# Research Papers — 2026-02 · bundle page 172 of 210

> Full-text markdown of papers #8551–8600 of 10493 published in 2026-02, newest first. AI-readable mirrors of the full 10493-paper 2026-02 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2026-02/page/172
[← Previous bundle](https://aipod.dev/papers/archive/2026-02/page/171.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-02/page/173.md)

### Interval-Based AUC (iAUC): Extending ROC Analysis to Uncertainty-Aware Classification

Authors: Yuqi Li, Matthew M. Engelhard
Published: 2026-02-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2602.04775v1

In high-stakes risk prediction, quantifying uncertainty through interval-valued predictions is essential for reliable decision-making. However, standard evaluation tools like the receiver operating characteristic (ROC) curve and the area under the curve (AUC) are designed for point scores and fail to capture the impact of predictive uncertainty on ranking performance. We propose an uncertainty-aware ROC framework specifically for interval-valued predictions, introducing two new measures: $AUC_L$ and $AUC_U$. This framework enables an informative three-region decomposition of the ROC plane, partitioning pairwise rankings into correct, incorrect, and uncertain orderings. This approach naturally supports selective prediction by allowing models to abstain from ranking cases with overlapping intervals, thereby optimizing the trade-off between abstention rate and discriminative reliability. We prove that under valid class-conditional coverage, $AUC_L$ and $AUC_U$ provide formal lower and upper bounds on the theoretical optimal AUC ($AUC^*$), characterizing the physical limit of achievable discrimination. The proposed framework applies broadly to interval-valued prediction models, regardless of the interval construction method. Experiments on real-world benchmark datasets, using bootstrap-based intervals as one instantiation, validate the framework's correctness and demonstrate its practical utility for uncertainty-aware evaluation and decision-making.

---

### Theory of Optimal Learning Rate Schedules and Scaling Laws for a Random Feature Model

Authors: Blake Bordelon, Francesco Mori
Published: 2026-02-04
Categories: cond-mat.dis-nn, cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2602.04774v2

Setting the learning rate (LR) for a deep learning model is a critical part of successful training. Choosing LRs is often done empirically with trial and error. In this work, we explore a solvable model of optimal LR schedules for a powerlaw random feature model trained with stochastic gradient descent (SGD). We consider the optimal schedule $η_T^\star(t)$ where $t$ is the current iterate and $T$ is the training horizon. This schedule is computed both as a numerical optimization problem and also analytically using optimal control theory. Our analysis reveals two regimes which we term the easy phase and hard phase. In the easy phase the optimal schedule is a polynomial decay $η_T^\star(t) \simeq T^{-ξ} (1-t/T)^δ$ where $ξ$ and $δ$ depend on the properties of the features and task. In the hard phase, the optimal schedule resembles warmup-stable-decay with constant initial LR and annealing performed over a vanishing fraction of training steps. We investigate joint optimization of LR and batch size and find batch ramps can improve the wall-clock time in the easy phase. Beyond SGD, we derive optimal schedules for momentum parameter $β(t)$ and show that it improves the loss-scaling exponent in the hard phase. We compare our optimal schedule to various benchmarks including (1) optimal constant learning rates $η_T(t) \sim T^{-ξ}$ (2) optimal power laws $η_T(t) \sim T^{-ξ} t^{-χ}$, finding that our schedule achieves better rates than either of these. Our theory suggests that LR transfer across training horizon depends on the structure of the model and task. For ResNet image classification on CIFAR-5M, the learning curves exhibit hard-phase behavior where optimal base LRs are constant under sufficient annealing. GPT-2 style transformers trained in language modeling exhibit easy-phase behavior where optimal LRs shift even under annealing.

---

### Generative Modeling via Drifting

Authors: Mingyang Deng, He Li, Tianhong Li, Yilun Du, Kaiming He
Published: 2026-02-04
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2602.04770v2

Generative modeling can be formulated as learning a mapping f such that its pushforward distribution matches the data distribution. The pushforward behavior can be carried out iteratively at inference time, for example in diffusion and flow-based models. In this paper, we propose a new paradigm called Drifting Models, which evolve the pushforward distribution during training and naturally admit one-step inference. We introduce a drifting field that governs the sample movement and achieves equilibrium when the distributions match. This leads to a training objective that allows the neural network optimizer to evolve the distribution. In experiments, our one-step generator achieves state-of-the-art results on ImageNet at 256 x 256 resolution, with an FID of 1.54 in latent space and 1.61 in pixel space. We hope that our work opens up new opportunities for high-quality one-step generation.

---

### NeuroCanvas: VLLM-Powered Robust Seizure Detection by Reformulating Multichannel EEG as Image

Authors: Yan Chen, Jie Peng, Moajjem Hossain Chowdhury, Tianlong Chen, Yunmei Liu
Published: 2026-02-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2602.04769v2

Accurate and timely seizure detection from Electroencephalography (EEG) is critical for clinical intervention, yet manual review of long-term recordings is labor-intensive. Recent efforts to encode EEG signals into large language models (LLMs) show promise in handling neural signals across diverse patients, but two significant challenges remain: (1) multi-channel heterogeneity, as seizure-relevant information varies substantially across EEG channels, and (2) computing inefficiency, as the EEG signals need to be encoded into a massive number of tokens for the prediction. To address these issues, we draw the EEG signal and propose the novel NeuroCanvas framework. Specifically, NeuroCanvas consists of two modules: (i) The Entropy-guided Channel Selector (ECS) selects the seizure-relevant channels input to LLM and (ii) the following Canvas of Neuron Signal (CNS) converts selected multi-channel heterogeneous EEG signals into structured visual representations. The ECS module alleviates the multi-channel heterogeneity issue, and the CNS uses compact visual tokens to represent the EEG signals that improve the computing efficiency. We evaluate NeuroCanvas across multiple seizure detection datasets, demonstrating a significant improvement of 20% in F1 score and reductions of 88% in inference latency. These results highlight NeuroCanvas as a scalable and effective solution for real-time and resource-efficient seizure detection in clinical practice.

---

### Billion-Scale Graph Foundation Models

Authors: Maya Bechler-Speicher, Yoel Gottlieb, Andrey Isakov, David Abensur, Ami Tavory, Daniel Haimovich, Ido Guy, Udi Weinsberg
Published: 2026-02-04
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2602.04768v2

Graph-structured data underpins many critical applications. While foundation models have transformed language and vision via large-scale pretraining and lightweight adaptation, extending this paradigm to general, real-world graphs is challenging. In this work, we present Graph Billion-Foundation-Fusion (GraphBFF): an end-to-end recipe for building billion-parameter Graph Foundation Models (GFMs) for large-scale heterogeneous graphs. Central to the recipe is the GraphBFF Transformer, a flexible and scalable architecture designed for practical billion-scale GFMs. Using the GraphBFF, we present neural scaling laws for heterogeneous graphs and show that loss decreases predictably as either model capacity or training data scales, depending on which factor is the bottleneck. The GraphBFF framework provides concrete methodologies for data batching, pretraining, and fine-tuning for building GFMs at scale. We demonstrate the effectiveness of the framework over a real-world billion-scale graph, with an evaluation of a billion-parameter GraphBFF Transformer following the proposed recipe. Across ten diverse, real-world downstream tasks on graphs unseen during training, spanning node- and link-level classification and regression, GraphBFF consistently outperforms baselines, with large margins of up to 31 PRAUC points, including in few-shot settings. Finally, we discuss key challenges and open opportunities for making GFMs a practical and principled foundation for graph learning at industrial scale.

---

### Beyond Many-Shot Translation: Scaling In-Context Demonstrations For Low-Resource Machine Translation

Authors: Luis Frentzen Salim, Esteban Carlin, Alexandre Morinvil, Xi Ai, Lun-Wei Ku
Published: 2026-02-04
Categories: cs.CL
arXiv: https://arxiv.org/abs/2602.04764v1

Building machine translation (MT) systems for low-resource languages is notably difficult due to the scarcity of high-quality data. Although Large Language Models (LLMs) have improved MT system performance, adapting them to lesser-represented languages remains challenging. In-context learning (ICL) may offer novel ways to adapt LLMs for low-resource MT by conditioning models on demonstration at inference time. In this study, we explore scaling low-resource machine translation ICL beyond the few-shot setting to thousands of examples with long-context models. We scale in-context token budget to 1M tokens and compare three types of training corpora used as in-context supervision: monolingual unsupervised data, instruction-style data, and parallel data (English--target and Indonesian--target). Our experiments on Javanese and Sundanese show that gains from additional context saturate quickly and can degrade near the maximum context window, with scaling behavior strongly dependent on corpus type. Notably, some forms of monolingual supervision can be competitive with parallel data, despite the latter offering additional supervision. Overall, our results characterize the effective limits and corpus-type sensitivity of long-context ICL for low-resource MT, highlighting that larger context windows do not necessarily yield proportional quality gains.

---

### Active Asymmetric Multi-Agent Multimodal Learning under Uncertainty

Authors: Rui Liu, Pratap Tokekar, Ming Lin
Published: 2026-02-04
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2602.04763v1

Multi-agent systems are increasingly equipped with heterogeneous multimodal sensors, enabling richer perception but introducing modality-specific and agent-dependent uncertainty. Existing multi-agent collaboration frameworks typically reason at the agent level, assume homogeneous sensing, and handle uncertainty implicitly, limiting robustness under sensor corruption. We propose Active Asymmetric Multi-Agent Multimodal Learning under Uncertainty (A2MAML), a principled approach for uncertainty-aware, modality-level collaboration. A2MAML models each modality-specific feature as a stochastic estimate with uncertainty prediction, actively selects reliable agent-modality pairs, and aggregates information via Bayesian inverse-variance weighting. This formulation enables fine-grained, modality-level fusion, supports asymmetric modality availability, and provides a principled mechanism to suppress corrupted or noisy modalities. Extensive experiments on connected autonomous driving scenarios for collaborative accident detection demonstrate that A2MAML consistently outperforms both single-agent and collaborative baselines, achieving up to 18.7% higher accident detection rate.

---

### Improved Dimension Dependence for Bandit Convex Optimization with Gradient Variations

Authors: Hang Yu, Yu-Hu Yan, Peng Zhao
Published: 2026-02-04
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2602.04761v2

Gradient-variation online learning has drawn increasing attention due to its deep connections to game theory and optimization. It has been studied extensively in the full-information setting, but is underexplored with bandit feedback. In this work, we focus on gradient variation in Bandit Convex Optimization (BCO) with two-point feedback. By proposing a refined analysis of the non-consecutive gradient variation, a fundamental quantity in gradient variation with bandit feedback, we improve the dimension dependence for both convex and strongly convex functions compared with the best known results (Chiang et al., 2013). Our improved analysis of the non-consecutive gradient variation also implies other favorable problem-dependent guarantees, such as gradient-variance and small-loss regret bounds. Beyond the two-point setup, we demonstrate the versatility of our technique by achieving the first gradient-variation bound for one-point bandit linear optimization over hyper-rectangular domains. Finally, we validate the effectiveness of our results in more challenging tasks such as dynamic and universal regret minimization, establishing the first gradient-variation dynamic and universal regret bounds for two-point BCO.

---

### Improved Dimension Dependence for Bandit Convex Optimization with Gradient Variations

Authors: Hang Yu, Yu-Hu Yan, Peng Zhao
Published: 2026-02-04
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2602.04761v1

Gradient-variation online learning has drawn increasing attention due to its deep connections to game theory, optimization, etc. It has been studied extensively in the full-information setting, but is underexplored with bandit feedback. In this work, we focus on gradient variation in Bandit Convex Optimization (BCO) with two-point feedback. By proposing a refined analysis on the non-consecutive gradient variation, a fundamental quantity in gradient variation with bandits, we improve the dimension dependence for both convex and strongly convex functions compared with the best known results (Chiang et al., 2013). Our improved analysis for the non-consecutive gradient variation also implies other favorable problem-dependent guarantees, such as gradient-variance and small-loss regrets. Beyond the two-point setup, we demonstrate the versatility of our technique by achieving the first gradient-variation bound for one-point bandit linear optimization over hyper-rectangular domains. Finally, we validate the effectiveness of our results in more challenging tasks such as dynamic/universal regret minimization and bandit games, establishing the first gradient-variation dynamic and universal regret bounds for two-point BCO and fast convergence rates in bandit games.

---

### A Dual-TransUNet Deep Learning Framework for Multi-Source Precipitation Merging and Improving Seasonal and Extreme Estimates

Authors: Yuchen Ye, Zixuan Qi, Shixuan Li, Wei Qi, Yanpeng Cai, Chaoxia Yuan
Published: 2026-02-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2602.04757v1

Multi-source precipitation products (MSPs) from satellite retrievals and reanalysis are widely used for hydroclimatic monitoring, yet spatially heterogeneous biases and limited skill for extremes still constrain their hydrologic utility. Here we develop a dual-stage TransUNet-based multi-source precipitation merging framework (DDL-MSPMF) that integrates six MSPs with four ERA5 near-surface physical predictors. A first-stage classifier estimates daily precipitation occurrence probability, and a second-stage regressor fuses the classifier outputs together with all predictors to estimate daily precipitation amount at 0.25 degree resolution over China for 2001-2020. Benchmarking against multiple deep learning and hybrid baselines shows that the TransUNet - TransUNet configuration yields the best seasonal performance (R = 0.75; RMSE = 2.70 mm/day) and improves robustness relative to a single-regressor setting. For heavy precipitation (>25 mm/day), DDL-MSPMF increases equitable threat scores across most regions of eastern China and better reproduces the spatial pattern of the July 2021 Zhengzhou rainstorm, indicating enhanced extreme-event detection beyond seasonal-mean corrections. Independent evaluation over the Qinghai-Tibet Plateau using TPHiPr further supports its applicability in data-scarce regions. SHAP analysis highlights the importance of precipitation occurrence probabilities and surface pressure, providing physically interpretable diagnostics. The proposed framework offers a scalable and explainable approach for precipitation fusion and extreme-event assessment.

---

### When Silence Is Golden: Can LLMs Learn to Abstain in Temporal QA and Beyond?

Authors: Xinyu Zhou, Chang Jin, Carsten Eickhoff, Zhijiang Guo, Seyed Ali Bahrainian
Published: 2026-02-04
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2602.04755v2

Large language models (LLMs) rarely admit uncertainty, often producing fluent but misleading answers, rather than abstaining (i.e., refusing to answer). This weakness is even evident in temporal question answering, where models frequently ignore time-sensitive evidence and conflate facts across different time-periods. In this paper, we present the first empirical study of training LLMs with an abstention ability while reasoning about temporal QA. Existing approaches such as calibration might be unreliable in capturing uncertainty in complex reasoning. We instead frame abstention as a teachable skill and introduce a pipeline that couples Chain-of-Thought (CoT) supervision with Reinforcement Learning (RL) guided by abstention-aware rewards. Our goal is to systematically analyze how different information types and training techniques affect temporal reasoning with abstention behavior in LLMs. Through extensive experiments studying various methods, we find that RL yields strong empirical gains on reasoning: a model initialized by Qwen2.5-1.5B-Instruct surpasses GPT-4o by $3.46\%$ and $5.80\%$ in Exact Match on TimeQA-Easy and Hard, respectively. Moreover, it improves the True Positive rate on unanswerable questions by $20\%$ over a pure supervised fine-tuned (SFT) variant. Beyond performance, our analysis shows that SFT induces overconfidence and harms reliability, while RL improves prediction accuracy but exhibits similar risks. Finally, by comparing implicit reasoning cues (e.g., original context, temporal sub-context, knowledge graphs) with explicit CoT supervision, we find that implicit information provides limited benefit for reasoning with abstention. Our study provides new insights into how abstention and reasoning can be jointly optimized, providing a foundation for building more reliable LLMs.

---

### Comparative Insights on Adversarial Machine Learning from Industry and Academia: A User-Study Approach

Authors: Vishruti Kakkad, Paul Chung, Hanan Hibshi, Maverick Woo
Published: 2026-02-04
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2602.04753v2

An exponential growth of Machine Learning and its Generative AI applications brings with it significant security challenges, often referred to as Adversarial Machine Learning (AML). In this paper, we conducted two comprehensive studies to explore the perspectives of industry professionals and students on different AML vulnerabilities and their educational strategies. In our first study, we conducted an online survey with professionals revealing a notable correlation between cybersecurity education and concern for AML threats. For our second study, we developed two CTF challenges that implement Natural Language Processing and Generative AI concepts and demonstrate a poisoning attack on the training data set. The effectiveness of these challenges was evaluated by surveying undergraduate and graduate students at Carnegie Mellon University, finding that a CTF-based approach effectively engages interest in AML threats. Based on the responses of the participants in our research, we provide detailed recommendations emphasizing the critical need for integrated security education within the ML curriculum.

---

### Decomposing Query-Key Feature Interactions Using Contrastive Covariances

Authors: Andrew Lee, Yonatan Belinkov, Fernanda Viégas, Martin Wattenberg
Published: 2026-02-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2602.04752v1

Despite the central role of attention heads in Transformers, we lack tools to understand why a model attends to a particular token. To address this, we study the query-key (QK) space -- the bilinear joint embedding space between queries and keys. We present a contrastive covariance method to decompose the QK space into low-rank, human-interpretable components. It is when features in keys and queries align in these low-rank subspaces that high attention scores are produced. We first study our method both analytically and empirically in a simplified setting. We then apply our method to large language models to identify human-interpretable QK subspaces for categorical semantic features and binding features. Finally, we demonstrate how attention scores can be attributed to our identified features.

---

### Exploiting contextual information to improve stance detection in informal political discourse with LLMs

Authors: Arman Engin Sucu, Yixiang Zhou, Mario A. Nascimento, Tony Mullen
Published: 2026-02-04
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2602.04750v1

This study investigates the use of Large Language Models (LLMs) for political stance detection in informal online discourse, where language is often sarcastic, ambiguous, and context-dependent. We explore whether providing contextual information, specifically user profile summaries derived from historical posts, can improve classification accuracy. Using a real-world political forum dataset, we generate structured profiles that summarize users' ideological leaning, recurring topics, and linguistic patterns. We evaluate seven state-of-the-art LLMs across baseline and context-enriched setups through a comprehensive cross-model evaluation. Our findings show that contextual prompts significantly boost accuracy, with improvements ranging from +17.5\% to +38.5\%, achieving up to 74\% accuracy that surpasses previous approaches. We also analyze how profile size and post selection strategies affect performance, showing that strategically chosen political content yields better results than larger, randomly selected contexts. These findings underscore the value of incorporating user-level context to enhance LLM performance in nuanced political classification tasks.

---

### Mitigating Long-Tail Bias via Prompt-Controlled Diffusion Augmentation

Authors: Buddhi Wijenayake, Nichula Wasalathilake, Roshan Godaliyadda, Vijitha Herath, Parakrama Ekanayake, Vishal M. Patel
Published: 2026-02-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2602.04749v3

Long-tailed class imbalance remains a fundamental obstacle in semantic segmentation of high-resolution remote-sensing imagery, where dominant classes shape learned representations and rare classes are systematically under-segmented. This challenge becomes more acute in cross-domain settings such as LoveDA, which exhibits an explicit Urban/Rural split with substantial appearance differences and inconsistent class-frequency statistics across domains. We propose a prompt-controlled diffusion augmentation framework that generates paired label-image samples with explicit control over semantic composition and domain, enabling targeted enrichment of underrepresented classes rather than indiscriminate dataset expansion. A domain-aware, masked, ratio-conditioned discrete diffusion model first synthesizes layouts that satisfy class-ratio targets while preserving realistic spatial co-occurrence, and a ControlNet-guided diffusion model then renders photorealistic, domain-consistent images from these layouts. When mixed with real data, the resulting synthetic pairs improve multiple segmentation backbones, especially on minority classes and under domain shift, showing that better downstream segmentation comes from adding the right samples in the right proportions. Source codes, pretrained models, and synthetic datasets are available at \href{https://buddhi19.github.io/SyntheticGen}{\texttt{buddhi19.github.io/SyntheticGen}}.

---

### Impact of diversity on bounded archives for multi-objective local search

Authors: Amadeu A. Coco, Cyprien Borée, Julien Baste, Laetitia Jourdan, Lucien Mousin
Published: 2026-02-04
Categories: cs.NE
arXiv: https://arxiv.org/abs/2602.04745v1

This work tackles two critical challenges related to the development of metaheuristics for Multi-Objective Optimization Problems (MOOPs): the exponential growth of non-dominated solutions and the tendency of metaheuristics to disproportionately concentrate their search on a subset of the Pareto Front. To counteract the first, bounded archives are employed as a strategic mechanism for effectively managing the increasing number of non-dominated solutions. Addressing the second challenge involves an in-depth exploration of solution diversity algorithms found in existing literature. Upon recognizing that current approaches predominantly center on diversity within the objective space, this research introduces innovative methods specifically designed to enhance diversity in the solution space. Results demonstrate the efficacy of the Hamming Distance Archiving Algorithm, one of the newly proposed algorithms for multi-objective local search, surpassing the performance of the Adaptive Grid Archiving and the Hypervolume Archiving, both drawn from the literature. This outcome suggests a promising avenue for enhancing the overall efficiency of metaheuristics employed for solving MOOPs.

---

### Inference-Time Reasoning Selectively Reduces Implicit Social Bias in Large Language Models

Authors: Molly Apsel, Michael N. Jones
Published: 2026-02-04
Categories: cs.CY, cs.CL
arXiv: https://arxiv.org/abs/2602.04742v1

Drawing on constructs from psychology, prior work has identified a distinction between explicit and implicit bias in large language models (LLMs). While many LLMs undergo post-training alignment and safety procedures to avoid expressions of explicit social bias, they still exhibit significant implicit biases on indirect tasks resembling the Implicit Association Test (IAT). Recent work has further shown that inference-time reasoning can impair LLM performance on tasks that rely on implicit statistical learning. Motivated by a theoretical link between implicit associations and statistical learning in human cognition, we examine how reasoning-enabled inference affects implicit bias in LLMs. We find that enabling reasoning significantly reduces measured implicit bias on an IAT-style evaluation for some model classes across fifteen stereotype topics. This effect appears specific to social bias domains, as we observe no corresponding reduction for non-social implicit associations. As reasoning is increasingly enabled by default in deployed LLMs, these findings suggest that it can meaningfully alter fairness evaluation outcomes in some systems, while also raising questions about how alignment procedures interact with inference-time reasoning to drive variation in bias reduction across model types. More broadly, this work highlights how theory from cognitive science and psychology can complement AI evaluation research by providing methodological and interpretive frameworks that reveal new insights into model behavior.

---

### Alignment Drift in Multimodal LLMs: A Two-Phase, Longitudinal Evaluation of Harm Across Eight Model Releases

Authors: Casey Ford, Madison Van Doren, Emily Dix
Published: 2026-02-04
Categories: cs.CL, cs.AI, cs.HC
arXiv: https://arxiv.org/abs/2602.04739v1

Multimodal large language models (MLLMs) are increasingly deployed in real-world systems, yet their safety under adversarial prompting remains underexplored. We present a two-phase evaluation of MLLM harmlessness using a fixed benchmark of 726 adversarial prompts authored by 26 professional red teamers. Phase 1 assessed GPT-4o, Claude Sonnet 3.5, Pixtral 12B, and Qwen VL Plus; Phase 2 evaluated their successors (GPT-5, Claude Sonnet 4.5, Pixtral Large, and Qwen Omni) yielding 82,256 human harm ratings. Large, persistent differences emerged across model families: Pixtral models were consistently the most vulnerable, whereas Claude models appeared safest due to high refusal rates. Attack success rates (ASR) showed clear alignment drift: GPT and Claude models exhibited increased ASR across generations, while Pixtral and Qwen showed modest decreases. Modality effects also shifted over time: text-only prompts were more effective in Phase 1, whereas Phase 2 produced model-specific patterns, with GPT-5 and Claude 4.5 showing near-equivalent vulnerability across modalities. These findings demonstrate that MLLM harmlessness is neither uniform nor stable across updates, underscoring the need for longitudinal, multimodal benchmarks to track evolving safety behaviour.

---

### Rationality Measurement and Theory for Reinforcement Learning Agents

Authors: Kejiang Qian, Amos Storkey, Fengxiang He
Published: 2026-02-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2602.04737v3

This paper proposes a suite of rationality measures and associated theory for reinforcement learning agents, a property increasingly critical yet rarely explored. We define an action in deployment to be perfectly rational if it maximises the hidden true value function in the steepest direction. The expected value discrepancy of a policy's actions against their rational counterparts, culminating over the trajectory in deployment, is defined to be expected rational risk; an empirical average version in training is also defined. Their difference, termed as rational risk gap, is decomposed into (1) an extrinsic component caused by environment shifts between training and deployment, and (2) an intrinsic one due to the algorithm's generalisability in a dynamic environment. They are upper bounded by, respectively, (1) the $1$-Wasserstein distance between transition kernels and initial state distributions in training and deployment, and (2) the empirical Rademacher complexity of the value function class. Our theory suggests hypotheses on the benefits from regularisers (including layer normalisation, $\ell_2$ regularisation, and weight normalisation) and domain randomisation, as well as the harm from environment shifts. Experiments are in full agreement with these hypotheses. The code is available at https://github.com/EVIEHub/Rationality.

---

### Conditional Counterfactual Mean Embeddings: Doubly Robust Estimation and Learning Rates

Authors: Thatchanon Anancharoenkij, Donlapark Ponnoprat
Published: 2026-02-04
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2602.04736v1

A complete understanding of heterogeneous treatment effects involves characterizing the full conditional distribution of potential outcomes. To this end, we propose the Conditional Counterfactual Mean Embeddings (CCME), a framework that embeds conditional distributions of counterfactual outcomes into a reproducing kernel Hilbert space (RKHS). Under this framework, we develop a two-stage meta-estimator for CCME that accommodates any RKHS-valued regression in each stage. Based on this meta-estimator, we develop three practical CCME estimators: (1) Ridge Regression estimator, (2) Deep Feature estimator that parameterizes the feature map by a neural network, and (3) Neural-Kernel estimator that performs RKHS-valued regression, with the coefficients parameterized by a neural network. We provide finite-sample convergence rates for all estimators, establishing that they possess the double robustness property. Our experiments demonstrate that our estimators accurately recover distributional features including multimodal structure of conditional counterfactual distributions.

---

### From Data to Behavior: Predicting Unintended Model Behaviors Before Training

Authors: Mengru Wang, Zhenqian Xu, Junfeng Fang, Yunzhi Yao, Shumin Deng, Huajun Chen, Ningyu Zhang
Published: 2026-02-04
Categories: cs.LG, cs.AI, cs.CL, cs.CY, cs.IR
arXiv: https://arxiv.org/abs/2602.04735v1

Large Language Models (LLMs) can acquire unintended biases from seemingly benign training data even without explicit cues or malicious content. Existing methods struggle to detect such risks before fine-tuning, making post hoc evaluation costly and inefficient. To address this challenge, we introduce Data2Behavior, a new task for predicting unintended model behaviors prior to training. We also propose Manipulating Data Features (MDF), a lightweight approach that summarizes candidate data through their mean representations and injects them into the forward pass of a base model, allowing latent statistical signals in the data to shape model activations and reveal potential biases and safety risks without updating any parameters. MDF achieves reliable prediction while consuming only about 20% of the GPU resources required for fine-tuning. Experiments on Qwen3-14B, Qwen2.5-32B-Instruct, and Gemma-3-12b-it confirm that MDF can anticipate unintended behaviors and provide insight into pre-training vulnerabilities.

---

### DMFlow: Disordered Materials Generation by Flow Matching

Authors: Liming Wu, Rui Jiao, Qi Li, Mingze Li, Songyou Li, Shifeng Jin, Wenbing Huang
Published: 2026-02-04
Categories: cs.LG, cond-mat.mtrl-sci
arXiv: https://arxiv.org/abs/2602.04734v1

The design of materials with tailored properties is crucial for technological progress. However, most deep generative models focus exclusively on perfectly ordered crystals, neglecting the important class of disordered materials. To address this gap, we introduce DMFlow, a generative framework specifically designed for disordered crystals. Our approach introduces a unified representation for ordered, Substitutionally Disordered (SD), and Positionally Disordered (PD) crystals, and employs a flow matching model to jointly generate all structural components. A key innovation is a Riemannian flow matching framework with spherical reparameterization, which ensures physically valid disorder weights on the probability simplex. The vector field is learned by a novel Graph Neural Network (GNN) that incorporates physical symmetries and a specialized message-passing scheme. Finally, a two-stage discretization procedure converts the continuous weights into multi-hot atomic assignments. To support research in this area, we release a benchmark containing SD, PD, and mixed structures curated from the Crystallography Open Database. Experiments on Crystal Structure Prediction (CSP) and De Novo Generation (DNG) tasks demonstrate that DMFlow significantly outperforms state-of-the-art baselines adapted from ordered crystal generation. We hope our work provides a foundation for the AI-driven discovery of disordered materials.

---

### Modular Expert Merging for Biomedical Retrieval

Authors: Sameh Khattab, Jean-Philippe Corbeil, Osman Alperen Çinar-Koraş, Amin Dada, Julian Friedrich, Jiawei He, Douglas Teodoro, Jens Kleesiek
Published: 2026-02-04
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2602.04731v3

Adapting general-purpose LLMs into domain-specialized dense retrievers typically requires large-scale training on mixed-domain data. We show that merging independently trained domain-specialized experts consistently exceeds this approach across four decoder-only LLM families (0.6B-7B), four merging methods, and twelve medical and general retrieval tasks from MTEB, suggesting that parameter-space composition captures complementary domain strengths that large-scale mixed-domain training averages out. To further maximize expert quality, we introduce Synthesize-Train-Merge (STM), a modular framework that synthesizes hard negatives with a top-tier LLM and fine-tunes domain-specialized experts via LoRA before merging them, without continual pre-training. Synthesized hard negatives yield the largest gains for smaller models, and STM achieves strong performance on biomedical retrieval tasks while maintaining competitive general-domain results across all four backbone families.

---

### Less Finetuning, Better Retrieval: Rethinking LLM Adaptation for Biomedical Retrievers via Synthetic Data and Model Merging

Authors: Sameh Khattab, Jean-Philippe Corbeil, Osman Alperen Koraş, Amin Dada, Julian Friedrich, François Beaulieu, Paul Vozila, Jens Kleesiek
Published: 2026-02-04
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2602.04731v1

Retrieval-augmented generation (RAG) has become the backbone of grounding Large Language Models (LLMs), improving knowledge updates and reducing hallucinations. Recently, LLM-based retriever models have shown state-of-the-art performance for RAG applications. However, several technical aspects remain underexplored on how to adapt general-purpose LLMs into effective domain-specific retrievers, especially in specialized domains such as biomedicine. We present Synthesize-Train-Merge (STM), a modular framework that enhances decoder-only LLMs with synthetic hard negatives, retrieval prompt optimization, and model merging. Experiments on a subset of 12 medical and general tasks from the MTEB benchmark show STM boosts task-specific experts by up to 23.5\% (average 7.5\%) and produces merged models that outperform both single experts and strong baselines without extensive pretraining. Our results demonstrate a scalable, efficient path for turning general LLMs into high-performing, domain-specialized retrievers, preserving general-domain capabilities while excelling on specialized tasks.

---

### "Be My Cheese?": Cultural Nuance Benchmarking for Machine Translation in Multilingual LLMs

Authors: Madison Van Doren, Casey Ford, Jennifer Barajas, Riley VanMeter, Cory Holland
Published: 2026-02-04
Categories: cs.CL
arXiv: https://arxiv.org/abs/2602.04729v2

We present a large-scale human evaluation benchmark for assessing cultural localisation in machine translation produced by state-of-the-art multilingual large language models (LLMs). Existing MT benchmarks emphasise token-level and grammatical accuracy, but often overlook the pragmatic and culturally grounded competencies required for real-world localisation. Building on a pilot study of 87 translations across 20 languages, we evaluate 7 multilingual LLMs across 15 target languages with 5 native-speaker raters per language. Each rater scored both full-text translations and segment-level instances of culturally nuanced language (idioms, puns, holidays, and culturally embedded concepts) on an ordinal 0-3 quality scale; segment ratings additionally included an NA option for untranslated segments. Across full-text evaluations, mean overall quality is modest (1.68/3): GPT-5 (2.10/3), Claude Sonnet 4 (1.97/3), and Mistral Medium 3.1 (1.84/3) form the strongest tier with fewer catastrophic failures. Segment-level results show sharp category effects: holidays (2.20/3) and cultural concepts (2.19/3) translate notably better than idioms (1.65/3) and puns (1.45/3), and idioms are most likely to be left untranslated. Inter-rater reliability was assessed using Krippendorff's α and Gwet's AC2, indicating moderate agreement overall (Krippendorff's α = 0.45) with the lowest agreement for puns. These findings demonstrate a persistent gap between grammatical adequacy and cultural resonance. To our knowledge, this is the first multilingual, human-annotated benchmark focused explicitly on cultural nuance in translation and localisation. The results highlight the need for culturally informed training data, improved cross-lingual pragmatics, and evaluation frameworks that support systematic benchmarking of culturally grounded translation.

---

### Scalable Cross-Attention Transformer for Cooperative Multi-AP OFDM Uplink Reception

Authors: Xavier Tardy, Grégoire Lefebvre, Apostolos Kountouris, Haïfa Fares, Amor Nafkha
Published: 2026-02-04
Categories: eess.SP, cs.IT, cs.LG
arXiv: https://arxiv.org/abs/2602.04728v3

We propose a cross-attention Transformer for joint decoding of uplink OFDM signals received by multiple coordinated access points. A shared per-receiver encoder learns the time-frequency structure of each grid, and a token-wise cross-attention module fuses the receivers to produce soft log-likelihood ratios for a standard channel decoder without explicit channel estimates. Trained with a bit-metric objective, the model adapts its fusion to per-receiver reliability and remains robust under degraded links, strong frequency selectivity, and sparse pilots. Over realistic Wi-Fi channels, it outperforms classical pipelines and strong neural baselines, often matching or surpassing a local perfect-CSI reference while remaining compact and computationally efficient on commodity hardware, making it suitable for next-generation coordinated Wi-Fi receivers.

---

### Supporting software engineering tasks with agentic AI: Demonstration on document retrieval and test scenario generation

Authors: Marian Kica, Lukas Radosky, David Slivka, Karin Kubinova, Daniel Dovhun, Tomas Uhercik, Erik Bircak, Ivan Polasek
Published: 2026-02-04
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2602.04726v1

The introduction of large language models ignited great retooling and rethinking of the software development models. The ensuing response of software engineering research yielded a massive body of tools and approaches. In this paper, we join the hassle by introducing agentic AI solutions for two tasks. First, we developed a solution for automatic test scenario generation from a detailed requirements description. This approach relies on specialized worker agents forming a star topology with the supervisor agent in the middle. We demonstrate its capabilities on a real-world example. Second, we developed an agentic AI solution for the document retrieval task in the context of software engineering documents. Our solution enables performing various use cases on a body of documents related to the development of a single software, including search, question answering, tracking changes, and large document summarization. In this case, each use case is handled by a dedicated LLM-based agent, which performs all subtasks related to the corresponding use case. We conclude by hinting at the future perspectives of our line of research.

---

### Benchmarking and Enhancing PPG-Based Cuffless Blood Pressure Estimation Methods

Authors: Neville Mathew, Yidan Shen, Renjie Hu, Maham Rahimi, George Zouridakis
Published: 2026-02-04
Categories: cs.LG, eess.SP
arXiv: https://arxiv.org/abs/2602.04725v1

Cuffless blood pressure screening based on easily acquired photoplethysmography (PPG) signals offers a practical pathway toward scalable cardiovascular health assessment. Despite rapid progress, existing PPG-based blood pressure estimation models have not consistently achieved the established clinical numerical limits such as AAMI/ISO 81060-2, and prior evaluations often lack the rigorous experimental controls necessary for valid clinical assessment. Moreover, the publicly available datasets commonly used are heterogeneous and lack physiologically controlled conditions for fair benchmarking. To enable fair benchmarking under physiologically controlled conditions, we created a standardized benchmarking subset NBPDB comprising 101,453 high-quality PPG segments from 1,103 healthy adults, derived from MIMIC-III and VitalDB. Using this dataset, we systematically benchmarked several state-of-the-art PPG-based models. The results showed that none of the evaluated models met the AAMI/ISO 81060-2 accuracy requirements (mean error $<$ 5 mmHg and standard deviation $<$ 8 mmHg). To improve model accuracy, we modified these models and added patient demographic data such as age, sex, and body mass index as additional inputs. Our modifications consistently improved performance across all models. In particular, the MInception model reduced error by 23\% after adding the demographic data and yielded mean absolute errors of 4.75 mmHg (SBP) and 2.90 mmHg (DBP), achieves accuracy comparable to the numerical limits defined by AAMI/ISO accuracy standards. Our results show that existing PPG-based BP estimation models lack clinical practicality under standardized conditions, while incorporating demographic information markedly improves their accuracy and physiological validity.

---

### How to rewrite the stars: Mapping your orchard over time through constellations of fruits

Authors: Gonçalo P. Matos, Carlos Santiago, João P. Costeira, Ricardo L. Saldanha, Ernesto M. Morgado
Published: 2026-02-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2602.04722v1

Following crop growth through the vegetative cycle allows farmers to predict fruit setting and yield in early stages, but it is a laborious and non-scalable task if performed by a human who has to manually measure fruit sizes with a caliper or dendrometers. In recent years, computer vision has been used to automate several tasks in precision agriculture, such as detecting and counting fruits, and estimating their size. However, the fundamental problem of matching the exact same fruits from one video, collected on a given date, to the fruits visible in another video, collected on a later date, which is needed to track fruits' growth through time, remains to be solved. Few attempts were made, but they either assume that the camera always starts from the same known position and that there are sufficiently distinct features to match, or they used other sources of data like GPS. Here we propose a new paradigm to tackle this problem, based on constellations of 3D centroids, and introduce a descriptor for very sparse 3D point clouds that can be used to match fruits across videos. Matching constellations instead of individual fruits is key to deal with non-rigidity, occlusions and challenging imagery with few distinct visual features to track. The results show that the proposed method can be successfully used to match fruits across videos and through time, and also to build an orchard map and later use it to locate the camera pose in 6DoF, thus providing a method for autonomous navigation of robots in the orchard and for selective fruit picking, for example.

---

### Superposition Without Interference? Towards Isolated Interventions via Almost Orthogonal Features in Language Models

Authors: Moritz Miller, Florent Draye, Bernhard Schölkopf
Published: 2026-02-04
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2602.04718v5

A central premise in mechanistic interpretability is that meaningful concepts in language models are represented by linear features in activation space. For such features to support reliable interventions, manipulating one feature should not substantially alter the effects of others. In practice, however, feature entanglement leads to interference such that localized interventions can have unintended downstream effects. Motivated by the _Independent Causal Mechanisms_ principle, we propose to constrain internal features to be almost orthogonal. We argue that this promotes modular representations amenable to causal intervention. We formalize this problem by characterizing the gap between an idealized isolated intervention and its realized effect on model outputs in terms of feature interference. We upper-bound the propagation of feature interference in terms of the self-coherence of the feature dictionary, and relate this discrepancy to an explicit orthogonality regularization on the dictionary itself. Empirically, we show that this regularization enables more isolated interventions on mathematical reasoning concepts while preserving model performance. Our code is available under https://github.com/mrtzmllr/sae-icm.

---

### Superposition Without Interference? Towards Isolated Interventions via Almost Orthogonal Features in Language Models

Authors: Moritz Miller, Florent Draye, Bernhard Schölkopf
Published: 2026-02-04
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2602.04718v4

A central premise in mechanistic interpretability is that meaningful concepts in language models are represented by linear features in activation space. For such features to support reliable interventions, manipulating one feature should not substantially alter the effects of others. In practice, however, feature entanglement leads to interference such that localized interventions can have unintended downstream effects. Motivated by the _Independent Causal Mechanisms_ principle, we propose to constrain internal features to be almost orthogonal. We argue that this promotes modular representations amenable to causal intervention. We formalize this problem by characterizing the gap between an idealized isolated intervention and its realized effect on model outputs in terms of feature interference. We upper-bound the propagation of feature interference in terms of the self-coherence of the feature dictionary, and relate this discrepancy to an explicit orthogonality regularization on the dictionary itself. Empirically, we show that this regularization enables more isolated interventions on mathematical reasoning concepts while preserving model performance. Our code is available under https://github.com/mrtzmllr/sae-icm.

---

### Evolutionary Mapping of Neural Networks to Spatial Accelerators

Authors: Alessandro Pierro, Jonathan Timcheck, Jason Yik, Marius Lindauer, Eyke Hüllermeier, Marcel Wever
Published: 2026-02-04
Categories: cs.NE
arXiv: https://arxiv.org/abs/2602.04717v1

Spatial accelerators, composed of arrays of compute-memory integrated units, offer an attractive platform for deploying inference workloads with low latency and low energy consumption. However, fully exploiting their architectural advantages typically requires careful, expert-driven mapping of computational graphs to distributed processing elements. In this work, we automate this process by framing the mapping challenge as a black-box optimization problem. We introduce the first evolutionary, hardware-in-the-loop mapping framework for neuromorphic accelerators, enabling users without deep hardware knowledge to deploy workloads more efficiently. We evaluate our approach on Intel Loihi 2, a representative spatial accelerator featuring 152 cores per chip in a 2D mesh. Our method achieves up to 35% reduction in total latency compared to default heuristics on two sparse multi-layer perceptron networks. Furthermore, we demonstrate the scalability of our approach to multi-chip systems and observe an up to 40% improvement in energy efficiency, without explicitly optimizing for it.

---

### Linguistically Informed Evaluation of Multilingual ASR for African Languages

Authors: Fei-Yueh Chen, Lateef Adeleke, C. M. Downey
Published: 2026-02-04
Categories: cs.CL
arXiv: https://arxiv.org/abs/2602.04716v1

Word Error Rate (WER) mischaracterizes ASR models' performance for African languages by combining phonological, tone, and other linguistic errors into a single lexical error. By contrast, Feature Error Rate (FER) has recently attracted attention as a viable metric that reveals linguistically meaningful errors in models' performance. In this paper, we evaluate three speech encoders on two African languages by complementing WER with CER, and FER, and add a tone-aware extension (TER). We show that by computing errors on phonological features, FER and TER reveal linguistically-salient error patterns even when word-level accuracy remains low. Our results reveal that models perform better on segmental features, while tones (especially mid and downstep) remain the most challenging features. Results on Yoruba show a striking differential in metrics, with WER=0.788, CER=0.305, and FER=0.151. Similarly for Uneme (an endangered language absent from pretraining data) a model with near-total WER and 0.461 CER achieves the relatively low FER of 0.267. This indicates model error is often attributable to individual phonetic feature errors, which is obscured by all-or-nothing metrics like WER.

---

### Bounded-Abstention Multi-horizon Time-series Forecasting

Authors: Luca Stradiotti, Laurens Devos, Anna Monreale, Jesse Davis, Andrea Pugnana
Published: 2026-02-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2602.04714v1

Multi-horizon time-series forecasting involves simultaneously making predictions for a consecutive sequence of subsequent time steps. This task arises in many application domains, such as healthcare and finance, where mispredictions can have a high cost and reduce trust. The learning with abstention framework tackles these problems by allowing a model to abstain from offering a prediction when it is at an elevated risk of making a misprediction. Unfortunately, existing abstention strategies are ill-suited for the multi-horizon setting: they target problems where a model offers a single prediction for each instance. Hence, they ignore the structured and correlated nature of the predictions offered by a multi-horizon forecaster. We formalize the problem of learning with abstention for multi-horizon forecasting setting and show that its structured nature admits a richer set of abstention problems. Concretely, we propose three natural notions of how a model could abstain for multi-horizon forecasting. We theoretically analyze each problem to derive the optimal abstention strategy and propose an algorithm that implements it. Extensive evaluation on 24 datasets shows that our proposed algorithms significantly outperforms existing baselines.

---

### Adaptive Prompt Elicitation for Text-to-Image Generation

Authors: Xinyi Wen, Lena Hegemann, Xiaofu Jin, Shuai Ma, Antti Oulasvirta
Published: 2026-02-04
Categories: cs.HC, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2602.04713v2

Aligning text-to-image generation with user intent remains challenging, as users frequently provide ambiguous inputs and struggle with model idiosyncrasies. We propose Adaptive Prompt Elicitation (APE), a technique that adaptively poses visual queries to help users refine prompts without extensive writing. Our technical contribution is a formulation of interactive intent inference under an information-theoretic framework. APE represents latent user intent as interpretable feature requirements using language model priors, adaptively generates visual queries, and compiles elicited requirements into effective prompts. Evaluation on IDEA-Bench and DesignBench shows that APE achieves stronger alignment with improved efficiency. A user study with 128 participants on user-defined tasks demonstrates 19.8% higher perceived alignment without increased workload. Our work contributes a principled approach to prompting that offers an effective and efficient complement to the prevailing prompt-based interaction paradigm with text-to-image models.

---

### SAR-RAG: ATR Visual Question Answering by Semantic Search, Retrieval, and MLLM Generation

Authors: David F. Ramirez, Tim Overman, Kristen Jaskie, Joe Marvin, Andreas Spanias
Published: 2026-02-04
Categories: cs.CV, cs.AI, eess.IV
arXiv: https://arxiv.org/abs/2602.04712v2

We present a visual-context image-retrieval-augmented generation (ImageRAG)- assisted AI agent for automatic target recognition (ATR) of synthetic aperture radar (SAR) imagery. SAR is a remote sensing method used in defense and security applications to detect and monitor the positions of military vehicles, which may appear indistinguishable in images. Researchers have extensively studied SAR ATR to improve the differentiation and identification of vehicle types, characteristics, and measurements. Test examples can be compared with known vehicle target types to improve recognition tasks. New methods enhance the capabilities of neural networks, transformer attention, and multimodal large language models. An agentic AI method may be developed to utilize a defined set of tools, such as searching through a library of similar examples. Our proposed method, SAR Retrieval-Augmented Generation (SAR-RAG), combines a multimodal large language model (MLLM) with a vector database of semantic embeddings to support contextual search for image exemplars with known qualities. By recovering past image examples of known true target types, our SAR-RAG system can compare similar vehicle categories, thereby improving ATR prediction accuracy. We evaluate this through search and retrieval metrics, categorical classification accuracy, and numeric regression of vehicle dimensions. These metrics all show improvements when SAR-RAG is added to an MLLM baseline method as an attached ATR memory bank.

---

### Addressing Corpus Knowledge Poisoning Attacks on RAG Using Sparse Attention

Authors: Sagie Dekel, Moshe Tennenholtz, Oren Kurland
Published: 2026-02-04
Categories: cs.IR, cs.AI
arXiv: https://arxiv.org/abs/2602.04711v3

Retrieval Augmented Generation (RAG) is a highly effective paradigm for keeping LLM-based responses up-to-date and reducing the likelihood of hallucinations. Yet, RAG was recently shown to be quite vulnerable to corpus knowledge poisoning: an attacker injects misleading documents to the corpus to steer an LLM's output to an undesired response. We argue that the standard causal attention mechanism in LLMs enables harmful cross-document interactions, specifically in cases of attacks. Accordingly, we introduce a novel defense approach for RAG: Sparse Document Attention RAG (SDAG). This is a block-sparse attention mechanism that disallows cross-attention between retrieved documents. SDAG requires a minimal inference-time change to the attention mask. We present an empirical evaluation of LLM-based question answering (QA) with a variety of attack strategies on RAG. We show that our SDAG method substantially outperforms the standard causal attention mechanism. We further demonstrate the clear merits of integrating SDAG with state-of-the-art RAG defense methods. Specifically, the integration results in performance that is statistically significantly better than the state-of-the-art.

---

### Addressing Corpus Knowledge Poisoning Attacks on RAG Using Sparse Attention

Authors: Sagie Dekel, Moshe Tennenholtz, Oren Kurland
Published: 2026-02-04
Categories: cs.IR, cs.AI
arXiv: https://arxiv.org/abs/2602.04711v2

Retrieval Augmented Generation (RAG) is a highly effective paradigm for keeping LLM-based responses up-to-date and reducing the likelihood of hallucinations. Yet, RAG was recently shown to be quite vulnerable to corpus knowledge poisoning: an attacker injects misleading documents to the corpus to steer an LLM's output to an undesired response. We argue that the standard causal attention mechanism in LLMs enables harmful cross-document interactions, specifically in cases of attacks. Accordingly, we introduce a novel defense approach for RAG: Sparse Document Attention RAG (SDAG). This is a block-sparse attention mechanism that disallows cross-attention between retrieved documents. SDAG requires a minimal inference-time change to the attention mask; furthermore, no fine-tuning or additional architectural changes are needed. We present an empirical evaluation of LLM-based question answering (QA) with a variety of attack strategies on RAG. We show that our SDAG method substantially outperforms the standard causal attention mechanism in terms of attack success rate. We further demonstrate the clear merits of integrating SDAG with state-of-the-art RAG defense methods. Specifically, the integration results in performance that is statistically significantly better than the state-of-the-art.

---

### Towards Understanding and Avoiding Limitations of Convolutions on Graphs

Authors: Andreas Roth
Published: 2026-02-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2602.04709v1

While message-passing neural networks (MPNNs) have shown promising results, their real-world impact remains limited. Although various limitations have been identified, their theoretical foundations remain poorly understood, leading to fragmented research efforts. In this thesis, we provide an in-depth theoretical analysis and identify several key properties limiting the performance of MPNNs. Building on these findings, we propose several frameworks that address these shortcomings. We identify two properties exhibited by many MPNNs: shared component amplification (SCA), where each message-passing iteration amplifies the same components across all feature channels, and component dominance (CD), where a single component gets increasingly amplified as more message-passing steps are applied. These properties lead to the observable phenomenon of rank collapse of node representations, which generalizes the established over-smoothing phenomenon. By generalizing and decomposing over-smoothing, we enable a deeper understanding of MPNNs, more targeted solutions, and more precise communication within the field. To avoid SCA, we show that utilizing multiple computational graphs or edge relations is necessary. Our multi-relational split (MRS) framework transforms any existing MPNN into one that leverages multiple edge relations. Additionally, we introduce the spectral graph convolution for multiple feature channels (MIMO-GC), which naturally uses multiple computational graphs. A localized variant, LMGC, approximates the MIMO-GC while inheriting its beneficial properties. To address CD, we demonstrate a close connection between MPNNs and the PageRank algorithm. Based on personalized PageRank, we propose a variant of MPNNs that allows for infinitely many message-passing iterations, while preserving initial node features. Collectively, these results deepen the theoretical understanding of MPNNs.

---

### LiteToken: Removing Intermediate Merge Residues From BPE Tokenizers

Authors: Yike Sun, Haotong Yang, Zhouchen Lin, Muhan Zhang
Published: 2026-02-04
Categories: cs.CL
arXiv: https://arxiv.org/abs/2602.04706v1

Tokenization is fundamental to how language models represent and process text, yet the behavior of widely used BPE tokenizers has received far less study than model architectures and training. In this paper, we investigate intermediate merge residues in BPE vocabularies: tokens that are frequent during merge learning so that retained in the final vocabulary, but are mostly further merged and rarely emitted when tokenizing the corpus during tokenizer usage. Such low-frequency tokens not only waste vocabulary capacity but also increase vulnerability to adversarial or atypical inputs. We present a systematic empirical characterization of this phenomenon across commonly used tokenizers and introduce LiteToken, a simple method for removing residue tokens. Because the affected tokens are rarely used, pretrained models can often accommodate the modified tokenizer without additional fine-tuning. Experiments show that LiteToken reduces token fragmentation, reduces parameters, and improves robustness to noisy or misspelled inputs, while preserving overall performance.

---

### ERNIE 5.0 Technical Report

Authors: Haifeng Wang, Hua Wu, Tian Wu, Yu Sun, Jing Liu, Dianhai Yu, Yanjun Ma, Jingzhou He, Zhongjun He, Dou Hong, Qiwen Liu, Shuohuan Wang et al.
Published: 2026-02-04
Categories: cs.CL
arXiv: https://arxiv.org/abs/2602.04705v1

In this report, we introduce ERNIE 5.0, a natively autoregressive foundation model desinged for unified multimodal understanding and generation across text, image, video, and audio. All modalities are trained from scratch under a unified next-group-of-tokens prediction objective, based on an ultra-sparse mixture-of-experts (MoE) architecture with modality-agnostic expert routing. To address practical challenges in large-scale deployment under diverse resource constraints, ERNIE 5.0 adopts a novel elastic training paradigm. Within a single pre-training run, the model learns a family of sub-models with varying depths, expert capacities, and routing sparsity, enabling flexible trade-offs among performance, model size, and inference latency in memory- or time-constrained scenarios. Moreover, we systematically address the challenges of scaling reinforcement learning to unified foundation models, thereby guaranteeing efficient and stable post-training under ultra-sparse MoE architectures and diverse multimodal settings. Extensive experiments demonstrate that ERNIE 5.0 achieves strong and balanced performance across multiple modalities. To the best of our knowledge, among publicly disclosed models, ERNIE 5.0 represents the first production-scale realization of a trillion-parameter unified autoregressive model that supports both multimodal understanding and generation. To facilitate further research, we present detailed visualizations of modality-agnostic expert routing in the unified model, alongside comprehensive empirical analysis of elastic training, aiming to offer profound insights to the community.

---

### Knowledge Distillation for mmWave Beam Prediction Using Sub-6 GHz Channels

Authors: Sina Tavakolian, Nhan Thanh Nguyen, Ahmed Alkhateeb, Markku Juntti
Published: 2026-02-04
Categories: eess.SP, cs.LG
arXiv: https://arxiv.org/abs/2602.04703v1

Beamforming in millimeter-wave (mmWave) high-mobility environments typically incurs substantial training overhead. While prior studies suggest that sub-6 GHz channels can be exploited to predict optimal mmWave beams, existing methods depend on large deep learning (DL) models with prohibitive computational and memory requirements. In this paper, we propose a computationally efficient framework for sub-6 GHz channel-mmWave beam mapping based on the knowledge distillation (KD) technique. We develop two compact student DL architectures based on individual and relational distillation strategies, which retain only a few hidden layers yet closely mimic the performance of large teacher DL models. Extensive simulations demonstrate that the proposed student models achieve the teacher's beam prediction accuracy and spectral efficiency while reducing trainable parameters and computational complexity by 99%.

---

### Annotation Free Spacecraft Detection and Segmentation using Vision Language Models

Authors: Samet Hicsonmez, Jose Sosa, Dan Pineau, Inder Pal Singh, Arunkumar Rathinam, Abd El Rahman Shabayek, Djamila Aouada
Published: 2026-02-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2602.04699v1

Vision Language Models (VLMs) have demonstrated remarkable performance in open-world zero-shot visual recognition. However, their potential in space-related applications remains largely unexplored. In the space domain, accurate manual annotation is particularly challenging due to factors such as low visibility, illumination variations, and object blending with planetary backgrounds. Developing methods that can detect and segment spacecraft and orbital targets without requiring extensive manual labeling is therefore of critical importance. In this work, we propose an annotation-free detection and segmentation pipeline for space targets using VLMs. Our approach begins by automatically generating pseudo-labels for a small subset of unlabeled real data with a pre-trained VLM. These pseudo-labels are then leveraged in a teacher-student label distillation framework to train lightweight models. Despite the inherent noise in the pseudo-labels, the distillation process leads to substantial performance gains over direct zero-shot VLM inference. Experimental evaluations on the SPARK-2024, SPEED+, and TANGO datasets on segmentation tasks demonstrate consistent improvements in average precision (AP) by up to 10 points. Code and models are available at https://github.com/giddyyupp/annotation-free-spacecraft-segmentation.

---

### Beyond Learning on Molecules by Weakly Supervising on Molecules

Authors: Gordan Prastalo, Kevin Maik Jablonka
Published: 2026-02-04
Categories: physics.chem-ph, cs.LG
arXiv: https://arxiv.org/abs/2602.04696v1

Molecular representations are inherently task-dependent, yet most pre-trained molecular encoders are not. Task conditioning promises representations that reorganize based on task descriptions, but existing approaches rely on expensive labeled data. We show that weak supervision on programmatically derived molecular motifs is sufficient. Our Adaptive Chemical Embedding Model (ACE-Mol) learns from hundreds of motifs paired with natural language descriptors that are cheap to compute, trivial to scale. Conventional encoders slowly search the embedding space for task-relevant structure, whereas ACE-Mol immediately aligns its representations with the task. ACE-Mol achieves state-of-the-art performance across molecular property prediction benchmarks with interpretable, chemically meaningful representations.

---

### LinGO: A Linguistic Graph Optimization Framework with LLMs for Interpreting Intents of Online Uncivil Discourse

Authors: Yuan Zhang, Thales Bertaglia
Published: 2026-02-04
Categories: cs.CL, cs.CY
arXiv: https://arxiv.org/abs/2602.04693v1

Detecting uncivil language is crucial for maintaining safe, inclusive, and democratic online spaces. Yet existing classifiers often misinterpret posts containing uncivil cues but expressing civil intents, leading to inflated estimates of harmful incivility online. We introduce LinGO, a linguistic graph optimization framework for large language models (LLMs) that leverages linguistic structures and optimization techniques to classify multi-class intents of incivility that use various direct and indirect expressions. LinGO decomposes language into multi-step linguistic components, identifies targeted steps that cause the most errors, and iteratively optimizes prompt and/or example components for targeted steps. We evaluate it using a dataset collected during the 2022 Brazilian presidential election, encompassing four forms of political incivility: Impoliteness (IMP), Hate Speech and Stereotyping (HSST), Physical Harm and Violent Political Rhetoric (PHAVPR), and Threats to Democratic Institutions and Values (THREAT). Each instance is annotated with six types of civil/uncivil intent. We benchmark LinGO using three cost-efficient LLMs: GPT-5-mini, Gemini 2.5 Flash-Lite, and Claude 3 Haiku, and four optimization techniques: TextGrad, AdalFlow, DSPy, and Retrieval-Augmented Generation (RAG). The results show that, across all models, LinGO consistently improves accuracy and weighted F1 compared with zero-shot, chain-of-thought, direct optimization, and fine-tuning baselines. RAG is the strongest optimization technique and, when paired with Gemini model, achieves the best overall performance. These findings demonstrate that incorporating multi-step linguistic components into LLM instructions and optimize targeted components can help the models explain complex semantic meanings, which can be extended to other complex semantic explanation tasks in the future.

---

### DRMOT: A Dataset and Framework for RGBD Referring Multi-Object Tracking

Authors: Sijia Chen, Lijuan Ma, Yanqiu Yu, En Yu, Liman Liu, Wenbing Tao
Published: 2026-02-04
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2602.04692v2

Referring Multi-Object Tracking (RMOT) aims to track specific targets based on language descriptions and is vital for interactive AI systems such as robotics and autonomous driving. However, existing RMOT models rely solely on 2D RGB data, making it challenging to accurately detect and associate targets characterized by complex spatial semantics (e.g., ``the person closest to the camera'') and to maintain reliable identities under severe occlusion, due to the absence of explicit 3D spatial information. In this work, we propose a novel task, RGBD Referring Multi-Object Tracking (DRMOT), which explicitly requires models to fuse RGB, Depth (D), and Language (L) modalities to achieve 3D-aware tracking. To advance research on the DRMOT task, we construct a tailored RGBD referring multi-object tracking dataset, named DRSet, designed to evaluate models' spatial-semantic grounding and tracking capabilities. Specifically, DRSet contains RGB images and depth maps from 187 scenes, along with 240 language descriptions, among which 56 descriptions incorporate depth-related information. Furthermore, we propose DRTrack, a MLLM-guided depth-referring tracking framework. DRTrack performs depth-aware target grounding from joint RGB-D-L inputs and enforces robust trajectory association by incorporating depth cues. Extensive experiments on the DRSet dataset demonstrate the effectiveness of our framework.

---

### Static and auto-regressive neural emulation of phytoplankton biomass dynamics from physical predictors in the global ocean

Authors: Mahima Lakra, Ronan Fablet, Lucas Drumetz, Etienne Pauthenet, Elodie Martinez
Published: 2026-02-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2602.04689v1

Phytoplankton is the basis of marine food webs, driving both ecological processes and global biogeochemical cycles. Despite their ecological and climatic significance, accurately simulating phytoplankton dynamics remains a major challenge for biogeochemical numerical models due to limited parameterizations, sparse observational data, and the complexity of oceanic processes. Here, we explore how deep learning models can be used to address these limitations predicting the spatio-temporal distribution of phytoplankton biomass in the global ocean based on satellite observations and environmental conditions. First, we investigate several deep learning architectures. Among the tested models, the UNet architecture stands out for its ability to reproduce the seasonal and interannual patterns of phytoplankton biomass more accurately than other models like CNNs, ConvLSTM, and 4CastNet. When using one to two months of environmental data as input, UNet performs better, although it tends to underestimate the amplitude of low-frequency changes in phytoplankton biomass. Thus, to improve predictions over time, an auto-regressive version of UNet was also tested, where the model uses its own previous predictions to forecast future conditions. This approach works well for short-term forecasts (up to five months), though its performance decreases for longer time scales. Overall, our study shows that combining ocean physical predictors with deep learning allows for reconstruction and short-term prediction of phytoplankton dynamics. These models could become powerful tools for monitoring ocean health and supporting marine ecosystem management, especially in the context of climate change.

---

### Investigating Disability Representations in Text-to-Image Models

Authors: Yang Tian, Yu Fan, Liudmila Zavolokina, Sarah Ebling
Published: 2026-02-04
Categories: cs.CL, cs.CV, cs.CY, cs.HC
arXiv: https://arxiv.org/abs/2602.04687v3

Text-to-image generative models have made remarkable progress in producing high-quality visual content from textual descriptions, yet concerns remain about how they represent social groups. While characteristics like gender and race have received increasing attention, disability representations remain underexplored. This study investigates how people with disabilities are represented in AI-generated images by analyzing outputs from Stable Diffusion XL and DALL-E 3 using a structured prompt design. We analyze disability representations by comparing image similarities between generic disability prompts and prompts referring to specific disability categories. Moreover, we evaluate how mitigation strategies influence disability portrayals, with a focus on assessing affective framing through sentiment polarity analysis, combining both automatic and human evaluation. Our findings reveal persistent representational imbalances and highlight the need for continuous evaluation and refinement of generative models to foster more diverse and inclusive portrayals of disability.

---

### Audio ControlNet for Fine-Grained Audio Generation and Editing

Authors: Haina Zhu, Yao Xiao, Xiquan Li, Ziyang Ma, Jianwei Yu, Bowen Zhang, Mingqi Yang, Xie Chen
Published: 2026-02-04
Categories: cs.SD, cs.AI, cs.CL, cs.MM
arXiv: https://arxiv.org/abs/2602.04680v1

We study the fine-grained text-to-audio (T2A) generation task. While recent models can synthesize high-quality audio from text descriptions, they often lack precise control over attributes such as loudness, pitch, and sound events. Unlike prior approaches that retrain models for specific control types, we propose to train ControlNet models on top of pre-trained T2A backbones to achieve controllable generation over loudness, pitch, and event roll. We introduce two designs, T2A-ControlNet and T2A-Adapter, and show that the T2A-Adapter model offers a more efficient structure with strong control ability. With only 38M additional parameters, T2A-Adapter achieves state-of-the-art performance on the AudioSet-Strong in both event-level and segment-level F1 scores. We further extend this framework to audio editing, proposing T2A-Editor for removing and inserting audio events at time locations specified by instructions. Models, code, dataset pipelines, and benchmarks will be released to support future research on controllable audio generation and editing.

---

### Let Experts Feel Uncertainty: A Multi-Expert Label Distribution Approach to Probabilistic Time Series Forecasting

Authors: Zhen Zhou, Zhirui Wang, Qi Hong, Yunyang Shi, Ziyuan Gu, Zhiyuan Liu
Published: 2026-02-04
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2602.04678v1

Time series forecasting in real-world applications requires both high predictive accuracy and interpretable uncertainty quantification. Traditional point prediction methods often fail to capture the inherent uncertainty in time series data, while existing probabilistic approaches struggle to balance computational efficiency with interpretability. We propose a novel Multi-Expert Learning Distributional Labels (LDL) framework that addresses these challenges through mixture-of-experts architectures with distributional learning capabilities. Our approach introduces two complementary methods: (1) Multi-Expert LDL, which employs multiple experts with different learned parameters to capture diverse temporal patterns, and (2) Pattern-Aware LDL-MoE, which explicitly decomposes time series into interpretable components (trend, seasonality, changepoints, volatility) through specialized sub-experts. Both frameworks extend traditional point prediction to distributional learning, enabling rich uncertainty quantification through Maximum Mean Discrepancy (MMD). We evaluate our methods on aggregated sales data derived from the M5 dataset, demonstrating superior performance compared to baseline approaches. The continuous Multi-Expert LDL achieves the best overall performance, while the Pattern-Aware LDL-MoE provides enhanced interpretability through component-wise analysis. Our frameworks successfully balance predictive accuracy with interpretability, making them suitable for real-world forecasting applications where both performance and actionable insights are crucial.

---