# Research Papers — 2026-04 · bundle page 27 of 233

> Full-text markdown of papers #1301–1350 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/27
[← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/26.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-04/page/28.md)

### PINNs in More General Geometry

Authors: Edward Hirst
Published: 2026-04-27
Categories: math.DG, cs.LG, hep-th
arXiv: https://arxiv.org/abs/2604.25020v1

Neural architectures trained with losses inspired by differential conditions are the basis for PINN models. Since many constructions in differential geometry may be framed as minimisation of a differential functional, these functionals can be coded as loss functions to align the AI loss-minimisation goal with that of solving the geometric problem. This contribution to the Recent Progress in Computational String Geometry workshop proceedings introduces the PINN architecture defining principles, motivates how they are well suited for problems in differential geometry, and demonstrates their use via summaries of three works at this intersection.

---

### Internet of Everything in the 6G Era: Paradigms, Enablers, Potentials and Future Directions

Authors: Driss Choukri, Essaid Sabir, Elmahdi Driouch, Abdelkrim Haqiq
Published: 2026-04-27
Categories: cs.ET, cs.AI, cs.DC, cs.NI
arXiv: https://arxiv.org/abs/2604.25018v2

The Internet of Everything (IoE) represents an evolution of the Internet of Things (IoT) by integrating people, data, processes, and things into a unified intelligent ecosystem. IoE aims to enhance automation, decision-making, and service efficiency across multiple application domains such as smart cities, healthcare, industry, and next-generation wireless networks. This paper provides a structured overview of the IoE concept, its core components, architectural foundations, enabling technologies, and major research challenges. Finally, open research directions toward 6G-enabled intelligent IoE systems are discussed, with emphasis on scalability, security, privacy, and energy efficiency.

---

### Why Search When You Can Transfer? Amortized Agentic Workflow Design from Structural Priors

Authors: Shiyi Du, Jiayuan Liu, Weihua Du, Yue Huang, Jiayi Li, Yingtao Luo, Xiangliang Zhang, Vincent Conitzer, Carl Kingsford
Published: 2026-04-27
Categories: cs.LG
arXiv: https://arxiv.org/abs/2604.25012v1

Automated agentic workflow design currently relies on per-task iterative search, which is computationally prohibitive and fails to reuse structural knowledge across tasks. We observe that optimized workflows converge to a small family of domain-specific topologies, suggesting that this combinatorial search is largely redundant. Building on this insight, we propose SWIFT (Synthesizing Workflows via Few-shot Transfer), a framework that amortizes workflow design into reusable structural priors. SWIFT first distills compositional heuristics and output-interface contracts from contrastive analysis of prior search trajectories across source tasks. At inference time, it conditions a single LLM generation pass on these priors together with cross-task workflow demonstrations to synthesize a complete, executable workflow for an unseen target task, bypassing iterative search entirely. On five benchmarks, SWIFT outperforms the state-of-the-art search-based method while reducing marginal per-task optimization cost by three orders of magnitude. It further generalizes to four additional unseen benchmarks and transfers successfully from GPT-4o-mini to three additional foundation models (Grok, Qwen, Gemma). Controlled ablations reveal that workflow demonstrations primarily transfer topological structure rather than surface semantics: replacing all operator names with random strings still retains over 93% of the full system's average performance.

---

### Why Does Reinforcement Learning Generalize? A Feature-Level Mechanistic Study of Post-Training in Large Language Models

Authors: Dan Shi, Zhuowen Han, Simon Ostermann, Renren Jin, Josef van Genabith, Deyi Xiong
Published: 2026-04-27
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.25011v1

Reinforcement learning (RL)-based post-training often improves the reasoning performance of large language models (LLMs) beyond the training domain, while supervised fine-tuning (SFT) frequently leads to general capabilities forgetting. However, the mechanisms underlying this contrast remain unclear. To bridge this gap, we present a feature-level mechanistic analysis methodology to probe RL generalization using a controlled experimental setup, where RL- and SFT-tuned models are trained from the same base model on identical data. Leveraging our interpretability framework, we align internal activations across models within a shared feature space and analyze how features evolve during post-training. We find that SFT rapidly introduces many highly specialized features that stabilize early in training, whereas RL induces more restrained and continually evolving feature changes that largely preserve base models' representations. Focusing on samples where RL succeeds but the base model fails, we identify a compact, task-agnostic set of features that directly mediate generalization across diverse tasks. Feature-level interventions confirm their causal role: disabling these features significantly degrades RL models' generalization performance, while amplifying them improves base models' performance. The code is available at https://github.com/danshi777/RL-generalization.

---

### EVT-Based Generative AI for Tail-Aware Channel Estimation

Authors: Parmida Valiahdi, Niloofar Mehrnia, Walid Saad, Sinem Coleri
Published: 2026-04-27
Categories: eess.SP, cs.AI, eess.SY
arXiv: https://arxiv.org/abs/2604.25008v1

Ultra-reliable and low-latency communication (URLLC) will play a key role in fifth-generation (5G) and beyond networks, enabling mission-critical applications. Meeting the stringent URLLC requirements, characterized by extremely low packet error rates and minimal latency, calls for advanced statistical modeling to accurately capture rare events in wireless channels. Traditional methods, such as those that rely on large datasets and computationally intensive estimation techniques, often fail in real-time scenarios. In this paper, a novel framework is proposed to meet URLLC requirements through a synergistic integration of extreme value theory (EVT) with generative artificial intelligence (AI). EVT is used to model channel tail distributions, providing an accurate characterization of rare events. Concurrently, generative AI enables data augmentation and channel parameter estimation from limited samples. The integration of EVT with generative AI can thus help overcome the limitations of generative models in capturing extreme events during channel characterization. Using an experimental dataset collected from an automotive environment, it is demonstrated that this integration enhances data augmentation for extreme quantiles, while requiring fewer samples than traditional analytical EVT methods and generative baselines in online estimation of channel distribution.

---

### Toward a Science of Intent: Closure Gaps and Delegation Envelopes for Open-World AI Agents

Authors: Maximiliano Armesto, Christophe Kolb
Published: 2026-04-27
Categories: cs.AI, cs.SE
arXiv: https://arxiv.org/abs/2604.25000v2

Recent work has framed intelligence in verifiable tasks as reducing time-to-solution through learned structure and test-time search, while systems work has explored learned runtimes in which computation, memory and I/O migrate into model state. These perspectives do not explain why capable models remain difficult to deploy in open institutions. We propose intent compilation: the transformation of partially specified human purpose into inspectable artifacts that bind execution. The relevant deployment distinction is closed-world solver versus open-world agent. In closed worlds, a checker is largely given; in open worlds, verification is distributed across semantic, evidentiary, procedural and institutional dimensions. Weformalize this residual openness as a closure-gap vector, define delegation envelopes as pre-authorized regions of action space, distinguish misclosure from undersearch, and outline benchmark metrics for testing when closure interventions outperform additional inference-time search.

---

### BifDet: A 3D Bifurcation Detection Dataset for Airway-Tree Modeling

Authors: Ali Keshavarzi, Quentin Bouniot, Benjamin M. Smith, Elsa Angelini
Published: 2026-04-27
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2604.24999v1

Thoracic Computed Tomography (CT) scans offer detailed insights into the intricate branching network of the airway tree, which is essential for understanding various respiratory diseases. Airway bifurcations, where airway branches split, are crucial landmarks for understanding lung physiology, disease mechanisms and lesion localization. Despite the significance of bifurcation analysis, a notable lack of datasets annotated for this task hinders the development of advanced automated specialized detection or segmentation tools. In this paper, we introduce BifDet, the first publicly-available dataset specialized for 3D airway bifurcation detection, filling a critical gap in existing resources. Our dataset comprises carefully annotated CT scans from the ATM22 open-access cohort with bifurcation bounding boxes covering the parent and daughter branches. As a use-case for demonstrating the potential of BifDet, we fine-tune and evaluate RetinaNet and DETR for 3D airway bifurcations detection on CT scans. We provide detailed pipelines, including preprocessing steps and specific implementation design choices. Results are detailed over various categories of minimal bounding box sizes to serve as baseline to benchmark future research.

---

### DouC: Dual-Branch CLIP for Training-Free Open-Vocabulary Segmentation

Authors: Mohamad Zamini, Diksha Shukla
Published: 2026-04-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.24997v1

Open-vocabulary semantic segmentation requires assigning pixel-level semantic labels while supporting an open and unrestricted set of categories. Training-free CLIP-based approaches preserve strong zero-shot generalization but typically rely on a single inference mechanism, limiting their ability to jointly address unreliable local tokens and insufficient spatial coherence. We propose DouC, a training-free dual-branch CLIP framework that decomposes dense prediction into two complementary components. OG-CLIP improves patch-level reliability via lightweight, inference-time token gating, while FADE-CLIP injects external structural priors through proxy attention guided by frozen vision foundation models. The two branches are fused at the logit level, enabling local token reliability and structure-aware patch interactions to jointly influence final predictions, with optional instance-aware correction applied as post-processing. DouC introduces no additional learnable parameters, requires no retraining, and preserves CLIP's zero-shot generalization. Extensive experiments across eight benchmarks and multiple CLIP backbones demonstrate that DouC consistently outperforms prior training-free methods and scales favorably with model capacity.

---

### Sparse Personalized Text Generation with Multi-Trajectory Reasoning

Authors: Bo Ni, Haowei Fu, Qinwen Ge, Franck Dernoncourt, Samyadeep Basu, Nedim Lipka, Seunghyun Yoon, Yu Wang, Nesreen K. Ahmed, Subhojyoti Mukherjee, Puneet Mathur, Ryan A. Rossi et al.
Published: 2026-04-27
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.24996v1

As Large Language Models (LLMs) advance, personalization has become a key mechanism for tailoring outputs to individual user needs. However, most existing methods rely heavily on dense interaction histories, making them ineffective in cold-start scenarios where such data is sparse or unavailable. While external signals (e.g., content of similar users) can offer a potential remedy, leveraging them effectively remains challenging: raw context is often noisy, and existing methods struggle to reason over heterogeneous data sources. To address these issues, we introduce PAT (Personalization with Aligned Trajectories), a reasoning framework for cold-start LLM personalization. PAT first retrieves information along two complementary trajectories: writing-style cues from stylistically similar users and topic-specific context from preference-aligned users. It then employs a reinforcement learning-based, iterative dual-reasoning mechanism that enables the LLM to jointly refine and integrate these signals. Experimental results across real-world personalization benchmarks show that PAT consistently improves generation quality and alignment under sparse-data conditions, establishing a strong solution to the cold-start personalization problem.

---

### Power Foam: Unifying Real-Time Differentiable Ray Tracing and Rasterization

Authors: Shrisudhan Govindarajan, Daniel Rebain, Dor Verbin, Kwang Moo Yi, Anish Prabhu, Andrea Tagliasacchi
Published: 2026-04-27
Categories: cs.GR, cs.CV
arXiv: https://arxiv.org/abs/2604.24994v1

We introduce a differentiable 3D representation that unifies the ray tracing capabilities of foam-based ray tracing with the efficiency of modern rasterization pipelines. While prior foam representations enable constant-time ray traversal through an explicit volumetric partition of space, their potentially unbounded cells hinder efficient tile-based rasterization. We address this limitation by generalizing Voronoi foams to bounded power diagrams with controllable cell extents, enabling spatially bounded primitives without requiring expensive Delaunay triangulations during training. We further introduce an oriented surface formulation that explicitly models interfaces between interior and exterior regions, and decouple geometry from appearance by embedding differentiable texture directly on these surfaces. Together, these contributions yield a representation that preserves state-of-the-art ray tracing efficiency while achieving rasterization performance competitive with current generation 3DGS, providing a practical path toward unified real-time differentiable rendering.

---

### Laplace-Bridged Randomized Smoothing for Fast Certified Robustness

Authors: Miao Lin, MD Saifur Rahman Mazumder, Feng Yu, Daniel Takabi, Rui Ning
Published: 2026-04-27
Categories: cs.LG
arXiv: https://arxiv.org/abs/2604.24993v1

Randomized Smoothing (RS) offers formal $\ell_2$ guarantees for arbitrary base classifiers but faces two key practical bottlenecks: (i) it often relies on noise-augmented training to achieve nontrivial certificates, which increases training cost, can reduce clean accuracy, and weakens RS as a genuinely post-hoc defense; and (ii) certification is computationally expensive, typically requiring tens of thousands of noisy forward passes per input, which hinders deployment, especially on resource-constrained edge devices. To address both limitations, we propose Laplace-Bridged Smoothing (LBS), an analytic reformulation of RS that replaces high-dimensional input-space Monte Carlo (MC) sampling with efficient computations in a low-dimensional probability space. LBS preserves formal robustness guarantees without requiring noise-augmented training while substantially reducing certification burden. On CIFAR-10 and ImageNet, LBS attains stronger certified robustness than RS and reduces per-sample certification cost by nearly an order of magnitude. Notably, on NVIDIA Jetson Orin Nano and Raspberry Pi 4, LBS achieves speedups of up to $494\times$, enabling practical certified deployment on real-world edge devices. Finally, we provide theoretical justification for the analytic formulation and certificate validity of LBS.

---

### A New Kind of Network? Review and Reference Implementation of Neural Cellular Automata

Authors: Martin Spitznagel, Janis Keuper
Published: 2026-04-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.24990v2

Stephen Wolfram proclaimed in his 2003 seminal work "A New Kind Of Science" that simple recursive programs in the form of Cellular Automata (CA) are a promising approach to replace currently used mathematical formalizations, e.g. differential equations, to improve the modeling of complex systems. Over two decades later, while Cellular Automata have still been waiting for a substantial breakthrough in scientific applications, recent research showed new and promising approaches which combine Wolfram's ideas with learnable Artificial Neural Networks: So-called Neural Cellular Automata (NCA) are able to learn the complex update rules of CA from data samples, allowing them to model complex, self-organizing generative systems. The aim of this paper is to review the existing work on NCA and provide a unified modular framework and notation, as well as a reference implementation in the open-source library NCAtorch. Supplementary materials, videos, and code are available at the project website: https://www.neural-cellular-automata.org/

---

### Assessing Y-Axis Influence: Bias in Multimodal Language Models on Chart-to-Table Translation

Authors: Seok Hwan Song, Azher Ahmed Efat, Wallapak Tavanapong
Published: 2026-04-27
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.24987v1

Chart-to-table translation converts chart images into structured tabular data. Accurate translation is crucial for Multimodal Language Model (MLM) to answer complex queries. We observe imbalances in the number of images across different aspects of the y-axis information in public chart datasets. Such imbalances can introduce unintended biases, causing uneven MLM performance. Previous works have not systematically examined these biases. To address this gap, we propose a new framework, FairChart2Table, for analyzing y-axis-related bias on five state-of-the-art models. Key Findings: (1) There are significant y-axis biases related to the digit length of the major tick values, the number of major ticks, the range of values, and the tick value format (e.g., abbreviation or scientific format). (2) The number of legends/entities in chart images impacts MLM performance. (3) Prompting MLM with y-axis information can significantly enhance the performance for some MLMs.

---

### Adaptive Prompt Embedding Optimization for LLM Jailbreaking

Authors: Miles Q. Li, Benjamin C. M. Fung, Boyang Li, Radin Hamidi Rad, Ebrahim Bagheri
Published: 2026-04-27
Categories: cs.AI
arXiv: https://arxiv.org/abs/2604.24983v1

Existing white-box jailbreak attacks against aligned LLMs typically append discrete adversarial suffixes to the user prompt, which visibly alters the prompt and operates in a combinatorial token space. Prior work has avoided directly optimizing the embeddings of the original prompt tokens, presumably because perturbing them risks destroying the prompt's semantic content. We propose Prompt Embedding Optimization (PEO), a multi-round white-box jailbreak that directly optimizes the embeddings of the original prompt tokens without appending any adversarial tokens, and show that the concern is unfounded: the optimized embeddings remain close enough to their originals that the visible prompt string is preserved exactly after nearest-token projection, and quantitative analysis shows the model's responses stay on topic for the large majority of prompts. PEO combines continuous embedding-space optimization with structured continuation targets and an adaptive failure-focused schedule. Counterintuitively, later PEO rounds can benefit from heuristic composite response scaffolds that are not natural standalone templates, yet ASR-Judge shows that the resulting gains are not merely empty formatting or scaffold-only outputs. Across two standard harmful-behavior benchmarks and competing white-box attacks spanning discrete suffix search, appended adversarial embeddings, and search-based adversarial generation, PEO outperforms all of them in our experiments.

---

### Dont Stop Early: Scalable Enterprise Deep Research with Controlled Information Flow and Evidence-Aware Termination

Authors: Prafulla Kumar Choubey, Kung-Hsiang Huang, Pranav Narayanan Venkit, Jiaxin Zhang, Vaibhav Vats, Yu Li, Xiangyu Peng, Chien-Sheng Wu
Published: 2026-04-27
Categories: cs.CL, cs.SE
arXiv: https://arxiv.org/abs/2604.24978v1

Enterprise deep research often fails to produce decision-ready reports due to uneven information coverage, context explosion, and premature stopping. We propose a scalable Enterprise Deep Research (EDR) architecture to address these failures. Our system (i) decomposes requests into coverage-driven objectives via outline generation with reflection, (ii) localizes context with dependency-guided execution and explicit information sharing, and (iii) enforces evidence-based completion criteria so agents iteratively collect information until sufficiency conditions are met. We evaluate on an internal sales enablement task and the public DeepResearch Bench benchmark, where our proposed system design achieves the strongest overall performance compared with competitive deep-research baselines. The results show that dependency-controlled context and explicit evidence sufficiency criteria reduce premature stopping and improve the consistency and depth of enterprise research outputs.

---

### A Survey on LLM-based Conversational User Simulation

Authors: Bo Ni, Leyao Wang, Yu Wang, Branislav Kveton, Franck Dernoncourt, Yu Xia, Hongjie Chen, Reuben Leura, Samyadeep Basu, Subhojyoti Mukherjee, Puneet Mathur, Nesreen Ahmed et al.
Published: 2026-04-27
Categories: cs.CL, cs.HC
arXiv: https://arxiv.org/abs/2604.24977v1

User simulation has long played a vital role in computer science due to its potential to support a wide range of applications. Language, as the primary medium of human communication, forms the foundation of social interaction and behavior. Consequently, simulating conversational behavior has become a key area of study. Recent advancements in large language models (LLMs) have significantly catalyzed progress in this domain by enabling high-fidelity generation of synthetic user conversation. In this paper, we survey recent advancements in LLM-based conversational user simulation. We introduce a novel taxonomy covering user granularity and simulation objectives. Additionally, we systematically analyze core techniques and evaluation methodologies. We aim to keep the research community informed of the latest advancements in conversational user simulation and to further facilitate future research by identifying open challenges and organizing existing work under a unified framework.

---

### Dynamic Decision Learning: Test-Time Evolution for Abnormality Grounding in Rare Diseases

Authors: Jun Li, Mingxuan Liu, Jiazhen Pan, Che Liu, Wenjia Bai, Cosmin I. Bercea, Julia A. Schnabel
Published: 2026-04-27
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.24972v1

Clinical abnormality grounding for rare diseases is often hindered by data scarcity, making supervised fine-tuning impractical and single-pass inference highly unstable. We propose Dynamic Decision Learning (DDL), a framework that enables frozen large vision-language models (LVLMs) to refine their decisions across both language and visual spaces by optimizing instructions and consolidating predictions under visual perturbations. This process improves localization quality and produces a consensus-based reliability score that quantifies model confidence. Results on brain imaging benchmarks, including a rare-disease dataset with 281 pathology types across models ranging from 3B to 72B parameters, show that DDL improves mAP@75 by up to 105% on rare-disease cases and outperforms adaptation baselines and supervised fine-tuning. Furthermore, DDL demonstrates stronger calibration between reliability scores and localization accuracy under severe distribution shifts and increasing task difficulty. Code is available at: https://lijunrio.github.io/DDL/

---

### PolyKV: A Shared Asymmetrically-Compressed KV Cache Pool for Multi-Agent LLM Inference

Authors: Ishan Patel, Ishan Joshi
Published: 2026-04-27
Categories: cs.LG, cs.CL, cs.DC
arXiv: https://arxiv.org/abs/2604.24971v1

We present PolyKV, a system in which multiple concurrent inference agents share a single, asymmetrically compressed KV cache pool. Rather than allocating a separate KV cache per agent -- the standard paradigm -- PolyKV writes a compressed cache once and injects it into N independent agent contexts via HuggingFace DynamicCache objects. Compression is asymmetric: Keys are quantized at int8 (q8_0) to preserve softmax stability, while Values are compressed using TurboQuant MSE -- a Fast Walsh-Hadamard Transform (FWHT) rotation followed by 3-bit Lloyd-Max quantization with centroids tuned to N(0,1). We evaluate across two model scales (SmolLM2-1.7B-Instruct and Llama-3-8B-Instruct), three context lengths (600-7,194 tokens), and up to 15 concurrent agents. PolyKV achieves a stable 2.91x compression ratio across all configurations. On Llama-3-8B with 15 agents sharing a 4K-token context, PolyKV reduces KV cache memory from 19.8 GB to 0.45 GB -- a 97.7% reduction -- while maintaining only +0.57% perplexity degradation and a mean BERTScore F1 of 0.928. PPL delta does not grow with agent count and improves as context length increases, inverting to -0.26% at 1,851 coherent tokens. To our knowledge, no prior work combines a single shared, lossy-compressed KV pool with multi-reader concurrent agent access.

---

### The Effects of Population Size on the Performance of BEAGLE GPU-Based Genetic Programming Runs

Authors: Nathan Haut, Ilya Basin, Ruchika Gupta, Marzieh Kianinejad, Zachary Perrico, Elijah Smith, Wolfgang Banzhaf
Published: 2026-04-27
Categories: cs.NE
arXiv: https://arxiv.org/abs/2604.24968v1

The Beagle framework, through GPU-based Genetic Programming, enables population dynamics previously unattainable (within practical time frames) by CPU-constrained Genetic Programming systems. This work explores how GPU-enabled population sizes impact the success of training for symbolic regression problems. Specifically, when using constant population sizes, we see benefits of using very narrow and deep searches (as narrow as 1000 individuals) for some problems, while other problems benefit from very broad and shallow searches (as broad as 10 million individuals). We also explore stepped population sizes that start with large populations and drop to small populations to balance the breadth and depth of search.

---

### Risk Reporting for Developers' Internal AI Model Use

Authors: Oscar Delaney, Sambhav Maheshwari, Joe O'Brien, Theo Bearman, Oliver Guest
Published: 2026-04-27
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2604.24966v1

Frontier AI companies first deploy their most advanced models internally, for weeks or months of safety testing, evaluation, and iteration, before a possible public release. For example, Anthropic recently developed a new class of model with advanced cyberoffense-relevant capabilities, Mythos Preview, which was available internally for at least six weeks before it was publicly announced. This internal use creates risks that external deployment frameworks may fail to address. Legal frameworks, notably California's Transparency in Frontier Artificial Intelligence Act (SB 53), New York's Responsible AI Safety And Education (RAISE) Act, and the EU's General-Purpose AI Code of Practice, all discuss risks from internal AI use. They require frontier developers to make and implement plans for how to manage risks from internal use, and to produce internal use risk reports describing their safeguards and any residual risks. This guide provides a harmonized standard for companies to produce internal use risk reports suitable for all three regulatory frameworks. It is addressed primarily to evaluation and safety teams at frontier AI developers, and secondarily to regulators and auditors seeking to understand what good reporting looks like. Given the pace of AI R&D automation and the limited external visibility into how companies use their most capable models internally, regular and detailed risk reporting may be one of the few mechanisms available to ensure that the risks from internal AI use are identified and managed before they materialize. Whenever a substantially more capable or riskier model is deployed internally, the developer should create a risk report and argue why the model is safe to deploy. We structure the reporting framework around two threat vectors -- autonomous AI misbehavior and insider threats -- and three risk factors for each: means, motive, and opportunity.

---

### Odysseys: Benchmarking Web Agents on Realistic Long Horizon Tasks

Authors: Lawrence Keunho Jang, Jing Yu Koh, Daniel Fried, Ruslan Salakhutdinov
Published: 2026-04-27
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2604.24964v1

Existing web agent benchmarks have largely converged on short, single-site tasks that frontier models are approaching saturation on. However, real world web use consists of long-horizon, multi-site workflows. Common web navigation tasks, such as comparing products across different domains, planning trips across multiple services, or summarizing information from multiple search queries, require sustained context and cross-site reasoning over potentially hours of browsing. To capture and evaluate such behaviors, we introduce Odysseys: a benchmark of 200 long-horizon web tasks derived from real world browsing sessions evaluated on the live Internet. We find that binary pass/fail evaluation is inadequate for long-horizon settings and introduce a rubric-based evaluation, annotating each Odysseys task with an average of 6.1 graded rubrics. We demonstrate that this yields higher agreement with humans and provides a more fine-grained signal than commonly used trajectory-level LLM-as-a-judge evaluation metrics. We tested several leading frontier models and find that the strongest models achieve a success rate of 44.5%, which leaves substantial room for future improvements. Beyond task success, we argue that efficiency is a first-class concern for long-horizon agents. We introduce a Trajectory Efficiency metric (rubric score per step) and find that even frontier agents achieve only 1.15%, marking an evident need for agents that can succeed efficiently and not simply eventually. Odysseys isolates the critical evaluation of long-horizon proficiency in open-web environments, providing a realistic benchmark to measure progress towards computer-use agents that can potentially productively operate for hours. We release our tasks, evaluation scripts, and other results at https://odysseys-website.pages.dev

---

### CoreFlow: Low-Rank Matrix Generative Models

Authors: Dongze Wu, Linglingzhi Zhu, Yao Xie
Published: 2026-04-27
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2604.24959v1

Learning matrix-valued distributions from high-dimensional and possibly incomplete training data is challenging: ambient-space generative modeling is computationally expensive and statistically fragile when the matrix dimension is large but the sample size is limited. We propose CoreFlow, a geometry-preserving low-rank flow model that learns shared row/column subspaces across the matrix distribution, and then trains a continuous normalizing flow only on the induced low-dimensional core. CoreFlow is designed for settings where shared low-rank matrix geometry is present, especially in high-dimensional limited-sample regimes. This separates shared matrix geometry from sample-specific variation, preserves matrix structure, and substantially improves training efficiency. The same framework also handles incomplete training matrices through masked Riemannian updates and iterative completion. Across real and synthetic benchmarks, CoreFlow substantially improves spectral and moment-level generation quality in few-sample regimes while remaining competitive in data-rich settings, even under compression to 9% of the ambient dimension and with up to 40% missing training entries.

---

### Compute Aligned Training: Optimizing for Test Time Inference

Authors: Adam Ousherovitch, Ambuj Tewari
Published: 2026-04-27
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2604.24957v2

Scaling test-time compute has emerged as a powerful mechanism for enhancing Large Language Model (LLM) performance. However, standard post-training paradigms, Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL), optimize the likelihood of individual samples under a base policy, creating a misalignment with test time procedures that rely on aggregated or filtered outputs. In this work, we propose Compute Aligned Training, which aligns training objectives with test-time strategies. By conceptualizing inference strategies as operators on the base policy, we derive new loss functions that maximize performance when said strategies are applied. We instantiate such loss functions for SFT and RL across common test time strategies. Finally, we provide empirical evidence that this training method substantially improves test time scaling over standard training.

---

### BenchGuard: Who Guards the Benchmarks? Automated Auditing of LLM Agent Benchmarks

Authors: Xinming Tu, Tianze Wang, Yingzhou, Lu, Kexin Huang, Yuanhao Qu, Sara Mostafavi
Published: 2026-04-27
Categories: cs.CL, cs.AI, cs.SE
arXiv: https://arxiv.org/abs/2604.24955v1

As benchmarks grow in complexity, many apparent agent failures are not failures of the agent at all - they are failures of the benchmark itself: broken specifications, implicit assumptions, and rigid evaluation scripts that penalize valid alternative approaches. We propose employing frontier LLMs as systematic auditors of evaluation infrastructure, and realize this vision through BenchGuard, the first automated auditing framework for task-oriented, execution-based agent benchmarks. BenchGuard cross-verifies all benchmark artifacts via structured LLM protocols, optionally incorporating agent solutions or execution traces as additional diagnostic evidence. Deployed on two prominent scientific benchmarks, BenchGuard identified 12 author-confirmed issues in ScienceAgentBench - including fatal errors rendering tasks unsolvable - and exactly matched 83.3% of expert-identified issues on the BIXBench Verified-50 subset, catching defects that prior human review missed entirely. A full audit of 50 complex bioinformatics tasks costs under USD 15, making automated benchmark auditing a practical and valuable complement to human review. These findings point toward AI-assisted benchmark development, where frontier models serve not only as subjects of evaluation but as active participants in validating the evaluation infrastructure itself.

---

### Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligence

Authors: NVIDIA, :, Amala Sanjay Deshmukh, Kateryna Chumachenko, Tuomas Rintamaki, Matthieu Le, Tyler Poon, Danial Mohseni Taheri, Ilia Karmanov, Guilin Liu, Jarno Seppanen, Arushi Goel et al.
Published: 2026-04-27
Categories: cs.LG, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2604.24954v2

We introduce Nemotron 3 Nano Omni, the latest model in the Nemotron multimodal series and the first to natively support audio inputs alongside text, images, and video. Nemotron 3 Nano Omni delivers consistent accuracy improvements over its predecessor, Nemotron Nano V2 VL, across all modalities, enabled by advances in architecture, training data and recipes. In particular, Nemotron 3 delivers leading results in real-world document understanding, long audio-video comprehension, and agentic computer use. Built on the highly efficient Nemotron 3 Nano 30B-A3B backbone, Nemotron 3 Nano Omni further incorporates innovative multimodal token-reduction techniques to deliver substantially lower inference latency and higher throughput than other models of similar size. We are releasing model checkpoints in BF16, FP8, and FP4 formats, along with portions of the training data and codebase to facilitate further research and development.

---

### ViPO: Visual Preference Optimization at Scale

Authors: Ming Li, Jie Wu, Justin Cui, Xiaojie Li, Rui Wang, Chen Chen
Published: 2026-04-27
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2604.24953v2

While preference optimization is crucial for improving visual generative models, how to effectively scale this paradigm remains largely unexplored. Current open-source preference datasets contain conflicting preference patterns, where winners excel in some dimensions but underperform in others. Naively optimizing on such noisy datasets fails to learn preferences, hindering effective scaling. To enhance robustness against noise, we propose Poly-DPO, which extends the DPO objective with an additional polynomial term that dynamically adjusts model confidence based on dataset characteristics, enabling effective learning across diverse data distributions. Beyond biased patterns, existing datasets suffer from low resolution, limited prompt diversity, and imbalanced distributions. To facilitate large-scale visual preference optimization by tackling data bottlenecks, we construct ViPO, a massive-scale preference dataset with 1M image pairs at 1024px across five categories and 300K video pairs at 720p+ across three categories. State-of-the-art generative models and diverse prompts ensure reliable preference signals with balanced distributions. Remarkably, when applying Poly-DPO to our high-quality dataset, the optimal configuration converges to standard DPO. This convergence validates dataset quality and Poly-DPO's adaptive nature: sophisticated optimization becomes unnecessary with sufficient data quality, yet remains valuable for imperfect datasets. We validate our approach across visual generation models. On noisy datasets like Pick-a-Pic V2, Poly-DPO achieves 6.87 and 2.32 gains over Diffusion-DPO on GenEval for SD1.5 and SDXL, respectively. For ViPO, models achieve performance far exceeding those trained on existing open-source preference datasets. These results confirm that addressing both algorithmic adaptability and data quality is essential for scaling visual preference optimization.

---

### Learning from Noisy Preferences: A Semi-Supervised Learning Approach to Direct Preference Optimization

Authors: Xinxin Liu, Ming Li, Zonglin Lyu, Yuzhang Shang, Chen Chen
Published: 2026-04-27
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2604.24952v2

Human visual preferences are inherently multi-dimensional, encompassing aesthetics, detail fidelity, and semantic alignment. However, existing datasets provide only single, holistic annotations, resulting in severe label noise: images that excel in some dimensions but are deficient in others are simply marked as winner or loser. We theoretically demonstrate that compressing multi-dimensional preferences into binary labels generates conflicting gradient signals that misguide Diffusion Direct Preference Optimization (DPO). To address this, we propose Semi-DPO, a semi-supervised approach that treats consistent pairs as clean labeled data and conflicting ones as noisy unlabeled data. Our method starts by training on a consensus-filtered clean subset, then uses this model as an implicit classifier to generate pseudo-labels for the noisy set for iterative refinement. Experimental results demonstrate that Semi-DPO achieves state-of-the-art performance and significantly improves alignment with complex human preferences, without requiring additional human annotation or explicit reward models during training. We will release our code and models at: https://github.com/L-CodingSpace/semi-dpo

---

### Learning from Noisy Preferences: A Semi-Supervised Learning Approach to Direct Preference Optimization

Authors: Xinxin Liu, Ming Li, Zonglin Lyu, Yuzhang Shang, Chen Chen
Published: 2026-04-27
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2604.24952v1

Human visual preferences are inherently multi-dimensional, encompassing aesthetics, detail fidelity, and semantic alignment. However, existing datasets provide only single, holistic annotations, resulting in severe label noise: images that excel in some dimensions but are deficient in others are simply marked as winner or loser. We theoretically demonstrate that compressing multi-dimensional preferences into binary labels generates conflicting gradient signals that misguide Diffusion Direct Preference Optimization (DPO). To address this, we propose Semi-DPO, a semi-supervised approach that treats consistent pairs as clean labeled data and conflicting ones as noisy unlabeled data. Our method starts by training on a consensus-filtered clean subset, then uses this model as an implicit classifier to generate pseudo-labels for the noisy set for iterative refinement. Experimental results demonstrate that Semi-DPO achieves state-of-the-art performance and significantly improves alignment with complex human preferences, without requiring additional human annotation or explicit reward models during training. We will release our code and models at: https://github.com/L-CodingSpace/semi-dpo

---

### Subjective Portrait Region Cropping in Landscape Videos with Temporal Annotation Smoothing

Authors: Cheng-Han Lee, Maniratnam Mandal, Neil Birkbeck, Yilin Wang, Balu Adsumilli, Alan C. Bovik
Published: 2026-04-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.24947v1

With the rise of mobile video consumption on diverse handheld display resolutions and orientation modes, altering videos to aspect ratios poses challenges. Static cropping and border padding often compromises visual quality, while warping may distort a video's intended meaning. Here we advocate for a more effective approach: cropping significant regions within video frames in a temporal manner, while minimizing distortion and preserving essential content. One barrier to solving this problem is the lack of sufficiently large-scale database devoted to informing these tasks. Towards filling this gap, we introduce the LIVE-YouTube Video Cropping (LIVE-YT VC) database, featuring 1800 videos, annotated by 90 human subjects. Using videos sourced from the YouTube-UGC and LSVQ Databases, this new resource is the largest publicly-available subjective video portrait region cropping database. We also introduce a post-processed version of the database, called LIVE-YT VC++, whereby a novel intra-frame temporal filter was deployed to smooth subjective annotations within each video. We demonstrate the usefulness of this new data resource using the SmartVidCrop algorithm and state-of-the-art video grounding models, in hopes of establishing our subjective dataset as a benchmark for future research. Our contributions offer a resource for advancing video aspect ratio transformation models towards ensuring that reshaped mobile-friendly video content retains its quality and meaning. Since our labels bear resemblances to video saliency annotations, we also conducted an additional analysis to explore the similarity between our labels and video saliency predictions. Finally, we repurposed state-of-the-art video grounding models for aspect ratio change tasks, and fine-tuned them on our dataset. As a service to the research community, we plan to open source the project.

---

### Independent-Component-Based Encoding Models of Brain Activity During Story Comprehension

Authors: Kamya Hari, Taha Binhuraib, Jin Li, Cory Shain, Anna A. Ivanova
Published: 2026-04-27
Categories: cs.CL, q-bio.NC
arXiv: https://arxiv.org/abs/2604.24942v2

Encoding models provide a powerful framework for linking continuous stimulus features to neural activity; however, traditional voxelwise approaches are limited by measurement noise, inter-subject variability, and redundancy arising from spatially correlated voxels encoding overlapping neural signals. Here, we propose an independent component (IC)-based encoding framework that dissociates stimulus-driven and noise-driven signals in fMRI data. We decompose continuous fMRI data from naturalistic story listening into ICs using one subset of the data, and train encoding models on independent data to predict IC time series from large language model representations of linguistic input. Across subjects, a subset of ICs exhibited consistently high predictivity. These ICs were spatially and temporally consistent across subjects and included cognitive networks known to respond during story listening (auditory and language). Auditory component time series were strongly correlated with acoustic stimulus features, highlighting the interpretability of identified component time series. Components identified as noise or motion-related artifacts by ICA-AROMA showed uniformly poor predictive performance, confirming that highly predicted components reflect genuine stimulus-related neural signals rather than confounds. Overall, IC-based encoding models enable analyses at the level of functional networks, accommodating the variability in network locations across individuals and providing interpretable results that are easy to compare across subjects. Code provided at: https://github.com/kamyahari/IC-Encoding-Models.git

---

### ADE: Adaptive Dictionary Embeddings -- Scaling Multi-Anchor Representations to Large Language Models

Authors: Orhan Demirci, Sezer Aptourachman, Aydın Kaya
Published: 2026-04-27
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2604.24940v2

Word embeddings are fundamental to natural language processing, yet traditional approaches represent each word with a single vector, creating representational bottlenecks for polysemous words and limiting semantic expressiveness. While multi-anchor representations have shown promise by representing words as combinations of multiple vectors, they have been limited to small-scale models due to computational inefficiency and lack of integration with modern transformer architectures. We introduce Adaptive Dictionary Embeddings (ADE), a framework that successfully scales multi-anchor word representations to large language models. ADE makes three key contributions: (1) Vocabulary Projection (VP), which transforms the costly two-stage anchor lookup into a single efficient matrix operation; (2) Grouped Positional Encoding (GPE), a novel positional encoding scheme where anchors of the same word share positional information, preserving semantic coherence while enabling anchor-level variation; and (3) context-aware anchor reweighting, which leverages self-attention to dynamically compose anchor contributions based on sequence context. We integrate these components into the Segment-Aware Transformer (SAT), which provides context-aware reweighting of anchor contributions at inference time. We evaluate ADE on AG News and DBpedia-14 text classification benchmarks. With 98.7% fewer trainable parameters than DeBERTa-v3-base, ADE surpasses DeBERTa on DBpedia-14 (98.06% vs. 97.80%) and approaches it on AG News (90.64% vs. 94.50%), while compressing the embedding layer over 40x -- demonstrating that multi-anchor representations are a practical and parameter-efficient alternative to single-vector embeddings in modern transformer architectures.

---

### Rethinking Layer Redundancy: Calibration Matters More Than Search in LLM Depth Pruning

Authors: Minkyu Kim, Vincent-Daniel Yun, Youngrae Kim, Suin Cho, Woosang Lim, Sunwoo Lee
Published: 2026-04-27
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2604.24938v3

Depth pruning improves the inference efficiency of large language models by removing Transformer blocks. Prior work typically treats layer redundancy as an inherent structural property of pretrained networks, emphasizing importance criteria and search algorithms to identify removable layers. In this study, we empirically investigate depth pruning from a functional perspective. Evaluating representative LLM families across diverse calibration configurations and multiple search algorithms, we show that different configurations produce different pruning patterns. Furthermore, under a fixed calibration configuration, complex search algorithms yield marginal performance improvements over simple one-shot methods, converging to similar pruned subsets. Overall, our results suggest that the calibration configuration plays a substantially larger role than the choice of search algorithm in shaping pruning patterns and calibration perplexity, while contributing comparably to variance in downstream reasoning accuracy. This indicates that future pruning efforts may benefit from prioritizing the calibration configuration over search complexity.

---

### A Unifying Framework for Unsupervised Concept Extraction

Authors: Chandler Squires, Pradeep Ravikumar
Published: 2026-04-27
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2604.24936v1

Techniques for concept extraction, such as sparse autoencoders and transcoders, aim to extract high-level symbolic concepts from low-level nonsymbolic representations. When these extracted concepts are used for downstream tasks such as model steering and unlearning, it is essential to understand their guarantees, or lack thereof. In this work, we present a unified theoretical framework for unsupervised concept extraction, in which we frame the task of concept extraction as identifying a generative model. We present a general meta-theorem for identifiability, which reduces the problem of establishing identifiability guarantees to the problem of characterizing the intersection of two sets. As we demonstrate on a range of widely-used approaches, this meta-theorem substantially simplifies the task of proving such guarantees, thus paving the way for the development of new, principled approaches for concept extraction.

---

### CAN-QA: A Question-Answering Benchmark for Reasoning over In-Vehicle CAN Traffic

Authors: Jing Chen, Abhijay Deevi, Onat Gungor, Tajana Rosing
Published: 2026-04-27
Categories: cs.CR, cs.LG
arXiv: https://arxiv.org/abs/2604.24935v1

The Controller Area Network (CAN) is a safety-critical in-vehicle communication protocol that lacks built-in security mechanisms, making intrusion detection essential. Existing approaches predominantly formulate CAN intrusion detection as a classification task, mapping complex traffic patterns to attack labels. However, this formulation abstracts away the temporal and relational structure of CAN traffic and misaligns with real-world forensic workflows, which require systematic reasoning about traffic behavior. To address this gap, we introduce CAN-QA, the first benchmark that reformulates CAN traffic analysis as a question-answering (QA) task. CAN-QA converts raw CAN logs into temporally segmented windows and applies deterministic rule-based templates to generate natural-language questions paired with automatically derived ground-truth answers. The resulting dataset comprises 33,128 QA pairs across 10 categories, each targeting distinct semantic and temporal properties of CAN traffic. Using CAN-QA, we evaluate large language models across both True/False and multiple-choice formats. Our results indicate that, although these models capture superficial statistical regularities, they struggle with temporal reasoning, multi-condition inference, and higher-level behavioral interpretation. Our code is available at https://github.com/Kriiiiss/CAN-QA.

---

### S-SONDO: Self-Supervised Knowledge Distillation for General Audio Foundation Models

Authors: Mohammed Ali El Adlouni, Aurian Quelennec, Pierre Chouteau, Geoffroy Peeters, Slim Essid
Published: 2026-04-27
Categories: cs.AI, cs.SD
arXiv: https://arxiv.org/abs/2604.24933v1

General audio foundation models have recently achieved remarkable progress, enabling strong performance across diverse tasks. However, state-of-the-art models remain extremely large, often with hundreds of millions of parameters, leading to high inference costs and limited deployability on edge devices. Knowledge distillation is a proven strategy for model compression, but prior work in audio has mostly focused on supervised settings, relying on class logits, intermediate features, or architecture-specific techniques. Such assumptions exclude models that output only embeddings, such as self-supervised or metric-learning models. We introduce S-SONDO (Self-Supervised KnOwledge DistillatioN for General AuDio FOundation Models), the first framework to distill general audio models using only their output embeddings. By avoiding the need for logits or layer-level alignment, S-SONDO is architecture-agnostic and broadly applicable to embedding-based teachers. We demonstrate its effectiveness by distilling two audio foundation models into three efficient students that are up to 61 times smaller while retaining up to 96% of teacher performance. We also provide practical insights on loss choice and clustering-based balanced data sampling. Code is available here: https://github.com/MedAliAdlouni/ssondo.

---

### GAIA-v2-LILT: Multilingual Adaptation of Agent Benchmark beyond Translation

Authors: Yunsu Kim, Kaden Uhlig, Joern Wuebker
Published: 2026-04-27
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2604.24929v1

Agent benchmarks remain largely English-centric, while their multilingual versions are often built with machine translation (MT) and limited post-editing. We argue that, for agentic tasks, this minimal workflow can easily break benchmark validity through query-answer misalignment or culturally off-target context. We propose a refined workflow for adapting English benchmarks into multiple languages with explicit functional alignment, cultural alignment, and difficulty calibration using both automated checks and human review. Using this workflow, we introduce GAIA-v2-LILT, a re-audited multilingual extension of GAIA covering five non-English languages. In experiments, our workflow improves agent success rates by up to 32.7% over minimally translated versions, bringing the closest audited setting to within 3.1% of English performance while substantial gaps remain in many other cases. This indicates that a substantial share of the multilingual performance gap is benchmark-induced measurement error, motivating task-level alignment when adapting English benchmarks across languages. The data is available as part of the MAPS package at https://huggingface.co/datasets/Fujitsu-FRE/MAPS/viewer/GAIA-v2-LILT. We also release the code used in our experiments at https://github.com/lilt/gaia-v2-lilt.

---

### Large Language Models Explore by Latent Distilling

Authors: Yuanhao Zeng, Ao Lu, Lufei Li, Zheng Zhang, Yexin Li, Kan Ren
Published: 2026-04-27
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2604.24927v2

Generating diverse responses is crucial for test-time scaling of large language models (LLMs), yet standard stochastic sampling mostly yields surface-level lexical variation, limiting semantic exploration. In this paper, we propose Exploratory Sampling (ESamp), a decoding approach that explicitly encourages semantic diversity during generation. ESamp is motivated by the well-known observation that neural networks tend to make lower-error predictions on inputs similar to those encountered before, and incur higher prediction error on novel ones. Building on this property, we train a lightweight Distiller at test time to predict deep-layer hidden representations of the LLM from its shallow-layer representations to model the LLM's depth-wise representation transitions. During decoding, the Distiller continuously adapts to the mappings induced by the current generation context. ESamp uses the prediction error as a novelty signal to reweight candidate token extensions conditioned on the current prefix, thereby biasing decoding toward less-explored semantic patterns. ESamp is implemented with an asynchronous training--inference pipeline, with less than 5% worst case overhead (1.2% in the optimized release). Empirical results show that ESamp significantly boosts the Pass@k efficiency of reasoning models, showing superior or comparable performance to strong stochastic and heuristic baselines. Notably, ESamp achieves robust generalization across mathematics, science, and code generation benchmarks and breaks the trade-off between diversity and coherence in creative writing. Our code has released at: https://github.com/LinesHogan/tLLM.

---

### Libra-VLA: Achieving Learning Equilibrium via Asynchronous Coarse-to-Fine Dual-System

Authors: Yifei Wei, Linqing Zhong, Yi Liu, Yuxiang Lu, Xindong He, Maoqing Yao, Guanghui Ren
Published: 2026-04-27
Categories: cs.RO, cs.AI, cs.CL, cs.CV
arXiv: https://arxiv.org/abs/2604.24921v1

Vision-Language-Action (VLA) models are a promising paradigm for generalist robotic manipulation by grounding high-level semantic instructions into executable physical actions. However, prevailing approaches typically adopt a monolithic generation paradigm, directly mapping visual-linguistic features to high-frequency motor commands in a flat, non-hierarchical fashion. This strategy overlooks the inherent hierarchy of robotic manipulation, where complex actions can be naturally modeled in a Hybrid Action Space, decomposing into discrete macro-directional reaching and continuous micro-pose alignment, severely widening the semantic-actuation gap and imposing a heavy representational burden on grounding high-level semantics to continuous actions. To address this, we introduce Libra-VLA, a novel Coarse-to-Fine Dual-System VLA architecture. We explicitly decouple the learning complexity into a coarse-to-fine hierarchy to strike a training equilibrium, while simultaneously leveraging this structural modularity to implement an asynchronous execution strategy. The Semantic Planner predicts discrete action tokens capturing macro-directional intent, while the Action Refiner conditions on coarse intent to generate high-frequency continuous actions for precise alignment. Crucially, our empirical analysis reveals that performance follows an inverted-U curve relative to action decomposition granularity, peaking exactly when the learning difficulty is balanced between the two sub-systems. With the asynchronous design, our approach offers a scalable, robust, and responsive solution for open-world manipulation.

---

### SUDP: Secret-Use Delegation Protocol for Agentic Systems

Authors: Xiaohang Yu, Hejia Geng, Xinmeng Zeng, William Knottenbelt
Published: 2026-04-27
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2604.24920v3

Agentic systems increasingly act with user secrets for APIs, messaging platforms, and cloud services. Today's agent runtimes typically implement authorization by exposure: enabling action often means placing a reusable secret, or a reusable artifact derived from it, inside the runtime, so a transient prompt-injection or tool-side compromise becomes durable account compromise. Existing defenses cover adjacent pieces such as secret storage, scoped delegation, sender-constrained tokens, and runtime monitoring, but leave the combined agentic obligation without a common specification: an untrusted autonomous requester should be able to cause a user-authorized secret-backed operation without gaining reusable authority over it. We formalize this as the Agent Secret Use (ASU) problem and identify seven security properties any solution must satisfy, spanning authorization integrity and secret confidentiality. We propose the Secret-Use Delegation Protocol (SUDP), in which a requester proposes a canonical operation, the user authorizes it with a fresh authenticator-backed grant, and a custodian redeems the grant to perform the bounded use; reusable authority never crosses the requester boundary. We specialize SUDP for LLM-driven agents, where it applies whenever a tool call would exercise user-enrolled authority-bearing material. Under standard cryptographic assumptions, SUDP satisfies all seven properties when integrated with a hardware-rooted runtime. A reference implementation is available at https://github.com/xhyumiracle/sudp.

---

### Agentic AI for Remote Sensing: Technical Challenges and Research Directions

Authors: Muhammad Akhtar Munir, Muhammad Umer Sheikh, Akashah Shabbir, Muhammad Haris Khan, Fahad Khan, Xiao Xiang Zhu, Begüm Demir, Salman Khan
Published: 2026-04-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.24919v3

Earth Observation (EO) is moving beyond static prediction toward multi-step analytical workflows that require coordinated reasoning over data, tools, and geospatial state. While foundation models and vision-language models have advanced representation learning and language-grounded interaction in remote sensing, and agentic AI has shown strong potential for long-horizon reasoning and tool use, EO is not a straightforward extension of generic agentic AI. EO workflows operate on georeferenced, multi-modal, and temporally structured data, where operations such as reprojection, resampling, compositing, and aggregation transform the underlying state and can constrain later analysis. As a result, errors may propagate silently across steps, and correctness depends not only on internal coherence but also on geospatial consistency, temporally valid comparisons, and physical validity. This position paper argues that these challenges are structural rather than incidental. We examine the assumptions commonly made in generic agentic systems, analyze how they break in geospatial workflows, and characterize failure modes in multi-step EO pipelines. We then outline design principles for EO-native agents centered on structured geospatial state, tool-aware reasoning, verifier-guided execution, and validity-aware learning and evaluation. Building reliable geospatial agents, therefore, requires rethinking agent design around the physical, geospatial, and workflow constraints that govern EO analysis.

---

### asRoBallet: Closing the Sim2Real Gap via Friction-Aware Reinforcement Learning for Underactuated Spherical Dynamics

Authors: Fang Wan, Guangyi Huang, Tianyu Wu, Zishang Zhang, Bangchao Huang, Haoran Sun, Mingdong Chen, Chaoyang Song
Published: 2026-04-27
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2604.24916v2

We introduce asRoBallet, to the best of our knowledge, the first end-to-end reinforcement learning (RL) locomotion policy deployed on a humanoid ballbot hardware platform. Historically, ballbots have served as a canonical benchmark for underactuated and nonholonomic control, which are characterized by a reality gap in complex friction models for wheel-ball-floor interactions. While current literature demonstrates successful handling of 3D balancing with LQR and MPC, transitioning to actual hardware for a humanoid ballbot using RL is currently hindered by critical gaps in contact modeling, actuator latency & jitter, and safe hardware exploration. This study proposes a high-fidelity MuJoCo simulation that explicitly models the discrete roller mechanics of ETH-type omni-wheels, thereby capturing parasitic vibrations and contact discontinuities that have previously been ignored. We also developed a Friction-Aware Reinforcement Learning framework that achieves zero-shot Sim2Real transfer by mastering the coupled rolling, lateral, and torsional friction channels at the wheel-ball and ball-floor interfaces. We designed asRoBallet through subtractive reconfiguration, repurposing key components from an overconstrained quadruped and integrating them into a newly designed structural frame to achieve a robust research platform at low cost. We also developed a generalized iOS ecosystem that transforms consumer electronics into a low-latency interface, enabling a single operator to orchestrate expressive humanoid maneuvers via intuitive natural motion.

---

### Generative diffusion models for spatiotemporal influenza forecasting

Authors: Joseph Lemaitre, Justin Lessler
Published: 2026-04-27
Categories: cs.LG, q-bio.PE
arXiv: https://arxiv.org/abs/2604.24913v2

Forecasting infectious disease incidence can provide important information to guide public health planning, yet it is difficult because epidemic dynamics are complex. Current mechanistic and statistical approaches often struggle to capture multimodal uncertainty or emergent trends. Influpaint adapts denoising diffusion probabilistic models to epidemic forecasting. By encoding influenza seasons as spatiotemporal images in which pixel intensity represents incidence, Influpaint learns a rich distribution of disease dynamics from a hybrid dataset of surveillance and simulated trajectories. Forecasting is formulated as a conditional generation (inpainting) task from partial observations. We show that Influpaint generates realistic, diverse epidemic trajectories and achieves forecast accuracy that is competitive with leading ensemble methods in retrospective evaluation. In real-time evaluation during the 2023--2025 U.S. CDC FluSight challenges, performance improved substantially across seasons, with highly accurate but somewhat overconfident projections in 2023--2025. The best performance was achieved with a training dataset containing 30% surveillance and 70% simulated trajectories. These results show that diffusion models can capture important spatiotemporal structure in influenza dynamics and provide a flexible framework for probabilistic infectious disease forecasting.

---

### Generative diffusion models for spatiotemporal influenza forecasting

Authors: Joseph Lemaitre, Justin Lessler
Published: 2026-04-27
Categories: cs.LG, q-bio.PE
arXiv: https://arxiv.org/abs/2604.24913v1

Forecasting infectious disease incidence can provide important information to guide public health planning, yet is difficult because epidemic dynamics are complex. Current mechanistic and statistical approaches often struggle to capture multimodal uncertainty or emergent trends. Influpaint adapts denoising diffusion probabilistic models to epidemic forecasting. By encoding influenza seasons as spatiotemporal images in which pixel intensity represents incidence, Influpaint learns a rich distribution of disease dynamics from a hybrid dataset of surveillance and simulated trajectories. Forecasting is formulated as a conditional generation (inpainting) task from partial observations. We show that Influpaint generates realistic, diverse epidemic trajectories and achieves forecast accuracy that is competitive with leading ensemble methods in retrospective evaluation. In real-time evaluation during the 2023--2025 U.S. CDC FluSight challenges, performance improved substantially across seasons, with highly accurate but somewhat overconfident projections in 2024--2025. The best performance was achieved with a training dataset containing 30% surveillance and 70% simulated trajectories. These results show that diffusion models can capture important spatiotemporal structure in influenza dynamics and provide a flexible framework for probabilistic infectious disease forecasting.

---

### Data-Driven Hamiltonian Reduction for Superconducting Qubits via Meta-Learning

Authors: Arielle Sanford, Andrew T. Kamen, Frederic T. Chong, Andy J. Goldschmidt
Published: 2026-04-27
Categories: quant-ph, cs.LG
arXiv: https://arxiv.org/abs/2604.24912v1

We introduce HAML (Hamiltonian Adaptation via Meta-Learning), a framework for fast online adaptation of effective Hamiltonian models of superconducting quantum processors. HAML proceeds in two phases. A supervised training phase uses an ensemble of simulated devices to learn an offline map from control inputs and device parameters to effective Hamiltonian coefficients. An online adaptation phase then uses a small number of hardware-accessible measurements to identify the unknown parameters of a new device. By training directly against effective two-qubit coefficients extracted from full multi-mode simulations, HAML implicitly learns the reduction from full multi-mode Hamiltonians to effective qubit descriptions without invoking perturbation theory. We further show that a variance-maximizing greedy selection of measurement configurations boosts online adaptation efficiency. We demonstrate HAML on a transmon-coupler-transmon system, recovering effective two-qubit coefficients across a wide range of operating regimes, including parameter regions where Schrieffer-Wolff perturbation theory (SWPT) breaks down. This establishes a scalable, sample-efficient approach to Hamiltonian reduction and characterization for near-term quantum processors, with direct implications for calibration, control, and error mitigation.

---

### Learning with Embedded Linear Equality Constraints via Variational Bayesian Inference

Authors: Matthew Marsh, Benoît Chachuat, Antonio del Rio Chanona
Published: 2026-04-27
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2604.24911v1

Machine Learning is becoming more prevalent in science and engineering, but many approaches do not provide meaningful uncertainty estimates and predictions may also violate known physical knowledge. We propose a Bayesian framework to embed linear relationships across inputs and outputs into the learning process, whilst characterizing full predictive uncertainty over both the model parameters and the domain knowledge. We evaluated our method on learning the single particle battery model subject to voltage and energy balances, showing its ability to provide reduced credible intervals and constraint violations compared to standard Bayesian neural networks based on variational inference.

---

### Contrastive Image-Metadata Pre-Training for Materials Transmission Electron Microscopy

Authors: Georgia Channing, Debora Keller, Marta D. Rossell, Philip Torr, Stig Helveg, Henrik Eliasson
Published: 2026-04-27
Categories: cs.LG, cs.CE
arXiv: https://arxiv.org/abs/2604.24909v2

The transmission electron microscope facilitates the highest-resolution imaging of any instrument ever created, and its limiting factor is no longer spatial resolution but dose efficiency. Low electron doses avoid sample damage but produce noisy images for which, unlike in classical computer vision, there is no ground truth. Autonomous materials experimentation poses a related problem, since closed-loop instruments need representations grounded in the microscope state at acquisition. Both demand representations grounded in how an image was acquired. We release 7,330 paired high-angle annular dark-field scanning-TEM (HAADF-STEM) images and their seven-dimensional acquisition metadata, and propose Contrastive Image-Metadata Pre-training (CIMP), a CLIP-style encoder that aligns the two modalities and reaches 84.4% Top-1 cross-modal retrieval on a held-out split. All seven parameters are individually recoverable from the frozen visual embedding through a linear probe, and we use the embedding to condition a metadata-conditioned style-transfer model that re-renders experimental images under different acquisition parameters. Virtually scaling dwell time and beam current of low-dose images turns this model into a physics-informed denoiser; in a blind user study, experimental microscopists prefer it over the current state-of-the-art denoiser for STEM imagery on 70.2% of trials.

---

### An analysis of sensor selection for fruit picking with suction-based grippers

Authors: Eva Krueger, Marcus Rosette, Joseph R. Davidson
Published: 2026-04-27
Categories: cs.RO, cs.LG, eess.SY
arXiv: https://arxiv.org/abs/2604.24906v1

Robotic fruit harvesting often fails to reliably detect whether a fruit has been successfully picked, limiting efficiency and increasing crop damage. This problem is difficult due to compliant fruit and grippers, variable stem attachment, and occlusions in orchard environments. Prior work has explored vision-based perception and multi-sensor learning approaches for pick state estimation. However, minimal sensor sets and phase-dependent sensing strategies for accurate pick and slip detection remain largely unexplored. In this work, we design and evaluate a multimodal sensing suite integrated into a compliant suction-based apple gripper. Our approach is unique because it identifies which sensors are most informative at different phases of the pick, enabling predictive detection of failures before they occur. The contributions of this paper are a phase-dependent evaluation of multimodal sensors and the identification of minimal sensor sets for reliable pick state classification. Experiments in a real apple orchard show that Random Forest and Multilayer Perceptron classifiers detect successful picks and impending failures with over 90% accuracy, and Random Forest predicts pick/slip events within 0.09 s of human-annotated ground truth.

---

### MultiHedge: Adaptive Coordination via Retrieval-Augmented Control

Authors: Feliks Bańka, Jarosław A. Chudziak
Published: 2026-04-27
Categories: cs.MA, cs.AI
arXiv: https://arxiv.org/abs/2604.24905v1

Decision-making under changing conditions remains a fundamental challenge in many real-world systems. Existing approaches often fail to generalize across shifting regimes and exhibit unstable behavior under uncertainty. This raises the research question: can retrieval-augmented LLM coordination improve the robustness of modular decision pipelines? We propose MultiHedge, a hybrid architecture where an LLM produces structured allocation decisions conditioned on retrieved historical precedents, and execution is grounded in canonical option strategies. In a controlled evaluation using U.S. equities, we compare MultiHedge to rule-based and learning-based baselines. The key result is that memory-augmented retrieval confers greater robustness and stability than increasing model scale alone. Our paper contributes a controlled computational study showing that memory and architectural design play a central role in robustness in modular decision systems.

---

### VISION-SLS: Safe Perception-Based Control from Learned Visual Representations via System Level Synthesis

Authors: Antoine P. Leeman, Shuyu Zhan, Melanie N. Zeilinger, Glen Chou
Published: 2026-04-27
Categories: cs.RO, cs.CV, cs.LG, eess.SY, math.OC
arXiv: https://arxiv.org/abs/2604.24894v1

We propose VISION-SLS, a method for nonlinear output-feedback control from high-resolution RGB images which provides robust constraint satisfaction guarantees under calibrated uncertainty bounds despite partial observability, sensor noise, and nonlinear dynamics. To enable scalability while retaining guarantees, we propose: (i) a learned low-dimensional observation map from pretrained visual features with state-dependent error bounds, and (ii) a causal affine time-varying output-feedback policy optimized via System Level Synthesis (SLS). We develop a scalable, novel solver for the resulting nonconvex program that leverages sequential convex programming coupled with efficient Riccati recursions. On two simulated visuomotor tasks (a 4D car and a 10D quadrotor) with >= 512 x 512 pixels and a 59D humanoid task with partial observability, our method enables safe, information-gathering behavior that reduces uncertainty while guaranteeing constraint satisfaction with empirically-calibrated error bounds. We also validate our method on hardware, safely controlling a ground vehicle from onboard images, outperforming baselines in safety rate and solve times. Together, these results show that learned visual abstractions coupled with an efficient solver make SLS-based safe visuomotor output-feedback practical at scale. The code implementation of our method is available at https://github.com/trustworthyrobotics/VISION-SLS.

---

### Interactive Episodic Memory with User Feedback

Authors: Nikesh Subedi, Loris Bazzani, Ziad Al-Halah
Published: 2026-04-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.24893v1

In episodic memory with natural language queries (EM-NLQ), a user may ask a question (e.g., "Where did I place the mug?") that requires searching a long egocentric video, captured from the user's perspective, to find the moment that answers it. However, queries can be ambiguous or incomplete, leading to incorrect responses. Current methods ignore this key aspect and address EM-NLQ in a one-shot setup, limiting their applicability in real-world scenarios. In this work, we address this gap and introduce the Episodic Memory with Questions and Feedback task (EM-QnF). Here, the user can provide feedback on the model's initial prediction or add more information (e.g., "Before this. I'm looking for the big blue mug not the white one"), helping the model refine its predictions interactively. To this end, we collect datasets for feedback-based interaction and propose a lightweight training scheme that avoids expensive sequential optimization. We also introduce a plug-and-play Feedback ALignment Module (FALM) that enables existing EM-NLQ models to incorporate user feedback effectively. Our approach significantly improves over the state of the art on three challenging benchmarks and is better than or competitive with commercial large vision-language models while remaining efficient. Evaluation with human-generated feedback shows that it generalizes well to real-world scenarios.

---