# Research Papers — 2026-04 · bundle page 131 of 233 > Full-text markdown of papers #6501–6550 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/131 [← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/130.md) [Next bundle →](https://aipod.dev/papers/archive/2026-04/page/132.md) ### MEME-Fusion@CHiPSAL 2026: Multimodal Ablation Study of Hate Detection and Sentiment Analysis on Nepali Memes Authors: Samir Wagle, Reewaj Khanal, Abiral Adhikari Published: 2026-04-13 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.14218v1 Hate speech detection in Devanagari-scripted social media memes presents compounded challenges: multimodal content structure, script-specific linguistic complexity, and extreme data scarcity in low-resource settings. This paper presents our system for the CHiPSAL 2026 shared task, addressing both Subtask A (binary hate speech detection) and Subtask B (three-class sentiment classification: positive, neutral, negative). We propose a hybrid cross-modal attention fusion architecture that combines CLIP (ViT-B/32) for visual encoding with BGE-M3 for multilingual text representation, connected through 4-head self-attention and a learnable gating network that dynamically weights modality contributions on a per-sample basis. Systematic evaluation across eight model configurations demonstrates that explicit cross-modal reasoning achieves a 5.9% F1-macro improvement over text-only baselines on Subtask A, while uncovering two unexpected but critical findings: English-centric vision models exhibit near-random performance on Devanagari script, and standard ensemble methods catastrophically degrade under data scarcity (N nearly equal to 850 per fold) due to correlated overfitting. The code can be accessed at https://github.com/Tri-Yantra-Technologies/MEME-Fusion/ --- ### Generalization Guarantees on Data-Driven Tuning of Gradient Descent with Langevin Updates Authors: Saumya Goyal, Rohith Rongali, Ritabrata Ray, Barnabás Póczos Published: 2026-04-13 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2604.13130v2 We study learning to learn through the lens of hyperparameter tuning. We propose the Langevin Gradient Descent Algorithm (LGD), which approximates the mean of the posterior distribution defined by the loss function and regularizer of a regression task with convex objective. For classification tasks, the LGD algorithm estimates the posterior probabilities of each class on the test set. We prove the existence of an optimal hyperparameter configuration for which the LGD algorithm achieves the Bayes' optimal solution for squared loss on regression tasks, and for which LGD closely approximates the posterior probabilities for well-specified classification tasks. Subsequently, we study generalization guarantees on meta learning optimal hyperparameters for the LGD algorithm from a given set of tasks in the data-driven setting. For a number of parameters $d$ and hyperparameter dimension $h$, we show a pseudo-dimension bound of $O(dh)$, up to logarithmic terms under mild assumptions on LGD. This matches the dependence of the bounds on number of parameters obtained in prior work for linear regression using the elastic net, which only allows for $h=2$ hyperparameters, and extends their bounds to regression on convex loss. Compared to bounds on regularized logistic regression that allow for only $h=1$ hyperparameter, our bounds improve greatly on the dependence on samples per task at the cost of worse dependence on the number of parameters by accounting for hardware-aware procedures. Finally, we show empirical evidence of the success of LGD and the meta learning procedure for few-shot learning on linear and logistic regression using synthetically created datasets. --- ### Generalization Guarantees on Data-Driven Tuning of Gradient Descent with Langevin Updates Authors: Saumya Goyal, Rohith Rongali, Ritabrata Ray, Barnabás Póczos Published: 2026-04-13 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2604.13130v1 We study learning to learn for regression problems through the lens of hyperparameter tuning. We propose the Langevin Gradient Descent Algorithm (LGD), which approximates the mean of the posterior distribution defined by the loss function and regularizer of a convex regression task. We prove the existence of an optimal hyperparameter configuration for which the LGD algorithm achieves the Bayes' optimal solution for squared loss. Subsequently, we study generalization guarantees on meta-learning optimal hyperparameters for the LGD algorithm from a given set of tasks in the data-driven setting. For a number of parameters $d$ and hyperparameter dimension $h$, we show a pseudo-dimension bound of $O(dh)$, upto logarithmic terms under mild assumptions on LGD. This matches the dimensional dependence of the bounds obtained in prior work for the elastic net, which only allows for $h=2$ hyperparameters, and extends their bounds to regression on convex loss. Finally, we show empirical evidence of the success of LGD and the meta-learning procedure for few-shot learning on linear regression using a few synthetically created datasets. --- ### Learning Probabilistic Responsibility Allocations for Multi-Agent Interactions Authors: Isaac Remy, Caleb Chang, Karen Leung Published: 2026-04-13 Categories: cs.MA, cs.LG, cs.RO, eess.SY arXiv: https://arxiv.org/abs/2604.13128v1 Human behavior in interactive settings is shaped not only by individual objectives but also by shared constraints with others, such as safety. Understanding how people allocate responsibility, i.e., how much one deviates from their desired policy to accommodate others, can inform the design of socially compliant and trustworthy autonomous systems. In this work, we introduce a method for learning a probabilistic responsibility allocation model that captures the multimodal uncertainty inherent in multi-agent interactions. Specifically, our approach leverages the latent space of a conditional variational autoencoder, combined with techniques from multi-agent trajectory forecasting, to learn a distribution over responsibility allocations conditioned on scene and agent context. Although ground-truth responsibility labels are unavailable, the model remains tractable by incorporating a differentiable optimization layer that maps responsibility allocations to induced controls, which are available. We evaluate our method on the INTERACTION driving dataset and demonstrate that it not only achieves strong predictive performance but also provides interpretable insights, through the lens of responsibility, into patterns of multi-agent interaction. --- ### Graph Propagated Projection Unlearning: A Unified Framework for Vision and Audio Discriminative Models Authors: Shreyansh Pathak, Jyotishman Das Published: 2026-04-13 Categories: cs.CV, cs.AI, cs.SD arXiv: https://arxiv.org/abs/2604.13127v2 The need to selectively and efficiently erase learned information from deep neural networks is becoming increasingly important for privacy, regulatory compliance, and adaptive system design. We introduce Graph-Propagated Projection Unlearning (GPPU), a unified and scalable algorithm for class-level unlearning that operates across both vision and audio models. GPPU employs graph-based propagation to identify class-specific directions in the feature space and projects representations onto the orthogonal subspace, followed by targeted fine-tuning, to ensure that target class information is effectively and irreversibly removed. Through comprehensive evaluations on six vision datasets and two large-scale audio benchmarks spanning a variety of architectures including CNNs, Vision Transformers, and Audio Transformers, we demonstrate that GPPU achieves highly efficient unlearning, realizing 10-20x speedups over prior methodologies while preserving model utility on retained classes. Our framework provides a principled and modality-agnostic approach to machine unlearning, evaluated at a scale that has received limited attention in prior work, contributing toward more efficient and responsible deep learning. --- ### Synthetic Tabular Generators Fail to Preserve Behavioral Fraud Patterns: A Benchmark on Temporal, Velocity, and Multi-Account Signals Authors: Bhavana Sajja Published: 2026-04-13 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.13125v1 We introduce behavioral fidelity -- a third evaluation dimension for synthetic tabular data that measures whether generated data preserves the temporal, sequential, and structural behavioral patterns that distinguish real-world entity activity. Existing frameworks evaluate statistical fidelity (marginal distributions and correlations) and downstream utility (classifier AUROC on synthetic-trained models), but neither tests for the behavioral signals that operational detection and analysis systems actually rely on. We formalize a taxonomy of four behavioral fraud patterns (P1-P4) covering inter-event timing, burst structure, multi-account graph motifs, and velocity-rule trigger rates; define a degradation ratio metric calibrated to a real-data noise floor (1.0 = matches real variability, k = k-times worse); and prove that row-independent generators -- the dominant paradigm -- are structurally incapable of reproducing P3 graph motifs (Proposition 1) and produce non-positive within-entity IET autocorrelation (Proposition 2), making the positive burst fingerprint of fraud sequences unachievable regardless of architecture or training data size. We benchmark CTGAN, TVAE, GaussianCopula, and TabularARGN on IEEE-CIS Fraud Detection and the Amazon Fraud Dataset. All four fail severely: on IEEE-CIS composite degradation ratios range from 24.4x (TVAE) to 39.0x (GaussianCopula); on Amazon FDB, row-independent generators score 81.6-99.7x, while TabularARGN achieves 17.2x. We document generator-specific failure modes and their resolutions. The P1-P4 framework extends to any domain with entity-level sequential tabular data, including healthcare and network security. We release our evaluation framework as open source. --- ### Spectral Entropy Collapse as a Phase Transition in Delayed Generalisation: An Interventional and Predictive Framework for Grokkin Authors: Truong Xuan Khanh, Truong Quynh Hoa, Luu Duc Trung, Phan Thanh Duc Published: 2026-04-13 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.13123v2 Grokking - the delayed transition from memorisation to generalisation in neural networks - remains poorly understood. We study this phenomenon through the geometry of learned representations and identify a consistent empirical signature preceding generalisation: collapse of the spectral entropy of the representation covariance matrix. Across modular arithmetic tasks and multiple random seeds, spectral entropy decreases gradually during training and crosses a stable task-specific threshold before test accuracy rises. A representation-mixing intervention that delays this collapse also delays grokking, including under norm-matched controls, indicating that the effect is not explained by parameter norm alone. We further show that the entropy gap predicts the remaining time until grokking with useful out-of-sample accuracy. To probe the structure underlying this transition, we introduce a Fourier-alignment observable for cyclic-group tasks. Entropy collapse is strongly coupled to the emergence of Fourier-aligned representations, suggesting that spectral entropy tracks concentration of the representation into task-structured directions rather than generic compression alone. The same qualitative dynamics appear in non-abelian group composition tasks, while MLP controls show that entropy collapse by itself is insufficient for grokking in the absence of appropriate inductive bias. Taken together, the results support a view of grokking as a representational phase transition with an observable geometric signature. We discuss the scope and limitations of this interpretation, connections to recent feature-learning and spectral-dynamics work, and directions for testing whether similar transitions appear in larger-scale learning systems. --- ### AgentForge: Execution-Grounded Multi-Agent LLM Framework for Autonomous Software Engineering Authors: Rajesh Kumar, Waqar Ali, Junaid Ahmed, Najma Imtiaz Ali, Shaban Usman Published: 2026-04-13 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2604.13120v1 Large language models generate plausible code but cannot verify correctness. Existing multi-agent systems simulate execution or leave verification optional. We introduce execution-grounded verification as a first-class principle: every code change must survive sandboxed execution before propagation. We instantiate this principle in AGENTFORGE, a multi-agent framework where Planner, Coder, Tester, Debugger, and Critic agents coordinate through shared memory and a mandatory Docker sandbox. We formalize software engineering with LLMs as an iterative decision process over repository states, where execution feedback provides a stronger supervision signal than next-token likelihood. AGENTFORGE achieves 40.0\% resolution on SWE-BENCH Lite, outperforming single-agent baselines by 26--28 points. Ablations confirm that execution feedback and role decomposition each independently drive performance. The framework is open-source at https://github.com/raja21068/AutoCodeAI. --- ### LASA: Language-Agnostic Semantic Alignment at the Semantic Bottleneck for LLM Safety Authors: Junxiao Yang, Haoran Liu, Jinzhe Tu, Jiale Cheng, Zhexin Zhang, Shiyao Cui, Jiaqi Weng, Jialing Tao, Hui Xue, Hongning Wang, Han Qiu, Minlie Huang Published: 2026-04-13 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.12710v2 Large language models (LLMs) often demonstrate strong safety performance in high-resource languages, yet exhibit severe vulnerabilities when queried in low-resource languages. We attribute this gap to a mismatch between language-agnostic semantic understanding ability and language-dominant safety alignment biased toward high-resource languages. Consistent with this hypothesis, we empirically identify the semantic bottleneck in LLMs, an intermediate layer in which the geometry of model representations is governed primarily by shared semantic content rather than language identity. Building on this observation, we propose Language-Agnostic Semantic Alignment (LASA), which anchors safety alignment directly in semantic bottlenecks. Experiments show that LASA substantially improves safety across all languages: average attack success rate (ASR) drops from 24.7% to 2.8% on LLaMA-3.1-8B-Instruct and remains around 3-4% across Qwen2.5 and Qwen3 Instruct models (7B-32B). Together, our analysis and method offer a representation-level perspective on LLM safety, suggesting that safety alignment requires anchoring safety understanding not in surface text, but in the model's language-agnostic semantic space. --- ### ViLL-E: Video LLM Embeddings for Retrieval Authors: Rohit Gupta, Jayakrishnan Unnikrishnan, Fan Fei, Sheng Liu, Son Tran, Mubarak Shah Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12148v1 Video Large Language Models (VideoLLMs) excel at video understanding tasks where outputs are textual, such as Video Question Answering and Video Captioning. However, they underperform specialized embedding-based models in Retrieval tasks, such as Text-toVideo Retrieval and Moment Retrieval. We introduce ViLL-E (Video-LLM-Embed), a unified VideoLLM architecture endowed with a novel embedding generation mechanism that allows the model to "think longer" for complex videos and stop early for easy ones. We train this model with a three-stage training methodology combining generative and contrastive learning: initial large-scale pre-training with video-caption pairs; followed by continual training on a smaller, detailed-caption dataset; and concluding with task-specific fine-tuning on a novel multi-task dataset covering Video QA, Temporal Localization, Video Retrieval, and Video-Text Matching. Our model significantly improves temporal localization (on avg. 7% over other VideoLLMs) and video retrieval (up to 4% over dual encoder models), achieving performance comparable to state-of-the-art specialized embedding models while remaining competitive on VideoQA tasks. Furthermore, our joint contrastive-generative training unlocks new zero-shot capabilities, significantly outperforming state-of-the-art methods in composed video retrieval (+5% over SotA) and retrieval from long text (+2% over SotA). --- ### From Plan to Action: How Well Do Agents Follow the Plan? Authors: Shuyang Liu, Saman Dehghan, Jatin Ganhotra, Martin Hirzel, Reyhaneh Jabbarvand Published: 2026-04-13 Categories: cs.SE, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.12147v3 Agents are commonly instructed to follow a task-specific plan for guidance. However, it is unknown to what extent agents actually follow instructed plans. Without such an analysis, determining the extent agents comply with a given plan, it is impossible to assess whether a solution was reached through correct strategic reasoning or through other means, e.g., data contamination or overfitting to a benchmark. This paper presents the first extensive, systematic analysis of plan compliance in programming agents, examining 21,120 trajectories from SWE-agent across four LLMs on SWE-bench Verified and SWE-bench Pro under eight plan variations. Without an explicit plan, agents fall back on internalized workflows during training, which are often incomplete, overfit, or inconsistently applied. Providing the standard plan improves issue resolution, and we observe that periodic plan reminders can mitigate plan violations and improve task success. A subpar plan hurts performance even more than no plan at all. Surprisingly, inserting additional task-relevant phases in the early stage can degrade performance, particularly when these phases do not align with the model's internal problem-solving strategy. These findings call for fine-tuning paradigms that teach models to follow instructed plans, rather than encoding task-specific plans in them, so that they reason and act adaptively, rather than memorizing workflows. --- ### Evaluating Plan Compliance in Autonomous Programming Agents Authors: Shuyang Liu, Saman Dehghan, Jatin Ganhotra, Martin Hirzel, Reyhaneh Jabbarvand Published: 2026-04-13 Categories: cs.SE, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.12147v2 Agents aspire to eliminate the need for task-specific prompt crafting through autonomous reason-act-observe loops. Still, they are commonly instructed to follow a task-specific plan for guidance, e.g., to resolve software issues following phases for navigation, reproduction, patch, and validation. Unfortunately, it is unknown to what extent agents actually follow such instructed plans. Without such an analysis, determining the extent agents comply with a given plan, it is impossible to assess whether a solution was reached through correct strategic reasoning or through other means, e.g., data contamination or overfitting to a benchmark. This paper presents the first extensive, systematic analysis of plan compliance in programming agents, examining 16,991 trajectories from SWE-agent across four LLMs on SWE-bench Verified and SWE-bench Pro under eight plan variations. Without an explicit plan, agents fall back on workflows internalized during training, which are often incomplete, overfit, or inconsistently applied. Providing the standard plan improves issue resolution, and we observe that periodic plan reminders can mitigate plan violations and improve task success. A subpar plan hurts performance even more than no plan at all. Surprisingly, augmenting a plan with additional task-relevant phases in the early stage can degrade performance, particularly when these phases do not align with the model's internal problem-solving strategy. These findings highlight a research gap: fine-tuning paradigms that teach models to follow instructed plans, rather than encoding task-specific plans in them. This requires teaching models to reason and act adaptively, rather than memorizing workflows. --- ### XANE(3): An E(3)-Equivariant Graph Neural Network for Accurate Prediction of XANES Spectra from Atomic Structures Authors: Vitor F. Grizzi, Luke N. Pretzie, Jiayi Xu, Cong Liu Published: 2026-04-13 Categories: cs.LG, cond-mat.mtrl-sci, physics.chem-ph arXiv: https://arxiv.org/abs/2604.12140v1 We present XANE(3), a physics-based E(3)-equivariant graph neural network for predicting X-ray absorption near-edge structure (XANES) spectra directly from atomic structures. The model combines tensor-product message passing with spherical harmonic edge features, absorber-query attention pooling, custom equivariant layer normalization, adaptive gated residual connections, and a spectral readout based on a multi-scale Gaussian basis with an optional sigmoidal background term. To improve line-shape fidelity, training is performed with a composite objective that includes pointwise spectral reconstruction together with first- and second-derivative matching terms. We evaluate the model on a dataset of 5,941 FDMNES simulations of iron oxide surface facets and obtain a spectrum mean squared error of $1.0 \times 10^{-3}$ on the test set. The model accurately reproduces the main edge structure, relative peak intensities, pre-edge features, and post-edge oscillations. Ablation studies show that the derivative-aware objective, custom equivariant normalization, absorber-conditioned attention pooling, adaptive gated residual mixing, and global background term each improve performance. Interestingly, a capacity-matched scalar-only variant achieves comparable pointwise reconstruction error but reduced derivative-level fidelity, indicating that explicit tensorial channels are not strictly required for low intensity error on this dataset, although they remain beneficial for capturing finer spectral structure. These results establish XANE(3) as an accurate and efficient surrogate for XANES simulation and offer a promising route toward accelerated spectral prediction, ML-assisted spectroscopy, and data-driven materials discovery. --- ### Retrieval-Augmented Generation Must Move Beyond Factual Grounding to Represent Diverse Opinions Authors: Aditya Agrawal, Alwarappan Nakkiran, Darshan Fofadiya, Alex Karlsson, Harsha Aduri, Aman Singh Thakur Published: 2026-04-13 Categories: cs.AI, cs.CL, cs.IR arXiv: https://arxiv.org/abs/2604.12138v3 This position paper argues that Retrieval-Augmented Generation (RAG) systems exhibit a factual bias-optimizing for epistemic uncertainty reduction while ignoring the aleatoric uncertainty inherent in opinion-rich content. This misalignment demands a paradigm shift in RAG system design. A survey of 34 major RAG benchmarks reveals that only one addresses opinion synthesis, confirming that the bias is structural and embedded in datasets, retrieval-generation objectives, and evaluation metrics alike. Beyond technical limitations, this bias poses risks to transparent and accountable AI. Namely, echo chamber effects that amplify dominant viewpoints, which can lead to opinion manipulation and under-representation of minority voices. We formalize the problem through the lens of uncertainty quantification, showing that factual queries should minimize posterior entropy while opinion queries must preserve it. We derive a unified objective over coverage, fidelity, and fairness using the Wasserstein distance. As an existence proof, we present Opinion-Aware RAG (O-RAG), an architecture featuring LLM-based opinion extraction and entity-linked opinion metadata. We evaluate it across two domains -- e-commerce seller forums and public hotel reviews. Experiments demonstrate 18-48% reduction in Wasserstein distance to corpus-level sentiment distributions, +26.8% sentiment diversity, and +42.7% entity match rate. Human evaluators preferred opinion-enriched generation 79.2% of the time. We propose a research agenda and argue that as RAG systems increasingly mediate access to information, their ability to represent diverse perspectives is of the essence. --- ### Observing the unobserved confounding through its effects: toward randomized trial-like estimates from real-world survival data Authors: Vasiliki Stoumpou, Dimitris Bertsimas, Samuel Singer, Georgios Antonios Margonis Published: 2026-04-13 Categories: stat.AP, cs.AI, stat.ME arXiv: https://arxiv.org/abs/2604.12137v1 Background: Randomized controlled trials (RCTs) are costly, time-consuming, and often infeasible, while treatment-effect estimation from observational data is limited by unobserved confounding. Methods: We developed a three-step framework to address unobserved confounding in observational survival data. First, we infer a latent prognostic factor (U) from restricted mean survival time (RMST) discrepancies between patients with similar observed factors, the same treatment, and divergent outcomes, leveraging the idea that the aggregate effect of unmeasured factors can be inferred even if individual factors cannot. Second, we balance U with observed baseline covariates using prognostic matching, entropy balancing, or inverse probability of treatment weighting. Third, we apply multivariable survival analysis to estimate hazard ratios (HRs). We evaluated the framework in three observational cohorts with RCT benchmarks, two RCT cohorts, and six multicenter observational cohorts. Results: In three observational cohorts (nine comparisons), balancing U improved agreement with trial HRs in all cases; in the strongest settings, it reduced absolute log-HR error by approximately ten-fold versus using observed covariates alone (mean reduction 0.344; p=0.001). In two RCT cohorts, U was balanced across arms (most SMDs <0.1) and adjustment had minimal impact on log-HRs (mean absolute change 0.08). Across six multicenter cohorts, balancing U within centers reduced cross-center dispersion in chemotherapy log-HR estimates (mean reduction 0.147; p=0.016); when populations were directly balanced across centers to account for case-mix differences, cross-center survival differences were narrowed in 75%-100% of comparisons. Conclusions: Inferring and balancing a latent prognostic signal may reduce unobserved confounding and improve treatment-effect estimation from real-world data. --- ### Towards Platonic Representation for Table Reasoning: A Foundation for Permutation-Invariant Retrieval Authors: Willy Carlos Tchuitcheu, Tan Lu, Ann Dooms Published: 2026-04-13 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.12133v1 Historical approaches to Table Representation Learning (TRL) have largely adopted the sequential paradigms of Natural Language Processing (NLP). We argue that this linearization of tables discards their essential geometric and relational structure, creating representations that are brittle to layout permutations. This paper introduces the Platonic Representation Hypothesis (PRH) for tables, positing that a semantically robust latent space for table reasoning must be intrinsically Permutation Invariant (PI). To ground this hypothesis, we first conduct a retrospective analysis of table-reasoning tasks, highlighting the pervasive serialization bias that compromises structural integrity. We then propose a formal framework to diagnose this bias, introducing two principled metrics based on Centered Kernel Alignment (CKA): (i) PI, which measures embedding drift under complete structural derangement, and (ii) rho, a Spearman-based metric that tracks the convergence of latent structures toward a canonical form as structural information is incrementally restored. Our empirical analysis quantifies an expected flaw in modern Large Language Models (LLMs): even minor layout permutations induce significant, disproportionate semantic shifts in their table embeddings. This exposes a fundamental vulnerability in RAG systems, in which table retrieval becomes fragile to layout-dependent noise rather than to semantic content. In response, we present a novel, structure-aware TRL encoder architecture that explicitly enforces the cognitive principle of cell header alignment. This model demonstrates superior geometric stability and moves towards the PI ideal. Our work provides both a foundational critique of linearized table encoders and the theoretical scaffolding for semantically stable, permutation invariant retrieval, charting a new direction for table reasoning in information systems. --- ### Aethon: A Reference-Based Replication Primitive for Constant-Time Instantiation of Stateful AI Agents Authors: Swanand Rao, Kiran Kashalkar, Parvathi Somashekar, Priya Krishnan Published: 2026-04-13 Categories: cs.AI, cs.AR, cs.DC, cs.MA arXiv: https://arxiv.org/abs/2604.12129v1 The transition from stateless model inference to stateful agentic execution is reshaping the systems assumptions underlying modern AI infrastructure. While large language models have made persistent, tool-using, and collaborative agents technically viable, existing runtime architectures remain constrained by materialization-heavy instantiation models that impose significant latency and memory overhead. This paper introduces Aethon, a reference-based replication primitive for near-constant-time instantiation of stateful AI agents. Rather than reconstructing agents as fully materialized objects, Aethon represents each instance as a compositional view over stable definitions, layered memory, and local contextual overlays. By shifting instantiation from duplication to reference, Aethon decouples creation cost from inherited structure. We present the conceptual framework, system architecture, and memory model underlying Aethon, including layered inheritance and copy-on-write semantics. We analyze its implications for complexity, scalability, multi-agent orchestration, and enterprise governance. We argue that reference-based instantiation is not merely an optimization, but a more appropriate systems abstraction for production-scale agentic software. Aethon points toward a new class of AI infrastructure in which agents become lightweight, composable execution identities that can be spawned, specialized, and governed at scale. --- ### When Self-Reference Fails to Close: Matrix-Level Dynamics in Large Language Models Authors: Ji Ho Bae Published: 2026-04-13 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.12128v1 We investigate how self-referential inputs alter the internal matrix dynamics of large language models. Measuring 106 scalar metrics across up to 7 analysis passes on four models from three architecture families -- Qwen3-VL-8B, Llama-3.2-11B, Llama-3.3-70B, and Gemma-2-9B -- over 300 prompts in a 14-level hierarchy at three temperatures ($T \in \{0.0, 0.3, 0.7\}$), we find that self-reference alone is not destabilizing: grounded self-referential statements and meta-cognitive prompts are markedly more stable than paradoxical self-reference on key collapse-related metrics, and on several such metrics can be as stable as factual controls. Instability concentrates in prompts inducing non-closing truth recursion (NCTR) -- truth-value computations with no finite-depth resolution. NCTR prompts produce anomalously elevated attention effective rank -- indicating attention reorganization with global dispersion rather than simple concentration collapse -- and key metrics reach Cohen's $d = 3.14$ (attention effective rank) to $3.52$ (variance kurtosis) vs. stable self-reference in the 70B model; 281/397 metric-model combinations differentiate NCTR from stable self-reference after FDR correction ($q < 0.05$), 198 with $|d| > 0.8$. Per-layer SVD confirms disruption at every sampled layer ($d > +1.0$ in all three models analyzed), ruling out aggregation artifacts. A classifier achieves AUC $0.81$-$0.90$; 30 minimal pairs yield 42/387 significant combinations; 43/106 metrics replicate across all four models. We connect these observations to three classical matrix-semigroup problems and propose, as a conjecture, that NCTR forces finite-depth transformers toward dynamical regimes where these problems concentrate. NCTR prompts also produce elevated contradictory output ($+34$-$56$ percentage points vs. controls), suggesting practical relevance for understanding self-referential failure modes. --- ### Long-Horizon Plan Execution in Large Tool Spaces through Entropy-Guided Branching Authors: Rongzhe Wei, Ge Shi, Min Cheng, Na Zhang, Pan Li, Sarthak Ghosh, Vaibhav Gorde, Leman Akoglu Published: 2026-04-13 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.12126v1 Large Language Models (LLMs) have significantly advanced tool-augmented agents, enabling autonomous reasoning via API interactions. However, executing multi-step tasks within massive tool libraries remains challenging due to two critical bottlenecks: (1) the absence of rigorous, plan-level evaluation frameworks and (2) the computational demand of exploring vast decision spaces stemming from large toolsets and long-horizon planning. To bridge these gaps, we first introduce SLATE (Synthetic Large-scale API Toolkit for E-commerce), a large-scale context-aware benchmark designed for the automated assessment of tool-integrated agents. Unlike static metrics, SLATE accommodates diverse yet functionally valid execution trajectories, revealing that current agents struggle with self-correction and search efficiency. Motivated by these findings, we next propose Entropy-Guided Branching (EGB), an uncertainty-aware search algorithm that dynamically expands decision branches where predictive entropy is high. EGB optimizes the exploration-exploitation trade-off, significantly enhancing both task success rates and computational efficiency. Extensive experiments on SLATE demonstrate that our dual contribution provides a robust foundation for developing reliable and scalable LLM agents in tool-rich environments. --- ### Beyond Perception Errors: Semantic Fixation in Large Vision-Language Models Authors: Md Tanvirul Alam Published: 2026-04-13 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2604.12119v1 Large vision-language models (VLMs) often rely on familiar semantic priors, but existing evaluations do not cleanly separate perception failures from rule-mapping failures. We study this behavior as semantic fixation: preserving a default interpretation even when the prompt specifies an alternative, equally valid mapping. To isolate this effect, we introduce VLM-Fix, a controlled benchmark over four abstract strategy games that evaluates identical terminal board states under paired standard and inverse rule formulations. Across 14 open and closed VLMs, accuracy consistently favors standard rules, revealing a robust semantic-fixation gap. Prompt interventions support this mechanism: neutral alias prompts substantially narrow the inverse-rule gap, while semantically loaded aliases reopen it. Post-training is strongly rule-aligned: training on one rule improves same-rule transfer but hurts opposite-rule transfer, while joint-rule training improves broader transfer. To test external validity beyond synthetic games, we evaluate analogous defamiliarization interventions on VLMBias and observe the same qualitative pattern. Finally, late-layer activation steering partially recovers degraded performance, indicating that semantic-fixation errors are at least partly editable in late representations. Project page, code, and dataset available at https://maveryn.github.io/vlm-fix/. --- ### The A-R Behavioral Space: Execution-Level Profiling of Tool-Using Language Model Agents in Organizational Deployment Authors: Shasha Yu, Fiona Carroll, Barry L. Bentley Published: 2026-04-13 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2604.12116v2 Large language models (LLMs) are increasingly deployed as tool-augmented agents capable of executing system-level operations. While existing benchmarks primarily assess textual alignment or task success, less attention has been paid to the structural relationship between linguistic signaling and executable behavior under varying autonomy scaffolds. This study introduces an execution-layer be-havioral measurement approach based on a two-dimensional A-R space defined by Action Rate (A) and Refusal Signal (R), with Divergence (D) capturing coor-dination between the two. Models are evaluated across four normative regimes (Control, Gray, Dilemma, and Malicious) and three autonomy configurations (di-rect execution, planning, and reflection). Rather than assigning aggregate safety scores, the method characterizes how execution and refusal redistribute across contextual framing and scaffold depth. Empirical results show that execution and refusal constitute separable behavioral dimensions whose joint distribution varies systematically across regimes and autonomy levels. Reflection-based scaffolding often shifts configurations toward higher refusal in risk-laden contexts, but redis-tribution patterns differ structurally across models. The A-R representation makes cross-sectional behavioral profiles, scaffold-induced transitions, and coordination variability directly observable. By foregrounding execution-layer characterization over scalar ranking, this work provides a deployment-oriented lens for analyzing and selecting tool-enabled LLM agents in organizational settings where execution privileges and risk tolerance vary. --- ### HTDC: Hesitation-Triggered Differential Calibration for Mitigating Hallucination in Large Vision-Language Models Authors: Xinyun Liu Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12115v1 Large vision-language models (LVLMs) achieve strong multimodal performance, but still suffer from hallucinations caused by unstable visual grounding and over-reliance on language priors. Existing training-free decoding methods typically apply calibration at every decoding step, introducing unnecessary computation and potentially disrupting stable predictions. We address this problem by identifying layer-wise hesitation, a simple signal of grounding instability reflected by fluctuations in token preference across intermediate layers. Based on this observation, we propose Hesitation-Triggered Differential Calibration (HTDC), a training-free decoding framework that preserves standard full-branch inference and activates calibration only at hesitation-prone steps. When triggered, HTDC contrasts the full branch with two lightweight probes, a visual-nullification probe and a semantic-nullification probe, to suppress hallucination-prone candidates while avoiding unnecessary intervention on stable steps. Experiments on representative hallucination benchmarks show that HTDC consistently reduces hallucinations while maintaining strong task accuracy, achieving a favorable trade-off between effectiveness and computational overhead. --- ### PR-MaGIC: Prompt Refinement Via Mask Decoder Gradient Flow For In-Context Segmentation Authors: Minjae Lee, Sungwoo Hur, Soojin Hwang, Won Hwa Kim Published: 2026-04-13 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.12113v1 Visual Foundation Models (VFMs) such as the Segment Anything Model (SAM) have significantly advanced broad use of image segmentation. However, SAM and its variants necessitate substantial manual effort for prompt generation and additional training for specific applications. Recent approaches address these limitations by integrating SAM into in-context (one/few shot) segmentation, enabling auto-prompting through semantic alignment between query and support images. Despite these efforts, they still generate sub-optimal prompts that degrade segmentation quality due to visual inconsistencies between support and query images. To tackle this limitation, we introduce PR-MaGIC (Prompt Refinement via Mask Decoder Gradient Flow for In-Context Segmentation), a training-free test-time framework that refines prompts via gradient flow derived from SAM's mask decoder. PR-MaGIC seamlessly integrates into in-context segmentation frameworks, being theoretically grounded yet practically stabilized through a simple top-1 selection strategy that ensures robust performance across samples. Extensive evaluations demonstrate that PR-MaGIC consistently improves segmentation quality across various benchmarks, effectively mitigating inadequate prompts without requiring additional training or architectural modifications. --- ### SOLARIS: Speculative Offloading of Latent-bAsed Representation for Inference Scaling Authors: Zikun Liu, Liang Luo, Qianru Li, Zhengyu Zhang, Wei Ling, Jingyi Shen, Zeliang Chen, Yaning Huang, Jingxian Huang, Abdallah Aboelela, Chonglin Sun, Feifan Gu et al. Published: 2026-04-13 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.12110v2 Recent advances in recommendation scaling laws have led to foundation models of unprecedented complexity. While these models offer superior performance, their computational demands make real-time serving impractical, often forcing practitioners to rely on knowledge distillation-compromising serving quality for efficiency. To address this challenge, we present SOLARIS (Speculative Offloading of Latent-bAsed Representation for Inference Scaling), a novel framework inspired by speculative decoding. SOLARIS proactively precomputes user-item interaction embeddings by predicting which user-item pairs are likely to appear in future requests, and asynchronously generating their foundation model representations ahead of time. This approach decouples the costly foundation model inference from the latency-critical serving path, enabling real-time knowledge transfer from models previously considered too expensive for online use. Deployed across Meta's advertising system serving billions of daily requests, SOLARIS achieves 0.67% revenue-driving top-line metrics gain, demonstrating its effectiveness at scale. --- ### LLM-Based Automated Diagnosis Of Integration Test Failures At Google Authors: Celal Ziftci, Ray Liu, Spencer Greene, Livio Dalloro Published: 2026-04-13 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2604.12108v1 Integration testing is critical for the quality and reliability of complex software systems. However, diagnosing their failures presents significant challenges due to the massive volume, unstructured nature, and heterogeneity of logs they generate. These result in a high cognitive load, low signal-to-noise ratio, and make diagnosis difficult and time-consuming. Developers complain about these difficulties consistently and report spending substantially more time diagnosing integration test failures compared to unit test failures. To address these shortcomings, we introduce Auto-Diagnose, a novel diagnosis tool that leverages LLMs to help developers efficiently determine the root cause of integration test failures. Auto-Diagnose analyzes failure logs, produces concise summaries with the most relevant log lines, and is integrated into Critique, Google's internal code review system, providing contextual and in-time assistance. Based on our case studies, Auto-Diagnose is highly effective. A manual evaluation conducted on 71 real-world failures demonstrated 90.14% accuracy in diagnosing the root cause. Following its Google-wide deployment, Auto-Diagnose was used across 52, 635 distinct failing tests. User feedback indicated that the tool was deemed "Not helpful" in only 5.8% of cases, and it was ranked #14 in helpfulness among 370 tools that post findings in Critique. Finally, user interviews confirmed the perceived usefulness of Auto-Diagnose and positive reception of integrating automatic diagnostic assistance into existing workflows. We conclude that LLMs are highly successful in diagnosing integration test failures due to their capacity to process and summarize complex textual data. Integrating such AI-powered tooling automatically into developers' daily workflows is perceived positively, with the tool's accuracy remaining a critical factor in shaping developer perception and adoption. --- ### Parametric Interpolation of Dynamic Mode Decomposition for Predicting Nonlinear Systems Authors: Ananda Chakrabarti, Haitham H. Saleh, Indranil Nayak, Balasubramaniam Shanker, Fernando L. Teixeira, Debdipta Goswami Published: 2026-04-13 Categories: eess.SY, cs.LG arXiv: https://arxiv.org/abs/2604.12103v1 We present parameter-interpolated dynamic mode decomposition (piDMD), a parametric reduced-order modeling framework that embeds known parameter-affine structure directly into the DMD regression step. Unlike existing parametric DMD methods which interpolate modes, eigenvalues, or reduced operators and can be fragile with sparse training data or multi-dimensional parameter spaces, piDMD learns a single parameter-affine Koopman surrogate reduced order model (ROM) across multiple training parameter samples and predicts at unseen parameter values without retraining. We validate piDMD on fluid flow past a cylinder, electron beam oscillations in transverse magnetic fields, and virtual cathode oscillations -- the latter two being simulated using an electromagnetic particle-in-cell (EMPIC) method. Across all benchmarks, piDMD achieves accurate long-horizon predictions and improved robustness over state-of-the-art interpolation-based parametric DMD baselines, with less training samples and with multi-dimensional parameter spaces. --- ### Spatial Atlas: Compute-Grounded Reasoning for Spatial-Aware Research Agent Benchmarks Authors: Arun Sharma Published: 2026-04-13 Categories: cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2604.12102v2 We introduce compute-grounded reasoning (CGR), a design paradigm for spatial-aware research agents in which every answerable sub-problem is resolved by deterministic computation before a language model is asked to generate. Spatial Atlas instantiates CGR as a single Agent-to-Agent (A2A) server that handles two challenging benchmarks: FieldWorkArena, a multimodal spatial question-answering benchmark spanning factory, warehouse, and retail environments, and MLE-Bench, a suite of 75 Kaggle machine learning competitions requiring end-to-end ML engineering. A structured spatial scene graph engine extracts entities and relations from vision descriptions, computes distances and safety violations deterministically, then feeds computed facts to large language models, thereby avoiding hallucinated spatial reasoning. Entropy-guided action selection maximizes information gain per step and routes queries across a three-tier frontier model stack (OpenAI + Anthropic). A self-healing ML pipeline with strategy-aware code generation, a score-driven iterative refinement loop, and a prompt-based leak audit registry round out the system. We evaluate across both benchmarks and show that CGR yields competitive accuracy while maintaining interpretability through structured intermediate representations and deterministic spatial computations. --- ### PC-MIL: Decoupling Feature Resolution from Supervision Scale in Whole-Slide Learning Authors: Syed Fahim Ahmed, Gnanesh Rasineni, Florian Koehler, Abu Zahid Bin Aziz, Mei Wang, Attila Gyulassy, Brian Summa, J. Quincy Brown, Valerio Pascucci, Shireen Y. Elhabian Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12100v1 Whole-slide image (WSI) classification in computational pathology is commonly formulated as slide-level Multiple Instance Learning (MIL) with a single global bag representation. However, slide-level MIL is fundamentally underconstrained: optimizing only global labels encourages models to aggregate features without learning anatomically meaningful localization. This creates a mismatch between the scale of supervision and the scale of clinical reasoning. Clinicians assess tumor burden, focal lesions, and architectural patterns within millimeter-scale regions, whereas standard MIL is trained only to predict whether "somewhere in the slide there is cancer." As a result, the model's inductive bias effectively erases anatomical structure. We propose Progressive-Context MIL (PC-MIL), a framework that treats the spatial extent of supervision as a first-class design dimension. Rather than altering magnification, patch size, or introducing pixel-level segmentation, we decouple feature resolution from supervision scale. Using fixed 20x features, we vary MIL bag extent in millimeter units and anchor supervision at a clinically motivated 2mm scale to preserve comparable tumor burden and avoid confounding scale with lesion density. PC-MIL progressively mixes slide- and region-level supervision in controlled proportions, enabling explicit train-context x test-context analysis. On 1,476 prostate WSIs from five public datasets for binary cancer detection, we show that anatomical context is an independent axis of generalization in MIL, orthogonal to feature resolution: modest regional supervision improves cross-context performance, and balanced multi-context training stabilizes accuracy across slide and regional evaluation without sacrificing global performance. These results demonstrate that supervision extent shapes MIL inductive bias and support anatomically grounded WSI generalization. --- ### The Effect of Document Selection on Query-focused Text Analysis Authors: Sandesh S Rangreji, Mian Zhong, Anjalie Field Published: 2026-04-13 Categories: cs.IR, cs.CL arXiv: https://arxiv.org/abs/2604.12099v1 Analyses of document collections often require selecting what data to analyze, as not all documents are relevant to a particular research question and computational constraints preclude analyzing all documents, yet little work has examined effects of selection strategy choices. We systematically evaluate seven selection methods (from random selection to hybrid retrieval) on outputs from four text analyses methods (LDA, BERTopic, TopicGPT, HiCode) over two datasets with 26 open-ended queries. Our evaluation reveals practice guidance: semantic or hybrid retrieval offer strong go-to approaches that avoid the pitfalls of weaker selection strategies and the unnecessary compute overhead of more complicated ones. Overall, our evaluation framework establishes data selection as a methodological decision, rather than a practical necessity, inviting the development of new strategies. --- ### Temporal Flattening in LLM-Generated Text: Comparing Human and LLM Writing Trajectories Authors: Zhanwei Cao, YeoJin Go, Yifan Hu, Shanu Sushmita Published: 2026-04-13 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.12097v1 Large language models (LLMs) are increasingly used in daily applications, from content generation to code writing, where each interaction treats the model as stateless, generating responses independently without memory. Yet human writing is inherently longitudinal: authors' styles and cognitive states evolve across months and years. This raises a central question: can LLMs reproduce such temporal structure across extended time periods? We construct and publicly release a longitudinal dataset of 412 human authors and 6,086 documents spanning 2012--2024 across three domains (academic abstracts, blogs, news) and compare them to trajectories generated by three representative LLMs under standard and history-conditioned generation settings. Using drift and variance-based metrics over semantic, lexical, and cognitive-emotional representations, we find temporal flattening in LLM-generated text. LLMs produce greater lexical diversity but exhibit substantially reduced semantic and cognitive-emotional drift relative to humans. These differences are highly predictive: temporal variability patterns alone achieve 94% accuracy and 98% ROC-AUC in distinguishing human from LLM trajectories. Our results demonstrate that temporal flattening persists regardless of whether LLMs generate independently or with access to incremental history, revealing a fundamental property of current deployment paradigms. This gap has direct implications for applications requiring authentic temporal structure, such as synthetic training data and longitudinal text modeling. --- ### LLM-HYPER: Generative CTR Modeling for Cold-Start Ad Personalization via LLM-Based Hypernetworks Authors: Luyi Ma, Wanjia Sherry Zhang, Zezhong Fan, Shubham Thakur, Kai Zhao, Kehui Yao, Ayush Agarwal, Rahul Iyer, Jason Cho, Jianpeng Xu, Evren Korpeoglu, Sushant Kumar et al. Published: 2026-04-13 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.12096v1 On online advertising platforms, newly introduced promotional ads face the cold-start problem, as they lack sufficient user feedback for model training. In this work, we propose LLM-HYPER, a novel framework that treats large language models (LLMs) as hypernetworks to directly generate the parameters of the click-through rate (CTR) estimator in a training-free manner. LLM-HYPER uses few-shot Chain-of-Thought prompting over multimodal ad content (text and images) to infer feature-wise model weights for a linear CTR predictor. By retrieving semantically similar past campaigns via CLIP embeddings and formatting them into prompt-based demonstrations, the LLM learns to reason about customer intent, feature influence, and content relevance. To ensure numerical stability and serviceability, we introduce normalization and calibration techniques that align the generated weights with production-ready CTR distributions. Extensive offline experiments show that LLM-HYPER significantly outperforms cold-start baselines in NDCG$@10$ by 55.9\%. Our real-world online A/B test on one of the top e-commerce platforms in the U.S. demonstrates the strong performance of LLM-HYPER, which drastically reduces the cold-start period and achieves competitive performance. LLM-HYPER has been successfully deployed in production. --- ### A Nonparametric Adaptive EWMA Control Chart for Binary Monitoring of Multiple Stream Processes Authors: Faruk Muritala, Austin Brown, Dhrubajyoti Ghosh, Sherry Ni Published: 2026-04-13 Categories: stat.ML, cs.LG, stat.AP, stat.ME arXiv: https://arxiv.org/abs/2604.12095v1 Monitoring binomial proportions across multiple independent streams is a critical challenge in Statistical Process Control (SPC), with applications from manufacturing to cybersecurity. While EWMA charts offer sensitivity to small shifts, existing implementations rely on asymptotic variance approximations that fail during early-phase monitoring. We introduce a Cumulative Standardized Binomial EWMA (CSB-EWMA) chart that overcomes this limitation by deriving the exact time-varying variance of the EWMA statistic for binary multiple-stream data, enabling adaptive control limits that ensure statistical rigor from the first sample. Through extensive simulations, we identify optimal smoothing (λ) and limit (L) parameters to achieve target in-control average run length (ARL0) of 370 and 500. The CSB-EWMA chart demonstrates rapid shift detection across both ARL0 targets, with out-of-control average run length (ARL1) dropping to 3-7 samples for moderate shifts (δ=0.2), and exhibits exceptional robustness across different data distributions, with low ARL1 Coefficients of Variation (CV < 0.10 for small shifts) for both ARL0 = 370 and 500. This work provides practitioners with a distribution-free, sensitive, and theoretically sound tool for early change detection in binomial multiple-stream processes. --- ### Robust Optimization for Mitigating Reward Hacking with Correlated Proxies Authors: Zixuan Liu, Xiaolin Sun, Zizhan Zheng Published: 2026-04-13 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.12086v1 Designing robust reinforcement learning (RL) agents in the presence of imperfect reward signals remains a core challenge. In practice, agents are often trained with proxy rewards that only approximate the true objective, leaving them vulnerable to reward hacking, where high proxy returns arise from unintended or exploitative behaviors. Recent work formalizes this issue using r-correlation between proxy and true rewards, but existing methods like occupancy-regularized policy optimization (ORPO) optimize against a fixed proxy and do not provide strong guarantees against broader classes of correlated proxies. In this work, we formulate reward hacking as a robust policy optimization problem over the space of all r-correlated proxy rewards. We derive a tractable max-min formulation, where the agent maximizes performance under the worst-case proxy consistent with the correlation constraint. We further show that when the reward is a linear function of known features, our approach can be adapted to incorporate this prior knowledge, yielding both improved policies and interpretable worst-case rewards. Experiments across several environments show that our algorithms consistently outperform ORPO in worst-case returns, and offer improved robustness and stability across different levels of proxy-true reward correlation. These results show that our approach provides both robustness and transparency in settings where reward design is inherently uncertain. The code is available at https://github.com/ZixuanLiu4869/reward_hacking. --- ### INST-Align: Implicit Neural Alignment for Spatial Transcriptomics via Canonical Expression Fields Authors: Bonian Han, Cong Qi, Przemyslaw Musialski, Zhi Wei Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12084v1 Spatial transcriptomics (ST) measures mRNA expression while preserving spatial organization, but multi-slice analysis faces two coupled difficulties: large non-rigid deformations across slices and inter-slice batch effects when alignment and integration are treated independently. We present INST-Align, an unsupervised pairwise framework that couples a coordinate-based deformation network with a shared Canonical Expression Field, an implicit neural representation mapping spatial coordinates to expression embeddings, for joint alignment and reconstruction. A two-phase training strategy first establishes a stable canonical embedding space and then jointly optimizes deformation and spatial-feature matching, enabling mutually constrained alignment and representation learning. Cross-slice parameter sharing of the canonical field regularizes ambiguous correspondences and absorbs batch variation. Across nine datasets, INST-Align achieves state-of-the-art mean OT Accuracy (0.702), NN Accuracy (0.719), and Chamfer distance, with Chamfer reductions of up to 94.9\% on large-deformation sections relative to the strongest baseline. The framework also yields biologically meaningful spatial embeddings and coherent 3D tissue reconstruction. The code will be released after review phase. --- ### Human-Inspired Context-Selective Multimodal Memory for Social Robots Authors: Hangyeol Kang, Slava Voloshynovskiy, Nadia Magnenat Thalmann Published: 2026-04-13 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.12081v1 Memory is fundamental to social interaction, enabling humans to recall meaningful past experiences and adapt their behavior accordingly based on the context. However, most current social robots and embodied agents rely on non-selective, text-based memory, limiting their ability to support personalized, context-aware interactions. Drawing inspiration from cognitive neuroscience, we propose a context-selective, multimodal memory architecture for social robots that captures and retrieves both textual and visual episodic traces, prioritizing moments characterized by high emotional salience or scene novelty. By associating these memories with individual users, our system enables socially personalized recall and more natural, grounded dialogue. We evaluate the selective storage mechanism using a curated dataset of social scenarios, achieving a Spearman correlation of 0.506, surpassing human consistency ($ρ=0.415$) and outperforming existing image memorability models. In multimodal retrieval experiments, our fusion approach improves Recall@1 by up to 13\% over unimodal text or image retrieval. Runtime evaluations confirm that the system maintains real-time performance. Qualitative analyses further demonstrate that the proposed framework produces richer and more socially relevant responses than baseline models. This work advances memory design for social robots by bridging human-inspired selectivity and multimodal retrieval to enhance long-term, personalized human-robot interaction. --- ### Robust Reasoning and Learning with Brain-Inspired Representations under Hardware-Induced Nonlinearities Authors: William Youngwoo Chung, Hamza Errahmouni Barkam, Tamoghno Das, Mohsen Imani Published: 2026-04-13 Categories: cs.ET, cs.LG arXiv: https://arxiv.org/abs/2604.12079v1 Traditional machine learning depends on high-precision arithmetic and near-ideal hardware assumptions, which is increasingly challenged by variability in aggressively scaled semiconductor devices. Compute-in-memory (CIM) architectures alleviate data-movement bottlenecks and improve energy efficiency yet introduce nonlinear distortions and reliability concerns. We address these issues with a hardware-aware optimization framework based on Hyperdimensional Computing (HDC), systematically compensating for non-ideal similarity computations in CIM. Our approach formulates encoding as an optimization problem, minimizing the Frobenius norm between an ideal kernel and its hardware-constrained counterpart, and employs a joint optimization strategy for end-to-end calibration of hypervector representations. Experimental results demonstrate that our method when applied to QuantHD achieves 84\% accuracy under severe hardware-induced perturbations, a 48\% increase over naive QuantHD under the same conditions. Additionally, our optimization is vital for graph-based HDC reliant on precise variable-binding for interpretable reasoning. Our framework preserves the accuracy of RelHD on the Cora dataset, achieving a 5.4$\times$ accuracy improvement over naive RelHD under nonlinear environments. By preserving HDC's robustness and symbolic properties, our solution enables scalable, energy-efficient intelligent systems capable of classification and reasoning on emerging CIM hardware. --- ### Narrative over Numbers: The Identifiable Victim Effect and its Amplification Under Alignment and Reasoning in Large Language Models Authors: Syed Rifat Raiyan Published: 2026-04-13 Categories: cs.CL, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2604.12076v1 The Identifiable Victim Effect (IVE) $-$ the tendency to allocate greater resources to a specific, narratively described victim than to a statistically characterized group facing equivalent hardship $-$ is one of the most robust findings in moral psychology and behavioural economics. As large language models (LLMs) assume consequential roles in humanitarian triage, automated grant evaluation, and content moderation, a critical question arises: do these systems inherit the affective irrationalities present in human moral reasoning? We present the first systematic, large-scale empirical investigation of the IVE in LLMs, comprising N=51,955 validated API trials across 16 frontier models spanning nine organizational lineages (Google, Anthropic, OpenAI, Meta, DeepSeek, xAI, Alibaba, IBM, and Moonshot). Using a suite of ten experiments $-$ porting and extending canonical paradigms from Small et al. (2007) and Kogut and Ritov (2005) $-$ we find that the IVE is prevalent but strongly modulated by alignment training. Instruction-tuned models exhibit extreme IVE (Cohen's d up to 1.56), while reasoning-specialized models invert the effect (down to d=-0.85). The pooled effect (d=0.223, p=2e-6) is approximately twice the single-victim human meta-analytic baseline (d$\approx$0.10) reported by Lee and Feeley (2016) $-$ and likely exceeds the overall human pooled effect by a larger margin, given that the group-victim human effect is near zero. Standard Chain-of-Thought (CoT) prompting $-$ contrary to its role as a deliberative corrective $-$ nearly triples the IVE effect size (from d=0.15 to d=0.41), while only utilitarian CoT reliably eliminates it. We further document psychophysical numbing, perfect quantity neglect, and marginal in-group/out-group cultural bias, with implications for AI deployment in humanitarian and ethical decision-making contexts. --- ### OpenTME: An Open Dataset of AI-powered H&E Tumor Microenvironment Profiles from TCGA Authors: Maaike Galama, Nina Kozar-Gillan, Christina Embacher, Todd Dembo, Cornelius Böhm, Evelyn Ramberger, Julika Ribbat-Idel, Rosemarie Krupar, Verena Aumiller, Miriam Hägele, Kai Standvoss, Gerrit Erdmann et al. Published: 2026-04-13 Categories: cs.CV, cs.AI, cs.LG, q-bio.QM arXiv: https://arxiv.org/abs/2604.12075v1 The tumor microenvironment (TME) plays a central role in cancer progression, treatment response, and patient outcomes, yet large-scale, consistent, and quantitative TME characterization from routine hematoxylin and eosin (H&E)-stained histopathology remains scarce. We introduce OpenTME, an open-access dataset of pre-computed TME profiles derived from 3,634 H&E-stained whole-slide images across five cancer types (bladder, breast, colorectal, liver, and lung cancer) from The Cancer Genome Atlas (TCGA). All outputs were generated using Atlas H&E-TME, an AI-powered application built on the Atlas family of pathology foundation models, which performs tissue quality control, tissue segmentation, cell detection and classification, and spatial neighborhood analysis, yielding over 4,500 quantitative readouts per slide at cell-level resolution. OpenTME is available for non-commercial academic research on Hugging Face. We will continue to expand OpenTME over time and anticipate it will serve as a resource for biomarker discovery, spatial biology research, and the development of computational methods for TME analysis. --- ### Robust Explanations for User Trust in Enterprise NLP Systems Authors: Guilin Zhang, Kai Zhao, Jeffrey Friedman, Xu Chu, Amine Anoun, Jerry Ting Published: 2026-04-13 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.12069v4 Robust explanations are increasingly required for user trust in enterprise NLP, yet pre-deployment validation is difficult in the common case of black-box deployment (API-only access) where representation-based explainers are infeasible and existing studies provide limited guidance on whether explanations remain stable under real user noise, especially when organizations migrate from encoder classifiers to decoder LLMs. To close this gap, we propose a unified black-box robustness evaluation framework for token-level explanations based on leave-one-out occlusion, and operationalize explanation robustness with top-token flip rate under realistic perturbations (swap, deletion, shuffling, and back-translation) at multiple severity levels. Using this protocol, we conduct a systematic cross-architecture comparison across three benchmark datasets and six models spanning encoder and decoder families (BERT, RoBERTa, Qwen 7B/14B, Llama 8B/70B; 64,800 cases). We find that decoder LLMs produce substantially more stable explanations than encoder baselines (73% lower flip rates on average), and that stability improves with model scale (44% gain from 7B to 70B). Finally, we relate robustness improvements to inference cost, yielding a practical cost-robustness tradeoff curve that supports model and explanation selection prior to deployment in compliance-sensitive applications. --- ### Privacy-Preserving Structureless Visual Localization via Image Obfuscation Authors: Vojtech Panek, Patrik Beliansky, Zuzana Kukelova, Torsten Sattler Published: 2026-04-13 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12068v1 Visual localization is the task of estimating the camera pose of an image relative to a scene representation. In practice, visual localization systems are often cloud-based. Naturally, this raises privacy concerns in terms of revealing private details through the images sent to the server or through the representations stored on the server. Privacy-preserving localization aims to avoid such leakage of private details. However, the resulting localization approaches are significantly more complex, slower, and less accurate than their non-privacy-preserving counterparts. In this paper, we consider structureless localization methods in the context of privacy preservation. Structureless methods represent the scene through a set of reference images with known camera poses and intrinsics. In contrast to existing methods proposing representations that are as privacy-preserving as possible, we study a simple image obfuscation approach based on common image operations, e.g., replacing RGB images with (semantic) segmentations. We show that existing structureless pipelines do not need any special adjustments, as modern feature matchers can match obfuscated images out of the box. The results are easy-to-implement pipelines that can ensure both the privacy of the query images and the scene representations. Detailed experiments on multiple datasets show that the resulting methods achieve state-of-the-art pose accuracy for privacy-preserving approaches. --- ### Mathematics Teachers Interactions with a Multi-Agent System for Personalized Problem Generation Authors: Candace Walkington, Theodora Beauchamp, Fareya Ikram, Merve Koçyiğit Gürbüz, Fangli Xia, Margan Lee, Andrew Lan Published: 2026-04-13 Categories: cs.AI, cs.CY arXiv: https://arxiv.org/abs/2604.12066v1 Large language models can increasingly adapt educational tasks to learners characteristics. In the present study, we examine a multi-agent teacher-in-the-loop system for personalizing middle school math problems. The teacher enters a base problem and desired topic, the LLM generates the problem, and then four AI agents evaluate the problem using criteria that each specializes in (mathematical accuracy, authenticity, readability, and realism). Eight middle school mathematics teachers created 212 problems in ASSISTments using the system and assigned these problems to their students. We find that both teachers and students wanted to modify the fine-grained personalized elements of the real-world context of the problems, signaling issues with authenticity and fit. Although the agents detected many issues with realism as the problems were being written, there were few realism issues noted by teachers and students in the final versions. Issues with readability and mathematical hallucinations were also somewhat rare. Implications for multi-agent systems for personalization that support teacher control are given. --- ### Interpretable DNA Sequence Classification via Dynamic Feature Generation in Decision Trees Authors: Nicolas Huynh, Krzysztof Kacprzyk, Ryan Sheridan, David Bentley, Mihaela van der Schaar Published: 2026-04-13 Categories: cs.LG, cs.AI, q-bio.GN arXiv: https://arxiv.org/abs/2604.12060v1 The analysis of DNA sequences has become critical in numerous fields, from evolutionary biology to understanding gene regulation and disease mechanisms. While deep neural networks can achieve remarkable predictive performance, they typically operate as black boxes. Contrasting these black boxes, axis-aligned decision trees offer a promising direction for interpretable DNA sequence analysis, yet they suffer from a fundamental limitation: considering individual raw features in isolation at each split limits their expressivity, which results in prohibitive tree depths that hinder both interpretability and generalization performance. We address this challenge by introducing DEFT, a novel framework that adaptively generates high-level sequence features during tree construction. DEFT leverages large language models to propose biologically-informed features tailored to the local sequence distributions at each node and to iteratively refine them with a reflection mechanism. Empirically, we demonstrate that DEFT discovers human-interpretable and highly predictive sequence features across a diverse range of genomic tasks. --- ### LoSA: Locality Aware Sparse Attention for Block-Wise Diffusion Language Models Authors: Haocheng Xi, Harman Singh, Yuezhou Hu, Coleman Hooper, Rishabh Tiwari, Aditya Tomar, Minjae Lee, Wonjun Kang, Michael Mahoney, Chenfeng Xu, Kurt Keutzer, Amir Gholami Published: 2026-04-13 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.12056v1 Block-wise diffusion language models (DLMs) generate multiple tokens in any order, offering a promising alternative to the autoregressive decoding pipeline. However, they still remain bottlenecked by memory-bound attention in long-context scenarios. Naive sparse attention fails on DLMs due to a KV Inflation problem, where different queries select different prefix positions, making the union of accessed KV pages large. To address this, we observe that between consecutive denoising steps, only a small fraction of active tokens exhibit significant hidden-state changes, while the majority of stable tokens remain nearly constant. Based on this insight, we propose LOSA (Locality-aware Sparse Attention), which reuses cached prefix-attention results for stable tokens and applies sparse attention only to active tokens. This substantially shrinks the number of KV indices that must be loaded, yielding both higher speedup and higher accuracy. Across multiple block-wise DLMs and benchmarks, LOSA preserves near-dense accuracy while significantly improving efficiency, achieving up to +9 points in average accuracy at aggressive sparsity levels while maintaining 1.54x lower attention density. It also achieves up to 4.14x attention speedup on RTX A6000 GPUs, demonstrating the effectiveness of the proposed method. --- ### Leveraging Weighted Syntactic and Semantic Context Assessment Summary (wSSAS) Towards Text Categorization Using LLMs Authors: Shreeya Verma Kathuria, Nitin Mayande, Sharookh Daruwalla, Nitin Joglekar, Charles Weber Published: 2026-04-13 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.12049v1 The use of Large Language Models (LLMs) for reliable, enterprise-grade analytics such as text categorization is often hindered by the stochastic nature of attention mechanisms and sensitivity to noise that compromise their analytical precision and reproducibility. To address these technical frictions, this paper introduces the Weighted Syntactic and Semantic Context Assessment Summary (wSSAS), a deterministic framework designed to enforce data integrity on large-scale, chaotic datasets. We propose a two-phased validation framework that first organizes raw text into a hierarchical classification structure containing Themes, Stories, and Clusters. It then leverages a Signal-to-Noise Ratio (SNR) to prioritize high-value semantic features, ensuring the model's attention remains focused on the most representative data points. By incorporating this scoring mechanism into a Summary-of-Summaries (SoS) architecture, the framework effectively isolates essential information and mitigates background noise during data aggregation. Experimental results using Gemini 2.0 Flash Lite across diverse datasets - including Google Business reviews, Amazon Product reviews, and Goodreads Book reviews - demonstrate that wSSAS significantly improves clustering integrity and categorization accuracy. Our findings indicate that wSSAS reduces categorization entropy and provides a reproducible pathway for improving LLM based summaries based on a high-precision, deterministic process for large-scale text categorization. --- ### Empirical Evaluation of PDF Parsing and Chunking for Financial Question Answering with RAG Authors: Omar El Bachyr, Yewei Song, Saad Ezzini, Jacques Klein, Tegawendé F. Bissyandé, Anas Zilali, Ulrick Ble, Anne Goujon Published: 2026-04-13 Categories: cs.CL, cs.IR arXiv: https://arxiv.org/abs/2604.12047v1 PDF files are primarily intended for human reading rather than automated processing. In addition, the heterogeneous content of PDFs, such as text, tables, and images, poses significant challenges for parsing and information extraction. To address these difficulties, both practitioners and researchers are increasingly developing new methods, including the promising Retrieval-Augmented Generation (RAG) systems to automated PDF processing. However, there is no comprehensive study investigating how different components and design choices affect the performance of a RAG system for understanding PDFs. In this paper, we propose such a study (1) by focusing on Question Answering, a specific language understanding task, and (2) by leveraging two benchmarks from the financial domain, including TableQuest, our newly generated, publicly available benchmark. We systematically examine multiple PDF parsers and chunking strategies (with varied overlap), along with their potential synergies in preserving document structure and ensuring answer correctness. Overall, our results offer practical guidelines for building robust RAG pipelines for PDF understanding. --- ### Think Through Uncertainty: Improving Long-Form Generation Factuality via Reasoning Calibration Authors: Xin Liu, Lu Wang Published: 2026-04-13 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.12046v1 Large language models (LLMs) often hallucinate in long-form generation. Existing approaches mainly improve factuality through post-hoc revision or reinforcement learning (RL) with correctness-based rewards, but they do not teach the model to estimate which parts of its generation are reliable. As a result, models may still state incorrect claims confidently in their responses. Recent advances in reasoning have significantly improved LLM performance, and have been leveraged to estimate confidence by incorporating calibration into RL objectives. However, existing approaches remain limited to a single scalar confidence for the entire response, which is insufficient for long-form generation where uncertainty varies across individual claims. To mitigate this problem, we propose CURE, a framework that improves long-form factuality by teaching LLMs to reason about uncertainty at the claim level. We first introduce a Claim-Aware Reasoning Protocol, which structures outputs into atomic claims paired with explicit confidence estimates. We then develop a multi-stage training pipeline that aligns model confidence with claims' correctness and then optimizes on factuality. The resulting calibrated confidence further enables selective prediction, allowing the model to abstain from uncertain claims at inference time. Experiments on four long-form factuality benchmarks show that CURE consistently improves factual accuracy over competitive supervised and RL baselines, while maintaining factual recall. In particular, it improves claim-level accuracy by up to 39.9% on Biography generation. These gains are accompanied by improved calibration, as reflected by a 16.0% increase in AUROC on FactBench. --- ### VISTA: Validation-Informed Trajectory Adaptation via Self-Distillation Authors: Eli Corn, Daphna Weinshall Published: 2026-04-13 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.12044v2 Deep learning models may converge to suboptimal solutions despite strong validation accuracy, masking an optimization failure we term Trajectory Deviation. This is because as training proceeds, models can abandon high generalization states for specific data sub-populations, thus discarding previously learned latent features without triggering classical overfitting signals. To address this problem we introduce VISTA, an online self-distillation framework that enforces consistency along the optimization trajectory. Using a validation-informed Marginal Coverage score, VISTA identifies expert anchors, which are earlier model states that retain specialized competence over distinct data regions. A coverage-weighted ensemble of these anchors is integrated online during training, regularizing the loss landscape and preserving mastered knowledge. When evaluated across multiple benchmarks, VISTA demonstrates improved robustness and generalization over standard training and prior self-distillation methods, while a lightweight implementation reduces storage overhead by 90% without performance loss. --- ### VISTA: Validation-Informed Trajectory Adaptation via Self-Distillation Authors: Eli Corn, Daphna Weinshall Published: 2026-04-13 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.12044v1 Deep learning models may converge to suboptimal solutions despite strong validation accuracy, masking an optimization failure we term Trajectory Deviation. This is because as training proceeds, models can abandon high generalization states for specific data sub-populations, thus discarding previously learned latent features without triggering classical overfitting signals. To address this problem we introduce VISTA, an online self-distillation framework that enforces consistency along the optimization trajectory. Using a validation-informed Marginal Coverage score, VISTA identifies expert anchors, which are earlier model states that retain specialized competence over distinct data regions. A coverage-weighted ensemble of these anchors is integrated online during training, regularizing the loss landscape and preserving mastered knowledge. When evaluated across multiple benchmarks, VISTA demonstrates improved robustness and generalization over standard training and prior self-distillation methods, while a lightweight implementation reduces storage overhead by 90% without performance loss. --- ### On the continuum limit of t-SNE for data visualization Authors: Jeff Calder, Zhonggan Huang, Ryan Murray, Adam Pickarski Published: 2026-04-13 Categories: stat.ML, cs.LG, math.AP, math.ST arXiv: https://arxiv.org/abs/2604.12041v1 This work is concerned with the continuum limit of a graph-based data visualization technique called the t-Distributed Stochastic Neighbor Embedding (t-SNE), which is widely used for visualizing data in a variety of applications, but is still poorly understood from a theoretical standpoint. The t-SNE algorithm produces visualizations by minimizing the Kullback-Leibler divergence between similarity matrices representing the high dimensional data and its low dimensional representation. We prove that as the number of data points $n \to \infty$, after a natural rescaling and in applicable parameter regimes, the Kullback-Leibler divergence is consistent as the number of data points $n \to \infty$ and the similarity graph remains sparse with a continuum variational problem that involves a non-convex gradient regularization term and a penalty on the magnitude of the probability density function in the visualization space. These two terms represent the continuum limits of the attraction and repulsion forces in the t-SNE algorithm. Due to the lack of convexity in the continuum variational problem, the question of well-posedeness is only partially resolved. We show that when both dimensions are $1$, the problem admits a unique smooth minimizer, along with an infinite number of discontinuous minimizers (interpreted in a relaxed sense). This aligns well with the empirically observed ability of t-SNE to separate data in seemingly arbitrary ways in the visualization. The energy is also very closely related to the famously ill-posed Perona-Malik equation, which is used for denoising and simplifying images. We present numerical results validating the continuum limit, provide some preliminary results about the delicate nature of the limiting energetic problem in higher dimensions, and highlight several problems for future work. --- ### SIR-Bench: Evaluating Investigation Depth in Security Incident Response Agents Authors: Daniel Begimher, Cristian Leo, Jack Huang, Pat Gaw, Bonan Zheng Published: 2026-04-13 Categories: cs.CR, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2604.12040v1 We present SIR-Bench, a benchmark of 794 test cases for evaluating autonomous security incident response agents that distinguishes genuine forensic investigation from alert parroting. Derived from 129 anonymized incident patterns with expert-validated ground truth, SIR-Bench measures not only whether agents reach correct triage decisions, but whether they discover novel evidence through active investigation. To construct SIR-Bench, we develop Once Upon A Threat (OUAT), a framework that replays real incident patterns in controlled cloud environments, producing authentic telemetry with measurable investigation outcomes. Our evaluation methodology introduces three complementary metrics: triage accuracy (M1), novel finding discovery (M2), and tool usage appropriateness (M3), assessed through an adversarial LLM-as-Judge that inverts the burden of proof -- requiring concrete forensic evidence to credit investigations. Evaluating our SIR agent on the benchmark demonstrates 97.1% true positive (TP) detection, 73.4% false positive (FP) rejection, and 5.67 novel key findings per case, establishing a baseline against which future investigation agents can be measured. ---