# Research Papers — 2026-04 · bundle page 112 of 233

> Full-text markdown of papers #5551–5600 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/112
[← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/111.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-04/page/113.md)

### NaP-Control: Navigating Diffusion Prior for Versatile and Fast Character Control

Authors: Chia-Wen Chen, Yan Wu, Korrawe Karunratanakul, Siyu Tang
Published: 2026-04-15
Categories: cs.GR, cs.LG, cs.RO
arXiv: https://arxiv.org/abs/2605.20209v2

Achieving precise, versatile whole-body character control in physics-based animation remains challenging. Recent diffusion-based policies generate rich and expressive motions but typically rely on gradient-based test-time guidance to satisfy task objectives, which is slow and can reduce robustness. We introduce NaP-Control (Navigating Diffusion Prior for Versatile and Fast Character Control), abbreviated as NaP. Our method uses reinforcement learning to manipulate the latent noise of a task-agnostic diffusion policy prior, steering it toward task-specific behaviors for fast, robust control with high motion fidelity. In contrast to methods that rely solely on offline training, NaP interacts with the environment during training to correct motions and optimize task rewards, improving success rates and enabling adaptation to challenging scenarios. By directly predicting task-optimized diffusion noise, NaP eliminates iterative guidance during denoising and enables efficient inference. Experiments show that NaP attains higher success rates and faster inference while preserving natural motion across diverse tasks.

---

### Agentic Graph Retrieval-Augmented Generation for Auditable Commercial Registry Analysis

Authors: Arthur Capozzi, Dirk Helbing
Published: 2026-04-15
Categories: cs.IR, cs.AI
arXiv: https://arxiv.org/abs/2605.18770v2

Public commercial registries are formally open, yet their practical analysis remains difficult because relevant facts are scattered across millions of records that combine structured metadata, multilingual legal notices, temporal events, and entity aliases. This paper presents a controlled, tool-mediated agentic GraphRAG architecture for auditable natural-language analysis of such registries. The proposed pipeline transforms publications from the Swiss Official Gazette of Commerce into a Neo4j knowledge graph comprising over five million nodes and 4.7 million relationships. It combines deterministic ingestion of structured registry fields, LLM-assisted extraction of latent actors from unstructured notices, and a deterministic identity-resolution layer. An analytical agent operates on this graph through intent routing, restricted graph tools, bounded reflection, and state-machine-guided response synthesis. We evaluate the system using a multi-tier protocol covering answer quality, retrieval behavior, entity resolution, and multi-turn conversational performance. The complete architecture is compared with dense, lexical, and hybrid flat-retrieval baselines and with controlled architectural ablations. On a manually curated benchmark, graph-mediated retrieval increases factual correctness from 0.26 for the strongest flat-retrieval baseline to 0.83 for the complete system, with comparable improvements in relevance and completeness. Ablation results show that bounded reflection improves answer quality while intent routing and LLM-based graph enrichment improve reliability in difficult entity resolution tasks. An exploratory dashboard displays the graph evidence and execution traces underlying each response, allowing users to inspect how answers were produced.

---

### ANVIL: Analogies and Videos for Lecturers

Authors: Yuri Noviello, Anastasiia Birillo, Gosia Migut
Published: 2026-04-15
Categories: cs.CY, cs.AI, cs.CL, cs.GR, cs.HC, cs.MM
arXiv: https://arxiv.org/abs/2605.16295v1

We present ANVIL, a multimodal generative system that automates the production of analogy-based instructional animations for computer science topics. Given a concept definition, ANVIL generates a textual analogy, compiles it into a structured visual screenplay, and produces executable manim code to render an animation, with an automated repair mechanism to improve robustness. Evaluating such systems at scale requires balancing pedagogical validity with scalability. We begin with a teacher evaluation to ground the quality assessment and use its findings to guide automated screening. For textual analogies, we introduce an LLM-based evaluator for scalable quality screening; for videos, where subjective judgments are difficult to automate, we instead assess fidelity to the intended screenplay using an automated proxy for auditing and error analysis. We further conduct a user study with educators to examine adoption requirements and risks. Our findings suggest that ANVIL can produce materials that are frequently rated as adequate, and that educators respond positively to its perceived value and usability.

---

### AsymmetryZero: A Framework for Operationalizing Human Expert Preferences as Semantic Evals

Authors: Tadhg Looram, Lucas Nuzzi, Kyle Waters, Steven Dillmann
Published: 2026-04-15
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2605.04083v1

Much of the focus in RL today is on evaluation design: building meaningful evals that serve simultaneously as benchmarks and as well-defined reward signals for post-training. Yet, many real-world tasks are governed by subjective, procedural, and domain-specific requirements that are difficult to encode as exact-match targets or open-ended preference judgments frequently used in RL pipelines today. In this work, we present AsymmetryZero, a framework for operationalizing human expert preferences as semantic evals. AsymmetryZero represents each task as a stable evaluation contract that makes grading criteria explicit: what is being graded, how each criterion is judged, and how criterion-level decisions are aggregated into a task outcome. The same contract can be executed using Inspect for model-only evaluations, as well as the Harbor Framework for agentic evaluations, enabling comparable scores and shared audit artifacts across both settings. We argue that the central challenge in post-training today is the faithful encoding of expert requirements into the evaluation itself. To that end, we present a study using Harbor that holds task contracts fixed and compares a five-model frontier jury against a five-model compact jury across four frontier-class solvers (Claude Opus 4.6, GPT-5.4, Grok-4.20, Gemini-3.1-Pro). We find that criterion-level frontier-vs-compact agreement ranges from $75.9\%$ to $89.6\%$ (strict common-subset agreement: $77.8\%$ to $92.1\%$), while compact juries exhibit substantially higher internal dissent (3--2 split rate $28.7\%$--$32.4\%$) than frontier juries ($6.1\%$--$11.5\%$). Verifier traces further show that compact juries reduce per-criterion judging cost to roughly $4.2\%$--$5.6\%$ of frontier and latency to roughly $21.7\%$--$27.1\%$, even as aggregated task-level outcomes often remain comparatively stable.

---

### Enhancing the interpretability of spatially variable N2O model predictions with soft sensors during wastewater treatment

Authors: Mohammad Raeisi Gahrouei, Pedram Ramin, Vincenzo A. Riggio, Carlos Domingo-Felez
Published: 2026-04-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.04082v1

Model-based solutions for nitrous oxide (N2O) emissions from wastewater treatment plants (WWTP) are informed by operational datasets designed to control nutrient levels in liquid waste, coupled with dedicated campaigns for N2O measurements. We analysed how machine learning (ML) models predict disturbances to WWT operation and spatially variable N2O emissions. A real dataset was investigated to validate the modelling framework from N2O emissions predicted by four ML models (R2 = 0.79 - 0.89). Monitoring campaigns for N2O were simulated with a plant-wide mechanistic model to include additional sensors, site-level N2O datasets, and wastewater disturbances (n = 16). ML models were highly accurate (0.97 +- 0.02, n = 80), but the feature importance depended on the model, the scenario and the N2O measurement scale (reactor vs. WWTP). We argue that N2O soft sensor model predictions are limited to the measuring location and the methodological uncertainty of the dataset, which affect the interpretability of the model. Lastly, the analysis of the mechanistic model structure exposed interactions between autotrophic and heterotrophic pathways over nitric oxide which can overestimate aerobic nitrite production and bias the N2O pathway contributions.

---

### Time series causal discovery with variable lags

Authors: Bruno Petrungaro, Anthony C. Constantinou
Published: 2026-04-15
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2605.04081v2

Causal Bayesian Networks (CBNs) are a powerful tool for reasoning under uncertainty about complex real-world problems. Such problems evolve over time, responding to external shocks as they occur. To support decision-making, CBNs require a cause-and-effect map of the variables under consideration, known as the network's structure. Learning the graphical structure of a causal model from data remains challenging; learning it from time-series data is even harder because dependencies may arise at different time lags. Existing time-series causal discovery methods often assume a fixed lag window and do not explicitly optimise edge-specific lags. We propose a Tabu-based structure learning algorithm that searches for a time-ordered directed structure (i.e., where every edge respects time) while allowing edge-specific lags up to a specified maximum lag. The approach uses a decomposable BIC-based score with node-specific effective sample sizes and an explicit lag-length penalty encouraging parsimonious delay assignments while preserving efficient local score updates. We provide theoretical guarantees of validity and local optimality, and we also describe a parallel implementation for improved scalability. In simulations, the method recovered graph structure competitively and estimated lags accurately when true adjacencies were recovered. On a real-world UK COVID-19 policy dataset, the learnt structure was dominated by short delays while retaining a substantial minority of longer-lag dependencies, consistent with delayed behavioural and epidemiological effects.

---

### H-Probes: Extracting Hierarchical Structures From Latent Representations of Language Models

Authors: Cutter Dawes, Aryan Sharma, Angelos Ioannis Lagos, Shivam Raval
Published: 2026-04-15
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2605.00847v2

Representing and navigating hierarchy is a fundamental primitive of reasoning. Large language models have demonstrated proficiency in a wide variety of tasks requiring hierarchical reasoning, but there exists limited analysis on how the models geometrically represent the necessary latent constructions for such thinking. To this end, we develop H-probes, a collection of linear probes that extract hierarchical structure, specifically depth and pairwise distance, from latent representations. In synthetic tree traversal tasks, the H-probes robustly find the subspaces containing hierarchical structure necessary to complete the tasks; furthermore, in comprehensive ablation experiments, we show that these hierarchy-containing subspaces are low-dimensional, causally important for high task performance, and generalize within- and out-of-domain. Furthermore, we find analogous, though weaker, hierarchical structure in real-world hierarchical contexts such as mathematical reasoning traces. These results demonstrate that models represent hierarchy not only at the level of syntax and concepts, but at deeper levels of abstraction -- including the reasoning process itself.

---

### SAND: Spatially Adaptive Network Depth for Fast Sampling of Neural Implicit Surfaces

Authors: Chuanxiang Yang, Junhui Hou, Yuan Liu, Siyu Ren, Guangshun Wei, Taku Komura, Yuanfeng Zhou, Wenping Wang
Published: 2026-04-15
Categories: cs.GR, cs.CV, eess.IV
arXiv: https://arxiv.org/abs/2604.25936v1

Implicit neural representations are powerful for geometric modeling, but their practical use is often limited by the high computational cost of network evaluations. We observe that implicit representations require progressively lower accuracy as query points move farther from the target surface, and that even within the same iso-surface, representation difficulty varies spatially with local geometric complexity. However, conventional neural implicit models evaluate all query points with the same network depth and computational cost, ignoring this spatial variation and thereby incurring substantial computational waste. Motivated by this observation, we propose an efficient neural implicit geometry representation framework with spatially adaptive network depth (SAND). SAND leverages a volumetric network-depth map together with a tailed multi-layer perceptron (T-MLP) to model implicit representation. The volumetric depth map records, for each spatial region, the network depth required to achieve sufficient accuracy, while the T-MLP is a modified MLP designed to learn implicit functions such as signed distance functions, where an output branch, referred to as a tail, is attached to each hidden layer. This design allows network evaluation to terminate adaptively without traversing the full network and directs computational resources to geometrically important and complex regions, improving efficiency while preserving high-fidelity representations. Extensive experimental results demonstrate that our approach can significantly improve the inference-time query speed of implicit neural representations.

---

### Elderly-Contextual Data Augmentation via Speech Synthesis for Elderly ASR

Authors: Minsik Lee, Seoi Hong, Chongmin Lee, Sieun Choi, Jian Kim, Jua Han, Jihie Kim
Published: 2026-04-15
Categories: cs.CL, cs.SD
arXiv: https://arxiv.org/abs/2604.24770v1

Despite recent progress in automatic speech recognition (ASR), elderly ASR (EASR) remains challenging due to limited training data and the distinct acoustic and linguistic characteristics of elderly speech. In this work, we address data scarcity in EASR through a data augmentation pipeline that combines large language model (LLM)-based transcript paraphrasing with text-to-speech (TTS) synthesis. Given an elderly speech dataset, the LLM first generates elderly-contextual paraphrases of the original transcripts, and the TTS model then synthesizes corresponding speech using elderly reference speakers. The resulting synthetic audio-text pairs are merged with the original data to fine-tune Whisper without architectural modification. We further analyze the effects of augmentation ratio and reference-speaker composition in low-resource EASR. Experiments on English and Korean elderly speech datasets from speakers aged 70 and above show that the proposed method consistently improves performance over conventional augmentation baselines, achieving up to a 58.2% reduction in word error rate (WER) compared with the Whisper baseline.

---

### Comparative Study of Bending Analysis using Physics-Informed Neural Networks and Numerical Dynamic Deflection in Perforated nanobeam

Authors: Ramanath Garai, Iswari Sahu, S. Chakraverty
Published: 2026-04-15
Categories: cs.LG, cs.AI, physics.comp-ph
arXiv: https://arxiv.org/abs/2604.24768v1

In this chapter, we investigate the bending behavior of a perforated nanobeam subjected to sinusoidal loading using an efficient and computationally robust Physics-Informed Functional Link Constrained Framework with Domain Mapping (DFL-TFC) method. Our aim is to determine the relationship between static bending response and dynamic deflection of a perforated nanobeam for various perforation cases. The static bending is obtained using the FL-TFC with Domain mapped method, whereas dynamic deflection is determined using the Galerkin method. The proposed approach employs the theory of functional connections (TFC) to systematically embed governing differential equation constraints into a constrained expression (CE), which exactly satisfies all prescribed initial and boundary conditions (ICs and BCs) and domain of differential equation is mapped to domain of orthogonal polynomials. Within this framework, the free function appearing in the constrained expression is expressed through a functional link neural network (FLNN). The cost is minimized by the mean square residual of DE, allowing training without requiring complex deep network architectures. Relationship between static and dynamic defection of simply-supported (S-S) perforated nanobeams has been investigated here. FL-TFC with Domain mapped method eliminates the need for deep and complex neural network architectures while ensuring accuracy, efficiency, and strict satisfaction of boundary conditions as compared to standard PINN.

---

### Automated detection of pediatric congenital heart disease from phonocardiograms using deep and handcrafted feature fusion

Authors: Abdul Jabbar, Ethan Grooby, Yang Yi Poh, Khawza I. Ahmad, Md Hassanuzzaman, Raqibul Mostafa, Ahsan H. Khandoker, Faezeh Marzbanrad
Published: 2026-04-15
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2604.24767v1

Congenital heart disease (CHD) is the most common type of birth defect, impacting about 1% of live births worldwide. Echocardiography, the gold-standard diagnostic method, is costly and inaccessible in low-resource settings. Diagnosis is delayed due to limited skilled experts, whose ability to interpret pathological patterns varies significantly, causing inter- and intra-clinician variability. Therefore, we present a new method for a more accessible diagnostic modality, the digital stethoscope, to detect CHDs. Our method is based on deep feature fusion, integrating deep and handcrafted features for the automated early detection of CHDs. For this work, Phonocardiography (PCG) recordings were obtained from 751 pediatric subjects (Age:1 month- 16 years) in Bangladesh, ranging from infants to adults at four auscultation locations: mitral valve (MV), aortic valve (AV), pulmonary valve (PV), and tricuspid valve (TV). These recordings were labeled based on confirmed diagnoses by cardiologists as either cases of CHD or non-CHD. The results demonstrated that our proposed model achieved an accuracy of 92%, a sensitivity of 91%, and a specificity of 91%, based on a patient-wise split of 70% training, 20% validation, and 10% testing. Furthermore, the Area Under the Receiver Operating Characteristic curve (AUROC) of 96%, and an F1-score of 92%. This model promises efficient real-time remote detection of CHDs as a cost-effective screening tool for low-resource settings.

---

### Model Capability Assessment and Safeguards for Biological Weaponization

Authors: Michael Richter
Published: 2026-04-15
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2604.19811v2

AI leaders and safety reports increasingly warn that advances in model reasoning may enable biological misuse, including by low-expertise users, while major labs describe safeguards as expanding but still evolving rather than settled. This study benchmarks ChatGPT 5.2 Auto, Gemini 3 Pro Thinking, Claude Opus 4.5 and Meta's Muse Spark Thinking on 73 novice-framed, open-ended benign STEM prompts to measure operational intelligence. On benign quantitative tasks, both Gemini and Meta scored very high; ChatGPT was partially useful but text-thinned, and Claude was sparsest with some apparent false-positive refusals. A second test set detected subtle harmful intent: edge case prompts revealed Gemini's seeming lack of contextual awareness. These results warranted a focused weaponization analysis on Gemini as capability appeared to be outpacing moderation calibration. Gemini was tested across four access environments and reported cases include poison-ivy-to-crowded-transit escalation, poison production and extraction via international-anonymous logged-out AI Mode, and other concerning examples. Biological misuse may become more prevalent as a geopolitical tool, increasing the urgency of U.S. policy responses, especially if model outputs come to be treated as regulated technical data. Guidance is provided for 25 high-risk agents to help distinguish legitimate use cases from higher-risk ones.

---

### The Existential Theory of Research: Why Discovery Is Hard

Authors: Angshul Majumdar
Published: 2026-04-15
Categories: cs.AI, eess.SP
arXiv: https://arxiv.org/abs/2604.19810v1

Can scientific discovery be made arbitrarily easy by choosing the right representation, collecting enough data, and deploying sufficiently powerful algorithms? This paper argues that the answer is fundamentally negative. We introduce the Existential Theory of Research (ETR), a formal framework that models discovery as the recovery of structured explanations under constraints of representation, observation, and computation. Within this framework, we show that these three components cannot be simultaneously optimized: no method can guarantee universally simple explanations, arbitrarily compressed observations, and efficient exact inference. This limitation is not model-specific, but arises from a synthesis of uncertainty principles in sparse representation, sample complexity bounds in high-dimensional recovery, and the computational hardness of exact inference. We further show that representation mismatch alone can inflate intrinsic simplicity into apparent complexity, rendering otherwise tractable problems observationally and computationally prohibitive. To quantify these effects, we introduce an uncertainty functional that captures the joint difficulty of discovery. The results suggest that scientific difficulty is not accidental, but a structural consequence of the geometry and complexity of inference.

---

### MIRROR: A Hierarchical Benchmark for Metacognitive Calibration in Large Language Models

Authors: Jason Z Wang
Published: 2026-04-15
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2604.19809v1

We introduce MIRROR, a benchmark comprising eight experiments across four metacognitive levels that evaluates whether large language models can use self-knowledge to make better decisions. We evaluate 16 models from 8 labs across approximately 250,000 evaluation instances using five independent behavioral measurement channels. Core experiments are run across the full model roster; experiments with specialized infrastructure requirements report explicitly marked model subsets. We find two phenomena with direct implications for agentic deployment: (1) compositional self-prediction fails universally -- the Compositional Calibration Error ranges from 0.500 to 0.943 on the original 15-model Exp3-v1 set (and 0.434 to 0.758 on the balanced 16-model Exp3-v2 expansion), indicating that models cannot predict their own performance on multi-domain tasks, and (2) models exhibit above-chance but imperfect domain-specific self-knowledge yet systematically fail to translate even this partial awareness into appropriate agentic action-selection -- external metacognitive control reduces the Confident Failure Rate from 0.600 to 0.143 (76% reduction at temperature 0; mean 70% at temperature 0.7 across 5 models from 4 labs). Providing models with their own calibration scores produces no significant improvement (p > 0.05); only architectural constraint is effective. This suggests that external metacognitive scaffolding -- not improved self-knowledge -- is the path to safer autonomous AI systems. Code, data, and Croissant metadata will be released publicly with the benchmark.

---

### Positive-Only Drifting Policy Optimization

Authors: Qi Zhang
Published: 2026-04-15
Categories: cs.LG, cs.RO
arXiv: https://arxiv.org/abs/2604.16519v1

In the field of online reinforcement learning (RL), traditional Gaussian policies and flow-based methods are often constrained by their unimodal expressiveness, complex gradient clipping, or stringent trust-region requirements. Moreover, they all rely on post-hoc penalization of negative samples to correct erroneous actions. This paper introduces Positive-Only Drifting Policy Optimization (PODPO), a likelihood-free and gradient-clipping-free generative approach for online RL. By leveraging the drifting model, PODPO performs policy updates via advantage-weighted local contrastive drifting. Relying solely on positive-advantage samples, it elegantly steers actions toward high-return regions while exploiting the inherent local smoothness of the generative model to enable proactive error prevention. In doing so, PODPO opens a promising new pathway for generative policy learning in online settings.

---

### On-Orbit Space AI: Federated, Multi-Agent, and Collaborative Algorithms for Satellite Constellations

Authors: Ziyang Wang
Published: 2026-04-15
Categories: cs.RO, astro-ph.IM, cs.AI
arXiv: https://arxiv.org/abs/2604.16518v1

Satellite constellations are transforming space systems from isolated spacecraft into networked, software-defined platforms capable of on-orbit perception, decision making, and adaptation. Yet much of the existing AI studies remains centered on single-satellite inference, while constellation-scale autonomy introduces fundamentally new algorithmic requirements: learning and coordination under dynamic inter-satellite connectivity, strict SWaP-C limits, radiation-induced faults, non-IID data, concept drift, and safety-critical operational constraints. This survey consolidates the emerging field of on-orbit space AI through three complementary paradigms: (i) {federated learning} for cross-satellite training, personalization, and secure aggregation; (ii) {multi-agent algorithms} for cooperative planning, resource allocation, scheduling, formation control, and collision avoidance; and (iii) {collaborative sensing and distributed inference} for multi-satellite fusion, tracking, split/early-exit inference, and cross-layer co-design with constellation networking. We provide a system-level view and a taxonomy that unifies collaboration architectures, temporal mechanisms, and trust models. To support community development and keep this review actionable over time, we continuously curate relevant papers and resources at https://github.com/ziyangwang007/AI4Space.

---

### SmoGVLM: A Small, Graph-enhanced Vision-Language Model

Authors: Debjyoti Mondal, Rituraj Singh, Subhadarshi Panda
Published: 2026-04-15
Categories: cs.CV, cs.CL
arXiv: https://arxiv.org/abs/2604.16517v1

Large vision-language models (VLMs) achieve strong performance on multimodal tasks but often suffer from hallucination and poor grounding in knowledge-intensive reasoning. We propose SmoGVLM, a small, graph-enhanced VLM that integrates structured knowledge with visual and textual modalities, using Graph Neural Networks. We investigate the effects of our method across a range of model sizes, from tiny (1.3B) to large (13B) models. Our results demonstrate that, when trained using our approach, a small model can achieve performance gains upto 16.24%, and surpass its larger counterparts, outperforming larger VLMs and strong fine-tuned baselines. These results highlight the potential of structured knowledge augmentation for efficient, smaller-scale multimodal reasoning systems.

---

### Operationalizing Fairness in Text-to-Image Models: A Survey of Bias, Fairness Audits and Mitigation Strategies

Authors: Megan Smith, Venkatesh Thirugnana Sambandham, Florian Richter, Laura Crompton, Matthias Uhl, Torsten Schön
Published: 2026-04-15
Categories: cs.CV, cs.LG, cs.MM
arXiv: https://arxiv.org/abs/2604.16516v1

Text-to-Image (T2I) generation models have been widely adopted across various industries, yet are criticized for frequently exhibiting societal stereotypes. While a growing body of research has emerged to evaluate and mitigate these biases, the field at present contends with conceptual ambiguity, for example terms like "bias" and "fairness" are not always clearly distinguished and often lack clear operational definitions. This paper provides a comprehensive systematic review of T2I fairness literature, organizing existing work into a taxonomy of bias types and fairness notions. We critically assess the gap between "target fairness" (normative ideals in T2I outputs) and "threshold fairness" (normative standards with actionable decision rules). Furthermore, we survey the landscape of mitigation strategies, ranging from prompt engineering to diffusion process manipulation. We conclude by proposing a new framework for operationalizing fairness that moves beyond descriptive metrics towards rigorous, target-based testing, offering an approach for more accountable generative AI development.

---

### Penny Wise, Pixel Foolish: Bypassing Price Constraints in Multimodal Agents via Visual Adversarial Perturbations

Authors: Jiachen Qian, Zhaolu Kang
Published: 2026-04-15
Categories: cs.CV, cs.CR, cs.LG
arXiv: https://arxiv.org/abs/2604.16515v1

The rapid proliferation of Multimodal Large Language Models (MLLMs) has enabled mobile agents to execute high-stakes financial transactions, but their adversarial robustness remains underexplored. We identify Visual Dominance Hallucination (VDH), where imperceptible visual cues can override textual price evidence in screenshot-based, price-constrained settings and lead agents to irrational decisions. We propose PriceBlind, a stealthy white-box adversarial attack framework for controlled screenshot-based evaluation. PriceBlind exploits the modality gap in CLIP-based encoders via a Semantic-Decoupling Loss that aligns the image embedding with low-cost, value-associated anchors while preserving pixel-level fidelity. On E-ShopBench, PriceBlind achieves around 80% ASR in white-box evaluation; under a simplified single-turn coordinate-selection protocol, Ensemble-DI-FGSM transfers with roughly 35-41% ASR across GPT-4o, Gemini-1.5-Pro, and Claude-3.5-Sonnet. We also show that robust encoders and Verify-then-Act defenses reduce ASR substantially, though with some clean-accuracy trade-off.

---

### BARD: Bridging AutoRegressive and Diffusion Vision-Language Models Via Highly Efficient Progressive Block Merging and Stage-Wise Distillation

Authors: Baoyou Chen, Hanchen Xia, Peng Tu, Haojun Shi, Liwei Zhang, Yuxuan Yao, Weihao Yuan, Siyu Zhu
Published: 2026-04-15
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2604.16514v5

Autoregressive vision-language models (VLMs) deliver strong multimodal capability, but their token-by-token decoding imposes a fundamental inference bottleneck. Diffusion VLMs offer a more parallel decoding paradigm, yet directly converting a pretrained autoregressive VLM into a large-block diffusion VLM (dVLM) often leads to substantial quality degradation. In this work, we present BARD, a simple and effective bridging framework that converts a pretrained autoregressive VLM into a same-architecture, decoding-efficient dVLM. Our approach combines progressive supervised block merging, which gradually enlarges the decoding block size, with stage-wise intra-dVLM distillation from a fixed small-block diffusion anchor to recover performance lost at larger blocks. We further incorporate a mixed noise scheduler to improve robustness and token revision during denoising, and memory-friendly training to enable efficient training on long multimodal sequences. A key empirical finding is that direct autoregressive-to-diffusion distillation is poorly aligned and can even hurt performance, whereas distillation within the diffusion regime is consistently effective. Experimental results show that, with $\leq$ 4.4M data, BARD-VL transfers strong multimodal capability from Qwen3-VL to a large-block dVLM. Remarkably, BARD-VL establishes a new SOTA among comparable-scale open dVLMs on our evaluation suite at both 4B and 8B scales. At the same time, BARD-VL achieves up to 3$\times$ decoding throughput speedup compared to the source model. Code is available at https://github.com/fudan-generative-vision/Bard-VL.

---

### SynthPID: P&ID digitization from Topology-Preserving Synthetic Data

Authors: Suraj Prasad, Pinak Mahapatra
Published: 2026-04-15
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2604.16513v1

Automating the digitization of Piping and Instrumentation Diagrams (P&IDs) into structured process graphs would unlock significant value in plant operations, yet progress is bottlenecked by a fundamental data problem: engineering drawings are proprietary, and the entire community shares a single public benchmark of just 12 annotated images. Prior attempts at synthetic augmentation have fallen short because template-based generators scatter symbols at random, producing graphs that bear little resemblance to real process plants and, accordingly, yield only approximately 33% edge detection accuracy under synth-only training. We argue the failure is structural rather than visual and address it by introducing SynthPID, a corpus of 665 synthetic P&IDs whose pipe topology is seeded directly from real drawings. Paired with a patch-based Relationformer adapted for high-resolution diagrams, a model trained on SynthPID alone achieves 63.8 +/- 3.1% edge mAP on PID2Graph OPEN100 without seeing a single real P&ID during training, closing within 8 pp of the real-data oracle. These gains hold up under a controlled comparison against the template-based regime, confirming that generation quality drives performance rather than model choice. A scaling study reveals that gains flatten beyond roughly 400 synthetic images, pointing to seed diversity as the binding constraint.

---

### Medial Axis Aware Learning of Signed Distance Functions

Authors: Samuel Weidemaier, Christoph Norden-Smoch, Martin Rumpf
Published: 2026-04-15
Categories: cs.CV, cs.CG, cs.GR, cs.LG, math.NA
arXiv: https://arxiv.org/abs/2604.16512v2

We propose a novel variational method to compute a highly accurate global signed distance function (SDF) to a given point cloud. To this end, the jump set of the gradient of the SDF, which coincides with the medial axis of the surface, is explicitly taken into account through a higher-order variational formulation that enforces linear growth along the gradient direction away from this discontinuity set. The eikonal equation and the zero-level set of the SDF are enforced as constraints. To make this variational problem computationally tractable, a phase field approximation of Ambrosio-Tortorelli type is employed. The associated phase field function implicitly describes the medial axis. The method is implemented for surfaces represented by unoriented point clouds using neural network approximations of both the SDF and the phase field. Experiments demonstrate the method's accuracy both in the near field and globally. Quantitative and qualitative comparisons with other approaches show the advantages of the proposed method.

---

### SQL Query Engine: A Self-Healing LLM Pipeline for Natural Language to PostgreSQL Translation

Authors: Muhammad Adeel Ijaz
Published: 2026-04-15
Categories: cs.DB, cs.CL
arXiv: https://arxiv.org/abs/2604.16511v1

We present SQL Query Engine, an open-source, self-hosted service that translates natural language questions into validated PostgreSQL queries through a two-stage LLM pipeline. The first stage performs automatic schema introspection and SQL generation; a multi-strategy response parser extracts SQL from any LLM output format (JSON, code blocks, or raw text) without requiring structured output APIs. The second stage executes the query against PostgreSQL and, upon failure or empty results, enters an iterative self-healing loop in which the LLM diagnoses the error using full SQLSTATE codes and PostgreSQL diagnostic messages. Two mechanisms prevent regressions: early-accept returns successful queries immediately without LLM re-evaluation, and best-result tracking preserves the best partial result across retries. Schema context is cached per session in Redis, progress events stream via Redis Pub/Sub and SSE, and an OpenAI-compatible /v1/chat/completions endpoint lets existing tools work without modification. All database connections are read-only at the driver level. We evaluate across five LLM backends on a synthetic benchmark (75 questions, three databases) where the self-healing loop yields up to +9.3pp accuracy gains with zero regressions on the best model (Llama 4 Scout 17B, 57.3%), and on BIRD (437 questions, 11 databases migrated from SQLite to PostgreSQL) where the full pipeline reaches 49.0% execution accuracy (GPT-OSS-120B, +4.6pp). Source code: https://github.com/codeadeel/sqlqueryengine.

---

### Learning-Guided Sparsification of Dynamic Graphs in Robotic Exploration

Authors: Adithya V. Sastry, Bibek Poudel, Weizi Li
Published: 2026-04-15
Categories: cs.RO, cs.LG
arXiv: https://arxiv.org/abs/2604.16509v2

Many robotic exploration algorithms rely on graph structures for frontier-based exploration and dynamic path planning. However, these graphs grow rapidly, accumulating redundant information and impacting performance. We present a hybrid transformer-based framework trained with Proximal Policy Optimization which complements exploration algorithms by pruning these graphs during exploration, limiting their growth and reducing the accumulation of excess information. The framework was evaluated on simulations of a robotic agent using Rapidly-Exploring Random Trees to carry out frontier-based exploration, where the learned policy reduces graph size by up to 96%. We find preliminary evidence that our framework enables effective, generalizable exploration under reduced information density, consistently outperforming the randomly pruned baseline and improving both exploration efficiency and computational efficiency beyond the unpruned baseline for complex long-horizon exploration tasks. To the best of our knowledge, these results are the first suggesting the viability of RL to prune dynamic graphs used in robotic exploration algorithms.

---

### Learning-Based Sparsification of Dynamic Graphs in Robotic Exploration Algorithms

Authors: Adithya V. Sastry, Bibek Poudel, Weizi Li
Published: 2026-04-15
Categories: cs.RO, cs.LG
arXiv: https://arxiv.org/abs/2604.16509v1

Many robotic exploration algorithms rely on graph structures for frontier-based exploration and dynamic path planning. However, these graphs grow rapidly, accumulating redundant information and impacting performance. We present a transformer-based framework trained with Proximal Policy Optimization (PPO) to prune these graphs during exploration, limiting their growth and reducing the accumulation of excess information. The framework was evaluated on simulations of a robotic agent using Rapidly Exploring Random Trees (RRT) to carry out frontier-based exploration, where the learned policy reduces graph size by up to 96%. We find preliminary evidence that our framework learns to associate pruning decisions with exploration outcomes despite sparse, delayed reward signals. We also observe that while intelligent pruning achieves a lower rate of exploration compared to baselines, it yields the lowest standard deviation, producing the most consistent exploration across varied environments. To the best of our knowledge, these results are the first suggesting the viability of RL in sparsification of dynamic graphs used in robotic exploration algorithms.

---

### Exascale Multi-Task Graph Foundation Models for Imbalanced, Multi-Fidelity Atomistic Data

Authors: Massimiliano Lupo Pasini, Jong Youl Choi, Kshitij Mehta, Richard Messerly, Rylie Weaver, Linda Ungerboeck, Isaac Lyngaas, Benajmin Stump, Ashwin M. Aji, Karl W. Schulz, Jorda Polo
Published: 2026-04-15
Categories: cond-mat.mtrl-sci, cs.AI
arXiv: https://arxiv.org/abs/2604.15380v1

We present an exascale workflow for materials discovery using atomistic graph foundation models built on HydraGNN. We jointly train on 16 open first-principles datasets (544+ million structures covering 85+ elements) using a multi-task architecture with per-dataset heads and a scalable ADIOS2/DDStore data pipeline. On Frontier, we execute six large-scale DeepHyper hyperparameter optimization campaigns in FP64 and promote the top-performing message-passing models to sustained 2,048-node training, yielding a PaiNN-based lead model. The resulting model enables billion-scale screening, evaluating 1.1 billion atomistic structures in 50 seconds, compressing a workload that would require years of first-principles computation, and supports data-scarce fine-tuning across diverse downstream tasks. We quantify precision-performance tradeoffs (BF16/FP32/FP64), demonstrate transfer across twelve chemically diverse downstream tasks, and establish seamless strong- and weak-scaling across Frontier, Aurora, and Perlmutter. This work allows fast and reliable exploration of vast chemical design spaces that are otherwise inaccessible to first-principles methods.

---

### M3R: Localized Rainfall Nowcasting with Meteorology-Informed MultiModal Attention

Authors: Sanjeev Panta, Rhett M Morvant, Xu Yuan, Li Chen, Nian-Feng Tzeng
Published: 2026-04-15
Categories: cs.LG, cs.CV, cs.MM
arXiv: https://arxiv.org/abs/2604.15377v1

Accurate and timely rainfall nowcasting is crucial for disaster mitigation and water resource management. Despite recent advances in deep learning, precipitation prediction remains challenging due to limitations in effectively leveraging diverse multimedia data sources. We introduce M3R, a Meteorology-informed MultiModal attention-based architecture for direct Rainfall prediction that synergistically combines visual NEXRAD radar imagery with numerical Personal Weather Station (PWS) measurements, using a comprehensive pipeline for temporal alignment of heterogeneous meteorological data. With specialized multimodal attention mechanisms, M3R novelly leverages weather station time series as queries to selectively attend to spatial radar features, enabling focused extraction of precipitation signatures. Experimental results for three spatial areas of 100 km * 100 km centered at NEXRAD radar stations demonstrate that M3R outperforms existing approaches, achieving substantial improvements in accuracy, efficiency, and precipitation detection capabilities. Our work establishes new benchmarks for multimedia-based precipitation nowcasting and provides practical tools for operational weather prediction systems. The source code is available at https://github.com/Sanjeev97/M3Rain

---

### Zoom Consistency: A Free Confidence Signal in Multi-Step Visual Grounding Pipelines

Authors: Keon Kim, Krish Chelikavada
Published: 2026-04-15
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2604.15376v1

Multi-step zoom-in pipelines are widely used for GUI grounding, yet the intermediate predictions they produce are typically discarded after coordinate remapping. We observe that these intermediate outputs contain a useful confidence signal for free: zoom consistency, the distance between a model's step-2 prediction and the crop center. Unlike log-probabilities or token-level uncertainty, zoom consistency is a geometric quantity in a shared coordinate space, making it directly comparable across architecturally different VLMs without calibration. We prove this quantity is a linear estimator of step-1 spatial error under idealized conditions (perfect step-2, target within crop) and show it correlates with prediction correctness across two VLMs (AUC = 0.60; Spearman rho = -0.14, p < 10^{-6} for KV-Ground-8B; rho = -0.11, p = 0.0003 for Qwen3.5-27B). The correlation is small but consistent across models, application categories, and operating systems. As a proof-of-concept, we use zoom consistency to route between a specialist and generalist model, capturing 16.5% of the oracle headroom between them (+0.8%, McNemar p = 0.19). Code is available at https://github.com/omxyz/zoom-consistency-routing.

---

### VeriCWEty: Embedding enabled Line-Level CWE Detection in Verilog

Authors: Prithwish Basu Roy, Zeng Wang, Anatolii Chuvashlov, Weihua Xiao, Johann Knechtel, Ozgur Sinanoglu, Ramesh Karri
Published: 2026-04-15
Categories: cs.AR, cs.AI, cs.CR
arXiv: https://arxiv.org/abs/2604.15375v1

Large Language Models (LLMs) have shown significant improvement in RTL code generation. Despite the advances, the generated code is often riddled with common vulnerabilities and weaknesses (CWEs) that can slip by untrained eyes. Attackers can often exploit these weaknesses to fulfill their nefarious motives. Existing RTL bug-detection techniques rely on rule-based checks, formal properties, or coarse-grained structural analysis, which either fail to capture semantic vulnerabilities or lack precise localization. In our work, we bridge this gap by proposing an embedding-based bug-detection framework that detects and classifies bugs at both module and line-level granularity. Our method achieves about 89% precision in identifying common CWEs such as CWE-1244 and CWE-1245, and 96% accuracy in detecting line-level bugs.

---

### Seeing the imagined: latent functional alignment in visual imagery decoding from fMRI data

Authors: Fabrizio Spera, Tommaso Boccato, Michal Olak, Nicola Toschi, Matteo Ferrante
Published: 2026-04-15
Categories: q-bio.NC, cs.AI, eess.IV
arXiv: https://arxiv.org/abs/2604.15374v2

Recent progress in visual brain decoding from fMRI has been enabled by large-scale datasets such as the Natural Scenes Dataset (NSD) and powerful diffusion-based generative models. While current pipelines are primarily optimized for perception, their performance under mental-imagery remains less well understood. In this work, we study how a state-of-the-art (SOTA) perception decoder (DynaDiff) can be adapted to reconstruct imagined content from the NSD-Imagery benchmark. We propose a latent functional alignment (LFA) approach that maps imagery-evoked activity to the pretrained model's semantic content-enriched conditioning space, by adding a simple alignment module, while keeping the original remaining components frozen. To mitigate the limited amount of matched imagery-perception supervision, we further introduce a neural retrieval-based augmentation strategy that selects semantically related NSD perception trials from the same participants. Across four subjects, LFA consistently improves high-level semantic reconstruction metrics relative to the frozen pretrained baseline and a voxel-space ridge alignment baseline, and enables above-chance decoding from multiple cortical regions. These results suggest that semantic structure learned from perception can be leveraged to stabilize and improve visual imagery decoding under out-of-distribution conditions.

---

### Seeing the imagined: a latent functional alignment in visual imagery decoding from fMRI data

Authors: Fabrizio Spera, Tommaso Boccato, Michal Olak, Sara Cammarota, Matteo Ciferri, Michelangelo Tronti, Nicola Toschi, Matteo Ferrante
Published: 2026-04-15
Categories: q-bio.NC, cs.AI, eess.IV
arXiv: https://arxiv.org/abs/2604.15374v1

Recent progress in visual brain decoding from fMRI has been enabled by large-scale datasets such as the Natural Scenes Dataset (NSD) and powerful diffusion-based generative models. While current pipelines are primarily optimized for perception, their performance under mental-imagery remains less well understood. In this work, we study how a state-of-the-art (SOTA) perception decoder (DynaDiff) can be adapted to reconstruct imagined content from the Imagery-NSD benchmark. We propose a latent functional alignment approach that maps imagery-evoked activity into the pretrained model's conditioning space, while keeping the remaining components frozen. To mitigate the limited amount of matched imagery-perception supervision, we further introduce a retrieval-based augmentation strategy that selects semantically related NSD perception trials. Across four subjects, latent functional alignment consistently improves high-level semantic reconstruction metrics relative to the frozen pretrained baseline and a voxel-space ridge alignment baseline, and enables above-chance decoding from multiple cortical regions. These results suggest that semantic structure learned from perception can be leveraged to stabilize and improve visual imagery decoding under out-of-distribution conditions.

---

### InfoChess: A Game of Adversarial Inference and a Laboratory for Quantifiable Information Control

Authors: Kieran A. Murphy
Published: 2026-04-15
Categories: cs.MA, cs.AI, cs.IT, cs.LG
arXiv: https://arxiv.org/abs/2604.15373v1

We propose InfoChess, a symmetric adversarial game that elevates competitive information acquisition to the primary objective. There is no piece capture, removing material incentives that would otherwise confound the role of information. Instead, pieces are used to alter visibility. Players are scored on their probabilistic inference of the opponent's king location over the duration of the game. To explore the space of strategies for playing InfoChess, we introduce a hierarchy of heuristic agents defined by increasing levels of opponent modeling, and train a reinforcement learning agent that outperforms these baselines. Leveraging the discrete structure of the game, we analyze gameplay through natural information-theoretic characterizations that include belief entropy, oracle cross entropy, and predictive log score under the action-induced observation channel. These measures disentangle epistemic uncertainty, calibration mismatch, and uncertainty induced by adversarial movement. The design of InfoChess renders it a testbed for studying multi-agent inference under partial observability. We release code for the environment and agents, and a public interface to encourage further study.

---

### The Synthetic Media Shift: Tracking the Rise, Virality, and Detectability of AI-Generated Multimodal Misinformation

Authors: Zacharias Chrysidis, Stefanos-Iordanis Papadopoulos, Symeon Papadopoulos
Published: 2026-04-15
Categories: cs.CR, cs.AI, cs.MM
arXiv: https://arxiv.org/abs/2604.15372v1

As generative AI advances, the distinction between authentic and synthetic media is increasingly blurred, challenging the integrity of online information. In this study, we present CONVEX, a large-scale dataset of multimodal misinformation involving miscaptioned, edited, and AI-generated visual content, comprising over 150K multimodal posts with associated notes and engagement metrics from X's Community Notes. We analyze how multimodal misinformation evolves in terms of virality, engagement, and consensus dynamics, with a focus on synthetic media. Our results show that while AI-generated content achieves disproportionate virality, its spread is driven primarily by passive engagement rather than active discourse. Despite slower initial reporting, AI-generated content reaches community consensus more quickly once flagged. Moreover, our evaluation of specialized detectors and vision-language models reveals a consistent decline in performance over time in distinguishing synthetic from authentic images as generative models evolve. These findings highlight the need for continuous monitoring and adaptive strategies in the rapidly evolving digital information environment.

---

### Applied Explainability for Large Language Models: A Comparative Study

Authors: Venkata Abhinandan Kancharla
Published: 2026-04-15
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2604.15371v1

Large language models (LLMs) achieve strong performance across many natural language processing tasks, yet their decision processes remain difficult to interpret. This lack of transparency creates challenges for trust, debugging, and deployment in real-world systems. This paper presents an applied comparative study of three explainability techniques: Integrated Gradients, Attention Rollout, and SHAP, on a fine-tuned DistilBERT model for SST-2 sentiment classification. Rather than proposing new methods, the focus is on evaluating the practical behavior of existing approaches under a consistent and reproducible setup. The results show that gradient-based attribution provides more stable and intuitive explanations, while attention-based methods are computationally efficient but less aligned with prediction-relevant features. Model-agnostic approaches offer flexibility but introduce higher computational cost and variability. This work highlights key trade-offs between explainability methods and emphasizes their role as diagnostic tools rather than definitive explanations. The findings provide practical insights for researchers and engineers working with transformer-based NLP systems. This is a preprint and has not undergone peer review.

---

### TopFeaRe: Locating Critical State of Adversarial Resilience for Graphs Regarding Topology-Feature Entanglement

Authors: Xinxin Fan, Wenxiong Chen, Quanliang Jing, Chi Lin, Shaoye Luo, Wenbo Song, Yunfeng Lu
Published: 2026-04-15
Categories: cs.CR, cs.LG
arXiv: https://arxiv.org/abs/2604.15370v1

Graph adversarial attacks are usually produced from the two perspectives of topology/structure and node feature, both of them represent the paramount characteristics learned by today's deep learning models. Although some defense countermeasures are proposed at present, they fails to disclose the intrinsic reasons why these two aspects necessitate and how they are adequately fused to co-learn the graph representation. Towards this question, we in this paper propose an adversarial defense approach through locating the graph's critical state of adversarial resilience, resorting to the equilibrium-point theory in the discipline of complex dynamic system (CDS). In brief, our work has three novelties: i) Adversarial-Attack Modeling, i.e. map a graph regime into CDS, and use the oscillation of dynamic system to model the behavior of adversarial perturbation; ii) 2D Topology-Feature-Entangled Function Design for Perturbed Graph, i.e. project graph topology and node feature as two characteristic spaces, and define two-dimensional entangled perturbation functions to represent the dynamic variance under adversarial attacks; and iii) Location of Critical State of Adversarial Resilience, i.e. utilize the equilibrium-point theory to locate the graph's critical state of attack resilience resorting to the perturbation-reflected 2D function. Finally, multi-facet experiments on five commonly-used realistic datasets validate the effectiveness of our proposed approach, and the results show our approach can significantly outperform the state-of-the-art baselines under four representative graph adversarial attacks.

---

### CobwebTM: Probabilistic Concept Formation for Lifelong and Hierarchical Topic Modeling

Authors: Karthik Singaravadivelan, Anant Gupta, Zekun Wang, Christopher J. MacLellan
Published: 2026-04-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.14489v2

Topic modeling seeks to uncover latent semantic structure in text corpora with minimal supervision. Neural approaches achieve strong performance but require extensive tuning and struggle with lifelong learning due to catastrophic forgetting and fixed capacity, while classical probabilistic models lack flexibility and adaptability to streaming data. We introduce CobwebTM, a low-parameter lifelong hierarchical topic model based on incremental probabilistic concept formation. By adapting the Cobweb algorithm to continuous document embeddings, CobwebTM constructs semantic hierarchies online, enabling unsupervised topic discovery, dynamic topic creation, and hierarchical organization without predefining the number of topics. Across diverse datasets, CobwebTM achieves strong topic coherence, stable topics over time, and high-quality hierarchies, demonstrating that incremental symbolic concept formation combined with pretrained representations is an efficient approach to topic modeling.

---

### Controlling Authority Retrieval: A Missing Retrieval Objective for Authority-Governed Knowledge

Authors: Andre Bacellar
Published: 2026-04-15
Categories: cs.IR, cs.CL
arXiv: https://arxiv.org/abs/2604.14488v3

In law, regulatory regimes for pharmaceuticals and software security, newer authorities can revoke older established ones even when semantically distant. We call this CAR: retrieving the currently active authority frontier for a semantic anchor q, that is, front(cl(A_k(q))). This differs from finding the most similar document by relevance score: argmax_d s(q, d). Theorem 4 characterizes when a set R truly covers the active authority set for q with TCA(R, q)=1, providing conditions necessary and sufficient for any retrieved set R: frontier inclusion (front(cl(A_k(q))) contained in R) and no-ignored-superseder (no superseding document exists in the corpus outside R). Proposition 2 shows that TCA@k <= phi(q) * R_anchor(q) in the worst case over any scope-indexed algorithm, proved by an adversarial permutation argument. We evaluated on three real-world datasets: security advisories (Dense TCA@5=0.270, two-stage 0.975), SCOTUS overruling pairs (Dense TCA=0.172, two-stage 0.926), and FDA drug records (Dense TCA=0.064, two-stage 0.774). A GPT-4o-mini experiment shows Dense RAG produces explicit "not patched" claims for 39% of queries where a patch exists; two-stage cuts this to 16%. Four benchmark datasets, domain adapters, and a single-command scorer are released at https://github.com/andremir/car-retrieval.

---

### Quantization of Spiking Neural Networks Beyond Accuracy

Authors: Evan Gibson Smith, Jacob Whitehill, Fatemeh Ganji
Published: 2026-04-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2604.14487v1

Quantization is a natural complement to the sparse, event-driven computation of Spiking Neural Networks, reducing memory bandwidth and arithmetic cost for deployment on resource-constrained hardware. However, existing SNN quantization evaluation focuses almost exclusively on accuracy, overlooking whether a quantized network preserves the firing behavior of its full-precision counterpart. We demonstrate that quantization method, clipping range, and bit-width can produce substantially different firing distributions at equivalent accuracy, differences invisible to standard metrics but relevant to deployment, where firing activity governs effective sparsity, state storage, and event-processing load. To capture this gap, we propose Earth Mover's Distance as a diagnostic metric for firing distribution divergence, and apply it systematically across weight and membrane quantization on SEW-ResNet architectures trained on CIFAR-10 and CIFAR-100. We find that uniform quantization induces distributional drift even when accuracy is preserved, while LQ-Net style learned quantization maintains firing behavior close to the full-precision baseline. Our results suggest that behavior preservation should be treated as an evaluation criterion alongside accuracy, and that EMD provides a principled tool for assessing it.

---

### Behavior Cloning Under PD Control: A Finite-Horizon Theory of Gain-Dependent Error Amplification

Authors: Junghoon Seo
Published: 2026-04-15
Categories: cs.RO, cs.AI, math.OC
arXiv: https://arxiv.org/abs/2604.14484v3

Behavior cloning (BC) on position-controlled robots is shaped by the PD loop that executes policy actions. We give a finite-horizon, nonasymptotic analysis of how controller gains affect BC failure. Independent sub-Gaussian action errors propagate through gain-dependent closed-loop dynamics into sub-Gaussian position errors. The resulting failure tail is controlled by controller amplification multiplied by validation loss and generalization slack, so validation loss alone can mis-rank gains. Under shape-preserving upper-bound assumptions, the analysis separates label difficulty, injection strength, and contraction, ranking compliant-overdamped gains as tightest and stiff-underdamped gains as loosest, with the mixed regimes system-dependent. In the canonical scalar second-order PD system, stationary position-error variance increases with stiffness and decreases with damping over the stable range, and exact zero-order-hold discretization inherits the ordering to leading order. This extends the error-attenuation explanation of bronars et al. (2026) to finite-horizon failure bounds.

---

### Seeing Through Circuits: Faithful Mechanistic Interpretability for Vision Transformers

Authors: Nina Żukowska, Wolfgang Stammer, Bernt Schiele, Jonas Fischer
Published: 2026-04-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.14477v1

Transparency of neural networks' internal reasoning is at the heart of interpretability research, adding to trust, safety, and understanding of these models. The field of mechanistic interpretability has recently focused on studying task-specific computational graphs, defined by connections (edges) between model components. Such edge-based circuits have been defined in the context of large language models, yet vision-based approaches so far only consider neuron-based circuits. These tell which information is encoded, but not how it is routed through the complex wiring of a neural network. In this work, we investigate whether useful mechanistic circuits can be identified through computational graphs in vision transformers. We propose an effective method for Automatic Visual Circuit Discovery (Vi-CD) that recovers class-specific circuits for classification, identifies circuits underlying typographic attacks in CLIP, and discovers circuits that lend themselves for steering to correct harmful model behavior. Overall, we find that insightful and actionable edge-based circuits can be recovered from vision transformers, adding transparency to the internal computations of these models.

---

### Evo-MedAgent: Beyond One-Shot Diagnosis with Agents That Remember, Reflect, and Improve

Authors: Weixiang Shen, Bailiang Jian, Jun Li, Che Liu, Johannes Moll, Xiaobin Hu, Daniel Rueckert, Hongwei Bran Li, Jiazhen Pan
Published: 2026-04-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.14475v1

Tool-augmented large language model (LLM) agents can orchestrate specialist classifiers, segmentation models, and visual question-answering modules to interpret chest X-rays. However, these agents still solve each case in isolation: they fail to accumulate experience across cases, correct recurrent reasoning mistakes, or adapt their tool-use behavior without expensive reinforcement learning. While a radiologist naturally improves with every case, current agents remain static. In this work, we propose Evo-MedAgent, a self-evolving memory module that equips a medical agent with the capacity for inter-case learning at test time. Our memory comprises three complementary stores: (1)~\emph{Retrospective Clinical Episodes} that retrieve problem-solving experiences from similar past cases, (2)~an \emph{Adaptive Procedural Heuristics} bank curating priority-tagged diagnostic rules that evolves via reflection, much like a physician refining their internal criteria, and (3)~a \emph{Tool Reliability Controller} that tracks per-tool trustworthiness. On ChestAgentBench, Evo-MedAgent raises multiple-choice question (MCQ) accuracy from 0.68 to 0.79 on GPT-5-mini, and from 0.76 to 0.87 on Gemini-3 Flash. With a strong base model, evolving memory improves performance more effectively than orchestrating external tools on qualitative diagnostic tasks. Because Evo-MedAgent requires no training, its per-case overhead is bounded by one additional retrieval pass and a single reflection call, making it deployable on top of any frozen model.

---

### Scouting By Reward: VLM-TO-IRL-Driven Player Selection For Esports

Authors: Qing Yan, Wenyu Yang, Yufei Wang, Wenhao Ma, Linchong Hu, Yifei Jin, Anton Dahbura
Published: 2026-04-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2604.14474v1

Traditional esports scouting workflows rely heavily on manual video review and aggregate performance metrics, which often fail to capture the nuanced decision-making patterns necessary to determine if a prospect fits a specific tactical archetype. To address this, we reframe style-based player evaluation in esports as an Inverse Reinforcement Learning (IRL) problem. In this paper, we introduce a novel player selection framework that learns professional-specific reward functions from logged gameplay demonstrations, allowing organizations to rank candidates by their stylistic alignment with a target star player. Our proposed architecture utilizes a multimodal, two-branch intake: one branch encodes structured state-action trajectories derived from high-resolution in-game telemetry, while the second encodes temporally aligned tactical pseudo-commentary generated by Vision-Language Models (VLMs) from broadcast footage. These representations are fused and evaluated via a Generative Adversarial Imitation Learning (GAIL) objective, where a discriminator learns to capture the unique mechanical and tactical signatures of elite professionals. By transitioning from generic skill estimation to scouting "by reward," this framework provides a scalable, workflow-aware digital twin system that enables data-driven roster construction and targeted talent discovery across massive candidate pools.

---

### Response-Aware User Memory Selection for LLM Personalization

Authors: Jillian Fisher, Jennifer Neville, Chan Young Park
Published: 2026-04-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.14473v1

A common approach to personalization in large language models (LLMs) is to incorporate a subset of the user memory into the prompt at inference time to guide the model's generation. Existing methods select these subsets primarily using similarity between user memory items and input queries, ignoring how features actually affect the model's response distribution. We propose Response-Utility optimization for Memory Selection (RUMS), a novel method that selects user memory items by measuring the mutual information between a subset of memory and the model's outputs, identifying items that reduce response uncertainty and sharpen predictions beyond semantic similarity. We demonstrate that this information-theoretic foundation enables more principled user memory selection that aligns more closely with human selection compared to state-of-the-art methods, and models $400\times$ larger. Additionally, we show that memory items selected using RUMS result in better response quality compared to existing approaches, while having up to $95\%$ reduction in computational cost.

---

### Auxiliary Finite-Difference Residual-Gradient Regularization for PINNs

Authors: Stavros Kassinos
Published: 2026-04-15
Categories: cs.LG, cs.AI, cs.CE, physics.comp-ph
arXiv: https://arxiv.org/abs/2604.14472v1

Physics-informed neural networks (PINNs) are often selected by a single scalar loss even when the quantity of interest is more specific. We study a hybrid design in which the governing PDE residual remains automatic-differentiation (AD) based, while finite differences (FD) appear only in a weak auxiliary term that penalizes gradients of the sampled residual field. The FD term regularizes the residual field without replacing the PDE residual itself. We examine this idea in two stages. Stage 1 is a controlled Poisson benchmark comparing a baseline PINN, the FD residual-gradient regularizer, and a matched AD residual-gradient baseline. Stage 2 transfers the same logic to a three-dimensional annular heat-conduction benchmark (PINN3D), where baseline errors concentrate near a wavy outer wall and the auxiliary grid is implemented as a body-fitted shell adjacent to the wall. In Stage 1, the FD regularizer reproduces the main effect of residual-gradient control while exposing a trade-off between field accuracy and residual cleanliness. In Stage 2, the shell regularizer improves the application-facing quantities, namely outer-wall flux and boundary-condition behavior. Across seeds 0-5 and 100k epochs, the most reliable tested configuration is a fixed shell weight of 5e-4 under the Kourkoutas-beta optimizer regime: relative to a matched run without the shell term, it reduces the mean outer-wall BC RMSE from 1.22e-2 to 9.29e-4 and the mean wall-flux RMSE from 9.21e-3 to 9.63e-4. Adam with beta2=0.999 becomes usable when the initial learning rate is reduced to 1e-3, although its shell benefit is less robust than under Kourkoutas-beta. Overall, the results support a targeted view of hybrid PINNs: an auxiliary-only FD regularizer is most valuable when it is aligned with the physical quantity of interest, here the outer-wall flux.

---

### Improving Human Performance with Value-Aware Interventions: A Case Study in Chess

Authors: Saumik Narayanan, Raja Panjwani, Siddhartha Sen, Chien-Ju Ho
Published: 2026-04-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.14465v1

AI systems are increasingly used to assist humans in sequential decision-making tasks, yet determining when and how an AI assistant should intervene remains a fundamental challenge. A potential baseline is to recommend the optimal action according to a strong model. However, such actions assume optimal follow-up actions, which human decision makers may fail to execute, potentially reducing overall performance. In this work, we propose and study value-aware interventions, motivated by a basic principle in reinforcement learning: under the Bellman equation, the optimal policy selects actions that maximize the immediate reward plus the value function. When a decision maker follows a suboptimal policy, this policy-value consistency no longer holds, creating discrepancies between the actions taken by the policy and those that maximize the immediate reward plus the value of the next state. We show that these policy-value inconsistencies naturally identify opportunities for intervention. We formalize this problem in a Markov decision process where an AI assistant may override human actions under an intervention budget. In the single-intervention regime, we show that the optimal strategy is to recommend the action that maximizes the human value function. For settings with multiple interventions, we propose a tractable approximation that prioritizes interventions based on the magnitude of the policy-value discrepancy. We evaluate these ideas in the domain of chess by learning models of humans from large-scale gameplay data. In simulation, our approach consistently outperforms interventions based on the strongest chess engine (Stockfish) in a wide range of settings. A within-subject human study with 20 players and 600 games further shows that our interventions significantly improve performance for low- and mid-skill players while matching expert-engine interventions for high-skill players.

---

### Psychological Steering of Large Language Models

Authors: Leonardo Blas, Robin Jia, Emilio Ferrara
Published: 2026-04-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.14463v1

Large language models (LLMs) emulate a consistent human-like behavior that can be shaped through activation-level interventions. This paradigm is converging on additive residual-stream injections, which rely on injection-strength sweeps to approximate optimal intervention settings. However, existing methods restrict the search space and sweep in uncalibrated activation-space units, potentially missing optimal intervention conditions. Thus, we introduce a psychological steering framework that performs unbounded, fluency-constrained sweeps in semantically calibrated units. Our method derives and calibrates residual-stream injections using psychological artifacts, and we use the IPIP-NEO-120, which measures the OCEAN personality model, to compare six injection methods. We find that mean-difference (MD) injections outperform Personality Prompting (P$^2$), an established baseline for OCEAN steering, in open-ended generation in 11 of 14 LLMs, with gains of 3.6\% to 16.4\%, overturning prior reports favoring prompting and positioning representation engineering as a new frontier in open-ended psychological steering. Further, we find that a hybrid of P$^2$ and MD injections outperforms both methods in 13 of 14 LLMs, with gains over P$^2$ ranging from 5.6\% to 21.9\% and from 3.3\% to 26.7\% over MD injections. Finally, we show that MD injections align with the Linear Representation Hypothesis and provide reliable, approximately linear control knobs for psychological steering. Nevertheless, they also induce OCEAN trait covariance patterns that depart from the Big Two model, suggesting a gap between learned representations and human psychology.

---

### Bias in Surface Electromyography Features across a Demographically Diverse Cohort

Authors: Aditi Agrawal, Celine John Philip, Giancarlo K. Sagastume, Marcus A. Battraw, Wilsaan M. Joiner, Jonathon S. Schofield, Lee M. Miller, Richard S. Whittle
Published: 2026-04-15
Categories: cs.HC, cs.LG
arXiv: https://arxiv.org/abs/2604.14460v2

Neuromotor decoding from upper-limb electromyography (sEMG) can enhance human-machine interfaces and offer a more natural means of controlling prosthetic limbs, virtual reality, and household electronics. Unfortunately, current sEMG technology does not always perform consistently across users because individual differences such as age and body mass index, among many others, can substantially alter signal quality. This variability makes sEMG characteristics highly idiosyncratic, often necessitating laborious personalization and iterative tuning to achieve reliable performance. This variability has particular import for sEMG-based assistive devices and neural interfaces, where demographic biases in sEMG features could undermine broad and fair deployment. In this study, we explore how demographic differences affect the sEMG signals produced and their implications for machine learning-based gesture decoding. We analyze the data set provided by, in which we derive 147 common sEMG features extracted from 81 demographically diverse individuals performing discrete hand gestures. Using mixed-effects linear models and partial least squares (PLS) analysis, which take into consideration demographic variables (including age, sex, height, weight, skin properties, subcutaneous fat, and hair density), we identify that 33\% (49 of 147) of commonly used sEMG features show significant associations with demographic characteristics. These results may help guide the development of fair and unbiased sEMG-based neural interfaces across a diverse population.

---

### Filling in the Mechanisms: How do LMs Learn Filler-Gap Dependencies under Developmental Constraints?

Authors: Atrey Desai, Sathvik Nair
Published: 2026-04-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.14459v1

For humans, filler-gap dependencies require a shared representation across different syntactic constructions. Although causal analyses suggest this may also be true for LLMs (Boguraev et al., 2025), it is still unclear if such a representation also exists for language models trained on developmentally feasible quantities of data. We applied Distributed Alignment Search (DAS, Geiger et al. (2024)) to LMs trained on varying amounts of data from the BabyLM challenge (Warstadt et al., 2023), to evaluate whether representations of filler-gap dependencies transfer between wh-questions and topicalization, which greatly vary in terms of their input frequency. Our results suggest shared, yet item-sensitive mechanisms may develop with limited training data. More importantly, LMs still require far more data than humans to learn comparable generalizations, highlighting the need for language-specific biases in models of language acquisition.

---

### FocalLens: Visualizing Narratives through Focalization

Authors: S M Raihanul Alam, Md Dilshadur Rahman, Md Naimul Hoque
Published: 2026-04-15
Categories: cs.HC, cs.AI
arXiv: https://arxiv.org/abs/2604.14456v1

Visualizing narratives is useful to writers to reflect on unfinished drafts and identify unintentional biases and inconsistencies. Literary scholars can use the visualizations to identify nuanced patterns and literary styles from written text. Current narrative visualization is limited to representing character and location co-occurrences in a timeline, omitting important and complex narrative components such as focalization, causality, and speech. This paper aims to capture and visualize underexplored, complex narrative components as a basis for narrative visualization. As a starting point, we propose a new narrative visualization, named FocalLens, that uses focalization, the component that establishes who sees or perceives the events in a narrative, for representing the narrative. We provide the theoretical foundation of focalization and describe various types and facets of focalization. The details are incorporated in the novel visualization that captures how different characters perceive an event, who directly participate in an event, who indirectly observe the event, and who narrate the event. We also developed a tool that provides fluid interaction between the text and the proposed visualization. The tool was evaluated with four writers and scholars in a qualitative study, where writers analyzed their draft stories and scholars analyzed well-known stories. The findings suggest the tool added a new dimension to the workflow for writers and scholars, an analytical lens that is not available otherwise. We conclude by identifying design implications and future directions.

---

### AIBuildAI: An AI Agent for Automatically Building AI Models

Authors: Ruiyi Zhang, Peijia Qin, Qi Cao, Li Zhang, Pengtao Xie
Published: 2026-04-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.14455v1

AI models underpin modern intelligent systems, driving advances across science, medicine, finance, and technology. Yet developing high-performing AI models remains a labor-intensive process that requires expert practitioners to iteratively design architectures, engineer representations, implement training pipelines and refine approaches through empirical evaluation. Existing AutoML methods partially alleviate this burden but remain limited to narrow aspects such as hyperparameter optimization and model selection within predefined search spaces, leaving the full development lifecycle largely dependent on human expertise. To address this gap, we introduce AIBuildAI, an AI agent that automatically builds AI models from a task description and training data. AIBuildAI adopts a hierarchical agent architecture in which a manager agent coordinates three specialized sub-agents: a designer for modeling strategy, a coder for implementation and debugging, and a tuner for training and performance optimization. Each sub-agent is itself a large language model (LLM) based agent capable of multi-step reasoning and tool use, enabling end-to-end automation of the AI model development process that goes beyond the scope of existing AutoML approaches. We evaluate AIBuildAI on MLE-Bench, a benchmark of realistic Kaggle-style AI development tasks spanning visual, textual, time-series and tabular modalities. AIBuildAI ranks first on MLE-Bench with a medal rate of 63.1%, outperforming all existing baseline methods and matching the capability of highly experienced AI engineers. These results demonstrate that hierarchical agent systems can automate the full AI model development process from task specification to deployable model, suggesting a pathway toward broadly accessible AI development with minimal human intervention.

---