be retained. As a byproduct, a framework for reweighting data points in ERM-RER by strategically choosing both the reference measure and the regularization factor is obtained. In this framework, exact unlearning is the special case in which zero-weight is assigned to the contribution of the data points to be unlearned. More generally, depending on the choice of certain parameters, data points can be up-weighted or down-weighted in ERM-RER problems for particular purposes, e.g., controlling the generalization error of Gibbs algorithms. This paves the way for new constructive or adversarial views on classical reweighting data points in ERM-RER.

---

### Transcript-Supervised Post-Training of Generative Speech Enhancement on Real Recordings via Reinforce Adjoint Matching

Authors: Julius Richter, Christoph Boeddeker, Yoshiki Masuyama, Kohei Saijo, Dominik Klement, Gordon Wichern, Jonathan Le Roux
Published: 2026-09-24
Categories: eess.AS, cs.LG
arXiv: https://arxiv.org/abs/2609.29405v1

We adapt Reinforce Adjoint Matching (RAM), a reward-based post-training method, to generative speech enhancement (SE). Starting from a pretrained SE model, RAM tilts the model's conditional distribution toward outputs with higher reward. During training, the current model generates enhanced speech on-policy, evaluates each generated endpoint with a potentially non-differentiable reward, and analytically re-noises the endpoint to construct inputs for a reward-guided regression objective. This enables post-training directly on real recordings using weak supervision, such as text transcripts, without requiring paired clean speech targets or reward gradients. We investigate word error rate (WER)-based post-training and whether recognition performance can be improved without compromising perceptual speech quality. Experiments on real CHiME-4 recordings reduce WER by 5.08 percentage points relative to pretrained FlowSE without reducing any of the reported non-intrusive speech quality metrics. A subjective listening test at the default reward scale finds no statistically significant preference between the post-trained and pretrained models.

---

### RD-JEPA: Predictive latent pretraining for few-trajectory transfer across reaction--diffusion equations

Authors: Chenhao Si, Ming Yan
Published: 2026-09-24
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2609.29403v1

Learning surrogates for time-dependent partial differential equations often requires a new simulation corpus when the governing operator changes. We introduce RD-JEPA, a joint-embedding predictive architecture for self-supervised pretraining on reaction-diffusion trajectories. A single model is pretrained on five parameterized systems and then adapted to three held-out systems whose reaction operators and trajectories are excluded from pretraining. Using one, five, or ten complete trajectories from a held-out system, RD-JEPA achieves lower mean relative discrete $\ell^2$ field error and mean absolute spatial first-difference error than five supervised surrogate baselines, an independently trained control that removes the trajectory-dependent predictive latent pathway, and an architecture-matched model trained from scratch. Within the evaluated equations, output resolution, forecast horizons, and choices of adaptation trajectories, the results indicate that prediction of future-state representations can support data-efficient adaptation across related reaction-diffusion systems.

---

### ICE: Task-Aligned Clifford Latent Fields for Multimodal Graph Foundation Models

Authors: Xunkai Li, Xu Wang, Yinlin Zhu, Xiong Yongfu, Yi Liu, Rong-Hua Li, Guoren Wang
Published: 2026-09-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2609.29398v1

Multimodal attributed graphs connect entities, visual content, language, and observed relations. Learning one foundation across such graphs requires more than compressing each node into a fused Euclidean vector. The representation must preserve entity semantics, construct interaction state from graph neighborhoods, and expose that state to prediction units with different geometry. Our empirical study shows why these requirements are inseparable. Higher-grade channels recover pair relations across the foundation graphs, specialized queries reveal information hidden by a generic readout, and rigid blade isolation removes cross-grade capacity. We therefore introduce ICE (Interaction-aware Clifford Encoder), a multimodal graph foundation model built on a node-indexed Clifford latent field. Topology, text, and images enter explicit Cl(3) addresses. Edge-aware geometric products transform these directions into scalar, bivector, and trivector relations over observed neighborhoods. A protected Grade-1 route preserves entity semantics, while the full grade and depth bank remains available to fresh node and link heads. We establish exact cross-grade reachability, node-permutation equivariance, and a bound on the task residual around the semantic score. Experiments span one shared foundation over eleven graphs, six node-classification datasets, three link-prediction datasets, and matched few-shot tasks. ICE ranks first in all 30 reported supervised and few-shot comparisons. Core removals reduce every task summary, and mechanism controls connect the gains to higher-order transport, retained multidepth structure, semantic protection, and direct field access.

---

### Baseline Shape Decides the Verdict: A Controlled Re-Examination of Ternary Language Models at 60K Parameters

Authors: Gautam Veldanda
Published: 2026-09-24
Categories: cs.CL
arXiv: https://arxiv.org/abs/2609.29397v1

Ternary (1.58-bit) weights are attractive for microcontroller-class language models, but the sub-1M-parameter regime rests mainly on isolated, single-seed comparisons. One prominent example reports that a routed ternary block (convolution, diagonal SSM and sparse attention mixed by a per-token router) beats a parameter-matched full-precision transformer by 22% at 60K parameters, attributing this to inductive bias. We re-run it under one fixed recipe, three seeds per cell, 98 byte-level runs on one laptop. (i) Baseline shape dominates: at a 16M-byte budget, param-matched transformers span 22.6% in validation loss purely by depth/width choice - far more than any architecture effect we measure there - and the best-shaped transformer ties the routed model, so the published margin is at least partly a baseline-shape effect; the ordering of shapes reverses with budget, so no single fixed shape can be trusted. (ii) At 130M bytes the routed model does win, by 22.2-24.0% over the three transformer shapes we evaluate there - but a plain gated diagonal-SSM block beats it by a further 9.1%, and the routed model's own router puts most of its weight on its recurrent pathway, so the gain does not require routing. (iii) The ternary penalty differs by architecture at the larger budget (+5.3% best transformer vs. +19.5% routed, +28.1% gated SSM), but we cannot attribute that to architecture alone: our transformers keep learned positional embeddings in full precision, 11-22% of their parameters, so they are less quantized than the models they are compared with. (iv) A 90/10 full-precision-then-ternary schedule beats all-ternary training, but only at a stage-2 learning rate about 10x the pretraining peak; at a conventional fine-tuning rate it looks 15.3% worse, reversing the conclusion. The from-scratch baseline was not itself learning-rate tuned, which bounds (iii) and (iv). Code and run logs released.

---

### Wearable ECG Quality Assessment: A Deep Learning and Ambulatory Context-Awareness Approach

Authors: Xiaopeng Mao, Marike Weisbjerg, Sadasivan Puthusserypady
Published: 2026-09-24
Categories: cs.AI, eess.SP
arXiv: https://arxiv.org/abs/2609.29396v1

This paper presents and evaluates a Deep Learning-based (DL-based) Signal Quality Assessment (SQA) model to distinguish between clean and noisy ambulatory Electrocardiograms (ECG). The model is trained on Copenhagen Center for Health Technology-Contextualized Arrhythmia Database (CACHET-CADB), which, to the best of our knowledge, is the first ambulatory ECG database with both physical and patient-reported contextual data. The model shows stable performance on different databases such as MIT-databases and the latest PyhsioNet/Cinc Challenge 2021 databases. Subsequently, the paper demonstrates how complicated ECG noise can be investigated by the SQA model and the physical contextual data.

---

### Concurrent Split Learning Through Stable Client Clustering

Authors: Mohammad Kohankhaki, Valentin Rentschler, Anke Schmeink
Published: 2026-09-24
Categories: cs.DC, cs.LG
arXiv: https://arxiv.org/abs/2609.29395v1

Training with a fixed global batch limits how many distributed clients can provide examples in any one step. We examine a way to use additional server workers without increasing the batch processed by an individual workload. Global Clustered Parallel Split Learning (GCPSL) assigns clients to fixed clusters, executes a Parallel Split Learning with Global Sampling (GPSL) workload for each cluster concurrently, and periodically fuses the client and server model segments. In simulations with 256 logical clients, dividing the population across more workloads improves direct data participation, while smaller clusters can incur an accuracy cost. A four-H100 implementation of label-aware GCPSL reaches 85% CIFAR-10 validation accuracy in $6.13 \pm 0.15$ minutes over three matched runs, versus $19.09 \pm 0.45$ minutes when the same workloads are serialized. Within the four-GPU allocation, size-balanced and random fixed affiliations reach the target in similar mean times (5.70 and 5.66 minutes); size balancing increases direct participation by 3.25 percentage points. These measurements characterize a trade-off among execution concurrency, assignment information, participation, and accuracy for stable-client split learning.

---

### Likelihood Ranking doesn't Scale Like Prompting in LLMs

Authors: Alessandro Bondielli, Lucia Passaro, Davide Bacciu, Alessandro Lenci
Published: 2026-09-24
Categories: cs.CL
arXiv: https://arxiv.org/abs/2609.29390v2

LLM evaluation is commonly performed either by prompting models to produce answers or by scoring candidate outputs with likelihood-based metrics. In multiple-choice QA, however, standard likelihood-based scoring is still conditioned on the question and answer set, and can therefore leverage the same task-conditioned answer-selection interface used in prompting. We study a complementary protocol based on likelihood ranking of declarative statements constructed from the same question--answer pairs. Across 95 decoder-only models, ranging from 0.1B to 104B parameters, and 10 MCQA datasets, we find a systematic divergence between declarative-statement likelihood ranking and prompted answering. Statement-likelihood accuracy remains comparatively stable across scale, whereas prompted answering improves sharply with scale and instruction-tuning. These results suggest that likelihood preferences over controlled declarative alternatives and task-conditioned answer selection probe distinct aspects of model behavior, and should not be treated as interchangeable.

---

### When Misalignment Becomes Supervision: Structured Label Noise in Supervised Synthetic CT Generation

Authors: Valentin Boussot, Cedric Hemon, Caroline Lafond, Jean-Claude Nunes, Jean-Louis Dillenseger
Published: 2026-09-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.29387v1

Supervised synthetic CT (sCT) generation is commonly trained and evaluated as voxel-wise regression against registered reference CT images. In practice, MRI-CT and CBCT-CT pairs are aligned through registration procedures that leave residual misalignments. These residuals are not independent intensity noise but spatially coherent geometric discrepancies that act as structured label noise. We investigate how this registration-induced bias affects supervised MRI-to-CT and CBCT-to-CT synthesis on 1,784 paired patients covering five anatomical regions. Voxel-wise scores strongly depend on the consistency between the registration used to build the training targets and the one used for evaluation: models score best when both conventions match, showing that networks partly learn the geometric convention of the registration pipeline and that standard metrics reward it. Training on more anatomically consistent registrations reduces prediction variability and improves out-of-distribution robustness, and CT-only controls show that registration alone produces metric errors in the range of top challenge submissions. To mitigate the limits of voxel-wise supervision, we introduce a perceptual loss computed in the feature space of a pretrained Segment Anything encoder. Compared with MAE-only and VGG-based objectives, it improves downstream segmentation and yields sharper, more structurally coherent sCT. Perceptual and voxel-wise metrics disagree under imperfect alignment and agree when the evaluation geometry is reliable. These results identify registration-induced bias as a central confounder in supervised sCT generation and argue for complementing voxel-wise agreement with anatomy-oriented evaluation criteria.

---

### MORE-PLR: multi-output regression employed for partial label ranking

Authors: Santo M. A. R. Thies, Juan C. Alfaro, Viktor Bengs
Published: 2026-09-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2609.29386v1

The partial label ranking problem is a supervised learning scenario that aims to fit a preference model that predicts a bucket order defined over a set of labels for a given input instance. This problem generalizes the well-known label ranking problem, which, in practice, is limited to outputting total orders of labels. Existing partial label ranking methods have primarily extended label ranking approaches to handle ties in predictions. This paper proposes using multi-output regression to address the partial label ranking problem, introducing an encoder that, during the learning phase, transforms the (possibly incomplete) rankings with ties of labels to multivariate regression targets, an underexplored perspective in both label ranking and partial label ranking. Moreover, during the inference phase, we introduce several post-hoc layers that convert the multi-output regression results into the output bucket order to effectively implement this approach. This framework provides learning strategies that are competitive with the current state-of-the-art partial label ranking methods, as demonstrated through experimental evaluations.

---

### Segment-Level Risk Discovery in Online Handwriting for Alzheimer's Disease Detection

Authors: Changqing Gong, Huafeng Qin, Mounîm A. El-Yacoubi
Published: 2026-09-24
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2609.29384v1

Online handwriting provides a non-invasive and low-cost behavioral biomarker for Alzheimer's disease (AD) detection, as it reflects both cognitive planning and fine motor control. Existing handwriting-based AD detection methods usually rely on global trajectory features or whole-sample representations, which can be strongly affected by individual writing style, task-specific variation, and acquisition noise. In this paper, we propose NormPaST-Risk, a healthy-normative Paper-Air selective trajectory state-space risk network for interpretable AD detection from online handwriting. Instead of treating the entire trajectory as a single holistic representation, our method reformulates AD handwriting detection as local disease-relevant segment discovery. Specifically, a multi-scale temporal encoder captures stroke dynamics at different temporal resolutions, while a selective Paper-Air state-space encoder models long-range handwriting progression and distinguishes on-paper motor execution from in-air planning and transition behaviors. To explicitly characterize abnormal deviations, a healthy normative branch learns normal handwriting dynamics from healthy controls, and a task-aware multi-expert segment-risk module estimates segment-level AD risk calibrated by hidden-state changes and normative deviations. A weakly supervised segment-level objective further enables high-risk segment discovery without manual segment annotations. Experiments on the DARWIN benchmark demonstrate that the proposed framework achieves superior AD/HC classification performance compared with existing methods. Moreover, the discovered high-risk segments can be projected back to the original handwriting trajectory, providing interpretable evidence associated with AD-related handwriting variations.

---

### Lightweight Probabilistic Downscaling from a Deterministic Base Model

Authors: Joseph McLean, Tiffany Vlaar, Sigrid Passano Hellan, Linus Ericsson
Published: 2026-09-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2609.29383v1

Climate data downscaling is the task of increasing the spatial resolution of climate data, typically by generating fine-resolution regional climate data from coarse global model output. Recent machine learning (ML) work in the related task of weather forecasting has seen significant improvements due to newly devised training methods and architectural components, but these have not yet benefited downscaling. We adapt two of these methods to create a family of lightweight probabilistic ML downscaling models built on a modified U-Net backbone and evaluate them on the CORDEX-ML-Bench suite for daily maximum temperature and precipitation across three geographic regions: the Alps, New Zealand and South Africa. We find that a two-stage training curriculum, combining deterministic pretraining with probabilistic tuning, transfers well to downscaling, beating the state-of-the-art for RMSE. Our work provides an advancement towards lightweight, probabilistic downscaling models, reducing the current trade-off between computational intensity and distributional fit.

---

### Decoupled Early Exits for Task-Dependent Compute Allocation in Flow-Matching VLAs

Authors: Riccardo Andrea Izzo, Rimvydas Rubavicius, Gianluca Bardaro, Subramanian Ramamoorthy, Matteo Matteucci, Alessandro Suglia
Published: 2026-09-24
Categories: cs.RO, cs.LG
arXiv: https://arxiv.org/abs/2609.29382v2

Flow-matching Vision-Language-Action (VLA) models have emerged as a potential solution for generalist robot control, designed by combining a pretrained Vision-Language Model (VLM) backbone with an action expert that generates continuous robot actions. While these models exhibit impressive capabilities, due to their very high number of parameters, their computational requirements are often prohibitive for robotics control. To mitigate these inefficiencies, existing methods predominantly skip VLM backbone layers with early exits or reduce denoising steps, while leaving action expert depth untouched. We propose a framework that exposes backbone depth $V$, action expert depth $A$, and denoising steps $D$ as three jointly configurable compute axes in a VLA. Starting from a pretrained VLA, we attach lightweight Exit Transformers (ET) at intermediate depths in both the backbone and the action expert, trained to distil the last layer of the policy into each exit. Furthermore, we introduce a KV Cache synthesis mechanism that manages the missing keys and values of the skipped backbone layers, allowing the action expert to exit deeper than the backbone. Finally, we show that the optimal compute budget is task-dependent, with different tasks benefiting from different axes and depths. Notably, our method does not require training the original policy from scratch, and for each exit, it increases the number of parameters by only $2.1\%$ for SmolVLA and $4.1\%$ for $π_{0.5}$. We validate our approach across two flow-matching VLAs (SmolVLA, $π_{0.5}$) and two benchmarks (LIBERO, Meta-World), revealing complementary effects: $V$ and $A$ respectively reduce FLOPs and latency, while $D$ improves both. Our joint configurations $(V,A,D)$ reduce latency by $79.2\%$ and computation (FLOPs) by $31.8\%$, while improving mean success rate by $5.6\%$.

---

### An auditable conditional-strategy framework for open-ended decision-making in complex lung cancer

Authors: Daoyun Wang, Zhicheng Huang, Huaiyuan Sun, Jiaqi Xu, Xiaowei Xu, Zhibo Zheng, Zhongxing Bing, Yuxiao Lin, Yicheng Liang, Chao Gao, Bowen Xue, Kai Zhang et al.
Published: 2026-09-24
Categories: cs.AI
arXiv: https://arxiv.org/abs/2609.29381v1

Complex lung cancer decisions can involve several defensible pathways whose eligibility, sequencing and safety depend on unresolved information. Effective support must make explicit how patient conditions govern pathway eligibility, deferral and redirection. MedGPT Clinical Explorer (MCE) organizes alternatives, decision-changing unknowns, safety constraints and fallback into a conditional strategy for clinician review. To evaluate this representation in physician-authored strategies, multidisciplinary experts established case-specific references for 40 cases within a purposive 100-case corpus, and 250 physicians from 98 institutions produced 2,250 strategies under unaided, retrieval-reference and MCE-assisted conditions. MCE-assisted strategies expressed more applicable clinical requirements, measured by the Admissible Pathway Attainment Score (APAS; 0-100), than unaided strategies (adjusted difference, 12.87; 95% CI, 11.18-14.55) and retrieval-reference strategies (5.22; 3.52-6.93). With the same knowledge base available in the retrieval-reference and MCE-assisted conditions, the additional content centered on candidate pathways, decision-critical information and safety constraints. Physicians' whole-strategy acceptability judgments correlated with APAS (Spearman's rho = 0.671), while a complementary relationship audit assessed whether candidates, conditions and subsequent actions were coherently connected. Together, these findings identify two complementary dimensions of open-ended decision support: coverage of clinically relevant content and coherent links among pathways, conditions and subsequent actions. MCE provides a shared decision object that makes consequential omissions and pathway contingencies visible before action; prospective studies should evaluate its effects on clinical workflow and patient outcomes.

---

### On the second-order optimization for spiking neural networks

Authors: Ngoc Phu Doan, Ihsen Alouani
Published: 2026-09-24
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2609.29379v1

Spiking Neural Networks (SNNs) offer an energy-efficient alternative to conventional neural networks by exploiting sparse, binary spikes, and event-driven computation. However, the training of SNNs remains challenging, as spiking activations create a sharp loss landscape that hinders training, and diagonal-curvature optimizers such as the Adam family may fail to capture this geometry. The extension of curvature-based optimization methods to SNNs is further complicated by the sparse, discrete, and temporally recurrent nature of their underlying dynamics. To address these limitations, we propose SpiKFAX, a second-order optimization method that formulates a computationally tractable, Kronecker-factored approximation of the Fisher information matrix specifically adapted to the structure of SNNs. Empirical evaluation across five architectures and seven datasets demonstrates that SpiKFAX consistently yields improvements in test accuracy and training stability relative to other popular optimizers.

---

### A Hybrid CNN--State-Space--Attention Backbone with Joint-Embedding Predictive Pretraining for 12-Lead ECG Classification

Authors: Yakoub Bazi, Sarah Aljuhani, Mohamad M. Al Rahhal, Mansour Zuair, Naif Alajlan
Published: 2026-09-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.29376v1

Automatic 12-lead electrocardiogram (ECG) classification requires representations that jointly capture local waveform morphology, long-range temporal dynamics, and cross-lead dependencies, yet integrating these properties within a single efficient architecture remains challenging. This paper introduces a hybrid CNN-SSM-Attention backbone for 12-lead ECG classification. A convolutional stem performs early waveform tokenization and temporal reduction, mixed state-space and depthwise-convolutional blocks model temporal dynamics and local morphology, and a late self-attention stage enables global token interaction at reduced resolution. To improve transfer from unlabeled data, we further develop an ECG-oriented Joint-Embedding Predictive Pretraining (JEPA) framework. Unlike ViT-based JEPA methods that mask patch tokens before the encoder, the proposed method samples span masks at the latent temporal resolution and projects them back to the waveform domain, then predicts clean latent targets from a momentum encoder without waveform reconstruction. Experiments on CPSC2018, Chapman-Shaoxing, and PTB-XL, with pretraining on approximately 350K unlabeled CODE-15 recordings, show that the proposed backbone provides strong supervised baselines under a compact parameter budget. JEPA pretraining further improves transfer, particularly in reduced-label settings and under both full fine-tuning and LoRA-based adaptation. Code: https://github.com/yakoubbazi/Hybrid_ECG_Jepa

---

### Free-Init: Scan-Free, Motion-Free, and Correspondence-Free Initialization for Doppler LiDAR-Inertial Systems

Authors: Mingle Zhao, Jiahao Wang, Tianxiao Gao, Chengzhong Xu, Hui Kong
Published: 2026-09-24
Categories: cs.RO, cs.CV, eess.SY
arXiv: https://arxiv.org/abs/2609.29375v2

Robust initialization is crucial for online systems. In the letter, a high-frequency and resilient initialization framework is designed for LiDAR-inertial systems, leveraging both inertial sensors and Doppler LiDAR. The innovative FMCW Doppler LiDAR opens up a novel avenue for robotic sensing by capturing not only point range but also Doppler velocity via the intrinsic Doppler effect. By fusing point-wise Doppler velocity with inertial measurements under non-inertial kinematics, the proposed framework, Free-Init, eliminates reliance on motion undistortion of LiDAR scans, excitation motions, and map correspondences during the initialization phase. Free-Init is also plug-and-play compatible with typical LiDAR-inertial systems and is versatile to handle a wide range of initial motions when the system starts, including stationary, dynamic, and even violent motions. The embedded Doppler-inertial velocimeter ensures fast convergence and high-frequency performance, delivering outputs exceeding 10 kHz. Comprehensive experiments on diverse platforms and across myriad motion scenes validate the framework's effectiveness. The results demonstrate the superior performance of Free-Init, highlighting the necessity of fast, resilient, and dynamic initialization for online systems.

---

### FMCW-LIO: A Doppler LiDAR-Inertial Odometry

Authors: Mingle Zhao, Jiahao Wang, Tianxiao Gao, Chengzhong Xu, Hui Kong
Published: 2026-09-24
Categories: cs.RO, cs.CV, eess.SY
arXiv: https://arxiv.org/abs/2609.29374v2

Conventional LiDAR-inertial odometry (LIO) or simultaneous localization and mapping (SLAM) methods heavily rely on geometric features of environments, as LiDARs primarily provide range measurements instead of motion measurements. From now on, however, the situation changes thanks to the novel Frequency Modulated Continuous Wave (FMCW) Doppler LiDARs. FMCW Doppler LiDARs not only offer the point range with high resolution but also capture the instant point Doppler velocity through the Doppler effect. In the letter, we propose FMCW-LIO, a novel and robust LIO, leveraging intrinsic Doppler measurements from FMCW Doppler LiDARs. To correctly exploit Doppler velocities, a motion compensation method is designed, and a Doppler-aided observation model is applied for on-manifold state estimation. Then, dynamic points can be effectively removed by the Doppler criteria, deriving more consistent geometric observations. FMCW-LIO eventually achieves accurate state estimation and static mapping, even in structure-degenerated environments. Extensive experiments in diverse scenes are performed and FMCW-LIO outperforms other algorithms on both accuracy and robustness.

---

### Shadow Reduction in Ultrasound Imaging Using Differentiable Simulation and Radiance Field Decomposition

Authors: Valentin Bacher, Pak Hei Yeung, Bernhard Kainz, Madeleine K. Wyburd, Nicola K. Dinsdale, Michael Gray, Ana I. L. Namburete
Published: 2026-09-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.29373v1

Acoustic shadows from bone and other highly attenuating tissues obscure clinically important structures in ultrasound. In fetal brain imaging, skull-induced artefacts disproportionately degrade the hemisphere closer to the transducer (proximal), limiting symmetric assessment of the two hemispheres. Existing correction methods require raw scanner data, impose restrictive assumptions on tissue properties, or rely on generative models that may hallucinate anatomy. We present RFlash, a physics-informed post-processing method that decomposes beamformed ultrasound images into explicit attenuation and scatter-intensity maps using a differentiable radiance-field formulation of image formation. Attenuation-adaptive re-rendering then removes the dependence of the signal at each depth on the intervening tissue, equivalent to virtually advancing the transducer into the tissue. Across 1,261 3D fetal brain volumes, 143 real 2D curvilinear abdominal scans, and 1,200 simulated 2D linear-probe liver scans, RFlash reduces shadow-related intensity differences more effectively than classical Hughes-Duck attenuation correction. For a gestational-age model trained on the distal hemisphere (further from the transducer) and applied to the proximal hemisphere, prediction error decreases by 5.1 days (40%) relative to the original images. The estimated attenuation maps also yield shadow-confidence maps that improve random-forest bone-shadow segmentation over the image alone and receive greater SHAP importance than an existing neural confidence-map baseline, suggesting greater physical consistency. RFlash requires neither hardware modification nor access to raw scanner data and supports 2D and 3D acquisitions with linear and curvilinear probes, making it widely applicable allowing clinicians to use our method on their already acquired scanners and images.

---

### WST-Graph: Topology-Preserving Wavelet Scattering Front-End for Speech Deepfake Detection

Authors: Kwok-Ho Ng, Tingting Song, Bingwen Feng, Zhihua Xia
Published: 2026-09-24
Categories: eess.AS, cs.AI, cs.SD
arXiv: https://arxiv.org/abs/2609.29372v1

The acoustic front-end determines which forensic cues a speech deepfake detector can exploit. The wavelet scattering transform (WST) provides stable multiscale coefficients with explicit coordinates, yet direct flattening obscures the parent relation between paths. We introduce WST-Graph, reconstructing these paths as a sparse modulation-carrier grid for an AASIST graph backend. Modulation-level normalization and length-aware adaptive local attention pooling produce fixed relative-time representations while retaining the acoustic axes before learned adaptation. This yields a waveform-to-graph interface with a fixed, parameter-free WST. Our configurations remain competitive with AASIST while using approximately 60% fewer trainable parameters and show clear gains on selected out-of-domain benchmarks. These results underscore the value of preserving parent-child relations within the carrier-modulation topology when constructing a compact, physically grounded interface for graph-based speech deepfake detection. Code will be released at https://github.com/saki-ciallo/wst-graph.

---

### BanglaTurn: A Benchmark and Whisper-Based Model for End-of-Turn Detection in Bangla Speech

Authors: Mizbaul Haque Maruf
Published: 2026-09-24
Categories: cs.CL
arXiv: https://arxiv.org/abs/2609.29371v1

This paper presents BanglaTurn, a corpus for end-of-turn detection in Bangla conversational speech, and a model trained on it. The corpus holds 35,374 samples of 3 to 15 s of podcast speech, labelled for turn state by combining speaker diarization with an LLM pass, with every label then checked by a human annotator. The model pairs a Whisper encoder with task-specific classification heads. On a class-balanced test set drawn from a held-out podcast, it reaches 84.33% accuracy (95% CI 80.3 to 88.1) against 69.28% for the Smart-Turn v3 baseline, and lowers the false negative rate from 51.57% to 7.55% at the cost of a higher false positive rate. We report what encoder layer fine-tuning, multi-scale pooling and INT8 quantization each contribute, and latency stays within 165 to 191 ms end to end on CPU.

---

### From Policy Documents to Structured Survey Responses: Evaluating Large Language Models for Policy Monitoring

Authors: Carolyn Cole, Matthias Deschryvere, Toqeer Ehsan, Arash Hajikhani
Published: 2026-09-24
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2609.29370v1

Science, technology, and innovation policies are crucial for competitiveness, yet their diversity and scale make them difficult to map and monitor consistently. Existing approaches rely heavily on manual survey efforts, which are costly and challenging to scale across countries. Large language models (LLMs) enable new possibilities for extracting and structuring information from long and unstructured policy documents. This paper presents an application of LLMs as "AI respondents" for generating structured survey responses from policy texts. We develop a data extraction pipeline based on long-context in-context learning to map information from public web sources into predefined survey categories, including policy instruments, target groups, and thematic areas. The pipeline integrates a validation step using a secondary LLM to assess relevance and evidence, alongside comparisons with human-provided responses. Using a multi-country dataset, we evaluate the alignment between LLM-generated and human-generated outputs through overlap measures and cross-validation. Results show that LLMs achieve high agreement for structured indicators (84-95%), while differences remain in free-text fields, where models tend to provide more detailed procedural descriptions. These findings highlight the potential of hybrid human-AI workflows for policy monitoring, improving both efficiency and scalability while maintaining the need for human validation and contextual interpretation.

---

### Epistemic-Probabilistic Model for Guarded Multi-Agent LLM Coordination

Authors: Mehdi Nasiri, Mohammad Saeed Arvenaghi, Sadegh Vaezi, Ebrahim Ardeshir-Larijani
Published: 2026-09-24
Categories: cs.AI
arXiv: https://arxiv.org/abs/2609.29366v1

Multi-agent large language models (LLMs) have become ubiquitous in applied AI, yet their theoretical foundations remain surprisingly understudied. Viewed through the lens of multi-agent systems theory, several shortcomings come to light: a lack of social intelligence, the absence of coordination mechanisms among agents, unknown emergent behavior, and interactions between agents that are bounded by natural language. We address two of these gaps: the absence of social behavior and the lack of mechanisms for inter-agent coordination. We introduce Epistemic Probabilistic Language Agents (EPLA), a neuro-symbolic architecture for multi-agent coordination under uncertainty. A Symbolic Guard provides structured diagnostic feedback. The LLM generates typed actions, and the Guard controls their execution against an authoritative symbolic state. We formalize the epistemic layer in a gossip testbed through epistemic lottery gossip models, which combine view-based call histories with agent-indexed probability weights. We argue that implementing such a formalism can address shortcomings of agentic LLMs.

---

### Beyond Simple Input-Output Assessment Tasks: Leveraging Automated Programming Assessment for Non-Trivial Courses

Authors: Artur Jordao
Published: 2026-09-24
Categories: cs.AI
arXiv: https://arxiv.org/abs/2609.29363v1

The public visibility of Artificial Intelligence (AI) is growing rapidly, driven by the positive impact of its applications across diverse fields of knowledge. In this new chapter, courses that cover the foundations of AI and machine learning become essential for understanding their role and potential in contemporary society. Therefore, understanding fundamental concepts and elementary algorithms through the close integration of theory with practice is essential in AI courses. In this essay, we report our experience designing machine learning exercises for automated assessment tools in programming. It is worth mentioning that we are not developing a novel form of automated grading system. Instead, we propose a perspective that frames machine learning problems as input-output assessment tasks. From this perspective, each exercise admits a unique and deterministic answer and enables automated programming assessment tools (e.g., VPL for Moodle, Codeforces, and MOJ) to effectively support AI education. We believe this essay can encourage instructors to foster educational innovation by adopting more dynamic and interactive approaches to AI courses that integrate theory and practice. Importantly, this essay does not introduce an innovation in the use of AI for education; rather, it introduces an innovative approach to improving the learning of AI, particularly, machine learning.

---

### Parts-of-Speech as Emergent Categories in SAE Latent Space

Authors: Alessandro Bondielli, Lucia Passaro, Serena Auriemma, Alessandro Lenci
Published: 2026-09-24
Categories: cs.CL
arXiv: https://arxiv.org/abs/2609.29362v1

Sparse AutoEncoders (SAEs) offer a promising way to inspect language model representations, but it is still unclear what kind of linguistic structure their latents expose. We use part-of-speech (PoS) categories as a controlled test case to study whether morpho-syntactic information is encoded by individual latents or by structured groups of features. We find that PoS distinctions are highly recoverable from SAE activations, but do not align with one-to-one latent / category mappings. This recoverability is not reducible to lexical memorisation, and Open and Closed PoS classes differ substantially. Categories are supported by compact groups of sparse latents, with substantial variation across tags. These groups remain stable on held-out data, while also showing overlap between related categories. Our results show that SAEs localise morpho-syntactic information in a distributed and category-dependent form rather than through atomic grammatical features.

---

### Domain Recentering and Confidence-Weighted Prior Calibration for Vision-Language Models

Authors: Youngeun Seol, Jimin Shin, Heeseo Yoon, Uiwon Hwang
Published: 2026-09-24
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2609.29358v1

Vision-language models such as CLIP achieve strong zero-shot classification, yet under distribution shift, visual embeddings drift from fixed text embeddings. Training-free calibration avoids the per-sample optimization of prompt learning, but prior feature calibration gives each image the full bias of one hard cluster. We propose Domain Recentering with Confidence Calibration (DRC), a training-free method adapting CLIP from a set of unlabeled target images. DRC fits a Gaussian mixture once and subtracts from each embedding a posterior-weighted average of component means. It then removes residual class preference with a log-prior correction, estimating the prior from confidence-weighted predictions. Among compared methods, DRC achieves the highest average accuracy on cross-domain datasets, exceeding zero-shot CLIP by 4.13 and 5.07 points with ViT-B/16 and ResNet-50, with gains over CLIP also holding under ImageNet distribution shifts.

---

### Learning a Flow to Self-Supervised Representations

Authors: Yuling Jiao, Wensen Ma, Houduo Qi, Defeng Sun
Published: 2026-09-24
Categories: cs.CV, cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2609.29350v2

Explicit geometric references offer a direct way to structure self-supervised representations. Existing adversarial distribution-matching formulations, however, require costly encoder-critic optimization. We introduce Flow-Based Distribution Matching (FBDM), a non-adversarial framework that learns this reference-directed geometry through spherical conditional velocity regression. An ETF-inspired reference allows its number of components K' to exceed the auxiliary flow dimension d* while retaining structured geometric separation. We assign both augmented views of each image to the same target, while limiting how many images each reference center can receive. An explicit alignment loss further pulls the two views' representations closer together. Experiments across benchmarks ranging from CIFAR to ImageNet show that FBDM achieves performance nearly on par with DM and remains competitive with existing SSL methods. Matched training-cost comparisons show a 1.48- to 1.83-fold speedup over DM with a negligible increase in GPU memory usage. We also provide a theoretical explanation for the usefulness of the learned representations: under stated conditions, we bound the downstream misclassification rate in terms of the FBDM pretraining loss.

---

### ArGuard Shared Task: Harmful Content Detection in Arabic Memes and LLM Prompts

Authors: Firoj Alam, Md. Rafiul Biswas, Mohamed Bayan Kmainasi, Ali Ezzat Shahroor, Hamdy Mubarak, George Mikros, Abul Hasnat, Wajdi Zaghouani
Published: 2026-09-24
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2609.29349v2

ArGuard is a shared task on harmful content detection in Arabic memes and LLM prompts. It includes two tracks: Track A focuses on multimodal hate detection in Arabic memes, while Track B addresses harmful prompt detection for Arabic LLM safety evaluation. In total, 58 teams registered, 35 participated in the final evaluation, and 27 submitted system-description papers. Participating teams explored models such as AraBERT, Jais, and Qwen3-VL. The best systems achieved macro-F1 scores of 0.823 on A1, 0.419 on A2, 0.984 on B1, and 0.790 on B2. Fine-grained meme classification in A2 was the most challenging setting, partly due to sparse labels and train-test distribution shifts.

---

### SEE Challenge 2026: Event-Guided Brightness Adjustment Across a Broad Illumination Range

Authors: Yunfan Lu, Mingchao Xu, Hanyu Zhou, Shaoyu Liu, Haoyue Liu, Peiqi Duan, Shihan Peng, Yinqiang Zheng, Boxin Shi, Gim Hee Lee, Hui Xiong, Davide Scaramuzza
Published: 2026-09-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.29347v1

Event cameras provide a high dynamic range and preserve brightness-change cues in lighting conditions where conventional RGB frames may be noisy or saturated. To benchmark event-guided restoration across a broad illumination range, we organized the SEE Challenge 2026 with the Event-Based Multimodal Vision Workshop at ECCV 2026. The task conditions restoration on one or more RGB frames, synchronized events, and a scalar target-brightness statistic provided by the organizers. It uses SEE-600K, which contains 610,126 image-event observations from 202 real-world scenes spanning low-light, normal-light, and high-light conditions with illumination variations of up to 1,000$\times$. The challenge follows an open-system protocol: participants may use different temporal contexts, architectures, pretrained weights, test-time augmentation, and post-processing strategies. PSNR determines the ranking, and SSIM is reported as a secondary metric. Around 70 teams registered interest and 15 valid CodaBench submissions were received. Six distinct teams completed organizer-side identity and technical verification, provided method descriptions, checkpoints, inference code, and instructions, and are included in the verified open-system ranking reported here. Beyond the ranking, this report analyzes exposure subsets, semantically distinct test cases, a shared failure pattern, system design choices, and inference strategies. The top systems obtain closely spaced average scores, while the best-performing method varies across cases and metrics; under severe underexposure, all verified systems retain visible local errors.

---

### The Last Human Gate: Forward Deployed Engineering for Governance Automation

Authors: Jeremy Canale
Published: 2026-09-24
Categories: cs.AI, cs.SE
arXiv: https://arxiv.org/abs/2609.29345v1

Enterprise governance requires decisions, evidence, and accountable authority; it does not require every review task to retain its current human implementation. We develop a task-substitution framework for Digital Governance Frameworks (DGF), treating each gate as an executable contract. Substitution requires sufficient accessible information, valid decision and authority checks, and a reduction in total human work after exceptions, verification, correction, and maintenance are counted. We derive a residual-work threshold and show why automating most cases can still increase labor. Forward deployed engineering connects these conditions to an architecture for agents, rule engines, evidence services, and escalation. DGF-Bench supplies controlled evidence from 300 synthetic projects and 899 evaluable model-project runs. Gemini 3.8 Flash, GPT-5.6 Luna, and DeepSeek v4.1 Flash achieve strict gate success of 94.98%, 83.29%, and 74.18%; complete-route success is 76.92%, 42.33%, and 24.67%. A deterministic control passes all 1,700 gates given the supplied rules and structured facts, locating the comparison in execution of a supplied decision kernel. Evidence audits and 135 repeated runs distinguish correct decisions from reliable execution. A document counterexample establishes an information-sufficiency obstruction. These results support the technical feasibility of replacing human execution of specified governance-review tasks with agents and software. The framework specifies a workforce test based on the complete human effort required at fixed output and quality; the present measurements concern review performance. Sources, dossiers, traces, and analyses are public.

---

### SkinAgent AI: A Safety-Grounded Multimodal Agentic Framework for Non-Diagnostic Skincare Support

Authors: Muhammad Muhtasim Shahriar, Abdullah Mohammad Sayem, Tze Hui Liew, M. F. Mridha, Md. Mahiuddin
Published: 2026-09-24
Categories: cs.AI
arXiv: https://arxiv.org/abs/2609.29341v1

Consumer-facing skincare AI must coordinate visual evidence, product information, tool use, and user-facing actions within explicit evidence and safety boundaries. This study evaluates SkinAgent AI, a non-diagnostic multimodal framework that combines visual concern routing with grounded and auditable LLM-based orchestration. The architecture includes routing for Acne, Pores, and Wrinkles; photograph-based skin-type estimation; count-informed ordinal acne-severity support; typed tools; database-grounded recommendation and action functions; deterministic safety, privacy, and evidence checks; approval before state-changing actions; and structured trace and replay mechanisms. Visual-model performance and system-level agent behavior were evaluated separately. Across three seeds, the skin-condition routing model achieved 99.84% +/- 0.07% accuracy. Skin-type estimation achieved 88.85% accuracy, while count-informed acne-severity support achieved 84.59% accuracy with a quadratic weighted kappa of 0.9076. On a locked but non-independent 240-case system benchmark, intent accuracy was 80.00%, exact tool-set match was 62.92%, and strict task completion was 47.08%. No violations or successful cross-user leakage events were observed in the finite safety and privacy test suites. Tool-selection errors, incomplete grounding of product attributes, and unreliable failure fallback nevertheless remained. These findings support the feasibility of bounded, database-grounded, and traceable agent orchestration for non-diagnostic skincare assistance. They do not establish clinical readiness, external generalization, formal privacy guarantees, or universal safety. Independent validation, expert assessment, robustness and fairness testing, and prospective evaluation in real-world settings remain necessary.

---

### A Study of the Limits of Collaborative DCT-Based Image Denoising via Interpretable Neural Networks

Authors: Cristian Comellas, Julia Navarro, Antoni Buades
Published: 2026-09-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2609.29334v1

Image denoising remains a fundamental problem in image restoration, with applications in photography, biomedical, and scientific imaging. Modern deep neural networks achieve strong performance by learning powerful image priors, but often rely on large black-box models with limited interpretability. In contrast, DCT-based sliding-window and collaborative filtering methods such as BM3D offer clear algorithmic structure, but depend on handcrafted and non-differentiable operations. This work studies how far such structured collaborative filtering principles can be pushed when reformulated as trainable models. We introduce DeepBM3D, a compact fully differentiable architecture that combines non-local patch grouping, DCT-domain filtering, and multi-stage refinement within a BM3D-inspired pipeline. Lightweight convolutional feature extractors guide patch grouping, while filtering is performed through learned Wiener weights in the DCT domain. Experiments show that DeepBM3D improves over classical and hybrid baselines, remains competitive with FFDNet at low and moderate noise levels, and performs particularly well on repetitive textures.

---

### Where LLM Graders Succeed and Break: Evidence from Two Computer-Science Exams

Authors: Ali Habibullah, Yazan Alshoibi, Mohammad Alshiekh, Salman Khan, Naeemullah Khan
Published: 2026-09-24
Categories: cs.CL, cs.AI, cs.CY
arXiv: https://arxiv.org/abs/2609.29333v1

One long-form exam in a large course costs hundreds of grader-hours, and qualified graders are scarce; LLM graders are a tempting alternative. To show its pitfalls we grade a practical Computer Vision exam ($570$ dual-graded students) under $171$ configurations spanning closed and open-weights models; the best reaches mean absolute error $1.64/35$, below the $2.61/35$ two human graders achieve against each other. The catch is the prompt: a short ''strict grader'' preamble drives $14$ of $17$ open-weights models out of the graded band ($\text{MAE} \ge 8$), three stopping grading altogether. The damage traces to the preamble's two credit-withholding sentences, not to tone or model scale; one of them, ''never give partial credit'', alone makes two of three probed models stop grading. The closed flagships of three vendors shift calibration under it but stay in the band. In $162$ further configurations on a second, independent Machine Learning exam from another course ($1{,}038$ dual-graded students), the preamble worsens ten models, moving three out of the band into collapse and one into refusal, yet improves seven whose neutral prompts over-mark: the vulnerability replicates, but its direction is exam-specific. Light LoRA fine-tuning repairs it: one adapter on the two exams' pooled $\sim 3{,}900$ graded examples brings five small open models to parity or better with a human grader in agreement with the grader pair, and sensitivity to the three harsh personas nearly vanishes ($\le 0.32$ MAE). We release the anonymised dataset, full ablation grid, and grading, fine-tuning and analysis pipelines.

---

### FlowAtom: Atom-Based Evidence Aggregation for Multi-Label Website Fingerprinting

Authors: Chongru Fan, Wentao Huang, Wei Wang, Zhenquan Ding, Jinqiao Shi, Wei Cai, Zhiyu Hao
Published: 2026-09-24
Categories: cs.LG, cs.NI
arXiv: https://arxiv.org/abs/2609.29330v1

Identifying the set of monitored websites in mixed encrypted traffic is challenging because an individual flow often provides only partial evidence of website identity. To address this challenge, we propose FlowAtom, which constructs shared prototypes, called Atoms, from flow representations without website labels. Specifically, FlowAtom pretrains a flow encoder on external unlabeled traffic and aggregates Atom responses across flows within each observation window into a fixed-dimensional, permutation-invariant representation for monitored website-set prediction. Across Direct HTTPS, Trojan, and VMess, FlowAtom achieves micro-F1 scores of 97.82%, 94.43%, and 93.92% in closed-world evaluation, respectively, and consistently outperforms the evaluated baselines in open-world evaluation on windows containing monitored visits. The code is available at https://github.com/aimafan123/FlowAtom.

---

### Hyperbolic Multimodal Continual Learning: A Closest-Admissible Solution

Authors: Jiahong Liu, Ming Shen, Xiaohao Liu, Rex Ying, Menglin Yang, Tat-Seng Chua, Irwin King
Published: 2026-09-24
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2609.29329v1

Existing continual-learning methods protect parameters, replayed examples, or Euclidean feature subspaces. When applied to hyperbolic multimodal models, they do not explicitly preserve the Lorentz geometry that jointly encodes within-modality similarity, cross-modal correspondence, and semantic hierarchy; sequential updates can therefore retain task scores while still distorting previously learned relations. We address this gap with Hyperbolic Multimodal Continual Learning (HMCL). We show that preserving the old multimodal geometry amounts to restricting all modalities to one shared hyperbolic isometry, which induces a family of admissible first-order parameter changes. We formulate a joint closest-admissible (CA) correction that retains the shared rotation best matching the candidate modal updates; its minimal-rotation (MR) special case fixes this rotation to zero. Both variants correct the displacement realized by AdamW, and task anchoring bounds within-task accumulation while preserving learning freedom. Across a unified 16-task classification-retrieval stream with three hyperbolic backbones, HMCL improves final performance and backward transfer over sequential fine-tuning and four continual-learning baselines; HMCL-CA gives the highest Overall score on every backbone. A modality-extended stream confirms the retrieval gains. Representation analyses find 81.2 to 95.5 percent less radial, angular, cross-modal, and paired-distance drift; ImageNet-WordNet results show better semantic ancestry and radial hierarchy.

---

### Grammatical "grandmother neurons" are rare in LLMs

Authors: Linyang He, Nima Mesgarani
Published: 2026-09-24
Categories: cs.CL
arXiv: https://arxiv.org/abs/2609.29328v1

Understanding how Large Language Models (LLMs) encode linguistic structures remains a fundamental challenge in interpretability research. While diagnostic classifiers (or "probes") are widely used for this task, they face significant methodological criticism: training auxiliary classifiers introduces capacity confounds and calibration issues, often making it difficult to distinguish the model's intrinsic representations from the probe's ability to learn the task. To address these limitations, we introduce a probe-free framework for localizing linguistic selectivity at the individual neuron level. Leveraging the controlled contrasts of linguistic minimal pairs, we propose a Neuron Separability Index (NSI), a metric that directly quantifies how reliably single neurons differentiate grammatical from ungrammatical constructions without parameter updates. Applying NSI across 68 linguistic paradigms and seven checkpoints reveals three main patterns: 1) raw separability reaches near-peak levels earlier for morphological and syntactic distinctions than for syntax-semantics interface and conceptual distinctions. 2) after permutation normalization, single-unit selectivity is sparse, weak, and narrowly tuned: only a small fraction of units are sensitive to an average paradigm, and strongly selective "grandmother neurons" are rare. 3) whole-vector linear separability, single-neuron selectivity, and behavioral competence are largely dissociated, and targeted ablations further separate activation selectivity from causal reliance.

---