# Research Papers — 2026-04 · bundle page 73 of 233

> Full-text markdown of papers #3601–3650 of 11612 published in 2026-04, newest first. AI-readable mirrors of the full 11612-paper 2026-04 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2026-04/page/73
[← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/72.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-04/page/74.md)

### Mechanistic Anomaly Detection via Functional Attribution

Authors: Hugo Lyons Keenan, Christopher Leckie, Sarah Erfani
Published: 2026-04-21
Categories: cs.LG, cs.CR
arXiv: https://arxiv.org/abs/2604.18970v2

We can often verify the correctness of neural network outputs using ground truth labels, but we cannot reliably determine whether the output was produced by normal or anomalous internal mechanisms. Mechanistic anomaly detection (MAD) aims to flag these cases, but existing methods either depend on latent space analysis, which is vulnerable to obfuscation, or are specific to particular architectures and modalities. We reframe MAD as a functional attribution problem: asking to what extent samples from a trusted set can explain the model's output, where attribution failure signals anomalous behavior. We operationalize this using influence functions, measuring functional coupling between test samples and a small reference set via parameter-space sampling. We evaluate across multiple anomaly types and modalities. For backdoors in vision models, our method achieves state-of-the-art detection on BackdoorBench, with an average Defense Effectiveness Rating (DER) of 0.93 across seven attacks and four datasets (next best 0.83). For LLMs, we similarly achieve a significant improvement over baselines for several backdoor types, including on explicitly obfuscated models. Beyond backdoors, our method can detect adversarial and out-of-distribution samples, and distinguishes multiple anomalous mechanisms within a single model. Our results establish functional attribution as an effective, modality-agnostic tool for detecting anomalous behavior in deployed models.

---

### CXRMate-2: Structured Multimodal Temporal Embeddings and Tractable Reinforcement Learning for Clinically Acceptable Chest X-ray Radiology Report Generation

Authors: Aaron Nicolson, Elizabeth J. Cooper, Hwan-Jin Yoon, Claire McCafferty, Ramya Krishnan, Michelle Craigie, Nivene Saad, Jason Dowling, Ian A. Scott, Bevan Koopman
Published: 2026-04-21
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.18967v2

Chest X-ray (CXR) radiology report generation (RRG) models have shown rapid progress on automated metrics, yet their clinical utility remains uncertain due to limited qualitative evaluation by radiologists. We present CXRMate-2, a state-of-the-art CXR RRG model that enables tractable reinforcement learning (RL) through structured multimodal temporal embeddings and high-resolution visual feature compression, for efficient, unified conditioning of an LLM decoder on visual, textual, and temporal context from a study and its prior. This enables group relative policy optimisation (GRPO), where a proposed reward function is used to improve semantic alignment with radiologist reports. Across the MIMIC-CXR, CheXpert Plus, and ReXgradient datasets, CXRMate-2 achieves statistically significant improvements over strong benchmarks, including gains of 11.2% and 24.4% in GREEN and RadGraph-XL, respectively, on MIMIC-CXR relative to MedGemma 1.5 (4B). To directly compare CXRMate-2 against radiologist reporting, we conduct a blinded, randomised qualitative retrospective evaluation. Three consultant radiologists compare generated and radiologist reports across 120 studies from the MIMIC-CXR test set. Generated reports were deemed acceptable (defined as preferred or rated equally to radiologist reports) in 45% of ratings, with no statistically significant difference in preference rates for seven of the eight analysed findings. Preferences for radiologist reports were driven primarily by higher recall, while generated reports were consistently preferred for readability. Together, these results define a clear pathway to clinically acceptable CXR RRG. Improving recall and the detection of subtle findings represents the primary remaining barrier to non-inferiority with radiologist reporting, positioning CXR RRG for prospective evaluation in assistive, radiologist-led workflows.

---

### Self-Improving Tabular Language Models via Iterative Reward-Guided Post-Training

Authors: Yunbo Long, Tejumade Afonja, Guangya Hao, Alexandra Brintrup, Mario Fritz
Published: 2026-04-21
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2604.18966v2

Tabular language models can generate synthetic tables by modeling rows as token sequences, but they are typically trained once with supervised fine-tuning and then used as static synthesizers. This is limiting because next-token likelihood does not directly optimize the distributional, utility, and indistinguishability properties used to evaluate synthetic data. We study iterative reward-guided post-training for tabular language models through a generate--score--align protocol, where a generator samples synthetic rows, a task-specified reward ranks them, and the model is updated relative to a fixed supervised reference. Within this protocol, we propose \textbf{TabGRAA} (\textbf{Tab}ular \textbf{G}roup-\textbf{R}elative \textbf{A}dvantage \textbf{A}lignment), a group-relative alignment method that compares high- and low-reward generated groups using group-averaged policy/reference log-ratios rather than one-to-one preference pairs. Across five mixed-type benchmarks, TabGRAA improves a GReaT backbone beyond additional supervised fine-tuning and achieves the strongest average trade-off among adapted DPO, KTO, and NPO baselines on fidelity and downstream utility, while maintaining empirical privacy diagnostics near the supervised baseline. Ablations show that the gains depend on meaningful reward ranking and stable group-level updates rather than extra training alone. Reward-substitution and scorer-separation studies further show that the post-training loop can use both classifier-based and classifier-free rewards, and that proper scorer separation is important for preserving the fidelity--utility--privacy trade-off. These results position TabGRAA as a self-improving post-training method for tabular language-model generators, complementary to strong static tabular synthesizers.

---

### DW-Bench: Benchmarking LLMs on Data Warehouse Graph Topology Reasoning

Authors: Ahmed G. A. H Ahmed, C. Okan Sakar
Published: 2026-04-21
Categories: cs.AI, cs.DB
arXiv: https://arxiv.org/abs/2604.18964v1

This paper introduces DW-Bench, a new benchmark that evaluates large language models (LLMs) on graph-topology reasoning over data warehouse schemas, explicitly integrating both foreign-key (FK) and data-lineage edges. The benchmark comprises 1,046 automatically generated, verifiably correct questions across five schemas. Experiments show that tool-augmented methods substantially outperform static approaches but plateau on hard compositional subtypes.

---

### Distillation Traps and Guards: A Calibration Knob for LLM Distillability

Authors: Weixiao Zhan, Yongcheng Jing, Leszek Rutkowski, Dacheng Tao
Published: 2026-04-21
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2604.18963v1

Knowledge distillation (KD) transfers capabilities from large language models (LLMs) to smaller students, yet it can fail unpredictably and also underpins model leakage risks. Our analysis revealed several distillation traps: tail noise, off-policy instability, and, most fundamentally, the teacher-student gap, that distort training signals. These traps manifest as overconfident hallucinations, self-correction collapse, and local decoding degradation, causing distillation to fail. Motivated by these findings, we propose a post-hoc calibration method that, to the best of our knowledge, for the first time enables control over a teacher's distillability via reinforcement fine-tuning (RFT). Our objective combines task utility, KL anchor, and across-tokenizer calibration reward. This makes distillability a practical safety lever for foundation models, connecting robust teacher-student transfer with deployment-aware model protection. Experiments across math, knowledge QA, and instruction-following tasks show that students distilled from distillable calibrated teachers outperform SFT and KD baselines, while undistillable calibrated teachers retain their task performance but cause distilled students to collapse, offering a practical knob for both better KD and model IP protection.

---

### AI-Enabled Image-Based Hybrid Vision/Force Control of Tendon-Driven Aerial Continuum Manipulators

Authors: Shayan Sepahvand, Farrokh Janabi-Sharifi, Farhad Aghili
Published: 2026-04-21
Categories: cs.RO, cs.CV
arXiv: https://arxiv.org/abs/2604.18961v1

This paper presents an AI-enabled cascaded hybrid vision/force control framework for tendon-driven aerial continuum manipulators based on constant-strain modeling in $SE(3)$ as a coupled system. The proposed controller is designed to enable autonomous, physical interaction with a static environment while stabilizing the image feature error. The developed strategy combines the cascaded fast fixed-time sliding mode control and a radial basis function neural network to cope with the uncertainties in the image acquired by the eye-in-hand monocular camera and the measurements from the force sensing apparatus. This ensures rapid, online learning of the vision- and force-related uncertainties without requiring offline training. Furthermore, the features are extracted via a state-of-the-art graph neural network architecture employed by a visual servoing framework using line features, rather than relying on heuristic geometric line extractors, to concurrently contribute to tracking the desired normal interaction force during contact and regulating the image feature error. A comparative study benchmarks the proposed controller against established rigid-arm aerial manipulation methods, evaluating robustness across diverse scenarios and feature extraction strategies. The simulation and experimental results showcase the effectiveness of the proposed methodology under various initial conditions and demonstrate robust performance in executing manipulation tasks.

---

### Bridging Foundation Models and ASTM Metallurgical Standards for Automated Grain Size Estimation from Microscopy Images

Authors: Abdul Mueez, Shruti Vyas
Published: 2026-04-21
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.18957v1

Extracting standardized metallurgical metrics from microscopy images remains challenging due to complex grain morphology and the data demands of supervised segmentation. To bridge foundational computer vision with practical metallurgical evaluation, we propose an automated pipeline for dense instance segmentation and grain size estimation that adapts Cellpose-SAM to microstructures and integrates its topology-aware gradient tracking with an ASTM E112 Jeffries planimetric module. We systematically benchmark this pipeline against a classical convolutional network (U-Net), an adaptive-prompting vision foundation model (MatSAM) and a contemporary vision-language model (Qwen2.5-VL-7B). Our evaluations reveal that while the out-of-the-box vision-language model struggles with the localized spatial reasoning required for dense microscopic counting and MatSAM suffers from over-segmentation despite its domain-specific prompt generation, our adapted pipeline successfully maintains topological separation. Furthermore, experiments across progressively reduced training splits demonstrate exceptional few-shot scalability; utilizing only two training samples, the proposed system predicts the ASTM grain size number (G) with a mean absolute percentage error (MAPE) as low as 1.50%, while robustness testing across varying target grain counts empirically validates the ASTM 50-grain sampling minimum. These results highlight the efficacy of application-level foundation model integration for highly accurate, automated materials characterization. Our project repository is available at https://github.com/mueez-overflow/ASTM-Grain-Size-Estimator.

---

### Assessing Capabilities of Large Language Models in Social Media Analytics: A Multi-task Quest

Authors: Ramtin Davoudi, Kartik Thakkar, Nazanin Donyapour, Tyler Derr, Hamid Karimi
Published: 2026-04-21
Categories: cs.CL, cs.AI, cs.SI
arXiv: https://arxiv.org/abs/2604.18955v1

In this study, we present the first comprehensive evaluation of modern LLMs - including GPT-4, GPT-4o, GPT-3.5-Turbo, Gemini 1.5 Pro, DeepSeek-V3, Llama 3.2, and BERT - across three core social media analytics tasks on a Twitter (X) dataset: (I) Social Media Authorship Verification, (II) Social Media Post Generation, and (III) User Attribute Inference. For the authorship verification, we introduce a systematic sampling framework over diverse user and post selection strategies and evaluate generalization on newly collected tweets from January 2024 onward to mitigate "seen-data" bias. For post generation, we assess the ability of LLMs to produce authentic, user-like content using comprehensive evaluation metrics. Bridging Tasks I and II, we conduct a user study to measure real users' perceptions of LLM-generated posts conditioned on their own writing. For attribute inference, we annotate occupations and interests using two standardized taxonomies (IAB Tech Lab 2023 and 2018 U.S. SOC) and benchmark LLMs against existing baselines. Overall, our unified evaluation provides new insights and establishes reproducible benchmarks for LLM-driven social media analytics. The code and data are provided in the supplementary material and will also be made publicly available upon publication.

---

### FlowForge: A Staged Local Rollout Engine for Flow-Field Prediction

Authors: Xiaowen Zhang, Ziming Zhou, Fengnian Zhao, David L. S. Hung
Published: 2026-04-21
Categories: cs.LG
arXiv: https://arxiv.org/abs/2604.18953v2

Deep learning surrogates for CFD flow-field prediction often rely on large, complex models, which can be slow and fragile when data are noisy or incomplete. We introduce FlowForge, a staged local rollout engine that predicts future flow fields by compiling a locality-preserving update schedule and executing it with a shared lightweight local predictor. Rather than producing the next frame in a single global pass, FlowForge rewrites spatial sites stage by stage so that each update conditions only on bounded local context exposed by earlier stages. This compile-execute design aligns inference with short-range physical dependence, keeps latency predictable, and limits error amplification from global mixing. Across PDEBench, CFDBench, and BubbleML, FlowForge matches or improves upon strong baselines in pointwise accuracy, delivers consistently better robustness to noise and missing observations, and maintains stable multi-step rollout behavior while reducing per-step latency.

---

### Superficial Success vs. Internal Breakdown: An Empirical Study of Generalization in Adaptive Multi-Agent Systems

Authors: Namyoung So, Seokgyu Jang, Taeuk Kim
Published: 2026-04-21
Categories: cs.MA, cs.CL
arXiv: https://arxiv.org/abs/2604.18951v2

Adaptive multi-agent systems (MAS) are increasingly adopted to tackle complex problems. However, the narrow task coverage of their optimization raises the question of whether they can function as general-purpose systems. To address this gap, we conduct an extensive empirical study of adaptive MAS, revealing two key findings: (1) topological overfitting -- they fail to generalize across different domains; and (2) illusory coordination -- they achieve reasonable surface-level accuracy while the underlying agent interactions diverge from ideal MAS behavior, raising concerns about their practical utility. These findings highlight the pressing need to prioritize generalization in MAS development and motivate evaluation protocols that extend beyond simple final-answer correctness.

---

### Reasoning Structure Matters for Safety Alignment of Reasoning Models

Authors: Yeonjun In, Wonjoong Kim, Sangwu Park, Chanyoung Park
Published: 2026-04-21
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.18946v1

Large reasoning models (LRMs) achieve strong performance on complex reasoning tasks but often generate harmful responses to malicious user queries. This paper investigates the underlying cause of these safety risks and shows that the issue lies in the reasoning structure itself. Based on this insight, we claim that effective safety alignment can be achieved by altering the reasoning structure. We propose AltTrain, a simple yet effective post training method that explicitly alters the reasoning structure of LRMs. AltTrain is both practical and generalizable, requiring no complex reinforcement learning (RL) training or reward design, only supervised finetuning (SFT) with a lightweight 1K training examples. Experiments across LRM backbones and model sizes demonstrate strong safety alignment, along with robust generalization across reasoning, QA, summarization, and multilingual setting.

---

### A Mechanism and Optimization Study on the Impact of Information Density on User-Generated Content Named Entity Recognition

Authors: Jiang Xiaobo, Dinghong Lai, Song Qiu, Yadong Deng, Xinkai Zhan
Published: 2026-04-21
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.18944v1

Named Entity Recognition (NER) models trained on clean, high-resource corpora exhibit catastrophic performance collapse when deployed on noisy, sparse User-Generated Content (UGC), such as social media. Prior research has predominantly focused on point-wise symptom remediation -- employing customized fine-tuning to address issues like neologisms, alias drift, non-standard orthography, long-tail entities, and class imbalance. However, these improvements often fail to generalize because they overlook the structural sparsity inherent in UGC. This study reveals that surface-level noise symptoms share a unified root cause: low Information Density (ID). Through hierarchical confounding-controlled resampling experiments (specifically controlling for entity rarity and annotation consistency), this paper identifies ID as an independent key factor. We introduce Attention Spectrum Analysis (ASA) to quantify how reduced ID causally leads to ``attention blunting,'' ultimately degrading NER performance. Informed by these mechanistic insights, we propose the Window-Aware Optimization Module (WOM), an LLM-empowered, model-agnostic framework. WOM identifies information-sparse regions and utilizes selective back-translation to directionally enhance semantic density without altering model architecture. Deployed atop mainstream architectures on standard UGC datasets (WNUT2017, Twitter-NER, WNUT2016), WOM yields up to 4.5\% absolute F1 improvement, demonstrating robustness and achieving new state-of-the-art (SOTA) results on WNUT2017.

---

### Personalized Benchmarking: Evaluating LLMs by Individual Preferences

Authors: Cristina Garbacea, Heran Wang, Chenhao Tan
Published: 2026-04-21
Categories: cs.AI, cs.CL, cs.HC, cs.IR, cs.LG
arXiv: https://arxiv.org/abs/2604.18943v1

With the rise in capabilities of large language models (LLMs) and their deployment in real-world tasks, evaluating LLM alignment with human preferences has become an important challenge. Current benchmarks average preferences across all users to compute aggregate ratings, overlooking individual user preferences when establishing model rankings. Since users have varying preferences in different contexts, we call for personalized LLM benchmarks that rank models according to individual needs. We compute personalized model rankings using ELO ratings and Bradley-Terry coefficients for 115 active Chatbot Arena users and analyze how user query characteristics (topics and writing style) relate to LLM ranking variations. We demonstrate that individual rankings of LLM models diverge dramatically from aggregate LLM rankings, with Bradley-Terry correlations averaging only $ρ= 0.04$ (57\% of users show near-zero or negative correlation) and ELO ratings showing moderate correlation ($ρ= 0.43$). Through topic modeling and style analysis, we find users exhibit substantial heterogeneity in topical interests and communication styles, influencing their model preferences. We further show that a compact combination of topic and style features provides a useful feature space for predicting user-specific model rankings. Our results provide strong quantitative evidence that aggregate benchmarks fail to capture individual preferences for most users, and highlight the importance of developing personalized benchmarks that rank LLM models according to individual user preferences.

---

### Disparities In Negation Understanding Across Languages In Vision-Language Models

Authors: Charikleia Moraitaki, Sarah Pan, Skyler Pulling, Gwendolyn Flusche, Kumail Alhamoud, Marzyeh Ghassemi
Published: 2026-04-21
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.18942v1

Vision-language models (VLMs) exhibit affirmation bias: a systematic tendency to select positive captions ("X is present") even when the correct description contains negation ("no X"). While prior work has documented this failure mode in English and proposed solutions, negation manifests differently across languages through varying morphology, word order, and cliticization patterns, raising the question of whether these solutions serve all linguistic communities equitably. We introduce the first human-verified multilingual negation benchmark, spanning seven typologically diverse languages: English, Mandarin Chinese, Arabic, Greek, Russian, Tagalog, and Spanish. Evaluating three VLMs - CLIP, SigLIP, and MultiCLIP - we find that standard CLIP performs at or below chance on non-Latin-script languages, while MultiCLIP achieves the highest and most uniform accuracy. We also evaluate SpaceVLM, a proposed negation correction, and find that it produces substantial improvements for several languages - particularly English, Greek, Spanish, and Tagalog - while showing varied effectiveness across typologically different languages. This variation reveals that linguistic properties like morphology, script, and negation structure interact with model improvements in fairness-relevant ways. As VLMs are deployed globally, multilingual benchmarks are essential for understanding not just whether solutions work, but for whom.

---

### Localization-Guided Foreground Augmentation in Autonomous Driving

Authors: Jiawei Yong, Deyuan Qu, Qi Chen, Kentaro Oguchi, Shintaro Fukushima
Published: 2026-04-21
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2604.18940v1

Autonomous driving systems often degrade under adverse visibility conditions-such as rain, nighttime, or snow-where online scene geometry (e.g., lane dividers, road boundaries, and pedestrian crossings) becomes sparse or fragmented. While high-definition (HD) maps can provide missing structural context, they are costly to construct and maintain at scale. We propose Localization-Guided Foreground Augmentation (LG-FA), a lightweight and plug-and-play inference module that enhances foreground perception by enriching geometric context online. LG-FA: (i) incrementally constructs a sparse global vector layer from per-frame Bird's-Eye View (BEV) predictions; (ii) estimates ego pose via class-constrained geometric alignment, jointly improving localization and completing missing local topology; and (iii) reprojects the augmented foreground into a unified global frame to improve per-frame predictions. Experiments on challenging nuScenes sequences demonstrate that LG-FA improves the geometric completeness and temporal stability of BEV representations, reduces localization error, and produces globally consistent lane and topology reconstructions. The module can be seamlessly integrated into existing BEV-based perception systems without backbone modification. By providing a reliable geometric context prior, LG-FA enhances temporal consistency and supplies stable structural support for downstream modules such as tracking and decision-making.

---

### TabEmb: Joint Semantic-Structure Embedding for Table Annotation

Authors: Ehsan Hoseinzade, Ke Wang, Anandharaju Durai Raju
Published: 2026-04-21
Categories: cs.LG
arXiv: https://arxiv.org/abs/2604.18939v1

Table annotation is crucial for making web and enterprise tables usable in downstream NLP applications. Unlike textual data where learning semantically rich token or sentence embeddings often suffice, tables are structured combinations of columns wherein useful representations must jointly capture column's semantics and the inter-column relationships. Existing models learn by linearizing the 2D table into a 1D token sequence and encoding it with pretrained language models (PLMs) such as BERT. However, this leads to limited semantic quality and weaker generalization to unseen or rare values compared to modern LLMs, and degraded structural modeling due to 2D-to-1D flattening and context-length constraints. We propose TabEmb, which directly targets these limitations by decoupling semantic encoding from structural modeling. An LLM first produces semantically rich embeddings for each column, and a graph-based module over columns then injects relationships into the embeddings, yielding joint semantic-tructural representations for table annotation. Experiments show that TabEmb consistently outperforms strong baselines on different table annotation tasks. Source code and datasets are available at https://github.com/hoseinzadeehsan/TabEmb

---

### Fine-Tuning Small Reasoning Models for Quantum Field Theory

Authors: Nathaniel S. Woodward, Zhiqi Gao, Yurii Kvasiuk, Kendrick M. Smith, Frederic Sala, Moritz Münchmeyer
Published: 2026-04-21
Categories: cs.LG, cs.AI, hep-ph, hep-th
arXiv: https://arxiv.org/abs/2604.18936v1

Despite the growing application of Large Language Models (LLMs) to theoretical physics, there is little academic exploration into how domain-specific physics reasoning ability develops while training these models. To investigate this, we perform the first academic fine-tuning study of small (7B-parameter) reasoning models dedicated specifically to theoretical physics. Because open-source verifiable training data required to train such capabilities is scarce, we developed a robust data generation pipeline that can both create synthetic problems and make existing human-authored problems suitable for model training. Selecting Quantum Field Theory (QFT) as our primary domain, we generated over 2,500 synthetic problems alongside a curated collection of human-adapted problems sourced from arXiv and standard pedagogical resources. We conduct both Reinforcement Learning (RL) and Supervised Fine-Tuning (SFT) experiments, benchmarking performance gains as well as generalization to other physics domains. We perform an extensive analysis of model chains-of-though before and after fine-tuning, to understand how reasoning errors evolve during RL and SFT. Finally, we publicly release our data pipeline, verifiable QFT training data, and $\sim$200M tokens of QFT reasoning traces.

---

### AutomationBench

Authors: Daniel Shepard, Robin Salimans
Published: 2026-04-21
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.18934v1

Existing AI benchmarks for software automation rarely combine cross-application coordination, autonomous API discovery, and policy adherence. Real business workflows demand all three: a single task may span a CRM, inbox, calendar, and messaging platform - requiring the agent to find the right endpoints, follow a policy document, and write correct data to each system. To address this gap, we introduce AutomationBench, a benchmark for evaluating AI agents on cross-application workflow orchestration via REST APIs. Drawing on real workflow patterns from Zapier's platform, tasks span Sales, Marketing, Operations, Support, Finance, and HR domains. Agents must discover relevant endpoints themselves, follow layered business rules, and navigate environments with irrelevant and sometimes misleading records. Grading is programmatic and end-state only: whether the correct data ended up in the right systems. Even the best frontier models currently score below 10%. AutomationBench provides a challenging, realistic measure of where current models stand relative to the agentic capabilities businesses actually need.

---

### Gated Memory Policy: In-Context Memorization and Adaptation

Authors: Yihuai Gao, Jeff Jinyun Liu, Shuang Li, Shuran Song
Published: 2026-04-21
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2604.18933v2

Robotic manipulation tasks exhibit varying memory requirements, ranging from Markovian tasks that require no memory to non-Markovian tasks that demand in-context memorization of historical information within a single trial or in-context adaptation based on the outcomes of multiple past trials. Surprisingly, simply extending observation histories of a visuomotor policy often leads to a significant performance drop due to distribution shift and overfitting. To address these issues, we propose Gated Memory Policy (GMP), a visuomotor policy that learns both when to recall memory and what to recall. To learn when to recall memory, GMP employs a learned memory gate mechanism that selectively activates history context only when necessary, improving robustness and reactivity. To learn what to recall efficiently, GMP introduces a lightweight cross-attention module that constructs effective latent memory representations. To further enhance robustness, GMP injects diffusion noise into historical actions, mitigating sensitivity to noisy or inaccurate histories during both training and inference. On our proposed non-Markovian benchmark MemMimic, GMP achieves a 30.1% average success rate improvement over long-history baselines, while maintaining competitive performance on Markovian tasks in RoboMimic. All code, data and in-the-wild deployment instructions are available on our project website https://gated-memory-policy.github.io/.

---

### Tadabur: A Large-Scale Quran Audio Dataset

Authors: Faisal Alherran
Published: 2026-04-21
Categories: cs.SD, cs.AI
arXiv: https://arxiv.org/abs/2604.18932v1

Despite growing interest in Quranic data research, existing Quran datasets remain limited in both scale and diversity. To address this gap, we present Tadabur, a large-scale Quran audio dataset. Tadabur comprises more than 1400+ hours of recitation audio from over 600 distinct reciters, providing substantial variation in recitation styles, vocal characteristics, and recording conditions. This diversity makes Tadabur a comprehensive and representative resource for Quranic speech research and analysis. By significantly expanding both the total duration and variability of available Quran data, Tadabur aims to support future research and facilitate the development of standardized Quranic speech benchmarks.

---

### Towards a New Grammar of Reasoning for Artificial Legal Intelligence and the Mecelle as Its Semantic Protocol

Authors: Ali Goksu, F. Gozde Kardes, Mustafa Yaylali
Published: 2026-04-20
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2608.04011v1

This article examines the enduring epistemic and methodological crisis of traditional legal practice in light of the opportunities and constraints introduced by artificial intelligence. It proposes an ontologically grounded framework termed the Mecellem semantic protocol as a response to this crisis. The analysis focuses on the structural tension within law between maintaining normative coherence and adapting to evolving social and institutional conditions, and shows why approaches based solely on codification, positivist systematization, or quantitative methods such as jurimetrics are insufficient. The article argues that legal reasoning cannot be reduced to data retrieval or statistical pattern recognition. Instead, it is grounded in the premise that meaning is context-dependent and must be dynamically reconstructed through ontologically defined entity categories and differentiated layers of knowledge. Within this perspective, Mecellem reconceptualizes law not as a fixed system of rules, but as an ontodynamic architecture structured along the interconnected axes of ontology, epistemology, and methodology. The transition from jurimetrics to a semantic protocol is presented not merely as a technical shift, but as a transformation in the foundations of legal knowledge. The article further argues that neurosymbolic systems, knowledge graphs, and agentic artificial intelligence can effectively address persistent legal challenges only when embedded within such an ontodynamic framework. By understanding law as a domain of ongoing formation rather than a completed rational totality, Mecellem advances a context-sensitive, auditable, and coherent model for legal meaning production at both human and machine levels, offering a comprehensive framework for rethinking legal reasoning in the age of artificial intelligence.

---

### Topological Void Analysis A Mathematical Framework for Systematic Technical Innovation Discovery in Knowledge Spaces

Authors: Kris Pan
Published: 2026-04-20
Categories: cs.IR, cs.AI
arXiv: https://arxiv.org/abs/2607.00005v1

Identifying where to innovate in a dense technical domain - such as operating systems or hardware/software co-design - is fundamentally a search problem in a high-dimensional knowledge space. Existing approaches rely on keyword search, citation proximity, or human intuition, none of which formalise the notion of an unexplored region that is simultaneously relevant to a target goal and absent from prior art. We present Topological Void Analysis (TVA), a mathematical framework that defines topological voids as triads (A, B, C) in a dense-sparse hybrid embedding space. A void requires three conditions: (i) both concepts A and B are semantically cohesive with domain anchor C; (ii) their pairwise similarity falls within a calibrated marginality band - avoiding both obvious combinations and unrelated noise; and (iii) they share a sparse lexical bridge while the geodesic midpoint on the embedding hypersphere is unoccupied. Applied to ~140k indexed documents, TVA generates 2,128 invention candidates across 96 targets; 90% survive automated quality filtering, yielding 191 REVISE and 1 APPROVE verdict from four-specialist adversarial review (0.05% end-to-end). Two case studies demonstrate the framework surfaces non-obvious connective tissue rather than merely obvious related pairs.

---

### Why Advanced Encoders Lag on Sparse Retrieval? The Answer and an Approach to Bridging Vocabulary Gaps

Authors: Zhichao Geng, Yang Yang
Published: 2026-04-20
Categories: cs.IR, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2607.00004v1

While advanced foundation models like ModernBERT significantly outperform older architectures in dense retrieval, they surprisingly lag behind the aging BERT-base baseline in learned sparse retrieval (LSR). We identify the root cause as the \textit{Vocabulary Gap}: modern tokenizers utilize raw, case-sensitive vocabularies designed for lossless reconstruction, which map single semantic units to redundant surface forms, wasting model capacity on morphological noise and hindering lexical matching. We formalize this intuition through a theoretical framework, demonstrating that appropriate vocabulary coarse-graining can tighten the generalization bounds by reducing complexity of the hypothesis class, provided that semantic integrity is preserved. To resolve this, we propose \textbf{Vocabulary Transfer (VT)}, a model-agnostic framework that migrates advanced encoders to sparse-friendly, normalized vocabularies with minimal computational cost. VT utilizes a novel \textbf{Semantic Initialization} via spatial topology to preserve geometric structure and an \textbf{Activation Potential Calibration (APC)} mechanism to align pre-trained manifolds with sparsity constraints, preventing the dead neuron and dense collapse observed in standard fine-tuning. Empirically, VT is universally effective: it enables ModernBERT to achieve state-of-the-art performance on the BEIR benchmark (\textbf{52.4} nDCG, a \textbf{+4.7} improvement), resuscitates failing models like RoBERTa-large, and generalizes seamlessly to inference-free architectures and specialized domains. These results confirm that the performance lag is not an architectural deficiency but a solvable vocabulary mismatch. We've released our code and models.\footnote{https://anonymous.4open.science/r/vocab-transfer/. All details included.}

---

### AI-Native Network Controller: A Modular Framework for Safe Agentic Control of Multi-Domain Network Infrastructure

Authors: Merim Dzaferagic
Published: 2026-04-20
Categories: cs.NI, cs.AI
arXiv: https://arxiv.org/abs/2606.20565v1

The convergence of multiple network domains, including radio access, optical transport, and core networks, under unified intelligent control is a fundamental requirement for future 6G systems. This is important because existing network controllers remain largely domain-specific, such as the O-RAN RIC for radio, or they lack native support for AI-driven automation across heterogeneous infrastructure. As a result, safe and coordinated agentic control of multi-domain networks is still an open challenge. In this paper, we present the AI-Native Network Controller (AI-NNC), an open-source and modular framework that enables agentic AI control across diverse network domains. The framework is designed around a protocol-agnostic architecture in which each physical device is integrated through a lightweight Python adapter, while control logic is implemented through domain-specific control applications. Beyond closed-loop control, the framework also supports dataset collection, agentic AI experimentation, and coordinated testbed operation using the same validated control and measurement interfaces. This design enables a safer paradigm for autonomous network management, where AI agents operate through validated applications rather than issuing commands directly to network equipment.

---

### Correct When Paired, Wrong When Split: Decoupling and Editing Modality-Specific Neurons in MLLMs

Authors: Tingchao Fu, Wenkai Wang, Fanxiao Li, Huadong Zhang, Jinhong Zhang, Dayang Li, Yunyun Dong, Renyang Liu, Wei Zhou
Published: 2026-04-20
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2606.17057v1

Although Knowledge Editing provides an efficient mechanism for updating the knowledge of Multimodal Large Language Models (MLLMs), we find that current paradigms still suffer from an important yet remain underexplored issue : editing decoupling failure, where entity-related knowledge can be updated when the model is triggered by multimodal inputs (text--image query pairs), however, it often reverts to outdated pre-edit facts when the paired inputs are split into unimodal ones. Our in-depth empirical analysis reveals that the entity knowledge in MLLMs is not stored as a unified representation, but is instead distributed across disentangled modality-specific pathways. As a result, updates biased toward multimodal queries fail to propagate effectively to unimodal circuits. To bridge this gap, we propose DECODE, which explicitly disentangles and localizes modality-specific neuron groups for targeted knowledge. Extensive experiments demonstrate that DECODE consistently achieves effective knowledge updates under different modality triggers, thereby mitigating editing decoupling failures.

---

### Poster: EdgeCitadel -- Hybrid NATS-MQTT Orchestration for Edge Multi-Agent Systems

Authors: Zhonghao Zhan, Yefan Zhang, Hamed Haddadi
Published: 2026-04-20
Categories: cs.DC, cs.AI
arXiv: https://arxiv.org/abs/2606.14710v1

Edge-resident AI agents increasingly span home servers, IoT hubs, laptops, and phones, yet their coordination stacks still assume cloud-style transports or a central relay. We present EdgeCitadel, an edge multi-agent orchestration platform built around a single NATS 2.10 server with the built-in MQTT adapter. The design combines MQTT connectivity for heterogeneous agents, JetStream-backed persistence and replay for backend services, direct peer delegation over a shared subject namespace, and a passive aggregator that visualizes and stores traffic without sitting on the delivery path. Our poster highlights the migration from MQTT relay prototypes (common in IoT communication) to the current hybrid architecture and demonstrates a working cross-device testbed spanning ARM64, x64, and Android clients.

---

### MA-DLE: Speech-based Automatic Depression Level Estimation via Memory Augmentation

Authors: Xuzhi Wang, Xinran Wu, Ziping Zhao, Jianhua Tao, Björn W. Schuller
Published: 2026-04-20
Categories: eess.AS, cs.AI, cs.CL, cs.SD
arXiv: https://arxiv.org/abs/2606.11197v1

Speech-based automatic estimation of depression levels is essential for enabling early detection and timely intervention, particularly in resource-constrained mental health settings. In recent years, deep learning has demonstrated impressive success across various domains, including affective computing and mental health assessment. Most existing approaches rely on RNN-based architectures (such as LSTM and GRU) to model temporal information for depression estimation. However, the extracted features often emphasize only a few adjacent speech segments, limiting their ability to capture long-range dependencies. To overcome this limitation, we introduce a memory-based feature augmentation method that enhances the representational capacity of GRU-extracted features. Rather than indiscriminately incorporating historical data, our memory bank is designed to selectively integrate two types of components in order to reduce redundancy and irrelevance: (1) historical temporal features that closely resemble the current GRU output, offering complementary contextual information; and (2) dynamic memory features identified based on feature variability, which capture behavioral and emotional fluctuations indicative of depressive symptoms. To effectively fuse the memory-augmented features with GRU outputs, we further design a Hierarchical Attention Fusion (HAF) module. Our method is evaluated on the widely used DAIC-WOZ and E-DAIC datasets, achieving state-of-the-art performance.

---

### PoQ-Judge: A Multi-Architecture Evaluation Framework for Cost-Aware Proof-of-Quality in Decentralized LLM Inference

Authors: Arther Tian, Alex Ding, Frank Chen, Simon Wu, Aaron Chan
Published: 2026-04-20
Categories: cs.CL, cs.AI, cs.CR, cs.LG
arXiv: https://arxiv.org/abs/2606.11196v1

Decentralized LLM inference networks need lightweight, reference-free quality evaluation for Proof of Quality (PoQ). We present PoQ-Judge, a framework that trains dedicated judge models to score query-output pairs without ground-truth references. We study three architectures across the quality-cost tradeoff: a TextCNN judge, a MiniLM cross-encoder, and a DeBERTa judge. Using two-stage training on UltraFeedback plus GPT-labeled in-domain data, the best model reaches 0.747 Pearson correlation with the ground-truth proxy on a held-out test set, outperforming reference-based evaluators from prior work. As a reference-free component in composite scoring, it achieves 0.645 Pearson correlation, matching the best single reference-based evaluator while removing the need for reference answers. We also show that online calibration identifies semantic quality as the dominant dimension and that cascade evaluation reduces cost by 72.7 percent with only modest quality loss. Results are much stronger on QA than summarization, pointing to proxy quality as the main remaining limitation.

---

### CollabSkill: Evaluating Human-Agent Collaboration On Real-World Tasks

Authors: Yijia Shao, Zora Zhiruo Wang, Neel Ahuja, Yicheng Wang, Bowen Liu, Diyi Yang
Published: 2026-04-20
Categories: cs.HC, cs.AI, cs.CY
arXiv: https://arxiv.org/abs/2606.09833v2

AI agents are reshaping the workspace, leading to drastic change of how humans work. Despite the considerable potential of human-agent collaboration both in preserving human agency and generating economic value, this paradigm remains largely absent from occupational task evaluation, hindered by the difficulty of gathering real human data and accounting for inter-human variability. We introduce CollabSkill, a framework for evaluating human-agent collaboration on real-world occupational tasks. CollabSkill pairs real human workers with AI agents on tasks matched to their occupational background, collecting data that capture the complexity of economically valuable tasks and the usage patterns of real workers. To account for inter-human variability, CollabSkill employs a Bayesian skill rating system to disentangle and quantify the skill contributions of both humans and AI agents. Drawing on over 1,500 prompts from 386 working sessions contributed by 93 human workers, our analysis yields insights on two fronts: on the agent side, rankings on CollabSkill diverge meaningfully from those of existing fully autonomous benchmarks where Codex leads, with Claude Code ranking first; on the human side, CollabSkill reveals that practical experience emerges as the primary driver of collaboration skill, with hands-on collaboration meaningfully shifting workers' AI literacy. Together, we hope CollabSkill enables the community to invest in systematic evaluation of human-agent collaboration and spurs development efforts aimed at building AI agents that genuinely augment human workers.

---

### BEACON: Behavioral Entropy Aggregation for Cross-Model Hallucination Detection in Large Language Models

Authors: Naveen Bera, Pulijala Sai Nikhila, Kondaguduru Abhiram, Shaik Gayaz Ali, Shoaib Sadiq Salehmohamed, Shaik Mohammed Omar, Jinal Prashant Thakkar, Hansika Aredla, Shalmali Ayachit
Published: 2026-04-20
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2606.07528v1

Hallucination in large language models (LLMs), defined as the generation of factually incorrect or unsupported content, remains a critical barrier to reliable deployment. We present BEACON (Behavioral Entropy Aggregation for Cross-model hallucination detectiON), a black-box hallucination detection framework that operates purely on model outputs without requiring access to internal representations or external knowledge bases. BEACON extracts a 31-dimensional feature vector from structured multi-pass generation, integrating NLI-based semantic entropy, embedding geometry, chain-of-thought consistency, and paraphrase stability signals. A gradient-boosted classifier trained on 7,617 labeled examples across seven benchmarks achieves 0.8123 +/- 0.0102 AUROC (95% CI: 0.7632-0.8251), outperforming standalone semantic entropy (+0.2298) and SelfCheckGPT-style consistency baselines (+0.2457). Feature importance analysis shows that hallucination is inherently multi-dimensional, requiring combined uncertainty signals. An efficient 5-call variant achieves 0.7795 AUROC, enabling practical deployment across black-box LLM APIs.

---

### Post-training is (Massive) Supervised Learning

Authors: Michael Hassid, Yossi Adi, Roy Schwartz
Published: 2026-04-20
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2606.07527v1

The prevailing paradigm for training LLMs has evolved to rely on a massive post-training phase consisting of SFT and RL. In this position paper, we argue that this methodology effectively marks a reversion to the ``pre-train then fine-tune'' approach of the BERT era, explicitly tailoring models to the desired behaviors and specific benchmarks on which they are evaluated. We begin with a historical overview of LLMs, describing the different phases of the LLM evolution. We argue that the current landscape is remarkably similar to the early days of LLMs, where task performance heavily relied on fitting the models to in-distribution datasets. To empirically demonstrate this, we compare pre-trained models to randomly initialized ones, by fine-tuning both variants on modern reasoning datasets and evaluating them on competitive math and code benchmarks. We show that models post-trained from scratch yield highly non-trivial performance. Our findings suggest that current post-training methodologies function primarily as a distribution-fitting mechanism. We finish by positing that developing generally capable models and systems requires moving beyond extensive post-training for predefined behaviors, shifting instead toward training procedures where models ``learn how to learn''.

---

### GraphLoRA: Structure-Aware Low-Rank Adaptation for Large Language Model Recommendation

Authors: Lin Mu, Guoji Wang, Li Ni, Lei Sang, Zhize Wu, Peiquan Jin, Yiwen Zhang
Published: 2026-04-20
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2606.07526v1

Large Language Models (LLMs) have shown strong potential for recommendation (LLMRec) due to their powerful reasoning and generalization abilities. However, effectively aligning the textual semantics modeled by LLMs with the collaborative signals remains a key challenge. Existing methods either translate collaborative information into textual prompts or inject pre-trained embeddings into the LLM, both of which treat structural information as static input and fail to capture high-order relational dependencies. To bridge this gap, we propose GraphLoRA, a novel framework that generalizes low-rank adaptation from independent to structure-aware propagation. GraphLoRA embeds a trainable graph message-passing network within the low-rank adaptation pathway, enabling structural signals to propagate through the parameter space. This design allows collaborative topology to explicitly guide parameter updates, fostering deep integration between graph-structured and textual semantic information. Extensive experiments on multiple benchmarks demonstrate that GraphLoRA not only outperforms state-of-the-art LLM-based recommendation methods but also achieves superior generalization, effectively balancing structural reasoning capability with computational efficiency. Code is available at \href{https://github.com/wgj15965/GraphLoRA}{https://github.com/wgj15965/GraphLoRA}.

---

### Implicit Causal Graph Construction in Text via Chain Discovery

Authors: Liesbeth Allein, Marie-Francine Moens
Published: 2026-04-20
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2606.07525v1

Causal graphs in text are typically populated by observable, predefined events. In contrast, we study implicit causal graph construction from text by treating each described cause-effect pair as the begin- and endpoint of an underlying latent causal graph and using large language models (LLMs) to infer intermediate causal events. We compare end-to-end graph construction with methods that frame the task as causal chain discovery. In the latter, graphs are built either by aggregating inferred chains or by progressively expanding partial chains through an iterative search process. We further explore Wisdom of the Crowd extensions that access causal knowledge from multiple LLMs in post-hoc aggregation and collaborative inference settings. We analyze trade-offs among these approaches and evaluate the validity of inferred causal relations using a manually curated database of 1,560 scientifically validated causal pairs. This database-based evaluation is proposed as reliable, resource-efficient, and transferable to settings where ground-truth graphs are unavailable.

---

### ABLE: Representing and Mapping LLMs via Attribution-Based Large-model Embedding

Authors: Zirui Wang, Yusen Hou, Shaofeng Liang, Bowen Tian, Yanlin Zhang, Wenshuo Chen, Yutao Yue
Published: 2026-04-20
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2606.07524v2

The explosive growth of large language models (LLMs) has created a heterogeneous and poorly documented ecosystem, making systematic model comparison increasingly important for provenance auditing, security analysis, and model selection. Existing representation methods struggle to address this setting efficiently. Approaches analyzing internal parameters are powerful when architectures are compatible, but face scalability barriers under structural heterogeneity, while methods relying on external outputs may conflate models with similar behaviors and are difficult to align in richer output spaces across different tokenizers. To bridge this gap, we propose ABLE (Attribution-Based Large-model Embedding), a framework that leverages the interpretability space to construct model representations. By aggregating gradient-based feature attributions via a tokenizer-agnostic word-level alignment, ABLE captures model-specific input-sensitivity patterns rather than only surface-level outputs. Beyond empirical utility, we provide a stability analysis showing that, under standard regularity assumptions for differentiable Transformer-style models, ABLE induces a Lipschitz-continuous parameter-to-embedding map with finite-sample convergence guarantees. Extensive experiments on 239 open-source LLMs demonstrate that our training-free approach achieves competitive or superior performance in relation prediction, model routing, and benchmark score prediction.

---

### ABLE: Representing and Mapping LLMs via Attribution-Based Large-model Embedding

Authors: Zirui Wang, Yusen Hou, Shaofeng Liang, Bowen Tian, Yanlin Zhang, Wenshuo Chen, Yutao Yue
Published: 2026-04-20
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2606.07524v1

The explosive growth of large language models (LLMs) has created a heterogeneous and poorly documented ecosystem, making systematic model comparison increasingly important for provenance auditing, security analysis, and model selection. Existing representation methods struggle to address this setting efficiently. Approaches analyzing internal parameters are powerful when architectures are compatible, but face scalability barriers under structural heterogeneity, while methods relying on external outputs may conflate models with similar behaviors and are difficult to align in richer output spaces across different tokenizers. To bridge this gap, we propose ABLE (Attribution-Based Large-model Embedding), a framework that leverages the interpretability space to construct model representations. By aggregating gradient-based feature attributions via a tokenizer-agnostic word-level alignment, ABLE captures model-specific input-sensitivity patterns rather than only surface-level outputs. Beyond empirical utility, we provide a stability analysis showing that, under standard regularity assumptions for differentiable Transformer-style models, ABLE induces a Lipschitz-continuous parameter-to-embedding map with finite-sample convergence guarantees. Extensive experiments on 239 open-source LLMs demonstrate that our training-free approach achieves competitive or superior performance in relation prediction, model routing, and benchmark score prediction.

---

### Retrieval Augmented Generation Framework for the Nepali Legal Domain Question Answering

Authors: Samir Wagle, Abiral Adhikari, Reewaj Khanal, Batsal Bhandari, Prashant Manandhar, Praveen Acharya, Bal Krishna Bal
Published: 2026-04-20
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2606.07523v1

Legal domains in high-resource languages like English have widely adopted artificial intelligence for legal question answering. However, data scarcity in low resource languages such as Nepali has limited the training of large language models on Nepali legal texts. This study presents the first application of a Retrieval Augmented Generation based model for Nepali legal question answering using case laws extracted from the Nepal Kanun Patrika digital archive. Using BM25 on chunked documents, the approach achieved a top precision at one of 91 percent, and up to 75 percent with the multilingual E5 large model. Evaluation of generated answers showed 74 percent groundedness, 85 percent truthfulness according to an automated judge model, and 84 percent human evaluated truthfulness when using BM25 document retrieval, with a 92 percent successful answer generation rate. These results demonstrate that the RAG pipeline can effectively address the gap in legal question answering for low resource languages and provide a foundation for reliable AI systems in the Nepali legal domain.

---

### Informing AI Policy Assessment using Large-Scale Simulation of Interventions

Authors: Julia Barnett, Kimon Kieslich, Natali Helberger, Nicholas Diakopoulos
Published: 2026-04-20
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2605.27395v2

As the rapid proliferation of AI systems and harms spurs efforts in AI governance around the world, prioritizing among competing policy options has become increasingly challenging for policymakers and researchers. We introduce a methodology for identifying viable policy options to mitigate specified AI harms, helping policymakers and researchers target areas that warrant greater time and resource investment. This method combines participatory evaluation of policies, expert assessment of implementation costs, and an LLM-based assessment of perceived harm mitigation under each policy option. We leverage a genetic algorithm-based simulation study to explore a vast solution space of potential policy combinations, and examine how outcomes change under different weightings of cost, participatory input, and harm mitigation. We find that this method enables exploration of different balances between participatory and expert components, allowing policymakers and researchers to assess how much weight to assign to each. We argue that the diversity of viable policy combinations found by the genetic algorithm could be a useful starting point for deliberation. This method operationalizes existing work on participatory AI by integrating it directly into practical policy development pipelines.

---

### Edge AI Deployment Beyond Models: A BSP-Aware Systems Framework for Industrial Embedded Platforms

Authors: Pitchai Muthu M
Published: 2026-04-20
Categories: cs.DC, cs.AI
arXiv: https://arxiv.org/abs/2605.26119v1

Industrial Edge AI programs often begin with the model and only later confront the platform. That sequencing is attractive because it allows early demonstrations, but it breaks down when the deployment target is an embedded system with long product lifecycles, vendor-specific kernels, heterogeneous accelerators, safety constraints, and nontrivial I/O paths. In that environment, a model is only one component of a larger execution chain that begins at the sensor, traverses the board support package (BSP), and ends in a production service loop. This paper argues that robust Edge AI deployment must be treated as a systems problem rather than a late-stage application packaging exercise. The paper presents a BSP-aware framework for industrial embedded platforms organized around five layers: hardware, BSP/operating-system adaptation, runtime and acceleration, application/inference, and operations/validation. The discussion is grounded in vendor architecture documentation for Android, NXP i.MX, NVIDIA Jetson, ONNX Runtime, and TensorRT, and in systems literature on embedded AI benchmarking, device instability, and heterogeneous edge fleets. The result is a practical framework that connects low-level platform work to measurable deployment outcomes such as reproducibility, diagnosability, sustained throughput, and field reliability.

---

### Catching The Correct Answer Trap: Characterising AI Tutor Blind Spots When Analysing Student Reasoning

Authors: Moiz Imran, Sahan Bulathwela
Published: 2026-04-20
Categories: cs.CY, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2605.23925v1

Intelligent tutoring systems increasingly provide automated feedback on student work, but robust feedback requires assessing reasoning, not only final answers. We study a failure mode we call the correct answer trap (CAT): models under-detect misconceptions when students reach a correct answer via flawed reasoning. Analysing real student responses from the Eedi mathematics platform, we show that 71% of these failures concentrate in just two question types, both sharing a common structure where flawed reasoning happens to produce the correct numerical answer. Comparing a fine-tuned T5 with a frontier large language model, we find that improved capabilities reduce but do not eliminate the problem (84% vs 57% detection accuracy). Even the best-performing model generates roughly four false alarms for every genuine detection, making stand-alone screening impractical at realistic class sizes. Our findings demonstrate that high overall accuracy can mask critical failures in reasoning assessment, and that careful analysis of student reasoning still benefits from human judgment.

---

### Improving the Completeness and Comparability of Segment Disclosures: A Large Language Model Approach

Authors: Yue Liu, Zhiyuan Cheng, Longying Lai
Published: 2026-04-20
Categories: cs.CL, cs.IR, q-fin.GN
arXiv: https://arxiv.org/abs/2605.23924v1

Segment-level disclosures are a central component of financial reporting, providing insight into firms' internal organization and the allocation of economic activities across operating units. However, segment information is often presented in both qualitative and quantitative forms, dispersed across tables and narrative sections of Form 10-K filings. Empirical research relying on structured databases faces both completeness and comparability challenges, as some firm-year observations may be missing, nested segment disclosures are not captured, and support for longitudinal and cross-firm comparability is limited. This study develops a large language model-based framework to extract segment disclosures directly from Form 10-K filings and to preserve both reportable and nested segment information. We further design a retrieval augmented system that incorporates information across multiple filings to support comparability. We use two representative settings to demonstrate its application: longitudinal analysis within a firm to interpret segment changes over time, and cross firm alignment of geographic segments across firms with different reporting structures. The results indicate that the artifact accurately extracts segment-level information and effectively addresses questions that require cross-period knowledge, demonstrating the potential of LLM-based approaches to enhance the measurement and interpretation of segment disclosures.

---

### Leveraging Vision-Language Models to Detect Attention in Educational Videos

Authors: Gabriel Becquet, Sébastien Lallé, Vanda Luengo, Ali Abou-Hassan
Published: 2026-04-20
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2605.20211v1

Educational videos are a cornerstone of remote and blended learning. However, learners' fluctuating attention remains a significant barrier to effective information retention. Prior research has attempted to mitigate this by detecting and reacting to attention loss at runtime using eye tracking. Such detection has been based so far on classical machine learning classifiers trained on engineered features, such as summary statistics over learners' fixations and saccades. These methods have struggled to capture the complex, temporal nature of learner engagement, thus exhibiting moderate prediction performance. In this study, we aim to advance the detection of attention by shifting from standard engineered features to a multimodal foundation models. Using an educational eye-tracking dataset (N = 70), we investigate a novel methodology that utilizes a Vision-Language Model (VLM) to analyze video content directly with superimposed gaze data. This approach aims to leverage the semantic reasoning capabilities of foundation models to contextualize learner focus within the video stream. We evaluate the performance of this VLM-based approach using several prompting strategies with Gemini 3, but ultimately found that none of them could outperform statistical baselines. Our results provide new insights into the limitations of using VLMs for real-time educational diagnostics.

---

### Reducing Credit Assignment Variance via Counterfactual Reasoning Paths

Authors: Fei Ding, Yongkang Zhang, Youwei Wang, Zijian Zeng
Published: 2026-04-20
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2605.16302v2

Reinforcement learning for multi-step reasoning with large language models (LLMs) typically relies on sparse terminal rewards, which creates a poorly conditioned credit-assignment problem: the final feedback is propagated uniformly across all intermediate decisions. This leads to high gradient variance, unstable training, and many ineffective updates, ultimately limiting sustained model improvement. We propose a counterfactual-comparison framework for credit assignment. For each input, the framework samples multiple reasoning trajectories and treats their differences as implicit approximations to alternative decisions. This yields an implicit process-level advantage estimator that converts sparse terminal rewards into step-sensitive learning signals. Building on this framework, we introduce Implicit Behavior Policy Optimization (IBPO), which substantially improves training stability and the performance ceiling on mathematical and code-reasoning benchmarks. Our results point to a promising direction for unlocking the reasoning potential of LLMs.

---

### SDOF: Taming the Alignment Tax in Multi-Agent Orchestration with State-Constrained Dispatch

Authors: Zhantao Wang
Published: 2026-04-20
Categories: cs.AI
arXiv: https://arxiv.org/abs/2605.15204v1

Multi-agent orchestration frameworks such as LangChain, LangGraph, and CrewAI route tasks through graph-based pipelines but do not enforce the stage constraints that govern real business processes. We present SDOF, a framework that treats multi-agent execution as a constrained state machine. SDOF operates through two primary defensive layers, implemented by three components: (1) an Online-RLHF Specialized Intent Router trained via Generative Reward Modeling (GRPO) and (2) a StateAwareDispatcher with GoalStage finite-automaton checks and precondition/postcondition SkillRegistry validation for auditable execution control. On a recruitment system backed by the Beisen iTalent platform (6000+ enterprises), 185 expert-curated scenarios trigger 1671 live API calls. Our GSPO-aligned 7B Intent Router achieves higher joint accuracy than zero-shot GPT-4o on this FSM-constrained adversarial routing benchmark (80.9% versus 48.9%). In end-to-end execution, SDOF reaches 86.5% task completion (95% confidence interval 80.8 to 90.7) and blocks all 22 operations in the injection, illegal HR subset. Under a broader message-level blocking audit, SDOF attains precision 100% and recall 88%, expert agreement kappa=0.94. A separate evaluation on 960 SGD-derived dialogues spanning 8 service domains surfaces 201 stage-order conflicts under our FSM mapping, 41 of which arise in the normal split. This arXiv version reports the current validated scope; extended multi-seed training comparisons and deeper workflow evaluations will be released in a subsequent update.

---

### Earth System Foundation Model (ESFM): A unified framework for heterogeneous data integration and forecasting

Authors: Firat Ozdemir, Yun Cheng, Salman Mohebi, Fanny Lehmann, Simon Adamov, Zhenyi Zhang, Leonardo Trentini, Dana Grund, Oliver Fuhrer, Torsten Hoefler, Siddhartha Mishra, Sebastian Schemm et al.
Published: 2026-04-20
Categories: physics.ao-ph, cs.AI, cs.LG, eess.IV
arXiv: https://arxiv.org/abs/2605.00850v1

Foundation models (FMs) for the Earth system learn statistical relationships between physical variables across massive datasets to enable versatile downstream applications through finetuning, separating them from task-specific weather models. Here, we introduce Earth System Foundation Model (ESFM), a fully open model building on the 3D Swin UNet backbone of the pioneering Aurora model. ESFM introduces extensions that increase functionality and foster adoption in climate sciences. First, the encoding scheme and training protocols have been extended to handle diverse datasets, including those containing missing values across all spatio-temporal dimensions such as satellite data, as well as station data, all under one backbone. Axial attention is introduced to capture inter-variable dependencies. As a result ESFM skillfully predicts variables in regions or on pressure levels where no data is present at the initial time, while preserving inter-variable relationships, for example between temperature, pressure, and humidity. Individual variable tokenization enables different sets of variables to be shuffled during training and simplifies the process of building extensions for new downstream tasks. Adaptive layer norm-based ensembles allow for a simple yet effective way to transform deterministic ESFM to a probabilistic FM. We present findings using dense gridded data (ERA5, CMIP6), regionally masked dense data, sparse gridded MODIS satellite data, and station data. Results demonstrate competitive or superior performance relative to state-of-the-art benchmarks. Case studies of Super Typhoon Doksuri (2023) and 2024 sudden stratospheric warming events show accurate positional and magnitude estimations of extreme weather. ESFM retains the strengths of previous foundation models, such as long-term stability, but facilitates application to a variety of downstream tasks.

---

### Deep Learning for Multi-Antenna Modulation Recognition of Radio Signals

Authors: Tao Chen, Shilian Zheng, Jiepeng Chen, Zhangbin Pei, Qi Xuan, Xiaoniu Yang
Published: 2026-04-20
Categories: eess.SP, cs.LG, eess.SY
arXiv: https://arxiv.org/abs/2605.00849v1

Multi-antenna receiving systems have become a prevalent technical solution in communication systems. Meanwhile, deep learning has achieved significant progress in automatic modulation recognition tasks in single-antenna systems. However, the application of deep learning in multi-antenna modulation recognition (MAMR) tasks is still limited. In this paper, we propose an MAMR method namely MAMR-IQ to fully explore the diversity gain of a multi-antenna receiving system, which concatenates the raw received in-phase and quadrature (IQ) signals of multiple antennas and feeds them into a convolutional neural network. Simulation results show that the proposed MAMR-IQ method outperforms two existing deep learning-based MAMR methods which are based on direct voting (DV) and weight average (WA) in terms of both recognition accuracy and computational complexity. To address the problem of limited training data in few-shot scenarios, we further propose a data augmentation method that involves exchanging IQ sequences received by any two antennas to generate augmented samples. Simulation results show that with the proposed augmentation method, the recognition accuracy can be further improved.

---

### Putting HUMANS first: Efficient LAM Evaluation with Human Preference Alignment

Authors: Woody Haosheng Gan, William Held, Diyi Yang
Published: 2026-04-20
Categories: cs.CL, cs.AI, cs.SD
arXiv: https://arxiv.org/abs/2605.00022v2

The rapid proliferation of large audio models (LAMs) demands efficient approaches for model comparison, yet comprehensive benchmarks are costly. To fill this gap, we investigate whether minimal subsets can reliably evaluate LAMs while reducing costs and data redundancy. Analyzing 10 subset selection methods with 18 audio models across 40 tasks covering major LAM evaluation dimensions, we show that subsets of just 50 examples (0.3% of data) can achieve over 0.93 Pearson correlation with full benchmark scores. To understand how well these scores align with what practitioners ultimately care about, user satisfaction, we collect 776 human preference ratings from realistic voice assistant conversations, finding that both subsets and full benchmark achieve only 0.85 correlation with human. To better predict preferences, we trained regression models on these selected subsets, achieving 0.98 correlation -- outperforming regression models trained on both random subsets and the full benchmark. This demonstrates that in regression modeling, well-curated subsets outpredict the full benchmark, showing quality over quantity. We open-source these regression-weighted subsets as the HUMANS benchmark, an efficient proxy for LAM evaluation that captures both benchmark performance and user preferences.

---

### Putting HUMANS first: Efficient LAM Evaluation with Human Preference Alignment

Authors: Woody Haosheng Gan, William Held, Diyi Yang
Published: 2026-04-20
Categories: cs.CL, cs.AI, cs.SD
arXiv: https://arxiv.org/abs/2605.00022v1

The rapid proliferation of large audio models (LAMs) demands efficient approaches for model comparison, yet comprehensive benchmarks are costly. To fill this gap, we investigate whether minimal subsets can reliably evaluate LAMs while reducing costs and data redundancy. Analyzing 10 subset selection methods with 18 audio models across 40 tasks covering major LAM evaluation dimensions, we show that subsets of just 50 examples (0.3% of data) can achieve over 0.93 Pearson correlation with full benchmark scores. To understand how well these scores align with what practitioners ultimately care about, user satisfaction, we collect 776 human preference ratings from realistic voice assistant conversations, finding that both subsets and full benchmark achieve only 0.85 correlation with human. To better predict preferences, we trained regression models on these selected subsets, achieving 0.98 correlation -- outperforming regression models trained on both random subsets and the full benchmark. This demonstrates that in regression modeling, well-curated subsets outpredict the full benchmark, showing quality over quantity. We open-source these regression-weighted subsets as the HUMANS benchmark, an efficient proxy for LAM evaluation that captures both benchmark performance and user preferences.

---

### OAMVOS:2nd Report for 5th PVUW MOSE Track

Authors: Deshui Miao, Xingsen Huang, Yameng Gu, Xiaogang yu, Xin Li, Ming-Hsuan Yang
Published: 2026-04-20
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2604.22837v1

SAM-based dense trackers provide strong short-term mask propagation but remain fragile under long occlusion, fast motion, viewpoint change, and distractors. The problem is especially severe for small objects, where a few incorrect memory updates can dominate later predictions. This report presents an occlusion- and reappearance-aware extension of DAM4SAM that improves memory control rather than changing the backbone. The method augments the original SAM3 tracker with four ingredients: a reliability-aware tracking state machine, branch-based recovery, delayed DRM promotion, and a selective policy for native SAM3 memory selection. During stable tracking, the model follows the original single-path propagation process. Once confidence drops, the tracker enters an ambiguous or recovery mode, maintains a small set of candidate branches, and commits memory only after a branch is reconfirmed. For small-object disappearance and reappearance, native memory selection is temporarily bypassed so older anchors remain accessible. In addition, the first conditioning frame is explicitly preserved, and the conditioning-memory budget is moderately enlarged to improve long-gap recovery. The resulting design keeps DAM4SAM efficient in easy cases while improving robustness in sequences dominated by occlusion and reappearance.

---

### AgentRVOS for MeViS-Text Track of 5th PVUW Challenge: 3rd Method

Authors: Deshui Miao, Chao Yang, Chao Tian, Guoqing Zhu, Kai Yang, Zhifan Mo, Xin Li
Published: 2026-04-20
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.22836v1

This report describes a Ref-VOS pipeline centered on Sa2VA and organized with explicit agent roles. The key idea is that Sa2VA should provide the first dense semantic hypothesis, while an agent loop decides whether that hypothesis should be accepted, revised, or refined. The pipeline starts with a target-presence judgment stage. If the referred object does not exist in the video, the system directly outputs zero masks. Otherwise, Sa2VA receives the video and referring prompt and produces a coarse mask trajectory over the full video. This trajectory is treated as a semantic prior rather than a final answer. A planner agent decomposes the query, temporal partition agents identify informative blocks, scout agents search for anchor frames, and refinement agents convert reliable Sa2VA masks into boxes and points for SAM3 propagation. A critic scores candidate trajectories, a reflection controller repairs weak hypotheses, and a collaboration controller reconciles multiple agent branches. The result is a Ref-VOS system in which Sa2VA is responsible for dense grounded understanding, while the agent layer handles presence verification, temporal search, confidence-aware revision, and final mask refinement.

---

### ParkingScenes: A Structured Dataset for End-to-End Autonomous Parking in Simulation Scenes

Authors: Haonan Chen, Kaiwen Xiao, Bin Tian, Jun Fu
Published: 2026-04-20
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2604.22835v1

Autonomous parking remains a critical yet challenging task in intelligent driving systems, particularly within constrained urban environments where maneuvering space is limited and precise control is essential. While recent advances in end-to-end learning have shown great promise, the lack of high-quality, structured datasets tailored for parking scenarios remains a significant bottleneck.To address this gap, we present ParkingScenes, a comprehensive multimodal dataset specifically designed for end-to-end autonomous parking in simulated scenes. Built on the CARLA simulator, ParkingScenes features structured parking trajectories generated by a Hybrid A* planner and a Model Predictive Controller (MPC), providing accurate and reproducible supervision signals. The dataset includes 16 reverse-in and 6 parallel parking scenarios, each executed under two pedestrian conditions (present and absent), resulting in 704 structured episodes and approximately 105000 frames. Each scenario is repeated 16 times to ensure consistent coverage. Each frame contains synchronized data from four RGB cameras, four depth sensors, vehicle motion states, and Bird's-Eye View (BEV) representations, enabling rich multimodal fusion and context-aware learning. To demonstrate the utility of our dataset, we compare models trained on ParkingScenes with those trained on unstructured, manually collected simulation data under identical conditions. Results show significant improvements in performance, underscoring the effectiveness of structured supervision for robust and accurate parking policy learning. By releasing both the dataset and the collection framework, ParkingScenes establishes a scalable and reproducible benchmark for advancing learning-based autonomous parking systems. The dataset and collection framework will be released at: https://github.com/haonan-ai/ParkingScenes

---