# Research Papers — 2026-06 · bundle page 90 of 260

> Full-text markdown of papers #4451–4500 of 12993 published in 2026-06, newest first. AI-readable mirrors of the full 12993-paper 2026-06 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2026-06/page/90
[← Previous bundle](https://aipod.dev/papers/archive/2026-06/page/89.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-06/page/91.md)

### CORTIS: Text-Only Adaptation of Spoken Language Models for Task-Oriented Voice Agents

Authors: Youngwon Choi, Hyeonyu Kim, Taeyoun Kwon, Donghyuk Jung, Myeongkyun Cho
Published: 2026-06-19
Categories: cs.HC, cs.AI, cs.SD, eess.AS
arXiv: https://arxiv.org/abs/2606.21453v1

Task-oriented voice agents need to map spoken user requests to structured outputs such as semantic frames, executable actions, and function calls. A common approach is to cascade ASR with a text-based LLM, but transcription errors can propagate to downstream structured output generation, especially under noisy conditions. Spoken language models (SLMs) offer a direct speech-based alternative, yet adapting them to new tasks typically requires paired speech-target annotations. Motivated by this gap, we present CORTIS, a text-only adaptation framework for task-oriented voice agents. CORTIS fine-tunes SLMs using text-form task supervision, enabling speech-based structured output generation at inference time without task-specific speech-target annotations during adaptation. We evaluate CORTIS on two Qwen2.5-Omni backbones and three task-oriented speech datasets, including an in-house product dataset, and compare it with matched ASR-LLM cascades trained with the same text-form task supervision. Results show that CORTIS performs competitively with matched cascades and offers clearer advantages under acoustic degradation, particularly in preserving high-level task semantics. These findings suggest that text-only fine-tuning of SLMs can serve as a practical adaptation strategy for voice agents when paired speech-target data are costly to collect.

---

### Fast-TurboQuant: A Multiplier-Free Online Vector Quantization Approach

Authors: Pedro M. R. Pereira, Felipe A. P. de Figueiredo, Rausley A. A. de Souza
Published: 2026-06-19
Categories: cs.LG, cs.IT, eess.SP
arXiv: https://arxiv.org/abs/2606.21448v1

As large language models scale, memory bandwidth for key-value caches and retrieval-augmented generation systems becomes a critical bottleneck. While 1-bit quantization addresses this constraint, recent TurboQuant relies on dense random rotation matrices to condition the vector distribution before quantization. This projection demands millions of floating-point multiplications per embedding, making it difficult to deploy on constrained edge silicon. We introduce Fast-TurboQuant, a multiplier-free projection architecture that replaces the dense matrix with a structured fast Johnson-Lindenstrauss transform. By applying a Rademacher phase inversion followed by a fast Walsh-Hadamard transform (FWHT), the method leverages sub-Gaussian concentration to satisfy the prerequisites of scalar Lloyd-Max quantization without Gaussian projections. This substitution reduces the arithmetic complexity to only additions, eliminating hardware multipliers. Evaluation on DBpedia OpenAI-3 Large embeddings demonstrates a 19.7 times algorithmic speedup under sequential execution. Furthermore, the dimension expansion due to the FWHT zero-padding reduces the mean squared error and improves Recall@10.

---

### Precision Recall Controllable Radiology Report Generation via Hybrid Natural Language and Clinical Reward Learning

Authors: Ling Chen, Ruinan Jin, Jun Luo, Hanliang Chen, Quirin Strotzer, Rongkai Yan, Yuan Xue, Luciano Prevedello, Dufan Wu
Published: 2026-06-19
Categories: cs.CL, cs.CV
arXiv: https://arxiv.org/abs/2606.21447v3

Automated radiology report generation (RRG) has gained increasing attention because it can reduce the heavy workload of clinical report writing. However, most existing methods mainly optimize for natural language generation (NLG) metrics that focus on language fluency, while providing little control over clinically important factors such as precision and recall. As consequence, generated reports may be fluent but not well aligned with different clinical needs. To address this challenge, we propose a reinforcement learning framework for precision recall controllable RRG, where a control parameter explicitly adjusts the trade-off between clinical precision and recall during inference. This design allows the model to flexibly generate reports according to different clinical requirements. To ensure clinical correctness, we introduce a clinical reward into the training objective, which helps improve clinical efficacy (CE) beyond standard language-based optimization. In addition, we apply a group-relative training strategy that normalizes rewards within each training group, reducing reward variance and improving training stability. Extensive experiments on the MIMIC-CXR dataset show that our method consistently outperforms state-of-the-art approaches in both NLG and CE evaluation metrics, while providing reliable control over the CE precision recall trade-off.

---

### Synergistic Dual-Branch Adaptation for Multi-modal Generalized Category Discovery

Authors: Yuxun Qu, Minyu Zhou, Yongqiang Tang, Chenyang Zhang, Wensheng Zhang
Published: 2026-06-19
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.21446v1

Generalized Category Discovery (GCD) aims to classify old categories and discover new ones from unlabeled data. Recent multi-modal approaches introduce retrieved or synthesized texts into a dual-branch architecture to provide semantic cues complementary to visual features. However, the cross-modal synergy in existing dual-branch methods remains coarse and incomplete: the two modalities are encoded independently with the bias and noise in the derived text left unaddressed during encoding, and existing mutual learning strategies operate only on global class-level anchors, lacking fine-grained relational supervision. To address these limitations, we propose the Synergistic Dual-Branch Adaptation (SDBA) framework, which serves as a plug-and-play enhancement compatible with existing dual-branch methods such as GET and TextGCD. SDBA comprises two components: the cross-modal synergistic adapter inserts lightweight adapters into both branches and further injects visual information into the text adapter at each encoder layer to enhance text feature learning during encoding; the neighborhood mutual learning module enforces consistent local neighborhood distributions between the two branches via bidirectional KL divergence, providing fine-grained relational supervision for both old and new classes. Extensive experiments on six benchmarks demonstrate state-of-the-art performance, and consistent improvements on different baselines validate the broad scalability of the proposed framework.

---

### AutoRAS: Learning Robust Agentic Systems with Primitive Representations

Authors: Yang Yue, Xuancheng Zhu, Yuyang Ma, Guoshun Nan, Zihan Dou, Jingru Shan, Congyu Guo, Ji Zhang, Hua Wang, Jingfeng Zhang
Published: 2026-06-19
Categories: cs.AI
arXiv: https://arxiv.org/abs/2606.21445v1

The automated design of agentic systems offers a promising pathway for scaling large language models (LLMs) beyond single-agent reasoning. While prior work has advanced task performance through handcrafted or automatically generated multi-agent workflows, robustness is often treated as an afterthought, leaving systems vulnerable to external adversaries and internal failures. We propose AutoRAS, a framework for the Automated design of Robust Agentic Systems. AutoRAS formulates system design as generating a sequence of symbolic primitives that jointly encode structural connectivity and behavioral actions, and learns to optimize this sequence using execution-derived safety signals and flow-based sequence-level objectives. Extensive experiments show that AutoRAS achieves the best performance in both vanilla and adversarial settings, with the smallest performance degradation under attacks. Further analyses demonstrate strong transferability, stable optimization behavior, stability across primitive sets, and favorable cost trade-offs. Our code is available at $\href{https://github.com/guohezuy/AutoRAS}{\text{this https URL}}$.

---

### Fusing Backdoors, Machine Learning, and Optimization for Large-Scale Parametric Mixed-Integer Programs

Authors: El Mehdi Er Raqabi, Pascal Van Hentenryck
Published: 2026-06-19
Categories: cs.LG, math.OC
arXiv: https://arxiv.org/abs/2606.21440v1

Large-scale optimization problems are often solved repeatedly under similar structural conditions, leading to substantial computational overhead. This occurs in applications such as power systems, transportation, and supply chain networks, where the underlying structure is fixed while parameters frequently vary under perturbations. This paper proposes a Learning to Optimize (LTO) framework that accelerates the solution of large-scale general mixed-integer problems by leveraging the concept of a backdoor, i.e., a subset of variables that drive most of the computational complexity. The proposed BIPC framework consists of three phases. Phase I is an identification procedure that discovers a backdoor for a set of instances in the distribution. Phase II uses supervised learning to develop machine learning models that, given an instance, predict values for bounded-domain backdoor variables and intervals for wide-domain backdoor variables. These predictions define a reduced optimization problem where the predictions constrain the backdoor variables, while the other variables remain free. Phase III optimizes this reduced problem and, if necessary, applies a correction step to restore feasibility or the optimality guarantees. Experiments on real-world, large-scale problems show substantial reductions in solution time with only a limited loss in solution quality. The framework enables organizations to solve large-scale optimization problems efficiently in the presence of frequent perturbations, such as unexpected events, demand fluctuations, or operational changes. Because these changes affect parameters rather than the problem structure, BIPC can quickly provide high-quality, feasible solutions, offering a practical approach to integrating machine learning into existing optimization pipelines.

---

### Universal Encoders for Modular Relational Deep Learning

Authors: Jakub Peleška, Gustav Šír
Published: 2026-06-19
Categories: cs.LG, cs.DB
arXiv: https://arxiv.org/abs/2606.21434v1

Relational Deep Learning (RDL) models multi-tabular databases as temporal heterogeneous graphs for end-to-end representation learning. While RDL is evolving rapidly, existing approaches face significant generalization obstacles. They are either schema-specific, requiring training from scratch for every new database, or they rely on monolithic architectures that entangle feature encoding with graph message-passing. Analyzing these limitations, we establish four core pillars for building foundational relational models: semantic granularity, structural topology, temporal causality, and unified optimization. Addressing these pillars, we propose a modular approach that decouples row encoding from graph message-passing. We introduce the Universal Row Encoder, a transformer-based module that integrates raw cell data with schema metadata$-$including column semantics, table names, and global distribution statistics$-$to produce table-width invariant row embeddings. By explicitly feeding global statistics to an intra-row self-attention mechanism, the encoder natively contextualizes unseen features and handles sparse data. Serving as a flexible "backend" for any downstream graph architecture, our pretrained encoder enhances cross-database knowledge transfer on the established RelBench benchmarks while improving learning convergence and memory footprint.

---

### Central limit theorem for the averaged Adam optimizer

Authors: Steffen Dereich, Arnulf Jentzen
Published: 2026-06-19
Categories: math.PR, cs.LG, math.ST
arXiv: https://arxiv.org/abs/2606.21433v1

In this article, we analyse convergence of the averaged Adam optimizer to an attracting zero of the Adam vector field. We provide a central limit theorem that, in particular, quantifies exactly the speed of convergence. The order of convergence is $n^{-1/2}$ in the number of steps of the algorithm which coincides with the order observed for classical stochastic approximation algorithms. The covariance in the central limit theorem is given in terms of properties of the Adam algorithm in the state of the attractor.

---

### Soliton-like Waves in a Two-Dimensional Recurrent Spiking Neural Network with Weighted Spike-Timing-Dependent Plasticity

Authors: Ch. Meessen
Published: 2026-06-19
Categories: cs.NE, q-bio.NC
arXiv: https://arxiv.org/abs/2606.21432v1

We construct a minimal but biologically plausible spiking neuron model operating in discrete time, combining multiplicative spike-timing-dependent plasticity (WSTDP), divisive normalization of synaptic integration, homeostatic threshold adaptation, and a one-step refractory period. We show that this normalization admits a biologically plausible dendritic implementation in which each binary junction operates using only locally available information. Assembling excitatory-inhibitory pairs of such neurons into a two-dimensional recurrent network and applying periodic localized stimulation, we find that the network spontaneously gives rise to stable, self-propagating wave packets with the properties of dissipative solitons: they maintain a stable spatial profile, propagate at constant speed, and annihilate upon frontal collision. Their emergence requires a geometric asymmetry between excitatory and inhibitory connection radii, and initial inhibitory synapses stronger than excitatory ones. WSTDP engraves the direction of propagation into the synaptic weight profile, so that the network learns by itself to sustain propagation in one direction while suppressing the reverse. When two sources are active simultaneously, the resulting waves annihilate upon collision, defining a semi-persistent boundary whose position encodes the relative phase and frequency of the two sources. These results provide a minimal computational framework for studying the emergence of cortical traveling waves, activity zone delimitation, and spatial memory from local plasticity rules alone.

---

### Does Mixture-of-Experts Actually Help Inference on Consumer and Edge Hardware? An Empirical Study

Authors: Alfarizy Alfarizy, Hung Truong Thanh Nguyen, René Richard, Roozbeh Razavi-Far, Hung Cao
Published: 2026-06-19
Categories: cs.PF, cs.AI
arXiv: https://arxiv.org/abs/2606.21428v3

Mixture-of-Experts (MoE) language models are often described as ideal for resource-constrained inference. Each token activates only a small subset of experts, so the per-token compute cost, in floating-point operations (FLOPs), resembles that of a much smaller dense model. Whether that FLOP advantage survives in practice is far less clear. We ask whether MoE models actually run faster and cheaper than comparable dense models on consumer-grade and edge hardware. We benchmark OLMoE-1B-7B (1.3 B active of 6.9 B total) against three dense baselines on an Apple M2 Pro and an NVIDIA Jetson Orin Nano 8 GB through llama$.$cpp, measuring throughput, memory, and on-device energy. The answer is device-dependent: OLMoE's active-parameter advantage is only partly realised on the laptop (~10% behind the same-active Llama-3.2-1B) and erodes on the edge device (~31% behind, at 2.1$\times$ the energy per token, with peak memory at the 8 GB ceiling). Patching llama$.$cpp to time the decode graph node-by-node shows routing accounts for under 9% of MoE-block compute on the cleaner edge backend, so the gap reflects total-parameter memory footprint, expert dispatch, and KV-cache pressure rather than routing. The implication is that on bandwidth-bound edge hardware, inference cost tracks total parameters, not active ones, and sparse activation does not buy back what the device is constrained on. These findings are bounded to one MoE model at this parameter scale and two devices, and we release the full measurement harness and per-run data.

---

### Dual-Attention Convolution Experts for Sparse Tensor Completion

Authors: Yanlei Liu, Zhenyu Liao
Published: 2026-06-19
Categories: cs.LG, cs.IR
arXiv: https://arxiv.org/abs/2606.21427v1

Tensor factorization (TF) has been widely adopted for high-dimensional sparse data completion tasks. Despite significant progress, neural TF methods often struggle to capture complex cross-mode interactions and remain vulnerable to (extreme) data sparsity. To address these challenges, we propose a novel neural tensor factorization approach, termed Dual-Attention Convolution Expert Networks with Group-Level Contrastive Learning (DCGC). For the first problem, DCGC generates diverse non-linear alignment patterns of latent factors via a multi-channel convolution network, and leverages the gated dual-attention mechanism to drive the model to focus on more important output channels (i.e., convolution experts) and the aligned features. Furthermore, DCGC introduces a group-level contrastive learning strategy that aggregates positive samples with identical feedback levels while separating negative samples across different levels. This strategy injects high-quality self-supervised signals to mitigate data sparsity. Extensive experiments conducted on five datasets demonstrate that our DCGC outperforms the state-of-the-art methods in sparse tensor completion for traffic and recommendation applications. Code to reproduce the experimental results in the paper is available at https://github.com/ku1z/DCGC.

---

### Federated Temporal Attention Intelligence for Cyber-Resilient IoMT: Lightweight Digital Twins and PPO-Driven Honeypot Deception

Authors: Syed Zeeshan Haider, Anwar Shah, Muneeb Arif, Hamza Iftikhar, Waqas Ali
Published: 2026-06-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.21422v1

The rapid proliferation of Internet of Medical Things (IoMT) devices introduces critical cybersecurity vulnerabilities in healthcare environments where resource-constrained medical devices operate under strict latency requirements and stringent data-privacy regulations. To address these challenges, this paper presents the Lightweight Digital Twin and Federated Reinforcement Learning (LDT-FRL) framework, a privacy-preserving defense architecture integrating four complementary mechanisms: a Temporal Attention Encoder (TAE) built on a GRU backbone with learned temporal self-attention for flow-level threat classification; lightweight LSTM-based Digital Twins trained on normal-class traffic to generate per-device anomaly scores that gate the TAE classifier through a learned sigmoid coupling; a Federated Proximal Policy Optimization (PPO) agent selecting among ALLOW, ISOLATE, and HONEYPOT_REDIRECT actions based on a seven-dimensional state; and an intelligent honeypot layer that converts redirected suspicious traffic into actionable threat intelligence. A federated aggregation strategy employing EMA-smoothed per-client validation losses as inverse-weighted FedAvg coefficients stabilizes global model updates under non-IID client distributions. Evaluated on CICDDoS 2019 and TON-IoT benchmarks, LDT-FRL achieves 99.66% and 99.95% test accuracy respectively, with macro-F1 scores of 0.9913 and 0.9995, converging 81% faster than the DTFL-CD baseline while attaining perfect F1=1.000 on the severely imbalanced MITM class. Explainability analysis via SHAP, LIME, Grad-CAM, and counterfactual methods confirms that the TAE focuses on semantically meaningful flow features, providing interpretable evidence for each defense decision.

---

### MIRCaps: A Large-Scale Mixed-Domain Dataset with Image-Level and Region-Level Captions for Fine-Grained Vision-Language Learning

Authors: Arlindo Luciano Tulumba Roberto, Hyungjoon Kim
Published: 2026-06-19
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2606.21419v1

Despite recent progress in Vision-Language Models (VLMs), mixed-domain image-caption datasets for both general-purpose and CCTV-based video surveillance systems remain limited. To address this gap, we introduce a large-scale multimodal dataset comprising 141,364 images, 981,947 image-level captions, 1,742,264 region-level captions, and 1,391,779 bounding box annotations. Each image is associated with an average of seven image-level captions describing different aspects of the overall scene, as well as seven region-level captions for each annotated bounding box. These complementary caption types are designed to help VLMs learn fine-grained visual attributes, including object categories, estimated sizes, colors, actions, states, and surrounding environmental context. We demonstrate the effectiveness of the dataset on two important downstream tasks: image captioning and object detection. Experimental results show that lightweight VLMs, including SmolVLM-256M-Instruct, BLIP, BLIP2, and Qwen2.5-VL 3B-Instruct, can be effectively fine-tuned using our dataset. Our dataset and code are publicly available at https://zenodo.org/records/20418601.

---

### One Size does not Fit All: Heterogeneous Latent Space Alignment for Unsupervised Domain Adaptation

Authors: Evangelia Koskinioti, Yi Shen, Georgios Stamou, Michael M. Zavlanos
Published: 2026-06-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.21415v1

Domain shift remains a major obstacle to the reliable deployment of machine learning models in high-stakes environments such as healthcare. While Domain adaptation aims to mitigate these effects, existing approaches suffer from limited expressiveness of latent representations and a reliance on handcrafted, static augmentations. In this work, we address these limitations by proposing a novel deep learning architecture for Unsupervised Domain Adaptation (UDA), specifically optimized for medical image segmentation. Our framework, ADualVUOT, integrates a dual-encoder Variational Autoencoder (VAE) with Continuous Normalizing Flows (CNFs) to increase modeling flexibility and posterior expressiveness. To achieve domain alignment, we leverage Unbalanced Optimal Transport (UOT) through the Gaussian-Gromov-Wasserstein (GGW) distance, which handles structural and topological discrepancies between domains. Furthermore, we incorporate an adversarial augmentation scheme to synthesize worst-case compositions, thus enhancing model robustness. Extensive experiments on medical imaging benchmarks show significant gains over prior OT-based approaches.

---

### 2D Versus 3D Diffusion for In Silico Training of Interventional X-ray AI Models

Authors: Sampath Rapuri, Jeremy Ko, Benjamin D. Killeen, Russell H. Taylor, Mathias Unberath
Published: 2026-06-19
Categories: eess.IV, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2606.21414v1

The ability to synthesize realistic X-ray images has catalyzed the development of AI models for X-ray image-guided procedures, which otherwise suffer from a lack of available annotated data. Prior work has demonstrated the effectiveness of mechanistic simulation of digitally reconstructed radiographs (DRRs) as a training data source for a myriad of tasks, including segmentation and anatomical landmark detection, with comparable or superior performance to real data training. However, mechanistic DRR synthesis still relies on the availability of annotated high-resolution anatomical models. Deriving these from CT images of real patients or specimens imposes an undesirable bottleneck on data quantity and variability. In this work, we explore two methods for synthesizing training data: (1) a 3D conditional latent diffusion model that generates CT volumes to use as inputs for mechanistic DRR generation without real, 3D anatomical models, and (2) a view-conditioned 2D diffusion model that produces synthetic X-rays. In controlled experiments, we demonstrate that synthetic 2D diffusion-based X-rays can be used to train an anatomical landmark detection model that generalized to real X-ray images with performance rivaling that of a model trained on real X-ray images. Thus, we provide preliminary evidence that synthetic, 2D diffusion-based training data can substitute for real X-ray data, identifying a promising avenue towards generating large, diverse datasets for training robust AI models in interventional X-ray imaging.

---

### CAT-Translate: Building Compact Open-Source Models for Japanese-English Translation

Authors: Yuu Jinnai
Published: 2026-06-19
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2606.21413v1

Nowadays, large multilingual translation models demonstrate impressive translation capabilities in the machine translation benchmarks. This raises a practical question to the developers: is it worth developing translation models specialized for a particular language pair if you only need to support that language pair? To give an anecdotal answer to this question, we develop a family of small language models (0.8B, 1.4B, 3.3B, and 7B parameters) specialized for Japanese-English bidirectional translation. We employ a two-stage supervised fine-tuning approach followed by Multi-Objective GRPO (Ichihara et al. 2025) to train models on synthetically generated parallel corpora. We evaluate our models on WMT and real-world translation benchmarks across business, legal, medical, financial, and patent domains. While multilingual models achieve strong performance on WMT benchmarks, our compact models outperform them on real-world benchmarks, suggesting the practical utility of developing specialized translation models even in the era of large multilingual models.

---

### Don't Blindly Trust It: How Unreliable Feedback Breaks Tool-Using LLM Agents

Authors: Chubin Zhang, Zhenglin Wan, Xingrui Yu, Pengfei Zhou, Wangbo Zhao, Jingxuan Wu, Yaxin Zhou, Ivor Tsang
Published: 2026-06-19
Categories: cs.AI
arXiv: https://arxiv.org/abs/2606.21409v1

Tool-augmented agents are typically evaluated by their gains under reliable external feedback. Yet these gains leave open a key counterfactual: when feedback is unreliable, would the agent be better off receiving no task evidence? We study this question with a controlled matched-loop comparison that fixes the agent loop, prompt, action space, and decoding, while varying only the returned observation: faithful, misleading, or absent. Across question answering and fact verification, persistent misleading feedback produces a value inversion: agents that benefit from clean tools can perform worse than the matched no-feedback fallback. On HotpotQA, Qwen2.5-7B reaches 44.8 F1 with clean retrieval and 22.3 F1 with no feedback, but drops to 4.7 F1 under shuffled retrieval. The inversion persists under stronger clean retrieval and locally plausible distractors, but weakens when later clean evidence can repair the trajectory. Early trajectory signals predict many failures, yet simple repairs remain fallback-limited: rejecting bad evidence helps only when the exposed fallback is reliable. These results show that clean-tool gains can overstate tool value, and that matched no-feedback fallback controls are necessary for evaluating tool-augmented agents.

---

### Robot Self-Improvement via Human-Video Dynamics Models

Authors: Hanzhi Chen, Anran Zhang, Simon Schaefer, Kejia Chen, Shi Chen, Daniel Cremers, Oier Mees, Stefan Leutenegger
Published: 2026-06-19
Categories: cs.RO, cs.CV
arXiv: https://arxiv.org/abs/2606.21406v1

A central question in robot learning is how to acquire skills from the kinds of data that humans learn from: passive observation, embodied practice, and the experience of failure. Human videos provide the first of these in abundance, and prior work has shown they can initialize useful policies. Far less clear is whether they can support the second and third: whether priors extracted from human videos can ground a robot's own attempts well enough to evaluate them, correct them, and improve from them. In this work, we show that human videos can be used to learn embodiment-agnostic action, dynamics, and value representations that transfer across robot embodiments, providing the predictive foundation required for robots to autonomously improve from their own rollouts and failures. We introduce Dynamics-Guided Action Correction (DGAC), a training-free approach that leverages these adapted models to repair failed states: each failure becomes a query for which the learned models propose and rank corrective actions, turning failures into supervision for the next policy update. Across seven real-world manipulation tasks spanning both a mobile manipulator and a static manipulator arm, our approach improves success rates from 40% to 81% across multiple policy backbones, demonstrating cross-embodiment robot self-improvement from human-video priors. These results show that human priors and robot failures can be combined to enable scalable autonomous policy improvement. Project page: https://ethz-mrl.github.io/robot-self-improvement-website/.

---

### SwarmX: Agentic Scheduling for Low-Latency Agentic Systems

Authors: Yeqi Huang, Yanwei Ye, Guomin Chen, Wenhao Su, Bin Gong, Jialian Li, Zhan Lu, Yangshen Deng, Xuan Sun, Le Xu, Luo Mai
Published: 2026-06-19
Categories: cs.DC, cs.AI
arXiv: https://arxiv.org/abs/2606.21401v2

Agentic AI applications compose multiple model calls and tool executions, creating new scheduling challenges for GPU-CPU clusters. Their inference time and model-call structure often depend on prompt semantics, making conventional scheduling approaches ineffective for low-latency serving. This paper presents SwarmX, a system that implements agentic scheduling for low-latency agentic applications. SwarmX uses scheduling-specific neural predictors to capture prompt, device, runtime, and target-model features; exposes distributional predictions to routers and scalers for tail-aware decisions; and provides mechanisms for predictor training and online adaptation. These predictors and mechanisms are integrated into a scheduler-agent framework that provides a common substrate for integration with existing scheduling and model-serving infrastructure. We evaluate SwarmX using production deployment (nearly one thousand GPUs and one million CPU cores) and controlled experiments on a 128-GPU testbed. Across multi-agent code generation, deep research, and multimodal agentic workflows, SwarmX reduces tail latency by up to 61.5% compared to state-of-the-art schedulers and sustains up to 2x the throughput of production schedulers under the same SLO.

---

### Calibration Is Not Control: Why LLM-Agent Oversight Needs Intervention

Authors: Chubin Zhang, Zhenglin Wan, Xingrui Yu, Jingxuan Wu, Qi Wen, Pengfei Zhou, Wangbo Zhao, Ivor Tsang
Published: 2026-06-19
Categories: cs.AI
arXiv: https://arxiv.org/abs/2606.21399v1

Runtime oversight for LLM agents is commonly framed as scalar risk prediction: estimate failure likelihood, confidence, or uncertainty, then intervene once the score crosses a threshold. We argue that this framing targets the wrong object for control. The relevant question is not how likely the agent is to fail if it continues, but whether an available intervention would improve the outcome. Two trajectory prefixes can have the same risk estimate while requiring different actions, because one remains recoverable and the other does not. We formalize this mismatch as target error and identify intervention advantage, the expected utility gain from intervening rather than continuing, as the decision object for oversight. To measure this mismatch, we introduce prefix branching, a same-prefix counterfactual protocol that executes candidate actions from identical trajectory states. Across four benchmarks, action-conditioned control yields regime-dependent gains over scalar routing. In a calibration decomposition, recalibrating the same scalar score improves prediction metrics but leaves control regret unchanged, showing that calibration alone does not repair target error. A simple prefix-only action-conditioned controller substantially reduces regret in the strongest interactive regime, from 0.506 to 0.110 on ALFWorld. Gains shrink when interventions are weak or when scalar routing already preserves intervention-relevant information. These results suggest that LLM-agent oversight should move from calibrated risk scoring toward action-conditioned value estimation.

---

### Evaluating LLMs for Real-World Web Vulnerability Detection

Authors: Sebastian Neef, Luca Jungnickel, Antonio Benjamin Buchholz, Valene Spence, Vicente Birke Gonzalez
Published: 2026-06-19
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2606.21397v1

Large Language Models (LLMs) have emerged as a promising tool for automated vulnerability detection, yet their effectiveness on web-specific vulnerabilities remains to be explored. This work benchmarks six frontier (Claude Opus 4.6, Codex GPT-5.4, Gemini 3.1-pro-preview) and open-weight models (Qwen 3.5, Qwen 3 Coder Next, MiniMax M2.5) on their ability to detect real-world web vulnerabilities using static analysis in WordPress plugins, including SQL injection, stored cross-site scripting, path traversal, and remote code execution. Using five prompt designs of varying structure, scope, and complexity across three experiment iterations, we aim to answer how model and prompt choice affects vulnerability detection. Our results show that all models are capable of detecting valid security issues, but the detection rate varies depending on the model and prompt. For example, Claude Opus 4.6 achieved the highest web vulnerability detection rate (63%), while open-weight MiniMax M2.5 performs on par with other frontier models (48%), and self-hosted Qwen 3.5 only achieved 35%. We show that scoped prompts that narrow the vulnerability scope outperform open-ended ones, whereas the prompt complexity has little impact. Surprisingly, no model achieved full reporting consistency across three experiment iterations, with some as low as 50%. Our experiments demonstrate the opportunities and limits of LLM-based vulnerability detection, as no model correctly identified one baseline vulnerability in one of the plugins. Additionally, we derive practical lessons learned for security practitioners and publish all code and data to support future research.

---

### Atomistic Language Models Understand and Generate Materials

Authors: Sathya Edamadaka, Krithik Ramesh, Ju Li, Rafael Gómez-Bombarelli
Published: 2026-06-19
Categories: cs.LG, cond-mat.mtrl-sci
arXiv: https://arxiv.org/abs/2606.21395v1

Atomistic structure and natural language have long been modeled separately, with language models either calling atomistic models as tools or being fine-tuned on lossy textual encodings that discard atomistic information. We introduce Atomistic Language Models (ALMs) to pursue native multimodality, in which a single language backbone understands atomistic structures, generates materials from natural language, and optimizes crystal structures as instructed by text. By unifying a pretrained atomistic encoder, large language model, and denoising diffusion model through purely continuous projectors and staged training, ALMs achieve state-of-the-art results on crystal structure prediction and de novo generation. ALMs are enabled by a continuous bridge that maps language model embeddings directly into the steering space of atomistic diffusion, and are assisted by Text-to-Crystal Feynman-Kac (T2C-FK), a particle-based sampler that scores partial denoising trajectories to enforce stoichiometric targets at inference time. To evaluate the ability of ALMs to optimize and generate materials from natural-language prompts and 3D atom-coordinate inputs, we introduce ALM Bench, the first benchmark for text-conditioned crystal generation and optimization. Code, training data, and model weights will be released soon.

---

### VLA-FAIL: Efficient Task Failure Detection for Finetuned Vision-Language-Action Models

Authors: Florian Seligmann, Emiliyan Gospodinov, Enes Ulas Dincer, Gerhard Neumann
Published: 2026-06-19
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2606.21386v1

Vision-language-action models (VLAs) achieve state-of-the-art performance on many robotic manipulation tasks, yet they can still behave unpredictably in out-of-distribution scenarios. Runtime failure detection is therefore essential for the safe real-world deployment of VLAs. However, existing task failure detectors require computationally expensive action sampling, are based on architectural assumptions that limit their applicability to VLAs, or need access to failure rollouts. We propose VLA-FAIL, a lightweight and broadly applicable failure detection framework for VLAs that combines two novel failure detectors with minimal overhead, without requiring failure data. The first, last-layer Mahalanobis distance (LLMD), detects out-of-distribution states by measuring token-wise deviations in last-layer features relative to the training data. The second, action chunk consistency (ACC), exploits the temporal overlap induced by receding-horizon control and detects failures when consecutive action chunks become inconsistent. To capture the trade-off between detection accuracy and detection latency, we introduce AUCPDT, a threshold-independent metric that jointly evaluates precision, recall, and detection time. Through extensive real-world and simulation experiments, we demonstrate that LLMD and ACC capture complementary failure modes whose combination enables reliable and early failure detection across diverse tasks, frequently outperforming significantly more expensive baseline methods.

---

### Unsupervised Disentanglement Without Compromises : How Functional Orthogonality Enforces Identifiability

Authors: Mathieu Cyrille Simon, Pascal Frossard, Christophe De Vleeschouwer
Published: 2026-06-19
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.21385v2

This paper explores unsupervised disentangled representation learning from a functional perspective. We define latent concepts as factors that influence observations through locally orthogonal directions, formalized as an orthogonality constraint on the Jacobian of the generative mapping. We prove that this condition yields identifiability of general nonlinear generative models, without requiring statistical independence or causal assumptions, provided the latent domain admits all combinations of factor values. Experiments with orthogonality-regularized normalizing flows empirically confirm the theory, demonstrate reliable recovery of ground-truth factors, and shed light on the success of VAEs. These findings challenge the prevailing impossibility claims for unsupervised disentanglement and provide a principled alternative foundation.

---

### Unsupervised Disentanglement Without Compromises : How Functional Orthogonality Enforces Identifiability

Authors: Mathieu Cyrille Simon, Pascal Frossard, Christophe De Vleeschouwer
Published: 2026-06-19
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.21385v1

This paper explores unsupervised disentangled representation learning from a functional perspective. We define latent concepts as factors that influence observations through locally orthogonal directions, formalized as an orthogonality constraint on the Jacobian of the generative mapping. We prove that this condition yields identifiability of general nonlinear generative models, without requiring statistical independence or causal assumptions, provided the latent domain admits all combinations of factor values. Experiments with orthogonality-regularized normalizing flows empirically confirm the theory, demonstrate reliable recovery of ground-truth factors, and shed light on the success of VAEs. These findings challenge the prevailing impossibility claims for unsupervised disentanglement and provide a principled alternative foundation.

---

### EnTrust: Modeling Inter-Modal Conflict for Trustworthy Multimodal Medical Image Analysis

Authors: Dwarikanath Mahapatra, Abhijit Das, Behzad Bozorgtabar, Zongyuan Ge, Sudipta Roy, Deepak Nayak, Mauricio Reyes, Imran Razzak
Published: 2026-06-19
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2606.21384v1

Multimodal medical imaging fuses complementary anatomical and functional information, yet modalities frequently disagree in pathologically heterogeneous regions. Current segmentation models handle this in one of two inadequate ways: deterministic fusion that averages away disagreement, or post-hoc uncertainty estimation decoupled from the fusion process that produces it. Both obscure the clinically critical question: why is this prediction unreliable? We present EnTrust, a framework that treats inter-modal conflict as the primary source of predictive uncertainty. Our EnFuse module decomposes multimodal features into three disentangled components: shared anatomical consensus (F_c), modality-specific cues (F_{u,m}), and spatially localized conflict signals (F_{cf}), with independence enforced via a cross-covariance objective. This structured decomposition conditions SegDiff, a diffusion-based generative segmentation model whose sampled hypotheses diverge specifically in regions of modal disagreement. TrustMap then translates this hypothesis divergence into calibrated, pixel-wise uncertainty using ensemble entropy, conflict-guided perturbation probing, and a learned calibration head, enabling clinicians to understand not only where predictions are uncertain, but why. Across four benchmarks spanning brain, cardiac, lesion, and oncology domains, EnTrust achieves state-of-the-art segmentation accuracy while reducing calibration error by 40% compared to the strongest baseline. Notably, it outperforms 5x deep ensembles using a single model at roughly half the memory footprint. Code and checkpoints are available at https://github.com/GenMI-Lab/EnTrust.git.

---

### OSOG: A Differentiable, Physics-Informed Synthetic Data Engine for Micro-Optical Environments

Authors: Caio Silva
Published: 2026-06-19
Categories: cs.CV, cs.GR, physics.optics
arXiv: https://arxiv.org/abs/2606.21381v1

Deep learning in computational microscopy is severely constrained by the scarcity of densely annotated datasets. While synthetic data generation has bridged this gap in macroscopic computer vision, traditional graphics engines rely on geometric ray-tracing, failing to capture the micro-optical phenomena required for microscopy. Conversely, while wave-optics formulations exist, rendering them computationally tractable at the scale required for deep learning remains a massive systems challenge. To address this, we introduce the Optical Synthetic Object Generator (OSOG), a high-performance, fully differentiable forward-modeling engine. Drawing on established physical models of diffraction and phase retardation, OSOG maps continuous Optical Path Difference (OPD) calculations into a highly optimized, PyTorch-native Structure-of-Arrays (SoA) architecture. We validate this computational framework across three axes: First, object detection models (YOLOv11-OBB) trained purely on OSOG-generated data achieve robust zero-shot transfer to real-world highly occluded Lysozyme micrographs. Second, we introduce DiffOSOG, demonstrating that the engine's end-to-end differentiability allows for the exact recovery of continuous optical parameters via curriculum-guided inverse rendering. Finally, OSOG bypasses the $\mathcal{O}(N)$ bottlenecks of sequential ray-tracing, demonstrating sub-linear scaling by synthesizing 40,000 complex wave-optic particles in under 50 milliseconds (\>20 FPS). By providing a fast, scalable, and physically grounded tensor pipeline, OSOG enables true real-time, on-the-fly dataset generation.

---

### Enhancing Creativity in 3D Generative Design via a TRIZ-Inspired Text-to-CAD Framework

Authors: Dongeon Lee, Leekyo Jeong, Soyoung Yoo, Sunwoong Yang, Namwoo Kang
Published: 2026-06-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.21378v1

Recent advances in large language models (LLMs) have demonstrated significant potential in supporting engineering design tasks, including computer-aided design (CAD) automation. However, most existing LLM-based 3D CAD generation approaches primarily focus on geometric precision and instruction-following performance, often overlooking the fundamental aspect of creative design exploration. This study presents a TRIZ-inspired text-to-CAD framework that leverages LLMs to generate high-quality, editable CAD models while systematically exploring creative design alternatives. The framework integrates the Theory of Inventive Problem Solving (TRIZ)-embedding deep human insights from extensive patent records-into LLM prompting strategies, enabling autonomous generation of innovative CAD variants that address technical contradictions. Through a comprehensive three-stage pipeline of design generation, enhancement, and optimization, the framework produces structurally diverse CAD models from well-crafted prompts. The present study implements and evaluates the first two stages, while positioning the design optimization stage as future work. A product design case study (chair) demonstrates that the TRIZ-inspired text-to-CAD framework generates multiple creative design alternatives by systematically applying TRIZ inventive principles such as segmentation, anti-weight, dynamics, and composite materials, achieving 4.0-14.7% mass reduction across all enhanced designs while maintaining structural integrity. The key findings suggest that integrating systematic innovation methodologies with LLM-based 3D CAD generation bridges the gap between precision-focused synthesis and creativity-focused exploration, advancing toward autonomous design systems where AI makes design decisions independently, supporting human decision-making in human-AI collaborative design for engineering applications.

---

### FLM-Occ: Feed-forward Likelihood Maximization for Efficient Indoor Occupancy Prediction

Authors: Guangcheng Chen, Lihuang Fang, Huaqi Tao, Yicheng He, Li He, Hong Zhang
Published: 2026-06-19
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2606.21373v1

Recent indoor occupancy prediction methods adopt Gaussian primitives as a sparse 3D representation for computational efficiency. However, their training relies on voxel classification, which imposes only local constraints and lacks global supervision on the distribution of the primitives. Therefore, they inevitably predict spurious primitives in empty regions, undermining both representational and computational efficiency. To address this, we propose Feed-forward Likelihood Maximization (FLM), a novel framework that reformulates occupancy prediction as voxel distribution estimation. In FLM, a network is trained to predict a mixture model that maximizes the likelihood over ground-truth occupied voxels in a feed-forward manner. To enable end-to-end training of networks and voxelization of a standard mixture model, we define mixture weights as normalized primitive volumes to implicitly enforce simplex constraints and derive novel voxelization formulas. Based on FLM, our FLM-Occ, a novel method that is capable of relocating randomly initialized primitives over long distances to model a scene. On Occ-ScanNet, FLM-Occ achieves superior accuracy using only 32 superquadrics, 2.7% of the prior SoTA, while running 3.7 times faster.

---

### NAC: Neural Action Codec for Vision-Language-Action Models

Authors: Ahad Jawaid, Yu Xiang
Published: 2026-06-19
Categories: cs.RO, cs.LG
arXiv: https://arxiv.org/abs/2606.21372v2

Vision-language-action (VLA) models rely on discrete action tokenizers to bridge continuous robot control and autoregressive sequence modeling, yet existing tokenizers often trade off between compression, latency, and downstream performance. We revisit this design through the lens of neural audio codecs - convolutional encoder-decoder architectures with residual vector quantization that serve as the standard front end for audio foundation models. Motivated by their success, we introduce the Neural Action Codec (NAC), which treats short robot action trajectories as multi-channel 1D signals and compresses them using a multi-scale RVQGAN architecture. With adaptations to the action representation, compression rate, and reconstruction objective, audio-codec-style models can autoencode actions with high fidelity without substantial architectural changes. NAC provides a compact, ordered token space via offset codebooks, enabling standard autoregressive policies to operate over short, structured sequences. Meanwhile, a Vocos-style decoder with an ISTFT head and adversarial discriminators recovers action trajectories. Across LIBERO-10, RoboMimic, and a suite of real-world manipulation tasks, NAC achieves high reconstruction fidelity and higher average success rates than binning, FAST, and prior VQ-based tokenizers at comparable or better compression rates. These results demonstrate that repurposed neural audio codecs offer a strong, practical backbone for learned action tokenization in modern VLAs.

---

### NAC: Neural Action Codec for Vision-Language-Action Models

Authors: Ahad Jawaid, Yu Xiang
Published: 2026-06-19
Categories: cs.RO, cs.LG
arXiv: https://arxiv.org/abs/2606.21372v1

Vision-language-action (VLA) models rely on discrete action tokenizers to bridge continuous robot control and autoregressive sequence modeling, yet existing tokenizers often trade off between compression, latency, and downstream performance. We revisit this design through the lens of neural audio codecs-convolutional encoder-decoder architectures with residual vector quantization that serve as the standard front end for audio foundation models. Motivated by their success, we introduce the Neural Action Codec (NAC), which treats short robot action trajectories as multi-channel 1D signals and compresses them using a multi-scale RVQGAN architecture. We observe that audio-specific mel-spectrogram objectives are ill-suited for kinematic signals; however, by replacing them with simple time-domain and non-mel spectral reconstruction losses, audio-codec-style models can autoencode actions with high fidelity without substantial architectural changes. NAC provides a compact, ordered token space via offset codebooks, enabling standard autoregressive policies to operate over short, structured sequences. Meanwhile, a Vocos-style decoder with an ISTFT head and adversarial discriminators recovers smooth, detailed trajectories. Across LIBERO-10, RoboMimic, and a suite of real-world manipulation tasks, NAC achieves lower reconstruction error and higher success rates than binning, FAST, and prior VQ-based tokenizers at comparable or better compression rates. These results demonstrate that repurposed neural audio codecs offer a strong, practical backbone for learned action tokenization in modern VLAs.

---

### Graph-of-Differences: Anatomy-Structured Difference Alignment for Medical Image Re-Identification

Authors: Nichula Wasalathilaka, Abhijit Das, Imran Razzak, Dwarikanath Mahapatra
Published: 2026-06-19
Categories: cs.CV, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2606.21368v1

Medical image re-identification (MedReID) enables longitudinal patient linkage but remains vulnerable to shortcut learning and often produces decisions that clinicians cannot audit against named anatomy. We propose Graph-of-Differences (GoD), which grounds identity comparisons in explicit anatomical structure. Each image is represented as an anatomy graph whose nodes correspond to named anatomical regions; given an image pair, soft node correspondence is established, and differences are computed over matched anatomy. A graph-level difference alignment objective ties these anatomy-matched differences to the global backbone difference, ensuring the retrieval signal is anchored in homologous structures rather than arbitrary spatial tokens. Explanations are defined over named graph nodes and quantitatively audited via node insertion/deletion tests, replacing unstable pixel heatmaps with verifiable structure-level evidence. On internal benchmarks, GoD improves Rank-1 by +7.1 pp on fundus and +3.1 pp on CXR over a strong frozen-backbone baseline, with further gains on zero-shot external transfers confirming that anatomy grounding improves both accuracy and generalization. Code is available at https://github.com/GenMI-Lab/GoD.git.

---

### Sexualised synthetic personas encode and amplify gendered power asymmetries through voice

Authors: Alice Ross, Ariadna Sanchez, Elin Kanhov, Catherine Lai, Éva Székely
Published: 2026-06-19
Categories: eess.AS, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2606.21366v2

This work examines sexualised AI-generated English-speaking voices offered by a popular commercial platform. New technologies may enable sexual empowerment and greater diversity in gender expression, yet toxic masculinity, heteronormativity, and the abuse of women and LGBTQ+ people remain pervasive online. Drawing on a Feminist HCI perspective, we examine how commercial voice AI systems reproduce and circulate particular performances of gender. We conducted a listening experiment with a diverse group of listeners, combining quantitative adjective selection, qualitative free-text responses, and acoustic analysis. Participants evaluated male- and female-coded voices presented with either sexualised scripts or neutral text. Results reveal a narrow range of gender expression, largely binary and heteronormative. Female-coded voices are more frequently described using sexualised and submissive terms, while male-coded voices are more often associated with dominance and positive traits.

---

### LambdaMark: Semantic Audio Watermarking for Robustness and Radioactivity

Authors: Kexin Li, Xiao Hu, Ilya Grishchenko, David Lie
Published: 2026-06-19
Categories: cs.SD, cs.AI
arXiv: https://arxiv.org/abs/2606.21365v1

Recent advances in generative audio have made voice cloning increasingly effortless, enabling voice fraud, impersonation, and other forms of unauthorized use. A common attack finetunes a speech generation model on recordings of a target speaker, allowing the model to synthesize speech in that speaker's voice. Audio watermarking offers a promising defense by embedding detectable signals into audio. A practical watermark must satisfy two key properties: robustness and radioactivity. Existing audio watermarking methods typically embed signals into low-level representations, such as waveforms or spectrograms, which makes them vulnerable to signal-level manipulations and limits their transfer to downstream models. We introduce LambdaMark -- the first generic radioactive watermarking scheme. Unlike all previous approaches, LambdaMark achieves generic radioactivity by embedding multi-bit watermark information into semantic audio latent representations. Our watermarks have semantic interpretation and are thus more likely to be learned by a downstream model through finetuning. LambdaMark includes a lightweight watermark encoder to inject multi-bit message-dependent perturbations into semantic audio representations and a decoder to detect watermark presence and recover the embedded bit information. Encoder and decoder are trained using a custom multi-component loss that preserves fidelity of the watermarked audio, increases bit-level recovery rate, and improves robustness against common distortions and adversarial removal attempts. Experiments show that LambdaMark achieves near-perfect robustness under common distortions. LambdaMark is also the only watermark that is robust against all evaluated removal attacks. Furthermore, LambdaMark exhibits general and robust radioactivity and remains robust to distortions and adversarial removal attacks even on the generated outputs of those finetuned models.

---

### Predictive Repair Management Using a Multi-Head Attention Transformer and Online Learning

Authors: Xinyao Zhang, Willie Cade, Karl R. Haapala, Arun Natarajan, Sara Behdad
Published: 2026-06-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.21364v1

Accurate prediction of repair duration is an important challenge in product maintenance due to its implications for resource allocation, customer satisfaction, and operational performance. This study aims to develop a deep learning framework to help fleet repair shops accurately categorize repair time given product historical data. The study uses an automobile repair and maintenance dataset and creates an end-to-end predictive framework by employing a multi-head attention network designed for tabular data. The developed framework combines categorical information, transformed through embeddings and attention mechanisms, with numerical historical data to facilitate integration and learning from diverse data features. A weighted loss function is introduced to overcome class imbalance issues in large datasets. Moreover, an online learning strategy is used for continuous incremental model updates to maintain predictive accuracy in evolving operational environments. Our empirical findings demonstrate that the multi-head attention mechanism extracts meaningful interactions between vehicle identifiers and repair types compared to a feed-forward neural network and a random forest model. Also, combining historical maintenance data with an online learning strategy facilitates real-time adjustments to changing patterns and increases the model's predictive performance on new data. The model is tested on real-world repair data spanning 2013 to 2020 and achieves an accuracy of 78%, with attention weight analyses illustrating feature interactions.

---

### Does Finetuning with Scientific Data Increase Hallucinations? A Multi-domain Factuality Evaluation of LLMs

Authors: Raia Abu Ahmad, Nikolas Rauscher, Ekaterina Borisova, Fabio Barth, Georg Rehm, Sebastian Möller
Published: 2026-06-19
Categories: cs.CL
arXiv: https://arxiv.org/abs/2606.21359v2

Large language models (LLMs) are increasingly used to communicate and explain scientific concepts, yet their tendency to hallucinate poses significant risks in this high stakes use-case. Prior scientific hallucination evaluation work remains largely restricted to the biomedical domain, treats hallucination as a binary task, and has not examined the growing family of scientifically fine-tuned LLMs. We address these gaps with SciFactCheck, a benchmark of 2,500 prompts across five scientific domains, paired with a modular evaluation framework targeting three factuality hallucination types: unverifiability, overclaim, and attribution. Using a controlled minimal-pairing design, we evaluate 18 LLMs by comparing each scientifically fine-tuned model against its general-purpose base. Our results indicate that 1. Scientifically fine-tuned models exhibit degraded factual reliability across all hallucination types and scientific domains, and 2. Fine-tuned models are internally less confident yet linguistically more assertive. A human pilot study further reveals that current fact-checking tools show only modest agreement with expert judgments on scientific content, and that defining scientifically check-worthy claims remains contested even among human annotators. Our findings fundamentally challenge current methods of domain-specific fine-tuning for factuality and call for developing improved verification infrastructure for scientific content.

---

### Finetuning with Scientific Data Increases Hallucinations: A Multi-domain Factuality Evaluation of LLMs

Authors: Raia Abu Ahmad, Nikolas Rauscher, Ekaterina Borisova, Fabio Barth, Georg Rehm, Sebastian Möller
Published: 2026-06-19
Categories: cs.CL
arXiv: https://arxiv.org/abs/2606.21359v1

Large language models (LLMs) are increasingly used to communicate and explain scientific concepts, yet their tendency to hallucinate poses significant risks in this high stakes use-case. Prior hallucination evaluation work remains largely restricted to the biomedical domain, treats hallucination as a binary task, and has not examined the growing family of scientifically fine-tuned LLMs. We address these gaps with SciFactCheck, a benchmark of 2,500 prompts across five scientific domains, paired with a modular evaluation framework targeting three factuality hallucination types: unverifiability, overclaim, and attribution. Using a controlled minimal-pairing design, we evaluate 18 LLMs by comparing each scientifically fine-tuned model against its general-purpose base. Our results indicate that 1. Scientifically fine-tuned models exhibit degraded factual reliability across all hallucination types and scientific domains, and 2. Fine-tuned models are internally less confident yet linguistically more assertive. A human pilot study further reveals that current fact-checking tools show only modest agreement with expert judgments on scientific content, and that defining scientifically check-worthy claims remains contested even among human annotators. Our findings fundamentally challenge current methods of domain-specific fine-tuning for factuality and call for developing improved verification infrastructure for scientific content.

---

### LEViL: Label-Efficient Video Learning via Zero-Shot Distillation over VLM-Generated Pseudo-Label Spaces

Authors: Aslı Çelik
Published: 2026-06-19
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.21358v1

Supervised video pretraining is a common transfer learning practice for improving downstream action recognition performance. However, it requires large-scale labeled source datasets, and the effectiveness of the learned initialization is influenced by the similarity between the source and target domains. Constructing such labeled pretraining datasets for different target domains is costly and difficult to scale. To address these limitations, this study proposes a label-efficient video learning framework that combines annotation-free video pretraining with target-label-set-aware fine-tuning. During pretraining, a vision-language model (VLM) generates textual descriptions of unlabeled videos, which are processed to construct an interpretable semantic pseudo-label space. A frozen video-language model then produces zero-shot soft target distributions over this space, allowing a student video encoder to learn semantically rich representations without manual source annotations. During downstream adaptation, target-label-set-aware fine-tuning combines supervised learning from labeled target videos with zero-shot distillation over the actual target label set, helping preserve VLM-derived semantic guidance while adapting the pretrained encoder to the target task. Experiments on UCF101 and HMDB51 show that the proposed framework outperforms the compared semi-supervised video action recognition methods across all evaluated limited-label regimes. Moreover, the annotation-free pretraining stage learns transferable representations that provide an effective initialization for full-data fine-tuning, despite relying on a comparatively modest unlabeled pretraining pool.

---

### SOHET: Sequence Of Heterogeneous Events Transformer with Self-Supervised Pre-Training

Authors: Kees Jan de Vries, Mustafa Radha, Mathijs de Jong
Published: 2026-06-19
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.21356v1

Many machine learning applications rely on heterogeneous event streams to make predictions, either causally as events arrive or bidirectionally over complete sequences. We propose SOHET (Sequence Of Heterogeneous Events Transformer), a hierarchical architecture combining event-type-specific tabular encoders with temporal and type embeddings, processed by a causal or bidirectional transformer. We introduce three self-supervised pre-training objectives for the causal setting. On a proprietary large-scale real-world Booking.com fraud detection task with 17 event types, SOHET outperforms FlexTPP, NAPPT, and CIPPT by 5.8%. Pre-training yields an additional 2.6% gain and 2.4% faster convergence. On the EBES benchmark, bidirectional SOHET matches or exceeds the published best on 6 out of 8 tasks.

---

### Urban Power Grid Topology and Hierarchy Identification from Open Data

Authors: Shiliang Zhang, Sabita Maharjan
Published: 2026-06-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.21352v1

Understanding the complex topology and hierarchy of urban power grid is crucial for energy prognosis, power flow management, and system resilience analysis. However, detailed grid information remains largely proprietary. This creates significant barriers for research and innovation, especially when analyzing the last-mile distribution networks connecting individual buildings. This paper addresses this challenge by developing an open-data-driven framework for the complete identification of urban power grid topology, from high-voltage transmission down to individual building connections. Particularly, we fuse public infrastructure data (power-lines, substations, transformers, poles) to map the high and medium-voltage skeleton using graph-based algorithms. We then leverage geospatial machine learning on OpenStreetMap building data to group power demand clusters, and infer the physical topology of the final distribution lines linking the clustered buildings. We apply the developed framework to the district of Alna in Oslo, Norway, and we reconstruct the complete grid topology that connects 7,330 buildings and all major electricity infrastructure assets. With the research in this work, we provide a critical tool that facilitates power system analysis, e.g., power flow optimization, cascading failure simulation, and grid resilience against the penetration of distributed renewable generation.

---

### A Reward-Petri-Net Interpretation of Temporal Behavior Trees

Authors: Till Schmeil, Günther Waxenegger-Wilfing, Sebastian Schirmer
Published: 2026-06-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.21350v1

This paper introduces an interpretation of Temporal Behavior Trees (TBTs) as Reward-Petri-Nets (RPNs) for reinforcement learning (RL). Designing reward functions for complex, long-horizon robotic tasks is notoriously difficult, especially when tasks have hierarchical structure and temporal constraints. TBTs extend conventional behavior trees (BTs) used in robotic applications by incorporating temporal properties into their leaf nodes. This allows TBTs to represents not only the behavioral task structure defined by BT operators such as Sequence, Fallback, and Parallel, but also the task's temporal constraints. In this work, the constraints are specified in the leaf nodes using Linear Temporal Logic. In order to inform RL rewards using TBTs, we provide a translation from TBT into a Petri Net (PN) and show how rewards can be automatically assigned based on the TBT's structure, resulting in a RPN. In a series of increasingly challenging environments, we demonstrate how TBT-based rewards enable learning where vanilla RL fails, improve sample efficiency, and offer flexible, intuitive control over the learning progress. We showcase the learning impact by using different reward distribution schemes and TBT structures.

---

### Factual Retrieval in LLMs Is a Redundant, Distributed and Non-Contiguous Process

Authors: Hail Hochman, Natalie Shapira, Yoav Goldberg
Published: 2026-06-19
Categories: cs.CL
arXiv: https://arxiv.org/abs/2606.21345v1

Large language models (LLMs) store and recall factual knowledge, yet the precise mechanism of how entity representations are transformed to enable specific attribute retrieval remains underexplored. In this work, we investigate this mechanism through the lens of an "attribute-computation path"-a sequence of computational steps over the entity representation required to elicit a target attribute. We then propose an iterative patching protocol to identify a minimal subset of layers necessary for this computation. Applying our method to LLaMA 3.1 8B and Qwen3 8B, we find that these paths are non-contiguous, often skipping layers, and that models possess multiple, functionally-equivalent paths for the same entity and fact, highlighting a high degree of redundancy in attribute computation. This implies that knowledge computation is highly distributed, potentially explaining the localization-editing mismatch and suggesting that knowledge storage and retrieval in LLMs is far from being well understood.

---

### Mind the Noise: Sensitivity of Transformer-based Interaction-Aware Trajectory Prediction Models to Noisy Data

Authors: Shahab Salehi, Luca Lusvarghi, Miguel Sepulcre, Javier Gozalvez
Published: 2026-06-19
Categories: cs.AI, cs.LG, cs.RO
arXiv: https://arxiv.org/abs/2606.21344v1

Trajectory prediction allows autonomous vehicles to anticipate the future behavior of surrounding objects (or agents) and, accordingly, maximize the safety and efficiency of their driving. State-of-the-art Transformed-based interaction-aware trajectory prediction models, which rely on attention mechanisms to capture multi-agent interactions and maximize prediction accuracy, are commonly trained and evaluated on long-range high-quality datasets. These datasets are typically obtained by aggregating data from multiple vehicles or drones and removing any object detection or tracking noise offline. Yet, information about a surrounding object's state (its position, speed, heading) is far from being noiseless in real-world deployments. Object state estimation is affected by perception uncertainties and localization errors that can be particularly large for objects received via Vehicle-to-Everything (V2X) communications. In this paper, we analyze the impact of noisy object state information on the trajectory prediction accuracy of a state-of-the-art Transformer-based interaction-aware trajectory prediction model. Our study demonstrates that trajectory prediction accuracy can rapidly deteriorate as the noise intensity increases. Numerical results show that the prediction accuracy can reduce by a 1.3x factor under small noise levels and by as much as a 3.9x factor under the highest (yet realistic) noise conditions. These findings reveal the strong sensitivity of trajectory prediction models to noisy data, underscoring the need for more realistic training and evaluation datasets as well as noise mitigation strategies.

---

### An Evaluation Framework for Text-to-Speech Voice Reconstruction

Authors: Ariadna Sanchez, Christoph Minixhofer, Korin Richmond, Ondrej Klejch, Peter Bell, Simon King
Published: 2026-06-19
Categories: eess.AS, cs.CL, cs.SD
arXiv: https://arxiv.org/abs/2606.21343v1

Voice reconstruction using Text-to-Speech (TTS) offers a communication method for people with speech disorders, which aims to retain their speaker identity while improving intelligibility. Previous work generally relies on Mean Opinion Score (MOS) to evaluate naturalness and speaker similarity, but this has limited sensitivity and reliability. We propose an evaluation framework with subjective and objective components. Subjectively, we evaluate perceived intelligibility and speaker identity using Best Worst Scaling (BWS) with situational framing. Objectively, we demonstrate that standard measures fail to predict reconstruction success for highly unintelligible speakers, so we introduce a novel dual-reference distributional measure to assess the trade-off between intelligibility and speaker identity. By evaluating the output of 17 zero-shot TTS systems for 193 speakers, we show that our framework provides a reliable and task-aligned approach for assessing voice reconstruction.

---

### Synthetic Audio Generation Framework for Air Traffic Control Speech Recognition

Authors: Raphaël Bagat, Zhe Zhang, Junichi Yamagishi, Irina Illina, Emmanuel Vincent
Published: 2026-06-19
Categories: cs.CL
arXiv: https://arxiv.org/abs/2606.21340v1

Automatic Speech Recognition (ASR) systems, despite achieving remarkable accuracy in general-purpose domains with native speech (L1), struggle in domains like Air Traffic Control (ATC) due to strong channel noise, a presence of non-native (L2) English accents, and data scarcity. We propose a synthetic data generation pipeline with acoustical properties simulations specifically designed to address this lack of real data to improve recognition accuracy in the ATC domain. Our approach leverages a combination of neural generation techniques, including Text-to-Speech, Voice Conversion, L2-to-L1 accent conversion, and a novel controllable L1-to-L2 accent conversion framework built to simulate accented speech. Our experiments with the Whisper model on the ATCO2 corpus demonstrate that fine-tuning with either synthetic data alone, or a mix of real and synthetic data, significantly improves the word error rate over out-of-the-box and real data only baselines respectively.

---

### DataClaw0: Agentic Tailoring Multimodal Data from Raw Streams

Authors: Cong Wan, Zeyu Guo, Zijian Cai, Jiangyang Li, SongLin Dong, Lin Peng, Xiangyang Luo, Zhiheng Ma, Yihong Gong
Published: 2026-06-19
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.21337v2

Raw multimodal streams are abundant but noisy, redundant, and unaligned with any particular training objective. Turning them into supervision today means either brittle heuristics or repeatedly querying a proprietary vision-language model, a cost that recurs with every new sample. We ask whether this conversion can instead be learned once and reused, and formalise intent-conditioned Data Tailoring: given a raw stream and a high-level intent, a model must return schema-aligned, evidence-grounded training instances. Training DataClaw0 at 4B, 9B and 27B, we find that whether five heterogeneous domains should share one model depends on capacity. A jointly trained model is worse than per-domain experts at the two smaller scales and better at the largest, placing the crossover near 18B parameters. Matched-data comparisons, in which the joint model sees exactly the same data per domain as that domain's expert, attribute the reversal to cross-domain transfer rather than to data volume: it gains most where a domain is data-poor, and recovers 59\% of in-domain performance on domains withheld from training entirely. Downstream post-training reproduces this ordering on GUI navigation, action video generation and spatio-temporal VQA, and the joint configuration is also the cheaper to deploy, serving one model instead of five. Github: https://github.com/vancyland/DataClaw0

---

### Direct Raw Audio Signal Processing via Reservoir Computing: An Investigation into 'Feature-Free' Architectures

Authors: Rinku Sebastian, Simon O Keefe, Martin A Trefzer
Published: 2026-06-19
Categories: cs.SD, cs.LG
arXiv: https://arxiv.org/abs/2606.21335v1

This paper evaluates Reservoir Computing (RC) as an autonomous, 'feature-free' framework for audio processing, designed to eliminate traditional, handcrafted feature extraction stages. We investigate whether the high-dimensional temporal dynamics inherent in a reservoir can function as a robust end-to-end processor for the direct classification of raw acoustic signals. By bypassing computationally intensive representations like MFCCs, this approach seeks to mitigate significant intellectual and pre-processing bottlenecks in traditional signal pipelines. Our study evaluates and compares shallow, sequential, and parallel deep reservoir architectures to determine their capacity for hierarchical feature representation. Experimental results demonstrate that the proposed parallel approach consistently outperforms shallow and sequential baselines while maintaining low model complexity. These findings highlight the potential of RC as an efficient and scalable alternative for time-domain audio processing, offering a promising pathway toward deployable, low-power acoustic systems with minimal preprocessing requirements.

---

### Ramanujan Graph Rewiring with Non Negative Resistance Curvature

Authors: Hugo Attali, Rachid El Jouhri
Published: 2026-06-19
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.21333v2

Graph Neural Networks (GNNs) have emerged as a powerful paradigm for learning on graph-structured data by iteratively propagating and aggregating information across edges. However, conventional message passing schemes often suffer from over-squashing, whereby exponentially large neighborhoods are compressed into fixed-dimensional embeddings, impeding effective long-range dependency learning. In this work, we introduce Ramanujan Propagation, a graph rewiring strategy that leverages Ramanujan graphs to alleviate topological bottlenecks in GNNs. We first establish that suitably chosen Ramanujan graphs guarantee non-negative resistance curvature, which mitigates over-squashing and facilitates efficient information flow. We then propose an algorithmic framework to construct a Ramanujan rewired graph that preserves the local connectivity of the original graph. Our experiments demonstrate that our method outperforms nine state-of-the-art rewiring techniques. These results establish Ramanujan graphs as a rigorous structural prior for scalable, topology-aware message passing in GNNs.

---

### MedTS-TTT: Test-Time Training for Medical Time Series Classification

Authors: Mingzhi Chen, Yiyu Gui, Guibo Luo
Published: 2026-06-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.21329v1

Medical time series (MedTS) signals such as electroencephalography (EEG) and electrocardiography (ECG) support many clinical applications. However, substantial subject-level heterogeneity often induces subject-level distribution shift, causing a fixed parameter set to generalize poorly to unseen individuals. Compared with domain adaptation methods that often depend on extra adaptation components or target-batch statistics, Test-Time Training (TTT) provides a more practical solution for sequential clinical data by enabling online adaptation from unlabeled test samples. However, many representative TTT methods require iterative inner-loop optimization, increasing test-time overhead. In this paper, we propose MedTS-TTT, a test-time training framework for medical time series modeling. MedTS-TTT is built upon Closed-Loop Self-Alignment Test-Time Training (CLSA-TTT) and a Gated Convolutional Backbone (GCB). CLSA-TTT constructs a token-level self-supervised target and performs a single-step fast-weight update for intra-layer closed-loop alignment, enabling rapid sample-wise adaptation without iterative inner-loop optimization. GCB combines CLSA-TTT-based fast adaptation and token-level fusion with a gated convolutional branch to balance local dynamic modeling and information-flow control. On 4 public datasets (2 EEG and 2 ECG) with subject-independent splits, MedTS-TTT achieves 11 top-1 rankings out of 12 evaluations across 9 baselines and 3 metrics. The code is publicly available at https://github.com/mingzhi-c/MedTS-TTT.

---

### Sea-Scan: High-Accuracy, ML-based Dark Vessel Detection and Localisation via Weakly Supervised DAS Monitoring

Authors: Tian Tian, Agastya Raj, Lara Flanagan, John Kennedy, Marco Ruffini
Published: 2026-06-19
Categories: cs.SD, cs.LG, eess.SP
arXiv: https://arxiv.org/abs/2606.21326v1

We present an ML-based vessel detection and localization system, trained with weak supervision from imperfect AIS labels, that achieves a 97.8% detection rate at 1.98% false-trigger rate, successfully identifies dark-vessel events from unlabeled data.

---