# Research Papers — 2026-03 · bundle page 19 of 243

> Full-text markdown of papers #901–950 of 12123 published in 2026-03, newest first. AI-readable mirrors of the full 12123-paper 2026-03 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2026-03/page/19
[← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/18.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-03/page/20.md)

### Do Small Language Models Know When They're Wrong? Confidence-Based Cascade Scoring for Educational Assessment

Authors: Tyler Burleigh
Published: 2026-03-29
Categories: cs.CY, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2604.19781v1

Automated scoring of student work at scale requires balancing accuracy against cost and latency. In "cascade" systems, small language models (LMs) handle easier scoring tasks while escalating harder ones to larger LMs -- but the challenge is determining which cases to escalate. We explore verbalized confidence -- asking the LM to state a numerical confidence alongside its prediction -- as a routing signal. Using 2,100 expert-scored decisions from student-AI math conversations, we evaluate cascade systems built from GPT-5.4, Claude 4.5+, and Gemini 3.1 model pairs. We find that: (1) confidence discrimination varies widely across small LMs, with the best achieving AUROC 0.857 and the worst producing a near-degenerate confidence distribution; (2) confidence tracks human scoring difficulty, with lower LM confidence where annotators disagreed and took longer to score; (3) the best cascade approached large-LM accuracy (kappa 0.802 vs. 0.819) at 76% lower cost and 61% lower latency. Confidence discrimination is the bottleneck: the two small LMs with meaningful confidence variance yielded cascades with no statistically detectable kappa loss, while the third -- whose confidence was near-degenerate -- could not close the accuracy gap regardless of threshold. Small LMs with strong discrimination let practitioners trade cost for accuracy along the frontier; those without it do not.

---

### Avoiding Overthinking and Underthinking: Curriculum-Aware Budget Scheduling for LLMs

Authors: Amirul Rahman, Aisha Karim, Kenji Nakamura, Yi-Fan Ng
Published: 2026-03-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.19780v1

Scaling test-time compute via extended reasoning has become a key paradigm for improving the capabilities of large language models (LLMs). However, existing approaches optimize reasoning under fixed or uniformly sampled token budgets, ignoring the fundamental mismatch between problem difficulty and allocated compute. This leads to overthinking on easy problems and underthinking on hard ones, resulting in suboptimal token efficiency across diverse reasoning scenarios. In this paper, we propose Budget-Adaptive Curriculum Reasoning (BCAE), a unified framework that jointly optimizes reasoning quality and token efficiency through three synergistic components: (1) a \emph{budget-conditioned unified policy} that embeds the token budget as a continuous conditioning signal, eliminating the need for decoupled thinking and summarization strategies; (2) a \emph{curriculum-aware budget scheduler} that adaptively shifts the training budget distribution from easy to hard problems based on real-time learning progress; and (3) a \emph{truncation-aware dense reward} mechanism that provides fine-grained credit assignment at intermediate reasoning steps via process-level verification. We further introduce \emph{Budget-Conditioned Advantage Estimation} (BCAE), a novel variance reduction technique that conditions the advantage baseline on the sampled budget, yielding more stable policy gradients. Experiments on mathematical reasoning benchmarks (MATH, GSM8K, AIME, and Minerva Math) demonstrate that BACR consistently outperforms other strong baselines across all token budgets, achieving up to 8.3\% accuracy improvement under tight budgets while reducing average token consumption by 34\% compared to unconstrained reasoning.

---

### ESGLens: An LLM-Based RAG Framework for Interactive ESG Report Analysis and Score Prediction

Authors: Tsung-Yu Yang, Meng-Chi Chen
Published: 2026-03-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.19779v1

Environmental, Social, and Governance (ESG) reports are central to investment decision-making, yet their length, heterogeneous content, and lack of standardized structure make manual analysis costly and inconsistent. We present ESGLens, a proof-of-concept framework combining retrieval-augmented generation (RAG) with prompt-engineered extraction to automate three tasks: (1)~structured information extraction guided by Global Reporting Initiative (GRI) standards, (2)~interactive question-answering with source traceability, and (3)~ESG score prediction via regression on LLM-generated embeddings. ESGLens is purpose-built for the domain: a report-processing module segments heterogeneous PDF content into typed chunks (text, tables, charts); a GRI-guided extraction module retrieves and synthesizes information aligned with specific standards; and a scoring module embeds extracted summaries and feeds them to a regression model trained against London Stock Exchange Group (LSEG) reference scores. We evaluate the framework on approximately 300 reports from companies in the QQQ, S\&P~500, and Russell~1000 indices (fiscal year 2022). Among three embedding methods (ChatGPT, BERT, RoBERTa) and two regressors (Neural Network, LightGBM), ChatGPT embeddings with a Neural Network achieve a Pearson correlation of 0.48 ($R^{2} \approx 0.23$) against LSEG ground-truth scores -- a modest but statistically meaningful signal given the ${\sim}300$-report training set and restriction to the environmental pillar. A traceability audit shows that 8 of 10 extracted claims verify against the source document, with two failures attributable to few-shot example leakage. We discuss limitations including dataset size and restriction to environmental indicators, and release the code to support reproducibility.

---

### Instructor-Created Custom GPTs as Pedagogical Partners Fostering Immersion in Online Higher Education: Two Case Studies

Authors: Dennis Beck, Leonel Morgado
Published: 2026-03-29
Categories: cs.CY, cs.AI, cs.HC
arXiv: https://arxiv.org/abs/2604.16397v1

As online higher education expands, sustaining student engagement remains a critical challenge. This paper approaches immersive learning by investigating how custom GPTs foster immersion (as a state of deep mental involvement) for students and instructors. While large language models (LLMs) offer potential for enhancing feedback, little research has examined instructor-created custom GPTs designed to align with specific pedagogical goals. This paper addresses this gap, employing the Immersive Learning Cube framework, which conceptualizes immersion through three dimensions: system (envelopment by the environment), narrative (meaningful context), and agency (commitment to meaning-making). Through a qualitative analysis of two distinct case studies, an accelerated graduate grant writing course in the US and an undergraduate software engineering course in Portugal, we analyze course-embedded artifacts to map how custom GPTs influence these immersion dimensions. In the grant writing course, the custom GPT functioned as a feedback partner, fostering system immersion through its immediacy, narrative immersion by reinforcing the proposal's evolving story, and agency immersion by empowering students to negotiate feedback and take ownership of revisions. In the software engineering course, a diegetically-framed custom GPT acted as a metacognitive tutor, enhancing system immersion via its permanent availability, narrative immersion through its role-play function and agency immersion by scaffolding students' self- and co-regulated learning. Our findings demonstrate that thoughtfully integrated custom GPTs can act as powerful pedagogical partners that leverage all three dimensions of immersion. Rather than replacing human instructors, they can amplify immediacy, coherence, and learner autonomy, creating more engaging and immersive online learning environments.

---

### QU-NLP at QIAS 2026: Multi-Stage QLoRA Fine-Tuning for Arabic Islamic Inheritance Reasoning

Authors: Mohammad AL-Smadi
Published: 2026-03-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2604.16396v1

Islamic inheritance law (ilm al-mawarıth) presents a challenging domain for evaluating large language models' structured reasoning capabilities, requiring multi-step legal analysis, rule-based blocking decisions, and precise fractional calculations. We present QU-NLP's submission to the QIAS 2026 shared task on Arabic Islamic inheritance reasoning. Our approach employs a multi-stage Quantized Low-Rank Adaptation (QLoRA) fine-tuning strategy on Qwen3-4B: (1) domain adaptation on 3,166 Islamic fatwa records to acquire inheritance terminology and jurisprudential reasoning patterns, followed by (2) task-specific training on 12,000 structured inheritance cases to optimize JSON-formatted output generation. Using 4-bit NF4 quantization with rank-128 LoRA adapters, our model achieves 90% MIR-E (Mawarith Inheritance Reasoning Evaluation) score on the test set, demonstrating competitive performance while requiring minimal computational resources. Our results show that domain-specific pre-adaptation combined with structured output training enables small language models to perform complex legal reasoning tasks effectively comparing to commercial systems such as Gemini-2.5-flash.

---

### Stream2LLM: Overlap Context Streaming and Prefill for Reduced Time-to-First-Token (TTFT)

Authors: Rajveer Bachkaniwala, Chengqi Luo, Richard So, Divya Mahajan, Kexin Rong
Published: 2026-03-29
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2604.16395v3

Context retrieval systems for LLM inference face a critical challenge: high retrieval latency creates a fundamental tension between waiting for complete context (poor time-to-first-token) and proceeding without it (reduced quality). Streaming context incrementally--overlapping retrieval with inference--can mitigate this latency, but doing so with concurrent requests introduces new challenges: requests contend for GPU compute and memory, and scheduling must adapt to dynamic context arrivals. We present Stream2LLM, a streaming-aware LLM serving system for concurrent prefill-decode disaggregated deployments. Stream2LLM introduces adaptive scheduling and preemption for two distinct retrieval patterns: append-mode (progressive context accumulation) and update-mode (iterative refinement with cache invalidation). It decouples scheduling decisions from resource acquisition, enabling flexible preemption strategies guided by hardware-specific cost models, and uses longest common prefix matching to minimize redundant computation when input changes dynamically. To evaluate Stream2LLM, we collect two large-scale, real-world streaming workloads based on web crawling and approximate nearest neighbor search. Our evaluation demonstrates that streaming architecture delivers up to 11x TTFT improvements, with cost-aware scheduling providing critical benefits under memory pressure, all while maintaining throughput parity with non-streaming baselines. Code: https://github.com/rajveerb/stream2llm/tree/mlsys_artifact

---

### Dynamic Forecasting and Temporal Feature Evolution of Stock Repurchases in Listed Companies Using Attention-Based Deep Temporal Networks

Authors: Xiang Ao, Jingxuan Zhang, Xinyu Zhao
Published: 2026-03-29
Categories: q-fin.ST, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2604.09650v1

Accurately predicting stock repurchases is crucial for quantitative investment and risk management, yet traditional static models fail to capture the complex temporal dependencies of corporate financial conditions. This paper proposes a dynamic early warning system integrating economic theory with deep temporal networks. Using Chinese A-share panel data (2014-2024), we employ a hybrid Temporal Convolutional Network (TCN) and Attention-based LSTM to capture long- and short-term financial evolutionary patterns. Rolling-window cross-validation demonstrates our model significantly outperforms static baselines like Logistic Regression and XGBoost. Furthermore, utilizing Explainable AI (XAI), we reveal the temporal dynamics of repurchase decisions: prolonged "undervaluation" serves as the long-term underlying motive, while a sharp increase in "cash flow" acts as the decisive short-term trigger. This study provides a robust deep learning paradigm for financial forecasting and offers dynamic empirical support for classic corporate finance hypotheses.

---

### WearBCI Dataset: Understanding and Benchmarking Real-World Wearable Brain-Computer Interfaces Signals

Authors: Haoxian Liu, Hengle Jiang, Lanxuan Hong, Xiaomin Ouyang
Published: 2026-03-29
Categories: cs.HC, cs.AI, eess.SP
arXiv: https://arxiv.org/abs/2604.09649v1

Brain-computer interfaces (BCIs) have opened new platforms for human-computer interaction, medical diagnostics, and neurorehabilitation. Wearable BCI systems, which typically employ non-invasive electrodes for portable monitoring, hold great promise for real-world applications, but also face significant challenges of signal quality degradation caused by motion artifacts and environmental interferences. Most existing wearable BCI datasets are collected under stationary or controlled lab settings, limiting their utility for evaluating performance under body movement. To bridge this gap, we introduce WearBCI, the first dataset that comprehensively evaluates wearable BCI signals under different motion dynamics with synchronized multimodal recordings (EEG, IMU, and egocentric video), and systematic benchmark evaluations for studying impacts of motion artifact. Specifically, we collect data from 36 participants across different motion dynamics, including body movements, walking, and navigation. This dataset includes synchronized electroencephalography (EEG), inertial measurement unit (IMU) data, and egocentric video recordings. We analyze the collected wearable EEG signals to understand the impact of motion artifacts across different conditions, and benchmark representative EEG signal enhancement techniques on our dataset. Furthermore, we explore two new case studies: cross-modal EEG signal enhancement and multi-dimension human behavior understanding. These findings offer valuable insights into real-world wearable BCI deployment and new applications.

---

### Multivariate Time Series Anomaly Detection via Dual-Branch Reconstruction and Autoregressive Flow-based Residual Density Estimation

Authors: Jun Liu, Ying Chen, Ziqian Lu, Qinyue Tong, Jun Tang
Published: 2026-03-29
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2604.08582v1

Multivariate Time Series Anomaly Detection (MTSAD) is critical for real-world monitoring scenarios such as industrial control and aerospace systems. Mainstream reconstruction-based anomaly detection methods suffer from two key limitations: first, overfitting to spurious correlations induced by an overemphasis on cross-variable modeling; second, the generation of misleading anomaly scores by simply summing up multivariable reconstruction errors, which makes it difficult to distinguish between hard-to-reconstruct samples and genuine anomalies. To address these issues, we propose DBR-AF, a novel framework that integrates a dual-branch reconstruction (DBR) encoder and an autoregressive flow (AF) module. The DBR encoder decouples cross-variable correlation learning and intra-variable statistical property modeling to mitigate spurious correlations, while the AF module employs multiple stacked reversible transformations to model the complex multivariate residual distribution and further leverages density estimation to accurately identify normal samples with large reconstruction errors. Extensive experiments on seven benchmark datasets demonstrate that DBR-AF achieves state-of-the-art performance, with ablation studies validating the indispensability of its core components.

---

### Probabilistic Language Tries: A Unified Framework for Compression, Decision Policies, and Execution Reuse

Authors: Gregory Magarshak
Published: 2026-03-29
Categories: cs.LG, cs.AI, cs.CL, cs.DS, cs.IR, cs.IT
arXiv: https://arxiv.org/abs/2604.06228v1

We introduce probabilistic language tries (PLTs), a unified representation that makes explicit the prefix structure implicitly defined by any generative model over sequences. By assigning to each outgoing edge the conditional probability of the corresponding token or action, a PLT simultaneously serves as: (i) an optimal lossless compressor via frequency-weighted interval encoding, generalizing arithmetic coding to model-conditioned distributions; (ii) a policy representation for sequential decision problems including games, search, and robotic control; and (iii) a memoization index that lets repeated inference queries be answered by structured retrieval rather than full model execution. The central technical result is a prior-guided caching theorem: under a stationary generative distribution, a PLT-guided cache achieves strictly lower expected inference cost than any empirical-frequency cache for all query counts below a threshold that grows with the concentration of the prior. This converts O(n^2) transformer attention cost into an expected cost of p_r * O(log N) + (1 - p_r) * O(n^2), where p_r is the prior-estimated reuse probability and N is the artifact store size. We further introduce a hybrid compression architecture decomposing any dataset into a PLT-covered majority and a sparse residual store, connecting arithmetic coding with Kolmogorov-style program representations and rate-distortion theory. We instantiate the framework across chess, web search, robotics, organizational workflows, and LLM inference, demonstrating that compression, decision making, and computational reuse are all derived from a single probability measure on sequence space.

---

### Attention, May I Have Your Decision? Localizing Generative Choices in Diffusion Models

Authors: Katarzyna Zaleska, Łukasz Popek, Monika Wysoczańska, Kamil Deja
Published: 2026-03-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2604.06052v3

Text-to-image diffusion models exhibit remarkable generative capabilities, yet their internal operations remain opaque, particularly when handling prompts that are not fully descriptive. In such scenarios, models must make implicit decisions to generate details not explicitly specified in the text. This work investigates the hypothesis that this decision-making process is not diffuse but is computationally localized within the model's architecture. While existing localization techniques focus on prompt-related interventions, we notice that such explicit conditioning may differ from implicit decisions. Therefore, we introduce a probing-based localization technique to identify the layers with the highest attribute separability for concepts. Our findings indicate that the resolution of ambiguous concepts is governed principally by self-attention layers, identifying them as the most effective point for intervention. Based on this discovery, we propose ICM (Implicit Choice-Modification) - a precise steering method that applies targeted interventions to a small subset of layers. Extensive experiments confirm that intervening on these specific self-attention layers yields superior debiasing performance compared to existing state-of-the-art methods, minimizing artifacts common to less precise approaches. The code is available at https://github.com/kzaleskaa/icm.

---

### AIFS-COMPO: A Global Data-Driven Atmospheric Composition Forecasting System

Authors: Paula Harder, Johannes Flemming, Mihai Alexe, Gert Mertes, Baudouin Raoult, Matthew Chantry
Published: 2026-03-29
Categories: physics.ao-ph, cs.AI
arXiv: https://arxiv.org/abs/2604.03300v1

We introduce AIFS-COMPO, a skilful medium-range data-driven global forecasting system for aerosols and reactive gases. Building on the ECMWF Artificial Intelligence Forecast System (AIFS), AIFS-COMPO employs a transformer-based encoder-processor-decoder architecture to jointly model meteorological and atmospheric composition variables. The model is trained on Copernicus Atmosphere Monitoring Service (CAMS) reanalysis, analysis, and forecast data to learn the coupled dynamics of weather, emissions, transport, and atmospheric chemistry. We evaluate AIFS-COMPO against a range of atmospheric composition observations and compare its performance with the operational CAMS global forecasting system IFS-COMPO. The results show that AIFS-COMPO achieves comparable or improved forecast skill for several key species while requiring only a fraction of the computational resources. Furthermore, the efficiency of the approach enables forecasts beyond the current operational horizon, demonstrating the potential of AI-based systems for fast and accurate global atmospheric composition prediction.

---

### MoViD: View-Invariant 3D Human Pose Estimation via Motion-View Disentanglement

Authors: Yejia Liu, Hengle Jiang, Haoxian Liu, Runxi Huang, Xiaomin Ouyang
Published: 2026-03-29
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2604.03299v1

3D human pose estimation is a key enabling technology for applications such as healthcare monitoring, human-robot collaboration, and immersive gaming, but real-world deployment remains challenged by viewpoint variations. Existing methods struggle to generalize to unseen camera viewpoints, require large amounts of training data, and suffer from high inference latency. We propose MoViD, a viewpoint-invariant 3D human pose estimation framework that disentangles viewpoint information from motion features. The key idea is to extract viewpoint information from intermediate pose features and leverage it to enhance both the robustness and efficiency of pose estimation. MoViD introduces a view estimator that models key joint relationships to predict viewpoint information, and an orthogonal projection module to disentangle motion and view features, further enhanced through physics-grounded contrastive alignment across views. For real-time edge deployment, MoViD employs a frame-by-frame inference pipeline with a view-aware strategy that adaptively activates flip refinement based on the estimated viewpoint. Evaluations on nine public datasets and newly collected multiview UAV and gait analysis datasets show that MoViD reduces pose estimation error by over 24.2\% compared to state-of-the-art methods, maintains robust performance under severe occlusions with 60\% less training data, and achieves real-time inference at 15 FPS on NVIDIA edge devices.

---

### Computational Foundations for Strategic Coopetition: Formalizing Sequential Interaction and Reciprocity

Authors: Vik Pant, Eric Yu
Published: 2026-03-29
Categories: cs.MA, cs.AI, cs.CY, cs.GT, cs.SE
arXiv: https://arxiv.org/abs/2604.01240v1

Strategic coopetition in multi-stakeholder systems requires understanding how cooperation persists through time without binding contracts. This technical report extends computational foundations for strategic coopetition to sequential interaction dynamics, bridging conceptual modeling (i* framework) with game-theoretic reciprocity analysis. We develop: (1) bounded reciprocity response functions mapping partner deviations to finite conditional responses, (2) memory-windowed history tracking capturing cognitive limitations over k recent periods, (3) structural reciprocity sensitivity derived from interdependence matrices where behavioral responses are amplified by structural dependencies, and (4) trust-gated reciprocity where trust modulates reciprocity responses. The framework applies to both human stakeholder interactions and multi-agent computational systems. Comprehensive validation across 15,625 parameter configurations demonstrates robust reciprocity effects, with all six behavioral targets exceeding thresholds: cooperation emergence (97.5%), defection punishment (100%), forgiveness dynamics (87.9%), asymmetric differentiation (100%), trust-reciprocity interaction (100%), and bounded responses (100%). Empirical validation using the Apple iOS App Store ecosystem (2008-2024) achieves 43/51 applicable points (84.3%), reproducing documented cooperation patterns across five ecosystem phases. Statistical significance confirmed at p < 0.001 with Cohen's d = 1.57. This report concludes the Foundations Series (TR-1 through TR-4) adopting uniaxial treatment where agents choose cooperation levels along a single continuum. Companion work on interdependence (arXiv:2510.18802), trust (arXiv:2510.24909), and collective action (arXiv:2601.16237) has been prepublished. Extensions Series (TR-5 through TR-8) introduces biaxial treatment where cooperation and competition are independent dimensions.

---

### Large Language Models for Analyzing Enterprise Architecture Debt in Unstructured Documentation

Authors: Christin Pagels, Simon Hacks, Rob Henk Bemthuis
Published: 2026-03-29
Categories: cs.SE, cs.LG
arXiv: https://arxiv.org/abs/2604.00046v1

Enterprise Architecture Debt (EA Debt) arises from suboptimal design decisions and misaligned components that can degrade an organization's IT landscape over time. Early indicators, Enterprise Architecture Smells (EA Smells), are currently mainly detected manually or only from structured artifacts, leaving much unstructured documentation under-analyzed. This study proposes an approach using a large language model (LLM) to identify and quantify EA Debt in unstructured architectural documentation. Following a design science research approach, we design and evaluate an LLM-based prototype for automated EA Smell detection. The artifact ingests unstructured documents (e.g., process descriptions, strategy papers), applies fine-tuned detection models, and outputs identified smells. We evaluate the prototype through a case study using synthetic yet realistic business documents, benchmarking against a custom GPT-based model. Results show that LLMs can detect multiple predefined EA Smells in unstructured text, with the benchmark model achieving higher precision and processing speed, and the fine-tuned on-premise model offering data protection advantages. The findings highlight opportunities for integrating LLM-based smell detection into EA governance practice.

---

### DriftScript: A Domain-Specific Language for Programming Non-Axiomatic Reasoning Agents

Authors: Seamus Brady
Published: 2026-03-29
Categories: cs.PL, cs.AI
arXiv: https://arxiv.org/abs/2604.00043v1

Non-Axiomatic Reasoning Systems (NARS) provide a framework for building adaptive agents that operate under insufficient knowledge and resources. However, the standard input language, Narsese, poses a usability barrier: its dense symbolic notation, overloaded punctuation, and implicit conventions make programs difficult to read, write, and maintain. We present DriftScript, a Lisp-like domain-specific language that compiles to Narsese. DriftScript provides source-level constructs covering the major sentence and term forms used in Non-Axiomatic Logic (NAL) levels 1 through 8, including inheritance, temporal implication, variable quantification, sequential conjunction, and operation invocation, while replacing symbolic syntax with readable keyword-based S-expressions. The compiler is a zero-dependency, four-stage pipeline implemented in 1,941 lines of C99. When used with the DriftNARS engine, DriftScript programs connect to external systems through four structured callback types and an HTTP operation registry, enabling a sense-reason-act loop for autonomous agents. We describe the language design and formal grammar, detail the compiler architecture, and evaluate the compiler through a 106-case test suite, equivalence testing against hand-written Narsese, a NAL coverage analysis, structural readability metrics, and compilation benchmarks. The source code is available at https://github.com/seamus-brady/DriftNARS. This paper focuses on the design and implementation of the DriftScript language and its embedding into DriftNARS, rather than on new inference algorithms for NARS itself.

---

### Incentives, Equilibria, and the Limits of Healthcare AI: A Game-Theoretic Perspective

Authors: Ari Ercole
Published: 2026-03-29
Categories: cs.GT, cs.AI
arXiv: https://arxiv.org/abs/2603.28825v2

Using a stylised coordination problem drawn from inpatient capacity management, three archetypal forms of AI deployment are described: effort-reducing technologies, observability-oriented systems, and interventions that alter underlying incentive structures. Effort reduction and observability may improve performance within existing patterns of behaviour but do not, in general, change which actions are individually rational. As a result, such interventions are typically absorbed into existing equilibria. By contrast, interventions that modify how local actions map to downstream consequences by redistributing or bounding local risk can change stable system behaviour. These mechanism-level interventions differ not in technical sophistication but in their interaction with institutional incentives. The analysis suggests that expectations of system-level gains from AI should be conditioned on whether a deployment changes incentives rather than optimising tasks or information flows alone. For healthcare organisations and policymakers, this has practical implications for procurement, governance, and evaluation of digital technologies.

---

### SNEAKDOOR: Stealthy Backdoor Attacks against Distribution Matching-based Dataset Condensation

Authors: He Yang, Dongyi Lv, Song Ma, Wei Xi, Jizhong Zhao
Published: 2026-03-29
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2603.28824v1

Dataset condensation aims to synthesize compact yet informative datasets that retain the training efficacy of full-scale data, offering substantial gains in efficiency. Recent studies reveal that the condensation process can be vulnerable to backdoor attacks, where malicious triggers are injected into the condensation dataset, manipulating model behavior during inference. While prior approaches have made progress in balancing attack success rate and clean test accuracy, they often fall short in preserving stealthiness, especially in concealing the visual artifacts of condensed data or the perturbations introduced during inference. To address this challenge, we introduce Sneakdoor, which enhances stealthiness without compromising attack effectiveness. Sneakdoor exploits the inherent vulnerability of class decision boundaries and incorporates a generative module that constructs input-aware triggers aligned with local feature geometry, thereby minimizing detectability. This joint design enables the attack to remain imperceptible to both human inspection and statistical detection. Extensive experiments across multiple datasets demonstrate that Sneakdoor achieves a compelling balance among attack success rate, clean test accuracy, and stealthiness, substantially improving the invisibility of both the synthetic data and triggered samples while maintaining high attack efficacy. The code is available at https://github.com/XJTU-AI-Lab/SneakDoor.

---

### Time is Not Compute: Scaling Laws for Wall-Clock Constrained Training on Consumer GPUs

Authors: Yi Liu
Published: 2026-03-29
Categories: cs.PF, cs.AI
arXiv: https://arxiv.org/abs/2603.28823v1

Scaling laws relate model quality to compute budget (FLOPs), but practitioners face wall-clock time constraints, not compute budgets. We study optimal model sizing under fixed time budgets from 5 minutes to 24 hours on consumer GPUs (RTX 4090). Across 70+ runs spanning 50M--1031M parameters, we find: (1)~at each time budget a U-shaped curve emerges where too-small models overfit and too-large models undertrain; (2)~optimal model size follows $N^* \propto t^{0.60}$, growing \emph{faster} than Chinchilla's $N^* \propto C^{0.50}$, with $α= 0.60 \pm 0.07$ robustly exceeding compute-optimal across all sensitivity analyses; (3)~a \emph{dual U-shape mechanism}: short-budget U-curves arise from compute bottlenecks, while long-budget U-curves emerge from data bottlenecks (overfitting), with an intermediate regime where the U-curve temporarily disappears. These findings have immediate implications for researchers training on consumer hardware, where wall-clock time -- not FLOPs -- is the binding constraint. We release all code, logs, and 70+ experimental configurations.

---

### Spatial Orthogonal Refinement for Robust RGB-Event Visual Object Tracking

Authors: Dexing Huang, Shiao Wang, Fan Zhang, Xiao Wang
Published: 2026-03-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.27913v1

Robust visual object tracking (VOT) remains challenging in high-speed motion scenarios, where conventional RGB sensors suffer from severe motion blur and performance degradation. Event cameras, with microsecond temporal resolution and high dynamic range, provide complementary structural cues that can potentially compensate for these limitations. However, existing RGB-Event fusion methods typically treat event data as dense intensity representations and adopt black-box fusion strategies, failing to explicitly leverage the directional geometric priors inherently encoded in event streams to rectify degraded RGB features. To address this limitation, we propose SOR-Track, a streamlined framework for robust RGB-Event tracking based on Spatial Orthogonal Refinement (SOR). The core SOR module employs a set of orthogonal directional filters that are dynamically guided by local motion orientations to extract sharp and motion-consistent structural responses from event streams. These responses serve as geometric anchors to modulate and refine aliased RGB textures through an asymmetric structural modulation mechanism, thereby explicitly bridging structural discrepancies between two modalities. Extensive experiments on the large-scale FE108 benchmark demonstrate that SOR-Track consistently outperforms existing fusion-based trackers, particularly under motion blur and low-light conditions. Despite its simplicity, the proposed method offers a principled and physics-grounded approach to multi-modal feature alignment and texture rectification. The source code of this paper will be released on https://github.com/Event-AHU/OpenEvTracking

---

### GAAMA: Graph Augmented Associative Memory for Agents

Authors: Swarna Kamal Paul, Shubhendu Sharma, Nitin Sareen
Published: 2026-03-29
Categories: cs.AI, cs.IR, cs.MA
arXiv: https://arxiv.org/abs/2603.27910v2

AI agents that interact with users across multiple sessions require persistent long-term memory to maintain coherent, personalized behavior. Current approaches either rely on flat retrieval-augmented generation (RAG), which loses structural relationships among memories, or use entity-centric knowledge graphs that suffer from mega-hub effects in conversational data, diluting graph-based relevance propagation. We propose GAAMA, a graph-augmented associative memory for agents that constructs a concept-mediated knowledge graph through a three-step pipeline: (1)verbatim episode preservation, (2)LLM-based extraction of atomic facts and topic-level concept nodes, and (3)synthesis of higher-order reflections. The resulting graph uses four node types (episode, fact, reflection, concept) connected by five structural edge types, with concept nodes providing cross-cutting traversal paths that avoid the mega-hub problem of entity-centric designs. Retrieval combines cosine-similarity-based k-nearest neighbor search with edge-type-aware Personalized PageRank (PPR) through an additive scoring function. We further introduce GRAFT (Graph Repair by Augmenting Facts & Topology), a post-retrieval corrective layer that diagnoses retrieval failures and surgically repairs the knowledge graph. On LoCoMo-10 (1,540 questions, 10 multi-session conversations), GAAMA achieves 79.1% mean reward, a +4.2~pp improvement over a tuned RAG baseline, the strongest comparator. On MemoryArena, GAAMA outperforms full-context baselines across three tasks - Group Travel (+0.4~pp), Web Shopping (+3.4~pp), and Progressive Search (+0.7~pp) - with advantages growing monotonically with dialogue length. Notably, GAAMA delivers consistent performance across all categories, matching the best competing method in each, whereas every competitor degrades in at least one category.

---

### An Empirical Markov Chain Car-Following (MC-CF) Model

Authors: Sungyong Chung, Yanlin Zhang, Nachuan Li, Dana Monzer, Alireza Talebpour
Published: 2026-03-29
Categories: eess.SY, cs.LG, cs.RO
arXiv: https://arxiv.org/abs/2603.27909v2

Car-following behavior is fundamental to traffic flow theory, yet traditional models often fail to capture the stochasticity of naturalistic driving. This paper proposes an empirical probabilistic sampling approach to car-following modeling that bypasses conventional parametric assumptions. Under this approach, we introduce the Markov Chain Car-Following (MC-CF) model, which represents state transitions as a Markov process and predicts behavior by randomly sampling accelerations from empirical distributions within discretized state bins. Evaluation on the Waymo Open Motion Dataset (WOMD) demonstrates that MC-CF variants significantly outperform all physics-based baselines (IDM, Gipps, FVDM, and SIDM) across both one-step and open-loop trajectory prediction metrics, and remain competitive with modern data-driven baselines including neural network and Gaussian mixture model approaches. Zero-shot generalization on the Naturalistic Phoenix (PHX) dataset further confirms cross-domain transferability. Finally, microscopic ring road simulations validate the framework's scalability: by incrementally integrating unconstrained free-flow trajectories and high-speed freeway data (TGSIM) alongside a conservative inference strategy, the model substantially reduces collisions across most tested scenarios and successfully reproduces naturalistic and stochastic shockwave propagation, though crashes persist under severe shockwave conditions. Overall, the proposed MC-CF model provides a robust and scalable foundation for simulating population-level stochastic traffic behavior that requires no behavioral parameter calibration, making it well-suited for the data-rich future of intelligent transportation.

---

### Data is All You Need: Markov Chain Car-Following (MC-CF) Model

Authors: Sungyong Chung, Yanlin Zhang, Nachuan Li, Dana Monzer, Alireza Talebpour
Published: 2026-03-29
Categories: eess.SY, cs.LG, cs.RO
arXiv: https://arxiv.org/abs/2603.27909v1

Car-following behavior is fundamental to traffic flow theory, yet traditional models often fail to capture the stochasticity of naturalistic driving. This paper introduces a new car-following modeling category called the empirical probabilistic paradigm, which bypasses conventional parametric assumptions. Within this paradigm, we propose the Markov Chain Car-Following (MC-CF) model, which represents state transitions as a Markov process and predicts behavior by randomly sampling accelerations from empirical distributions within discretized state bins. Evaluation of the MC-CF model trained on the Waymo Open Motion Dataset (WOMD) demonstrates that its variants significantly outperform physics-based models including IDM, Gipps, FVDM, and SIDM in both one-step and open-loop trajectory prediction accuracy. Statistical analysis of transition probabilities confirms that the model-generated trajectories are indistinguishable from real-world behavior, successfully reproducing the probabilistic structure of naturalistic driving across all interaction types. Zero-shot generalization on the Naturalistic Phoenix (PHX) dataset further confirms the model's robustness. Finally, microscopic ring road simulations validate the framework's scalability. By incrementally integrating unconstrained free-flow trajectories and high-speed freeway data (TGSIM) alongside a conservative inference strategy, the model drastically reduces collisions, achieving zero crashes in multiple equilibrium and shockwave scenarios, while successfully reproducing naturalistic and stochastic shockwave propagation. Overall, the proposed MC-CF model provides a robust, scalable, and calibration-free foundation for high-fidelity stochastic traffic modeling, uniquely suited for the data-rich future of intelligent transportation.

---

### ATLAS-RTC: Closing the Loop on LLM Agent Output with Token-Level Runtime Control

Authors: Christopher Cruz
Published: 2026-03-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.27905v2

We present ATLAS-RTC, a runtime control system for autoregressive language models that enforces structured output during decoding. ATLAS-RTC monitors generation at each step, detects drift from output contracts using lightweight signals, and applies targeted interventions such as biasing, masking, and rollback. Unlike post-hoc validation or static constrained decoding, it operates in a closed loop, enabling correction before errors materialize. Across structured generation and tool-calling tasks, ATLAS-RTC improves first-attempt success rates by 20 to 37.8 percentage points, with up to 88% latency reduction in failure-dominated settings. Results show that many failures arise from decoding artifacts rather than task misunderstanding, motivating runtime control as a distinct layer in LLM systems.

---

### BINO: Encoder Centric Self Supervised Stereo With Native Pair Input

Authors: Haokun Zhou
Published: 2026-03-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.27904v1

Stereo needs features that preserve fine cross view correspondence rather than only semantic similarity. Recent self supervised vision models transfer well, but they are not built for this goal, and geometry focused methods often rely on a binocular decoder or another explicit linkage module during pretraining. BINO asks whether strong binocular structure can instead be learned inside a compact encoder. It does this by fusing the rectified pair at the input stage, forming stereo micro cell tokens, and using a row aware patch phase positional encoding. Training uses one view masked token only distillation together with occlusion and view specific appearance mismatch. In a strict low resource setting with pretraining only on KITTI object, BINO gives the best frozen descriptor results under a no linkage probe among all compared baselines on proxy dense stereo, hard negative retrieval, and KITTI Stereo~2012 disparity. With the same lightweight stereo head for every encoder, it stays near CroCo~v2 while using a much smaller encoder. Supplementary transfer experiments on KITTI Stereo~2015 show the same qualitative trend. These results suggest that much of the cross view reasoning often assigned to a separate linkage module can be learned inside a compact and reusable encoder.

---

### Persistence diagrams of random matrices via Morse theory: universality and a new spectral diagnostic

Authors: Matthew Loftus
Published: 2026-03-29
Categories: stat.ML, cs.LG, math-ph, math.AT
arXiv: https://arxiv.org/abs/2603.27903v1

We prove that the persistence diagram of the sublevel set filtration of the quadratic form f(x) = x^T M x restricted to the unit sphere S^{n-1} is analytically determined by the eigenvalues of the symmetric matrix M. By Morse theory, the diagram has exactly n-1 finite bars, with the k-th bar living in homological dimension k-1 and having length equal to the k-th eigenvalue spacing s_k = λ_{k+1} - λ_k. This identification transfers random matrix theory (RMT) universality to persistence diagram universality: for matrices drawn from the Gaussian Orthogonal Ensemble (GOE), we derive the closed-form persistence entropy PE = log(8n/π) - 1, and verify numerically that the coefficient of variation of persistence statistics decays as n^{-0.6}. Different random matrix ensembles (GOE, GUE, Wishart) produce distinct universal persistence diagrams, providing topological fingerprints of RMT universality classes. As a practical consequence, we show that persistence entropy outperforms the standard level spacing ratio \langle r \rangle for discriminating GOE from GUE matrices (AUC 0.978 vs. 0.952 at n = 100, non-overlapping bootstrap 95% CIs), and detects global spectral perturbations in the Rosenzweig-Porter model to which \langle r \rangle is blind. These results establish persistence entropy as a new spectral diagnostic that captures complementary information to existing RMT tools.

---

### Rényi Entropy: A New Token Pruning Metric for Vision Transformers

Authors: Wei-Yuan Su, Ruijie Zhang, Zheng Zhang
Published: 2026-03-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.27900v1

Vision Transformers (ViTs) achieve state-of-the-art performance but suffer from the $O(N^2)$ complexity of self-attention, making inference costly for high-resolution inputs. To address this bottleneck, token pruning has emerged as a critical technique to accelerate inference. Most existing methods rely on the (CLS) token to estimate patch importance. However, we argue that the (CLS) token can be unreliable in early layers where semantic representations are still immature. As a result, pruning in the early layer often leads to inaccurate importance estimation and unnecessary information loss. In this work, we propose a training-free token importance metric, namely Col-Ln, which is derived from Rényi entropy that enables the identification of informative tokens from the first layer of the network, thereby enabling more reliable pruning in token reduction. Extensive experiments on ViTs and Large Vision-Language Models (LVLMs) demonstrate that our approach consistently outperforms state-of-the-art pruning methods across diverse benchmarks.

---

### SAGE: Sink-Aware Grounded Decoding for Multimodal Hallucination Mitigation

Authors: Tripti Shukla, Zsolt Kira
Published: 2026-03-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.27898v1

Large vision-language models (VLMs) frequently suffer from hallucinations, generating content that is inconsistent with visual inputs. Existing methods typically address this problem through post-hoc filtering, additional training objectives, or external verification, but they do not intervene during the decoding process when hallucinations arise. In this work, we introduce SAGE, a Sink-Aware Grounded Decoding framework that mitigates hallucinations by dynamically modulating self-attention during generation. Hallucinations are strongly correlated with attention sink tokens - punctuation or function tokens that accumulate disproportionate attention despite carrying limited semantic content. SAGE leverages these tokens as anchors to monitor grounding reliability in real time. At each sink trigger, the method extracts semantic concepts from the generated sequence, estimates their visual grounding using both self-attention maps and gradient-based attribution, and measures their spatial agreement. Based on this signal, self-attention distributions are adaptively sharpened or broadened to reinforce grounded regions or suppress unreliable ones. Extensive experiments across diverse hallucination benchmarks demonstrate that SAGE consistently outperforms existing decoding strategies, achieving substantial reductions in hallucination while preserving descriptive coverage, without requiring model retraining or architectural modifications. Our method achieves an average relative improvement of 10.65% on MSCOCO and 7.19% on AMBER across diverse VLM architectures, demonstrating consistent gains in hallucination mitigation.

---

### Poppy: Polarization-based Plug-and-Play Guidance for Enhancing Monocular Normal Estimation

Authors: Irene Kim, Sai Tanmay Reddy Chakkera, Alexandros Graikos, Dimitris Samaras, Akshat Dave
Published: 2026-03-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.27891v1

Monocular surface normal estimators trained on large-scale RGB-normal data often perform poorly in the edge cases of reflective, textureless, and dark surfaces. Polarization encodes surface orientation independently of texture and albedo, offering a physics-based complement for these cases. Existing polarization methods, however, require multi-view capture or specialized training data, limiting generalization. We introduce Poppy, a training-free framework that refines normals from any frozen RGB backbone using single-shot polarization measurements at test time. Keeping backbone weights frozen, Poppy optimizes per-pixel offsets to the input RGB and output normal along with a learned reflectance decomposition. A differentiable rendering layer converts the refined normals into polarization predictions and penalizes mismatches with the observed signal. Across seven benchmarks and three backbone architectures (diffusion, flow, and feed-forward), Poppy reduces mean angular error by 23-26% on synthetic data and 6-16% on real data. These results show that guiding learned RGB-based normal estimators with polarization cues at test time refines normals on challenging surfaces without retraining.

---

### Article and Comment Frames Shape the Quality of Online Comments

Authors: Matteo Guida, Yulia Otmakhova, Eduard Hovy, Lea Frermann
Published: 2026-03-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.27889v2

Framing theory posits that how information is presented shapes audience responses, but computational work has largely ignored audience reactions. While recent work showed that article framing systematically shapes the content of reader responses, this paper asks: does framing also affect response quality? Analyzing 1M comments across 2.7K news articles, we operationalize quality as comment health. We find that article frames significantly predict comment health while controlling for topic, and that comments that adopt the article frame are healthier than those that depart from it. Further, unhealthy top-level comments tend to generate more unhealthy responses, independent of the frame being used in the comment. Our results establish a link between framing theory and discourse quality, laying the groundwork for downstream applications. We illustrate this potential with a pro-active frame-aware LLM- based system to mitigate unhealthy discourse.

---

### AI-ready design of realistic 2D materials and interfaces with Mat3ra-2D

Authors: Vsevolod Biryukov, Kamal Choudhary, Timur Bazhirov
Published: 2026-03-29
Categories: cond-mat.mtrl-sci, cs.AI, physics.comp-ph
arXiv: https://arxiv.org/abs/2603.27886v1

Artificial intelligence (AI) and machine learning (ML) models in materials science are predominantly trained on ideal bulk crystals, limiting their transferability to real-world applications where surfaces, interfaces, and defects dominate. We present Mat3ra-2D, an open-source framework for the rapid design of realistic two-dimensional materials and related structures, including slabs and heterogeneous interfaces, with support for disorder and defect-driven complexity. The approach combines: (1) well-defined standards for storing and exchanging materials data with a modular implementation of core concepts and (2) transformation workflows expressed as configuration-builder pipelines that preserve provenance and metadata. We implement typical structure generation tasks, such as constructing orientation-specific slabs or strain-matching interfaces, in reusable Jupyter notebooks that serve as both interactive documentation and templates for reproducible runs. To lower the barrier to adoption, we design the examples to run in any web browser and demonstrate how to incorporate these developments into a web application. Mat3ra-2D enables systematic creation and organization of realistic 2D- and interface-aware datasets for AI/ML-ready applications.

---

### Spectral Signatures of Data Quality: Eigenvalue Tail Index as a Diagnostic for Label Noise in Neural Networks

Authors: Matthew Loftus
Published: 2026-03-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.27885v1

We investigate whether spectral properties of neural network weight matrices can predict test accuracy. Under controlled label noise variation, the tail index alpha of the eigenvalue distribution at the network's bottleneck layer predicts test accuracy with leave-one-out R^2 = 0.984 (21 noise levels, 3 seeds per level), far exceeding all baselines: the best conventional metric (Frobenius norm of the optimal layer) achieves LOO R^2 = 0.149. This relationship holds across three architectures (MLP, CNN, ResNet-18) and two datasets (MNIST, CIFAR-10). However, under hyperparameter variation at fixed data quality (180 configurations varying width, depth, learning rate, and weight decay), all spectral and conventional measures are weak predictors (R^2 < 0.25), with simple baselines (global L_2 norm, LOO R^2 = 0.219) slightly outperforming spectral measures (tail alpha, LOO R^2 = 0.167). We therefore frame the tail index as a data quality diagnostic: a powerful detector of label corruption and training set degradation, rather than a universal generalization predictor. A noise detector calibrated on synthetic noise successfully identifies real human annotation errors in CIFAR-10N (9% noise detected with 3% error). We identify the information-processing bottleneck layer as the locus of this signature and connect the observations to the BBP phase transition in spiked random matrix models. We also report a negative result: the level spacing ratio <r> is uninformative for weight matrices due to Wishart universality.

---

### Near-Optimal Primal-Dual Algorithm for Learning Linear Mixture CMDPs with Adversarial Rewards

Authors: Kihyun Yu, Seoungbin Bae, Dabeen Lee
Published: 2026-03-29
Categories: cs.LG, math.OC
arXiv: https://arxiv.org/abs/2603.27884v2

We study safe reinforcement learning in finite-horizon linear mixture constrained Markov decision processes (CMDPs) with adversarial rewards under full-information feedback and an unknown transition kernel. We propose a primal-dual policy optimization algorithm that achieves regret and constraint violation bounds of $\widetilde{O}(\sqrt{d^2 H^3 K})$ under mild conditions, where $d$ is the feature dimension, $H$ is the horizon, and $K$ is the number of episodes. To the best of our knowledge, this is the first provably efficient algorithm for linear mixture CMDPs with adversarial rewards. In particular, our regret bound is near-optimal, matching the known minimax lower bound up to logarithmic factors. The key idea is to introduce a regularized dual update that enables a drift-based analysis. This step is essential, as strong duality-based analysis cannot be directly applied when reward functions change across episodes. In addition, we extend weighted ridge regression-based parameter estimation to the constrained setting, allowing us to construct tighter confidence intervals that are crucial for deriving the near-optimal regret bound.

---

### Near-Optimal Primal-Dual Algorithm for Learning Linear Mixture CMDPs with Adversarial Rewards

Authors: Kihyun Yu, Seoungbin Bae, Dabeen Lee
Published: 2026-03-29
Categories: cs.LG, math.OC
arXiv: https://arxiv.org/abs/2603.27884v1

We study safe reinforcement learning in finite-horizon linear mixture constrained Markov decision processes (CMDPs) with adversarial rewards under full-information feedback and an unknown transition kernel. We propose a primal-dual policy optimization algorithm that achieves regret and constraint violation bounds of $\widetilde{O}(\sqrt{d^2 H^3 K})$ under mild conditions, where $d$ is the feature dimension, $H$ is the horizon, and $K$ is the number of episodes. To the best of our knowledge, this is the first provably efficient algorithm for linear mixture CMDPs with adversarial rewards. In particular, our regret bound is near-optimal, matching the known minimax lower bound up to logarithmic factors. The key idea is to introduce a regularized dual update that enables a drift-based analysis. This step is essential, as strong duality-based analysis cannot be directly applied when reward functions change across episodes. In addition, we extend weighted ridge regression-based parameter estimation to the constrained setting, allowing us to construct tighter confidence intervals that are crucial for deriving the near-optimal regret bound.

---

### Kernel Dynamics under Path Entropy Maximization

Authors: Jnaneshwar Das
Published: 2026-03-29
Categories: cs.LG, cs.AI, cs.RO, math.DS
arXiv: https://arxiv.org/abs/2603.27880v1

We propose a variational framework in which the kernel function k : X x X -> R, interpreted as the foundational object encoding what distinctions an agent can represent, is treated as a dynamical variable subject to path entropy maximization (Maximum Caliber, MaxCal). Each kernel defines a representational structure over which an information geometry on probability space may be analyzed; a trajectory through kernel space therefore corresponds to a trajectory through a family of effective geometries, making the optimization landscape endogenous to its own traversal. We formulate fixed-point conditions for self-consistent kernels, propose renormalization group (RG) flow as a structured special case, and suggest neural tangent kernel (NTK) evolution during deep network training as a candidate empirical instantiation. Under explicit information-thermodynamic assumptions, the work required for kernel change is bounded below by delta W >= k_B T delta I_k, where delta I_k is the mutual information newly unlocked by the updated kernel. In this view, stable fixed points of MaxCal over kernels correspond to self-reinforcing distinction structures, with biological niches, scientific paradigms, and craft mastery offered as conjectural interpretations. We situate the framework relative to assembly theory and the MaxCal literature, separate formal results from structured correspondences and conjectural bridges, and pose six open questions that make the program empirically and mathematically testable.

---

### HumMusQA: A Human-written Music Understanding QA Benchmark Dataset

Authors: Benno Weck, Pablo Puentes, Andrea Poltronieri, Satyajeet Prabhu, Dmitry Bogdanov
Published: 2026-03-29
Categories: cs.CL, cs.SD
arXiv: https://arxiv.org/abs/2603.27877v1

The evaluation of music understanding in Large Audio-Language Models (LALMs) requires a rigorously defined benchmark that truly tests whether models can perceive and interpret music, a standard that current data methodologies frequently fail to meet. This paper introduces a meticulously structured approach to music evaluation, proposing a new dataset of 320 hand-written questions curated and validated by experts with musical training, arguing that such focused, manual curation is superior for probing complex audio comprehension. To demonstrate the use of the dataset, we benchmark six state-of-the-art LALMs and additionally test their robustness to uni-modal shortcuts.

---

### Stability and Sensitivity Analysis of Relative Temporal-Difference Learning: Extended Version

Authors: Masoud S. Sakha, Rushikesh Kamalapurkar, Sean Meyn
Published: 2026-03-29
Categories: cs.LG, math.OC
arXiv: https://arxiv.org/abs/2603.27874v2

Relative temporal-difference (TD) learning was introduced to mitigate the slow convergence of TD methods when the discount factor approaches one by subtracting a baseline from the temporal-difference update. While this idea has been studied in the tabular setting, stability guarantees with function approximation remain poorly understood. This paper analyzes relative TD learning with linear function approximation. We establish stability conditions for the algorithm and show that the choice of baseline distribution plays a central role. In particular, when the baseline is chosen as the empirical distribution of the state-action process, the algorithm is stable for any non-negative baseline weight and any discount factor. We also provide a sensitivity analysis of the resulting parameter estimates, characterizing both asymptotic bias and covariance. The asymptotic covariance and asymptotic bias are shown to remain uniformly bounded as the discount factor approaches one.

---

### Statistical Guarantees for Distributionally Robust Optimization with Optimal Transport and OT-Regularized Divergences

Authors: Jeremiah Birrell, Xiaoxi Shen
Published: 2026-03-29
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2603.27871v1

We study finite-sample statistical performance guarantees for distributionally robust optimization (DRO) with optimal transport (OT) and OT-regularized divergence model neighborhoods. Specifically, we derive concentration inequalities for supervised learning via DRO-based adversarial training, as commonly employed to enhance the adversarial robustness of machine learning models. Our results apply to a wide range of OT cost functions, beyond the $p$-Wasserstein case studied by previous authors. In particular, our results are the first to: 1) cover soft-constraint norm-ball OT cost functions; soft-constraint costs have been shown empirically to enhance robustness when used in adversarial training, 2) apply to the combination of adversarial sample generation and adversarial reweighting that is induced by using OT-regularized $f$-divergence model neighborhoods; the added reweighting mechanism has also been shown empirically to further improve performance. In addition, even in the $p$-Wasserstein case, our bounds exhibit better behavior as a function of the DRO neighborhood size than previous results when applied to the adversarial setting.

---

### A Revealed Preference Framework for AI Alignment

Authors: Elchin Suleymanov
Published: 2026-03-29
Categories: econ.TH, cs.AI, cs.GT
arXiv: https://arxiv.org/abs/2603.27868v1

Human decision makers increasingly delegate choices to AI agents, raising a natural question: does the AI implement the human principal's preferences or pursue its own? To study this question using revealed preference techniques, I introduce the Luce Alignment Model, where the AI's choices are a mixture of two Luce rules, one reflecting the human's preferences and the other the AI's. I show that the AI's alignment (similarity of human and AI preferences) can be generically identified in two settings: the laboratory setting, where both human and AI choices are observed, and the field setting, where only AI choices are observed.

---

### Wan-R1: Verifiable-Reinforcement Learning for Video Reasoning

Authors: Ming Liu, Yunbei Zhang, Shilong Liu, Liwen Wang, Wensheng Zhang
Published: 2026-03-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.27866v1

Video generation models produce visually coherent content but struggle with tasks requiring spatial reasoning and multi-step planning. Reinforcement learning (RL) offers a path to improve generalization, but its effectiveness in video reasoning hinges on reward design -- a challenge that has received little systematic study. We investigate this problem by adapting Group Relative Policy Optimization (GRPO) to flow-based video models and training them on maze-solving and robotic navigation tasks. We first show that multimodal reward models fail catastrophically in this setting. To address this, we design verifiable reward functions grounded in objective task metrics. For structured game environments, we introduce a multi-component trajectory reward. For robotic navigation, we propose an embedding-level verifiable reward. Our experiments show that RL fine-tuning with verifiable rewards improves generalization. For example, on complex 3D mazes, our model improves exact match accuracy by 29.1\% over the SFT baseline, and on trap-avoidance tasks by 51.4\%. Our systematic reward analysis reveals that verifiable rewards are critical for stable training, while multimodal reward models could lead to degenerate solutions. These findings establish verifiable reward design as a key enabler for robust video reasoning. Code will be publicly available.

---

### Vertical Consensus Inference for High-Dimensional Random Partition

Authors: Khai Nguyen, Yang Ni, Peter Mueller
Published: 2026-03-29
Categories: stat.ME, stat.CO, stat.ML
arXiv: https://arxiv.org/abs/2603.27864v1

We review recently proposed Bayesian approaches for clustering high-dimensional data. After identifying the main limitations of available approaches, we introduce an alternative framework based on vertical consensus inference (VCI) to mitigate the curse of dimensionality in high-dimensional Bayesian clustering. VCI builds on the idea of consensus Monte Carlo by dividing the data into multiple shards (smaller subsets of variables), performing posterior inference on each shard, and then combining the shard-level posteriors to obtain a consensus posterior. The key distinction is that VCI splits the data vertically, producing vertical shards that retain the same number of observations but have lower dimensionality. We use an entropic regularized Wasserstein barycenter to define a consensus posterior. The shard-specific barycenter weights are constructed to favor shards that provide meaningful partitions, distinct from a trivial single cluster or all singleton clusters, favoring balanced cluster sizes and precise shard-specific posterior random partitions. We show that VCI can be interpreted as a variational approximation to the posterior under a hierarchical model with a generalized Bayes prior. For relatively low-dimensional problems, experiments suggest that VCI closely approximates inference based on clustering the entire multivariate data. For high-dimensional data and in the presence of many noninformative dimensions, VCI introduces a new framework for model-based and principled inference on random partitions. Although our focus here is on random partitions, VCI can be applied to any dimension-independent parameters and serves as a bridge to emerging areas in statistics such as consensus Monte Carlo, optimal transport, variational inference, and generalized Bayes.

---

### ImagenWorld: Stress-Testing Image Generation Models with Explainable Human Evaluation on Open-ended Real-World Tasks

Authors: Samin Mahdizadeh Sani, Max Ku, Nima Jamali, Matina Mahdizadeh Sani, Paria Khoshtab, Wei-Chieh Sun, Parnian Fazel, Zhi Rui Tam, Thomas Chong, Edisy Kin Wai Chan, Donald Wai Tong Tsang, Chiao-Wei Hsu et al.
Published: 2026-03-29
Categories: cs.GR, cs.AI, cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2603.27862v1

Advances in diffusion, autoregressive, and hybrid models have enabled high-quality image synthesis for tasks such as text-to-image, editing, and reference-guided composition. Yet, existing benchmarks remain limited, either focus on isolated tasks, cover only narrow domains, or provide opaque scores without explaining failure modes. We introduce \textbf{ImagenWorld}, a benchmark of 3.6K condition sets spanning six core tasks (generation and editing, with single or multiple references) and six topical domains (artworks, photorealistic images, information graphics, textual graphics, computer graphics, and screenshots). The benchmark is supported by 20K fine-grained human annotations and an explainable evaluation schema that tags localized object-level and segment-level errors, complementing automated VLM-based metrics. Our large-scale evaluation of 14 models yields several insights: (1) models typically struggle more in editing tasks than in generation tasks, especially in local edits. (2) models excel in artistic and photorealistic settings but struggle with symbolic and text-heavy domains such as screenshots and information graphics. (3) closed-source systems lead overall, while targeted data curation (e.g., Qwen-Image) narrows the gap in text-heavy cases. (4) modern VLM-based metrics achieve Kendall accuracies up to 0.79, approximating human ranking, but fall short of fine-grained, explainable error attribution. ImagenWorld provides both a rigorous benchmark and a diagnostic tool to advance robust image generation.

---

### KazByte: Adapting Qwen models to Kazakh via Byte-level Adapter

Authors: Rauan Akylzhanov
Published: 2026-03-29
Categories: cs.CL, math.NA
arXiv: https://arxiv.org/abs/2603.27859v1

Large language models fragment Kazakh text into many more tokens than equivalent English text, because their tokenizers were built for high-resource languages. This tokenizer tax inflates compute, shortens the effective context window, and weakens the model's grip on Kazakh morphology. We propose to bypass the tokenizer entirely by feeding raw bytes through a small adapter that learns to speak the internal language of a frozen Qwen2.5-7B. Once the adapter is trained, we freeze it and fine-tune only the attention layers of Qwen on Kazakh text. Our central hypothesis is that this two-stage process -- first teach the interface, then adapt the model -- should match or exceed the accuracy of the original Qwen2.5-7B on standard Kazakh benchmarks. This report describes the ByteKaz architecture and training protocol. Empirical validation is ongoing; this version stakes the design and hypotheses for the record.

---

### CARGO: Carbon-Aware Gossip Orchestration in Smart Shipping

Authors: Alexandros S. Kalafatelis, Nikolaos Nomikos, Vasileios Nikolakakis, Nikolaos Tsoulakos, Panagiotis Trakadas
Published: 2026-03-29
Categories: cs.AI
arXiv: https://arxiv.org/abs/2603.27857v1

Smart shipping operations increasingly depend on collaborative AI, yet the underlying data are generated across vessels with uneven connectivity, limited backhaul, and clear commercial sensitivity. In such settings, server-coordinated FL remains a weak systems assumption, depending on a reachable aggregation point and repeated wide-area synchronization, both of which are difficult to guarantee in maritime networks. A serverless gossip approach therefore represents a more natural approach, but existing methods still treat communication mainly as an optimization bottleneck, rather than as a resource that must be managed jointly with carbon cost, reliability, and long-term participation balance. In this context, this paper presents CARGO, a carbon-aware gossip orchestration framework for smart-shipping. CARGO separates learning into a control and a data plane. The data plane performs local optimization with compressed gossip exchange, while the control plane decides, at each round, which vessels should participate, which communication edges should be activated, how aggressively updates should be compressed, and when recovery actions should be triggered. We evaluate CARGO under a predictive-maintenance scenario using operational bulk-carrier engine data and a trace-driven maritime communication protocol that captures client dropout, partial participation, packet loss, and multiple connectivity regimes, derived from mobility-aware vessel interactions. Across the tested stress settings, CARGO consistently remains in the high-accuracy regime while reducing carbon footprint and communication overheads, compared to accuracy-competitive decentralized baselines. Overall, the conducted performance evaluation demonstrates that CARGO is a feasible and practical solution for reliable and resource-conscious maritime AI deployment.

---

### What can LLMs tell us about the mechanisms behind polarity illusions in humans? Experiments across model scales and training steps

Authors: Dario Paape
Published: 2026-03-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.27855v1

I use the Pythia scaling suite (Biderman et al. 2023) to investigate if and how two well-known polarity illusions, the NPI illusion and the depth charge illusion, arise in LLMs. The NPI illusion becomes weaker and ultimately disappears as model size increases, while the depth charge illusion becomes stronger in larger models. The results have implications for human sentence processing: it may not be necessary to assume "rational inference" mechanisms that convert ill-formed sentences into well-formed ones to explain polarity illusions, given that LLMs cannot plausibly engage in this kind of reasoning, especially at the implicit level of next-token prediction. On the other hand, shallow, "good enough" processing and/or partial grammaticalization of prescriptively ungrammatical structures may both occur in LLMs. I propose a synthesis of different theoretical accounts that is rooted in the basic tenets of construction grammar.

---

### EffiSkill: Agent Skill Based Automated Code Efficiency Optimization

Authors: Zimu Wang, Yuling Shi, Mengfan Li, Zijun Liu, Jie M. Zhang, Chengcheng Wan, Xiaodong Gu
Published: 2026-03-29
Categories: cs.SE, cs.CL
arXiv: https://arxiv.org/abs/2603.27850v1

Code efficiency is a fundamental aspect of software quality, yet how to harness large language models (LLMs) to optimize programs remains challenging. Prior approaches have sought for one-shot rewriting, retrieved exemplars, or prompt-based search, but they do not explicitly distill reusable optimization knowledge, which limits generalization beyond individual instances. In this paper, we present EffiSkill, a framework for code-efficiency optimization that builds a portable optimization toolbox for LLM-based agents. The key idea is to model recurring slow-to-fast transformations as reusable agent skills that capture both concrete transformation mechanisms and higher-level optimization strategies. EffiSkill adopts a two-stage design: Stage I mines Operator and Meta Skills from large-scale slow/fast program pairs to build a skill library; Stage II applies this library to unseen programs through execution-free diagnosis, skill retrieval, plan composition, and candidate generation, without runtime feedback. Results on EffiBench-X show that EffiSkill achieves higher optimization success rates, improving over the strongest baseline by 3.69 to 12.52 percentage points across model and language settings. These findings suggest that mechanism-level skill reuse provides a useful foundation for execution-free code optimization, and that the resulting skill library can serve as a reusable resource for broader agent workflows.

---

### Model Capability Dominates: Inference-Time Optimization Lessons from AIMO 3

Authors: Natapong Nitarach
Published: 2026-03-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.27844v2

Majority voting over multiple LLM attempts improves mathematical reasoning, but correlated errors limit the effective sample size. A natural fix is to assign different reasoning strategies to different voters. The approach, Diverse Prompt Mixer, is tested on the AIMO 3 competition: 3 models, 23+ experiments, 50 IMO-level problems, one H100 80 GB, 5-hour limit. Every prompt-level intervention fails. High-temperature sampling already decorrelates errors; weaker strategies reduce accuracy more than they reduce correlation. Across an 8-point capability gap at equal N=8 and every optimization tested, model capability dominates. The gap between the best majority-vote score (42/50) and pass@20 (~45.5) is selection loss, not prompt loss. A verifier-based selector could close it. Prompt engineering cannot.

---

### ProText: A benchmark dataset for measuring (mis)gendering in long-form texts

Authors: Hadas Kotek, Margit Bowler, Patrick Sonnenberg, Yu'an Yang
Published: 2026-03-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.27838v1

We introduce ProText, a dataset for measuring gendering and misgendering in stylistically diverse long-form English texts. ProText spans three dimensions: Theme nouns (names, occupations, titles, kinship terms), Theme category (stereotypically male, stereotypically female, gender-neutral/non-gendered), and Pronoun category (masculine, feminine, gender-neutral, none). The dataset is designed to probe (mis)gendering in text transformations such as summarization and rewrites using state-of-the-art Large Language Models, extending beyond traditional pronoun resolution benchmarks and beyond the gender binary. We validated ProText through a mini case study, showing that even with just two prompts and two models, we can draw nuanced insights regarding gender bias, stereotyping, misgendering, and gendering. We reveal systematic gender bias, particularly when inputs contain no explicit gender cues or when models default to heteronormative assumptions.

---

### Q-Bridge: Code Translation for Quantum Machine Learning via LLMs

Authors: Runjia Zeng, Priyabrata Senapati, Ruixiang Tang, Dongfang Liu, Qiang Guan
Published: 2026-03-29
Categories: quant-ph, cs.CL
arXiv: https://arxiv.org/abs/2603.27836v1

Large language models have recently shown potential in bridging the gap between classical machine learning and quantum machine learning. However, the lack of standardized, high-quality datasets and robust translation frameworks limits progress in this domain. We introduce Q-Bridge, an LLM-guided code translation framework that systematically converts CML implementations into executable QML variants. Our approach builds on a self-involving pipeline that iteratively expands a verified seed codebase into a large-scale dataset, CML-2-QML, integrating verifiable and unverifiable code pairs. The Q-Bridge model is fine-tuned using supervised LoRA adaptation for scalable and memory-efficient training, achieving faithful and interpretable quantum code generation across diverse architectures. Empirical analysis confirms the feasibility of direct CML-to-QML translation and reveals consistent structural alignment between classical and quantum paradigms. Case studies further demonstrate that Q-Bridge can maintain deterministic correctness and also enable creative architectural exploration. This work establishes the first reproducible framework and dataset for LLM-driven quantum code translation, offering a foundation for scalable quantum AI development.

---

### 3-D Representations for Hyperspectral Flame Tomography

Authors: Nicolas Tricard, Zituo Chen, Sili Deng
Published: 2026-03-29
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2603.27832v1

Flame tomography is a compelling approach for extracting large amounts of data from experiments via 3-D thermochemical reconstruction. Recent efforts employing neural-network flame representations have suggested improved reconstruction quality compared with classical tomography approaches, but a rigorous quantitative comparison with the same algorithm using a voxel-grid representation has not been conducted. Here, we compare a classical voxel-grid representation with varying regularizers to a continuous neural representation for tomographic reconstruction of a simulated pool fire. The representations are constructed to give temperature and composition as a function of location, and a subsequent ray-tracing step is used to solve the radiative transfer equation to determine the spectral intensity incident on hyperspectral infrared cameras, which is then convolved with an instrument lineshape function. We demonstrate that the voxel-grid approach with a total-variation regularizer reproduces the ground-truth synthetic flame with the highest accuracy for reduced memory intensity and runtime. Future work will explore more representations and under experimental configurations.

---