# Research Papers — 2026-01 · bundle page 48 of 189 > Full-text markdown of papers #2351–2400 of 9403 published in 2026-01, newest first. AI-readable mirrors of the full 9403-paper 2026-01 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-01/page/48 [← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/47.md) [Next bundle →](https://aipod.dev/papers/archive/2026-01/page/49.md) ### Malicious Repurposing of Open Science Artefacts by Using Large Language Models Authors: Zahra Hashemi, Zhiqiang Zhong, Jun Pang, Wei Zhao Published: 2026-01-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.18998v1 The rapid evolution of large language models (LLMs) has fuelled enthusiasm about their role in advancing scientific discovery, with studies exploring LLMs that autonomously generate and evaluate novel research ideas. However, little attention has been given to the possibility that such models could be exploited to produce harmful research by repurposing open science artefacts for malicious ends. We fill the gap by introducing an end-to-end pipeline that first bypasses LLM safeguards through persuasion-based jailbreaking, then reinterprets NLP papers to identify and repurpose their artefacts (datasets, methods, and tools) by exploiting their vulnerabilities, and finally assesses the safety of these proposals using our evaluation framework across three dimensions: harmfulness, feasibility of misuse, and soundness of technicality. Overall, our findings demonstrate that LLMs can generate harmful proposals by repurposing ethically designed open artefacts; however, we find that LLMs acting as evaluators strongly disagree with one another on evaluation outcomes: GPT-4.1 assigns higher scores (indicating greater potential harms, higher soundness and feasibility of misuse), Gemini-2.5-pro is markedly stricter, and Grok-3 falls between these extremes. This indicates that LLMs cannot yet serve as reliable judges in a malicious evaluation setup, making human evaluation essential for credible dual-use risk assessment. --- ### Anatomically-aware conformal prediction for medical image segmentation with random walks Authors: Mélanie Gaillochet, Christian Desrosiers, Hervé Lombaert Published: 2026-01-26 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2601.18997v1 The reliable deployment of deep learning in medical imaging requires uncertainty quantification that provides rigorous error guarantees while remaining anatomically meaningful. Conformal prediction (CP) is a powerful distribution-free framework for constructing statistically valid prediction intervals. However, standard applications in segmentation often ignore anatomical context, resulting in fragmented, spatially incoherent, and over-segmented prediction sets that limit clinical utility. To bridge this gap, this paper proposes Random-Walk Conformal Prediction (RW-CP), a model-agnostic framework which can be added on top of any segmentation method. RW-CP enforces spatial coherence to generate anatomically valid sets. Our method constructs a k-nearest neighbour graph from pre-trained vision foundation model features and applies a random walk to diffuse uncertainty. The random walk diffusion regularizes the non-conformity scores, making the prediction sets less sensitive to the conformal calibration parameter $λ$, ensuring more stable and continuous anatomical boundaries. RW-CP maintains rigorous marginal coverage while significantly improving segmentation quality. Evaluations on multi-modal public datasets show improvements of up to $35.4\%$ compared to standard CP baselines, given an allowable error rate of $α=0.1$. --- ### FreeOrbit4D: Training-Free Arbitrary Camera Redirection for Monocular Videos via Foreground-Complete 4D Reconstruction Authors: Wei Cao, Hao Zhang, Fengrui Tian, Yulun Wu, Yingying Li, Shenlong Wang, Ning Yu, Yaoyao Liu Published: 2026-01-26 Categories: cs.CV, cs.AI, cs.GR arXiv: https://arxiv.org/abs/2601.18993v2 Camera redirection aims to replay a dynamic scene from a single monocular video under a user-specified camera trajectory. However, large-angle redirection is inherently ill-posed: a monocular video captures only a narrow spatio-temporal view of a dynamic 3D scene, providing severely limited observations of the underlying 4D world. The key challenge is therefore to recover a complete and coherent representation from this limited input, with consistent geometry and motion. While recent diffusion-based methods achieve impressive visual generation quality, they often break down under large-angle viewpoint changes far from the original trajectory, where missing visual grounding leads to severe geometric ambiguity and temporal inconsistency. We present FreeOrbit4D, an effective training-free framework that tackles this ambiguity by recovering a foreground-complete 4D proxy as structural grounding for video generation. We obtain this proxy by decoupling foreground and background reconstructions: we unproject the monocular video into a static background and partial foreground point clouds in a unified global space, then use an object-centric multi-view diffusion model to synthesize multi-view images and reconstruct complete foreground point clouds in canonical object space. By aligning the canonical foreground point cloud to the global scene space via dense pixel-synchronized 3D-3D correspondences and projecting the foreground-complete 4D proxy onto target camera viewpoints, we provide geometric scaffolds that guide a conditional video diffusion model. Extensive experiments show that FreeOrbit4D produces more faithful and temporally coherent redirected videos under challenging large-angle trajectories, and our proxy further enables applications such as edit propagation and 4D data generation. Project page: https://freeorbit4d.vision.ischool.illinois.edu/ --- ### LLMs versus the Halting Problem: Characterizing Program Termination Reasoning Authors: Oren Sultan, Jordi Armengol-Estape, Pascal Kesseli, Julien Vanegue, Dafna Shahaf, Yossi Adi, Peter O'Hearn Published: 2026-01-26 Categories: cs.CL, cs.AI, cs.PL arXiv: https://arxiv.org/abs/2601.18987v7 Determining whether a program terminates is a central problem in computer science. Turing's Halting Problem established termination as undecidable, showing that no algorithm can universally determine termination for all programs and inputs. Hence, verification tools approximate termination, sometimes failing to prove or disprove; these tools rely on problem-specific architectures and are usually tied to particular programming languages. Recent advances in LLMs raise a natural question: To what extent can they reason about program termination? We evaluate frontier LLMs on a diverse set of C programs from the International Competition on Software Verification (SV-Comp) 2025. Our results show that GPT-5 and Claude Sonnet-4.5 achieve scores comparable to top-ranked verification tools (with test-time scaling). However, while models often correctly infer whether programs terminate, they frequently fail to construct a witness as formal proof, revealing a gap between semantic recognition and symbolic proof generation. Performance further degrades as code length increases. Beyond witness automaton graphs, we introduce a divergence-precondition formulation that characterizes non-termination conditions as logical constraints. We hope these findings motivate future research on real-world termination benchmarks, neuro-symbolic approaches that combine LLMs with symbolic verification methods, and, more broadly, LLM reasoning on other undecidable problems. --- ### Save the Good Prefix: Precise Error Penalization via Process-Supervised RL to Enhance LLM Reasoning Authors: Haolin Liu, Dian Yu, Sidi Lu, Yujun Zhou, Rui Liu, Zhenwen Liang, Haitao Mi, Chen-Yu Wei, Dong Yu Published: 2026-01-26 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2601.18984v1 Reinforcement learning (RL) has emerged as a powerful framework for improving the reasoning capabilities of large language models (LLMs). However, most existing RL approaches rely on sparse outcome rewards, which fail to credit correct intermediate steps in partially successful solutions. Process reward models (PRMs) offer fine-grained step-level supervision, but their scores are often noisy and difficult to evaluate. As a result, recent PRM benchmarks focus on a more objective capability: detecting the first incorrect step in a reasoning path. However, this evaluation target is misaligned with how PRMs are typically used in RL, where their step-wise scores are treated as raw rewards to maximize. To bridge this gap, we propose Verifiable Prefix Policy Optimization (VPPO), which uses PRMs only to localize the first error during RL. Given an incorrect rollout, VPPO partitions the trajectory into a verified correct prefix and an erroneous suffix based on the first error, rewarding the former while applying targeted penalties only after the detected mistake. This design yields stable, interpretable learning signals and improves credit assignment. Across multiple reasoning benchmarks, VPPO consistently outperforms sparse-reward RL and prior PRM-guided baselines on both Pass@1 and Pass@K. --- ### Attention-Enhanced Graph Filtering for False Data Injection Attack Detection and Localization Authors: Ruslan Abdulin, Mohammad Rasoul Narimani Published: 2026-01-26 Categories: cs.LG, cs.CR, math.OC arXiv: https://arxiv.org/abs/2601.18981v1 The increasing deployment of Internet-of-Things (IoT)-enabled measurement devices in modern power systems has expanded the cyberattack surface of the grid. As a result, this critical infrastructure is increasingly exposed to cyberattacks, including false data injection attacks (FDIAs) that compromise measurement integrity and threaten reliable system operation. Existing FDIA detection methods primarily exploit spatial correlations and network topology using graph-based learning; however, these approaches often rely on high-dimensional representations and shallow classifiers, limiting their ability to capture local structural dependencies and global contextual relationships. Moreover, naively incorporating Transformer architectures can result in overly deep models that struggle to model localized grid dynamics. This paper proposes a joint FDIA detection and localization framework that integrates auto-regressive moving average (ARMA) graph convolutional filters with an Encoder-Only Transformer architecture. The ARMA-based graph filters provide robust, topology-aware feature extraction and adaptability to abrupt spectral changes, while the Transformer encoder leverages self-attention to capture long-range dependencies among grid elements without sacrificing essential local context. The proposed method is evaluated using real-world load data from the New York Independent System Operator (NYISO) applied to the IEEE 14- and 300-bus systems. Numerical results demonstrate that the proposed model effectively exploits both the state and topology of the power grid, achieving high accuracy in detecting FDIA events and localizing compromised nodes. --- ### Intent2QoS: Language Model-Driven Automation of Traffic Shaping Configurations Authors: Sudipta Acharya, Burak Kantarci Published: 2026-01-26 Categories: cs.NI, cs.CL arXiv: https://arxiv.org/abs/2601.18974v1 Traffic shaping and Quality of Service (QoS) enforcement are critical for managing bandwidth, latency, and fairness in networks. These tasks often rely on low-level traffic control settings, which require manual setup and technical expertise. This paper presents an automated framework that converts high-level traffic shaping intents in natural or declarative language into valid and correct traffic control rules. To the best of our knowledge, we present the first end-to-end pipeline that ties intent translation in a queuing-theoretic semantic model and, with a rule-based critic, yields deployable Linux traffic control configuration sets. The framework has three steps: (1) a queuing simulation with priority scheduling and Active Queue Management (AQM) builds a semantic model; (2) a language model, using this semantic model and a traffic profile, generates sub-intents and configuration rules; and (3) a rule-based critic checks and adjusts the rules for correctness and policy compliance. We evaluate multiple language models by generating traffic control commands from business intents that comply with relevant standards for traffic control protocols. Experimental results on 100 intents show significant gains, with LLaMA3 reaching 0.88 semantic similarity and 0.87 semantic coverage, outperforming other models by over 30\. A thorough sensitivity study demonstrates that AQM-guided prompting reduces variability threefold compared to zero-shot baselines. --- ### When Does Adaptation Win? Scaling Laws for Meta-Learning in Quantum Control Authors: Nima Leclerc, Chris Miller, Nicholas Brawand Published: 2026-01-26 Categories: cs.LG, cs.AI, eess.SY, quant-ph arXiv: https://arxiv.org/abs/2601.18973v4 Quantum hardware suffers from intrinsic device heterogeneity and environmental drift, forcing practitioners to choose between suboptimal non-adaptive controllers or costly per-device recalibration. We derive a scaling law lower bound for meta-learning showing that the adaptation gain (expected fidelity improvement from task-specific gradient steps) saturates exponentially with gradient steps and scales linearly with task variance, providing a quantitative criterion for when adaptation justifies its overhead. Validation on quantum gate calibration shows negligible benefits for low-variance tasks but >40% fidelity gains on two-qubit gates under extreme out-of-distribution conditions (10$\times$ the training noise), with implications for reducing per-device calibration time on cloud quantum processors. Further validation on classical linear-quadratic control confirms these laws emerge from general optimization geometry rather than quantum-specific physics. We further introduce a few-shot pre-adaptation protocol that estimates the optimal adaptation budget from $N{=}3$-5 probe steps within 3-19% relative error across out-of-distribution regimes. --- ### Towards Self-Optimizing Electron Microscope: Robust Tuning of Aberration Coefficients via Physics-Aware Multi-Objective Bayesian Optimization Authors: Utkarsh Pratiush, Austin Houston, Richard Liu, Gerd Duscher, Sergei Kalinin Published: 2026-01-26 Categories: cs.LG, cond-mat.mtrl-sci arXiv: https://arxiv.org/abs/2601.18972v1 Realizing high-throughput aberration-corrected Scanning Transmission Electron Microscopy (STEM) exploration of atomic structures requires rapid tuning of multipole probe correctors while compensating for the inevitable drift of the optical column. While automated alignment routines exist, conventional approaches rely on serial, gradient-free searches (e.g., Nelder-Mead) that are sample-inefficient and struggle to correct multiple interacting parameters simultaneously. Conversely, emerging deep learning methods offer speed but often lack the flexibility to adapt to varying sample conditions without extensive retraining. Here, we introduce a Multi-Objective Bayesian Optimization (MOBO) framework for rapid, data-efficient aberration correction. Importantly, this framework does not prescribe a single notion of image quality; instead, it enables user-defined, physically motivated reward formulations (e.g., symmetry-induced objectives) and uses Pareto fronts to expose the resulting trade-offs between competing experimental priorities. By using Gaussian Process regression to model the aberration landscape probabilistically, our workflow actively selects the most informative lens settings to evaluate next, rather than performing an exhaustive blind search. We demonstrate that this active learning loop is more robust than traditional optimization algorithms and effectively tunes focus, astigmatism, and higher-order aberrations. By balancing competing objectives, this approach enables "self-optimizing" microscopy by dynamically sustaining optimal performance during experiments. --- ### Pay Attention to Where You Looked Authors: Alex Berian, JhihYang Wu, Daniel Brignac, Natnael Daba, Abhijit Mahalanobis Published: 2026-01-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.18970v2 Novel view synthesis (NVS) has advanced with generative modeling, enabling photorealistic image generation. In few-shot NVS, where only a few input views are available, existing methods often assume equal importance for all input views relative to the target, leading to suboptimal results. We address this limitation by introducing a camera-weighting mechanism that adjusts the importance of source views based on their relevance to the target. We propose two approaches: a deterministic weighting scheme leveraging geometric properties like Euclidean distance and angular differences, and a cross-attention-based learning scheme that optimizes view weighting. Additionally, models can be further trained with our camera-weighting scheme to refine their understanding of view relevance and enhance synthesis quality. This mechanism is adaptable and can be integrated into various NVS algorithms, improving their ability to synthesize high-quality novel views. Our results demonstrate that adaptive view weighting enhances accuracy and realism, offering a promising direction for improving NVS. --- ### Fauna Sprout: A lightweight, approachable, developer-ready humanoid robot Authors: Fauna Robotics, :, Diego Aldarondo, Ana Pervan, Daniel Corbalan, Dave Petrillo, Bolun Dai, Aadhithya Iyer, Nina Mortensen, Erik Pearson, Sridhar Pandian Arunachalam, Emma Reznick et al. Published: 2026-01-26 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2601.18963v1 Recent advances in learned control, large-scale simulation, and generative models have accelerated progress toward general-purpose robotic controllers, yet the field still lacks platforms suitable for safe, expressive, long-term deployment in human environments. Most existing humanoids are either closed industrial systems or academic prototypes that are difficult to deploy and operate around people, limiting progress in robotics. We introduce Sprout, a developer platform designed to address these limitations through an emphasis on safety, expressivity, and developer accessibility. Sprout adopts a lightweight form factor with compliant control, limited joint torques, and soft exteriors to support safe operation in shared human spaces. The platform integrates whole-body control, manipulation with integrated grippers, and virtual-reality-based teleoperation within a unified hardware-software stack. An expressive head further enables social interaction -- a domain that remains underexplored on most utilitarian humanoids. By lowering physical and technical barriers to deployment, Sprout expands access to capable humanoid platforms and provides a practical basis for developing embodied intelligence in real human environments. --- ### Reinforcement Learning for Quantum Technology Authors: Marin Bukov, Florian Marquardt Published: 2026-01-26 Categories: quant-ph, cond-mat.quant-gas, cond-mat.stat-mech, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.18953v1 Many challenges arising in Quantum Technology can be successfully addressed using a set of machine learning algorithms collectively known as reinforcement learning (RL), based on adaptive decision-making through interaction with the quantum device. After a concise and intuitive introduction to RL aimed at a broad physics readership, we discuss the key ideas and core concepts in reinforcement learning with a particular focus on quantum systems. We then survey recent progress in RL in all relevant areas. We discuss state preparation in few- and many-body quantum systems, the design and optimization of high-fidelity quantum gates, and the automated construction of quantum circuits, including applications to variational quantum eigensolvers and architecture search. We further highlight the interactive capabilities of RL agents, emphasizing recent progress in quantum feedback control and quantum error correction, and briefly discuss quantum reinforcement learning as well as applications to quantum metrology. The review concludes with a discussion of open challenges -- such as scalability, interpretability, and integration with experimental platforms -- and outlines promising directions for future research. Throughout, we highlight experimental implementations that exemplify the increasing role of reinforcement learning in shaping the development of quantum technologies. --- ### Vector-Valued Distributional Reinforcement Learning Policy Evaluation: A Hilbert Space Embedding Approach Authors: Mehrdad Mohammadi, Qi Zheng, Ruoqing Zhu Published: 2026-01-26 Categories: cs.LG, stat.ME, stat.ML arXiv: https://arxiv.org/abs/2601.18952v3 We propose Kernel Embedding Distributional Reinforcement Learning (KE-DRL) an offline method to estimate kernel mean embeddings of conditional multivariate return distributions from observed offline trajectories with continuous state-action inputs and vector-valued rewards. KE-DRL approximates the conditional embedding and estimates its a finite-dictionary representation coefficients through a maximum mean discrepancy Bellman criterion. For a regular class of return distributions, we show that a Matérn embedding and a Bellman-invariant Sobolev-moment class identifies the unique distributional Bellman fixed point and and derive a Hölder-type bound that controls Wasserstein error by the population embedding Bellman residual. For directly observed responses, we provide finite-sample and uniform error bounds for a regularized conditional mean embedding estimator. Simulation experiments evaluate pointwise embedding recovery against Monte Carlo benchmarks across multiple behavior-target policy pairs. An application to Expedia hotel-search data illustrates conditional multivariate return evaluation under two policies. --- ### Collaborative Compressors in Distributed Mean Estimation with Limited Communication Budget Authors: Harsh Vardhan, Arya Mazumdar Published: 2026-01-26 Categories: stat.ML, cs.IT, cs.LG arXiv: https://arxiv.org/abs/2601.18950v1 Distributed high dimensional mean estimation is a common aggregation routine used often in distributed optimization methods. Most of these applications call for a communication-constrained setting where vectors, whose mean is to be estimated, have to be compressed before sharing. One could independently encode and decode these to achieve compression, but that overlooks the fact that these vectors are often close to each other. To exploit these similarities, recently Suresh et al., 2022, Jhunjhunwala et al., 2021, Jiang et al, 2023, proposed multiple correlation-aware compression schemes. However, in most cases, the correlations have to be known for these schemes to work. Moreover, a theoretical analysis of graceful degradation of these correlation-aware compression schemes with increasing dissimilarity is limited to only the $\ell_2$-error in the literature. In this paper, we propose four different collaborative compression schemes that agnostically exploit the similarities among vectors in a distributed setting. Our schemes are all simple to implement and computationally efficient, while resulting in big savings in communication. The analysis of our proposed schemes show how the $\ell_2$, $\ell_\infty$ and cosine estimation error varies with the degree of similarity among vectors. --- ### Tricky$^2$: Towards a Benchmark for Evaluating Human and LLM Error Interactions Authors: Cole Granger, Dipin Khati, Daniel Rodriguez-Cardenas, Denys Poshyvanyk Published: 2026-01-26 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2601.18949v1 Large language models (LLMs) are increasingly integrated into software development workflows, yet they often introduce subtle logic or data-misuse errors that differ from human bugs. To study how these two error types interact, we construct Tricky$^2$, a hybrid dataset that augments the existing TrickyBugs corpus of human-written defects with errors injected by both GPT-5 and OpenAI-oss-20b across C++, Python, and Java programs. Our approach uses a taxonomy-guided prompting framework to generate machine-originated bugs while preserving original human defects and program structure. The resulting corpus spans human-only, LLM-only, and human+LLM splits, enabling analysis of mixed-origin error behavior, multi-bug repair robustness, and reliability in hybrid human-machine code. This paper outlines the dataset construction pipeline and illustrates its use through small-scale baseline evaluations of classification, localization, and repair tasks. --- ### Smart Split-Federated Learning over Noisy Channels for Embryo Image Segmentation Authors: Zahra Hafezi Kafshgari, Ivan V. Bajic, Parvaneh Saeedi Published: 2026-01-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.18948v1 Split-Federated (SplitFed) learning is an extension of federated learning that places minimal requirements on the clients computing infrastructure, since only a small portion of the overall model is deployed on the clients hardware. In SplitFed learning, feature values, gradient updates, and model updates are transferred across communication channels. In this paper, we study the effects of noise in the communication channels on the learning process and the quality of the final model. We propose a smart averaging strategy for SplitFed learning with the goal of improving resilience against channel noise. Experiments on a segmentation model for embryo images shows that the proposed smart averaging strategy is able to tolerate two orders of magnitude stronger noise in the communication channels compared to conventional averaging, while still maintaining the accuracy of the final model. --- ### Neural Theorem Proving for Verification Conditions: A Real-World Benchmark Authors: Qiyuan Xu, Xiaokun Luan, Renxi Wang, Joshua Ong Jun Leang, Peixin Wang, Haonan Li, Wenda Li, Conrad Watt Published: 2026-01-26 Categories: cs.AI, cs.PL, cs.SE arXiv: https://arxiv.org/abs/2601.18944v2 Theorem proving is fundamental to program verification, where the automated proof of Verification Conditions (VCs) remains a primary bottleneck. Real-world program verification frequently encounters hard VCs that existing Automated Theorem Provers (ATPs) cannot prove, leading to a critical need for extensive manual proofs that burden practical application. While Neural Theorem Proving (NTP) has achieved significant success in mathematical competitions, demonstrating the potential of machine learning approaches to formal reasoning, its application to program verification--particularly VC proving--remains largely unexplored. Despite existing work on annotation synthesis and verification-related theorem proving, no benchmark has specifically targeted this fundamental bottleneck: automated VC proving. This work introduces Neural Theorem Proving for Verification Conditions (NTP4VC), presenting the first real-world multi-language benchmark for this task. From real-world projects such as Linux and Contiki-OS kernel, our benchmark leverages industrial pipelines (Why3 and Frama-C) to generate semantically equivalent test cases across formal languages of Isabelle, Lean, and Rocq. We evaluate large language models (LLMs), both general-purpose and those fine-tuned for theorem proving, on NTP4VC. Results indicate that although LLMs show promise in VC proving, significant challenges remain for program verification, highlighting a large gap and opportunity for future research. --- ### Configurable p-Neurons Using Modular p-Bits Authors: Saleh Bunaiyan, Mohammad Alsharif, Abdelrahman S. Abdelrahman, Hesham ElSawy, Suraj S. Cheema, Suhaib A. Fahmy, Kerem Y. Camsari, Feras Al-Dirini Published: 2026-01-26 Categories: cs.ET, cond-mat.dis-nn, cs.AI, cs.AR, cs.LG arXiv: https://arxiv.org/abs/2601.18943v1 Probabilistic bits (p-bits) have recently been employed in neural networks (NNs) as stochastic neurons with sigmoidal probabilistic activation functions. Nonetheless, there remain a wealth of other probabilistic activation functions that are yet to be explored. Here we re-engineer the p-bit by decoupling its stochastic signal path from its input data path, giving rise to a modular p-bit that enables the realization of probabilistic neurons (p-neurons) with a range of configurable probabilistic activation functions, including a probabilistic version of the widely used Logistic Sigmoid, Tanh and Rectified Linear Unit (ReLU) activation functions. We present spintronic (CMOS + sMTJ) designs that show wide and tunable probabilistic ranges of operation. Finally, we experimentally implement digital-CMOS versions on an FPGA, with stochastic unit sharing, and demonstrate an order of magnitude (10x) saving in required hardware resources compared to conventional digital p-bit implementations. --- ### A Few Bad Neurons: Isolating and Surgically Correcting Sycophancy Authors: Claire O'Brien, Jessica Seto, Dristi Roy, Aditya Dwivedi, Sunishchal Dev, Kevin Zhu, Sean O'Brien, Ashwinee Panda, Ryan Lagasse Published: 2026-01-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.18939v1 Behavioral alignment in large language models (LLMs) is often achieved through broad fine-tuning, which can result in undesired side effects like distributional shift and low interpretability. We propose a method for alignment that identifies and updates only the neurons most responsible for a given behavior, a targeted approach that allows for fine-tuning with significantly less data. Using sparse autoencoders (SAEs) and linear probes, we isolate the 3% of MLP neurons most predictive of a target behavior, decode them into residual space, and fine-tune only those neurons using gradient masking. We demonstrate this approach on the task of reducing sycophantic behavior, where our method matches or exceeds state-of-the-art performance on four benchmarks (Syco-Bench, NLP, POLI, PHIL) using Gemma-2-2B and 9B models. Our results show that sparse, neuron-level updates offer a scalable and precise alternative to full-model fine-tuning, remaining effective even in situations when little data is available --- ### FSD-CAP: Fractional Subgraph Diffusion with Class-Aware Propagation for Graph Feature Imputation Authors: Xin Qiao, Shijie Sun, Anqi Dong, Cong Hua, Xia Zhao, Longfei Zhang, Guangming Zhu, Liang Zhang Published: 2026-01-26 Categories: cs.LG, cs.SI, stat.ML arXiv: https://arxiv.org/abs/2601.18938v1 Imputing missing node features in graphs is challenging, particularly under high missing rates. Existing methods based on latent representations or global diffusion often fail to produce reliable estimates, and may propagate errors across the graph. We propose FSD-CAP, a two-stage framework designed to improve imputation quality under extreme sparsity. In the first stage, a graph-distance-guided subgraph expansion localizes the diffusion process. A fractional diffusion operator adjusts propagation sharpness based on local structure. In the second stage, imputed features are refined using class-aware propagation, which incorporates pseudo-labels and neighborhood entropy to promote consistency. We evaluated FSD-CAP on multiple datasets. With $99.5\%$ of features missing across five benchmark datasets, FSD-CAP achieves average accuracies of $80.06\%$ (structural) and $81.01\%$ (uniform) in node classification, close to the $81.31\%$ achieved by a standard GCN with full features. For link prediction under the same setting, it reaches AUC scores of $91.65\%$ (structural) and $92.41\%$ (uniform), compared to $95.06\%$ for the fully observed case. Furthermore, FSD-CAP demonstrates superior performance on both large-scale and heterophily datasets when compared to other models. --- ### Bi-Level Online Provisioning and Scheduling with Switching Costs and Cross-Level Constraints Authors: Jialei Liu, C. Emre Koksal, Ming Shi Published: 2026-01-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.18936v2 We study a bi-level online provisioning and scheduling problem motivated by network resource allocation, where provisioning decisions are made at a slow time scale while queue-/state-dependent scheduling is performed at a fast time scale. We model this two-time-scale interaction using an upper-level online convex optimization (OCO) problem and a lower-level constrained Markov decision process (CMDP). Existing OCO typically assumes stateless decisions and thus cannot capture MDP network dynamics such as queue evolution. Meanwhile, CMDP algorithms typically assume a fixed constraint threshold, whereas in provisioning-and-scheduling systems, the threshold varies with online budget decisions. To address these gaps, we study bi-level OCO-CMDP learning under switching costs (budget reprovisioning/system reconfiguration) and cross-level constraints that couple budgets to scheduling decisions. Our new algorithm solves this learning problem via several non-trivial developments, including a carefully designed dual feedback that returns the budget multiplier as sensitivity information for the upper-level update and a lower level that solves a budget-adaptive safe exploration problem via an extended occupancy-measure linear program. We establish near-optimal regret and high-probability satisfaction of the cross-level constraints. --- ### BabyReasoningBench: Generating Developmentally-Inspired Reasoning Tasks for Evaluating Baby Language Models Authors: Kaustubh D. Dhole Published: 2026-01-26 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.18933v2 Traditional evaluations of reasoning capabilities of language models are dominated by adult-centric benchmarks that presuppose broad world knowledge, complex instruction following, and mature pragmatic competence. These assumptions are mismatched to baby language models trained on developmentally plausible input such as child-directed speech and early-childhood narratives, and they obscure which reasoning abilities (if any) emerge under such constraints. We introduce BabyReasoningBench, a GPT-5.2 generated benchmark of 19 reasoning tasks grounded in classic paradigms from developmental psychology, spanning theory of mind, analogical and relational reasoning, causal inference and intervention selection, and core reasoning primitives that are known to be confounded by memory and pragmatics. We find that two GPT-2 based baby language models (pretrained on 10M and 100M of child-directed speech text) show overall low but uneven performance, with dissociations across task families: scaling improves several causal and physical reasoning tasks, while belief attribution and pragmatics-sensitive tasks remain challenging. BabyReasoningBench provides a developmentally grounded lens for analyzing what kinds of reasoning are supported by child-like training distributions, and for testing mechanistic hypotheses about how such abilities emerge. --- ### Advances in Diffusion-Based Generative Compression Authors: Yibo Yang, Stephan Mandt Published: 2026-01-26 Categories: eess.IV, cs.IT, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2601.18932v2 Popularized by their strong image generation performance, diffusion and related methods for generative modeling have found widespread success in visual media applications. In particular, diffusion methods have enabled new approaches to data compression, where realistic reconstructions can be generated at extremely low bit-rates. This article provides a unifying review of recent diffusion-based methods for generative lossy compression, with a focus on image compression. These methods generally encode the source into an embedding and use a diffusion model to iteratively refine it during decoding, so that the reconstruction approximately follows the true data distribution. The embedding can take various forms and is typically transmitted via an auxiliary entropy model, and recent methods also explore the use of diffusion models themselves for information transmission via channel simulation. We review representative approaches through the lens of rate-distortion-perception theory, highlighting the role of common randomness and connections to inverse problems, and identify open challenges. --- ### Toward Learning POMDPs Beyond Full-Rank Actions and State Observability Authors: Seiji Shaw, Travis Manderson, Chad Kessens, Nicholas Roy Published: 2026-01-26 Categories: cs.LG, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2601.18930v5 We are interested in enabling autonomous agents to learn and reason about systems with hidden states, such as locking mechanisms. We cast this problem as learning the parameters of a discrete Partially Observable Markov Decision Process (POMDP). The agent begins with knowledge of the POMDP's actions and observation spaces, but not its state space, transitions, or observation models. These properties must be constructed from a sequence of actions and observations. Spectral approaches to learning models of partially observable domains, such as Predictive State Representations (PSRs), learn representations of state that are sufficient to predict future outcomes. PSR models, however, do not have explicit transition and observation system models that can be used with different reward functions to solve different planning problems. Under a mild set of rankness assumptions on the products of transition and observation matrices, we show how PSRs learn POMDP matrices up to a similarity transform, and this transform may be estimated via tensor decomposition methods. Our method learns observation matrices and transition matrices up to a partition of states, where the states in a single partition have the same observation distributions corresponding to actions whose transition matrices are full-rank. Our numerical experiments suggest that explicit observation and transition likelihoods can be leveraged to generate new plans for different goals and reward functions after the model has been learned. We also show that learning a POMDP beyond a partition of states is impossible from sequential data by constructing two POMDPs that agree on all observation distributions but differ in their transition dynamics. --- ### On the Role of Depth in Surgical Vision Foundation Models: An Empirical Study of RGB-D Pre-training Authors: John J. Han, Adam Schmidt, Muhammad Abdullah Jamal, Chinedu Nwoye, Anita Rau, Jie Ying Wu, Omid Mohareri Published: 2026-01-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.18929v2 Vision foundation models (VFMs) have emerged as powerful tools for surgical scene understanding. However, current approaches predominantly rely on unimodal RGB pre-training, overlooking the complex 3D geometry inherent to surgical environments. Although several architectures support multimodal or geometry-aware inputs in general computer vision, the benefits of incorporating depth information in surgical settings remain underexplored. We conduct a large-scale empirical study comparing eight ViT-based VFMs that differ in pre-training domain, learning objective, and input modality (RGB vs. RGB-D). For pre-training, we use a curated dataset of 1.4 million robotic surgical images paired with depth maps generated from an off-the-shelf network. We evaluate these models under both frozen-backbone and end-to-end fine-tuning protocols across eight surgical datasets spanning object detection, segmentation, depth estimation, and pose estimation. Our experiments yield several consistent findings. Models incorporating explicit geometric tokenization, such as MultiMAE, substantially outperform unimodal baselines across all tasks. Notably, geometric-aware pre-training enables remarkable data efficiency: models fine-tuned on just 25% of labeled data consistently surpass RGB-only models trained on the full dataset. Importantly, these gains require no architectural or runtime changes at inference; depth is used only during pre-training, making adoption straightforward. These findings suggest that multimodal pre-training offers a viable path towards building more capable surgical vision systems. --- ### RIFT: Reordered Instruction Following Testbed To Evaluate Instruction Following in Singular Multistep Prompt Structures Authors: Andrew Jaffe, Noah Reicin, Jinho D. Choi Published: 2026-01-26 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.18924v1 Large Language Models (LLMs) are increasingly relied upon for complex workflows, yet their ability to maintain flow of instructions remains underexplored. Existing benchmarks conflate task complexity with structural ordering, making it difficult to isolate the impact of prompt topology on performance. We introduce RIFT, Reordered Instruction Following Testbed, to assess instruction following by disentangling structure from content. Using rephrased Jeopardy! question-answer pairs, we test LLMs across two prompt structures: linear prompts, which progress sequentially, and jumping prompts, which preserve identical content but require non-sequential traversal. Across 10,000 evaluations spanning six state-of-the-art open-source LLMs, accuracy dropped by up to 72% under jumping conditions (compared to baseline), revealing a strong dependence on positional continuity. Error analysis shows that approximately 50% of failures stem from instruction-order violations and semantic drift, indicating that current architectures internalize instruction following as a sequential pattern rather than a reasoning skill. These results reveal structural sensitivity as a fundamental limitation in current architectures, with direct implications for applications requiring non-sequential control flow such as workflow automation and multi-agent systems. --- ### DeFM: Learning Foundation Representations from Depth for Robotics Authors: Manthan Patel, Jonas Frey, Mayank Mittal, Fan Yang, Alexander Hansson, Amir Bar, Cesar Cadena, Marco Hutter Published: 2026-01-26 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2601.18923v1 Depth sensors are widely deployed across robotic platforms, and advances in fast, high-fidelity depth simulation have enabled robotic policies trained on depth observations to achieve robust sim-to-real transfer for a wide range of tasks. Despite this, representation learning for depth modality remains underexplored compared to RGB, where large-scale foundation models now define the state of the art. To address this gap, we present DeFM, a self-supervised foundation model trained entirely on depth images for robotic applications. Using a DINO-style self-distillation objective on a curated dataset of 60M depth images, DeFM learns geometric and semantic representations that generalize to diverse environments, tasks, and sensors. To retain metric awareness across multiple scales, we introduce a novel input normalization strategy. We further distill DeFM into compact models suitable for resource-constrained robotic systems. When evaluated on depth-based classification, segmentation, navigation, locomotion, and manipulation benchmarks, DeFM achieves state-of-the-art performance and demonstrates strong generalization from simulation to real-world environments. We release all our pretrained models, which can be adopted off-the-shelf for depth-based robotic learning without task-specific fine-tuning. Webpage: https://de-fm.github.io/ --- ### Accelerating Large-Scale Cheminformatics Using a Byte-Offset Indexing Architecture for Terabyte-Scale Data Integration Authors: Malikussaid, Septian Caesar Floresko, Sutiyo Published: 2026-01-26 Categories: cs.DB, cs.CE, cs.LG, q-bio.QM arXiv: https://arxiv.org/abs/2601.18921v2 The integration of large-scale chemical databases represents a critical bottleneck in modern cheminformatics research, particularly for machine learning applications requiring high-quality, multi-source validated datasets. This paper presents a case study of integrating three major public chemical repositories: PubChem (176 million compounds), ChEMBL, and eMolecules, to construct a curated dataset for molecular property prediction. We investigate whether byte-offset indexing can practically overcome brute-force scalability limits while preserving data integrity at hundred-million scale. Our results document the progression from an intractable brute-force search algorithm with projected 100-day runtime to a byte-offset indexing architecture achieving 3.2-hour completion - a 740-fold performance improvement through algorithmic complexity reduction from $O(N \times M)$ to $O(N + M)$. Systematic validation of 176 million database entries revealed hash collisions in InChIKey molecular identifiers, necessitating pipeline reconstruction using collision-free full InChI strings. We present performance benchmarks, quantify trade-offs between storage overhead and scientific rigor, and compare our approach with alternative large-scale integration strategies. The resulting system successfully extracted 435,413 validated compounds and demonstrates generalizable principles for large-scale scientific data integration where uniqueness constraints exceed hash-based identifier capabilities. --- ### One Global Model, Many Behaviors: Stockout-Aware Feature Engineering and Dynamic Scaling for Multi-Horizon Retail Demand Forecasting with a Cost-Aware Ordering Policy (VN2 Winner Report) Authors: Bartosz Szabłowski Published: 2026-01-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.18919v1 Inventory planning for retail chains requires translating demand forecasts into ordering decisions, including asymmetric shortages and holding costs. The VN2 Inventory Planning Challenge formalizes this setting as a weekly decision-making cycle with a two-week product delivery lead time, where the total cost is defined as the shortage cost plus the holding cost. This report presents the winning VN2 solution: a two-stage predict-then-optimize pipeline that combines a single global multi-horizon forecasting model with a cost-aware ordering policy. The forecasting model is trained in a global paradigm, jointly using all available time series. A gradient-boosted decision tree (GBDT) model implemented in CatBoost is used as the base learner. The model incorporates stockout-aware feature engineering to address censored demand during out-of-stock periods, per-series scaling to focus learning on time-series patterns rather than absolute levels, and time-based observation weights to reflect shifts in demand patterns. In the decision stage, inventory is projected to the start of the delivery week, and a target stock level is calculated that explicitly trades off shortage and holding costs. Evaluated by the official competition simulation in six rounds, the solution achieved first place by combining a strong global forecasting model with a lightweight cost-aware policy. Although developed for the VN2 setting, the proposed approach can be extended to real-world applications and additional operational constraints. --- ### GraIP: A Benchmarking Framework For Neural Graph Inverse Problems Authors: Semih Cantürk, Andrei Manolache, Arman Mielke, Chendi Qian, Antoine Siraudin, Christopher Morris, Mathias Niepert, Guy Wolf Published: 2026-01-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.18917v1 A wide range of graph learning tasks, such as structure discovery, temporal graph analysis, and combinatorial optimization, focus on inferring graph structures from data, rather than making predictions on given graphs. However, the respective methods to solve such problems are often developed in an isolated, task-specific manner and thus lack a unifying theoretical foundation. Here, we provide a stepping stone towards the formation of such a foundation and further development by introducing the Neural Graph Inverse Problem (GraIP) conceptual framework, which formalizes and reframes a broad class of graph learning tasks as inverse problems. Unlike discriminative approaches that directly predict target variables from given graph inputs, the GraIP paradigm addresses inverse problems, i.e., it relies on observational data and aims to recover the underlying graph structure by reversing the forward process, such as message passing or network dynamics, that produced the observed outputs. We demonstrate the versatility of GraIP across various graph learning tasks, including rewiring, causal discovery, and neural relational inference. We also propose benchmark datasets and metrics for each GraIP domain considered, and characterize and empirically evaluate existing baseline methods used to solve them. Overall, our unifying perspective bridges seemingly disparate applications and provides a principled approach to structural learning in constrained and combinatorial settings while encouraging cross-pollination of existing methods across graph inverse problems. --- ### ASEHybrid: When Geometry Matters Beyond Homophily in Graph Neural Networks Authors: Shalima Binta Manir, Tim Oates Published: 2026-01-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.18912v1 Standard message-passing graph neural networks (GNNs) often struggle on graphs with low homophily, yet homophily alone does not explain this behavior, as graphs with similar homophily levels can exhibit markedly different performance and some heterophilous graphs remain easy for vanilla GCNs. Recent work suggests that label informativeness (LI), the mutual information between labels of adjacent nodes, provides a more faithful characterization of when graph structure is useful. In this work, we develop a unified theoretical framework that connects curvature-guided rewiring and positional geometry through the lens of label informativeness, and instantiate it in a practical geometry-aware architecture, ASEHybrid. Our analysis provides a necessary-and-sufficient characterization of when geometry-aware GNNs can improve over feature-only baselines: such gains are possible if and only if graph structure carries label-relevant information beyond node features. Theoretically, we relate adjusted homophily and label informativeness to the spectral behavior of label signals under Laplacian smoothing, show that degree-based Forman curvature does not increase expressivity beyond the one-dimensional Weisfeiler--Lehman test but instead reshapes information flow, and establish convergence and Lipschitz stability guarantees for a curvature-guided rewiring process. Empirically, we instantiate ASEHybrid using Forman curvature and Laplacian positional encodings and conduct controlled ablations on Chameleon, Squirrel, Texas, Tolokers, and Minesweeper, observing gains precisely on label-informative heterophilous benchmarks where graph structure provides label-relevant information beyond node features, and no meaningful improvement in high-baseline regimes. --- ### How Is Uncertainty Propagated in Knowledge Distillation? Authors: Ziyao Cui, Jian Pei Published: 2026-01-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.18909v1 Knowledge distillation transfers behavior from a teacher to a student model, but the process is inherently stochastic: teacher outputs, student training, and student inference can all be random. Collapsing these uncertainties to a single point estimate can distort what is learned. We systematically study how uncertainty propagates through knowledge distillation across three representative model classes--linear regression, feed-forward neural networks, and large language models (LLMs)--and propose simple corrections. We distinguish inter-student uncertainty (variance across independently distilled students) from intra-student uncertainty (variance of a single student's predictive distribution), showing that standard single-response knowledge distillation suppresses intra-student variance while leaving substantial inter-student variability. To address these mismatches, we introduce two variance-aware strategies: averaging multiple teacher responses, which reduces noise at rate $O(1/k)$, and variance-weighting, which combines teacher and student estimates via inverse-variance weighting to yield a minimum-variance estimator. We provide formal guarantees in linear regression, validate the methods in neural networks, and demonstrate empirical gains in LLM distillation, including reduced systematic noise and hallucination. These results reframe knowledge distillation as an uncertainty transformation and show that variance-aware distillation produces more stable students that better reflect teacher uncertainty. --- ### Enhancing Speech Emotion Recognition using Dynamic Spectral Features and Kalman Smoothing Authors: Marouane El Hizabri, Abdelfattah Bezzaz, Ismail Hayoukane, Youssef Taki Published: 2026-01-26 Categories: cs.SD, cs.LG, eess.AS arXiv: https://arxiv.org/abs/2601.18908v1 Speech Emotion Recognition systems often use static features like Mel-Frequency Cepstral Coefficients (MFCCs), Zero Crossing Rate (ZCR), and Root Mean Square Energy (RMSE). Because of this, they can misclassify emotions when there is acoustic noise in vocal signals. To address this, we added dynamic features using Dynamic Spectral features (Deltas and Delta-Deltas) along with the Kalman Smoothing algorithm. This approach reduces noise and improves emotion classification. Since emotion changes over time, the Kalman Smoothing filter also helped make the classifier outputs more stable. Tests on the RAVDESS dataset showed that this method achieved a state-of-the-art accuracy of 87\% and reduced misclassification between emotions with similar acoustic features --- ### Implicit Q-Learning and SARSA: Liberating Policy Control from Step-Size Calibration Authors: Hwanwoo Kim, Eric Laber Published: 2026-01-26 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2601.18907v1 Q-learning and SARSA are foundational reinforcement learning algorithms whose practical success depends critically on step-size calibration. Step-sizes that are too large can cause numerical instability, while step-sizes that are too small can lead to slow progress. We propose implicit variants of Q-learning and SARSA that reformulate their iterative updates as fixed-point equations. This yields an adaptive step-size adjustment that scales inversely with feature norms, providing automatic regularization without manual tuning. Our non-asymptotic analyses demonstrate that implicit methods maintain stability over significantly broader step-size ranges. Under favorable conditions, it permits arbitrarily large step-sizes while achieving comparable convergence rates. Empirical validation across benchmark environments spanning discrete and continuous state spaces shows that implicit Q-learning and SARSA exhibit substantially reduced sensitivity to step-size selection, achieving stable performance with step-sizes that would cause standard methods to fail. --- ### MetaSICL: Globalizing Auditory LLMs for Underserved Speakers and Languages via Meta Speech In-Context Learning Authors: Haolong Zheng, Siyin Wang, Zengrui Jin, Mark Hasegawa-Johnson Published: 2026-01-26 Categories: cs.SD, cs.AI, cs.CL, eess.AS arXiv: https://arxiv.org/abs/2601.18904v3 Generative AI for speech and audio is increasingly expected to serve users across languages, cultures, and communities, yet current auditory Large Language Models (LLMs) are still largely trained and evaluated on high-resource data. Globalizing such systems requires handling low-resource settings, where the target speakers, languages, or tasks are poorly represented in training data. In these regimes, collecting enough labeled in-domain data is often impractical, and the small corpora available may still under-represent the test distribution, making direct fine-tuning brittle under domain shift. In-Context Learning (ICL) offers an alternative: instead of updating model parameters for every underserved community, an auditory LLM can adapt at inference time by conditioning on a few local demonstrations. However, vanilla speech ICL remains limited because most auditory LLMs are not explicitly trained to use such demonstrations effectively. We address this gap with Meta Speech In-Context Learning (MetaSICL), a post-training recipe that strengthens an auditory LLM's in-context adaptation ability using only abundant high-resource speech data. Although MetaSICL never trains on the target low-resource domains, it improves performance across two backbones on children's ASR, audio understanding/reasoning, and speech translation and ASR in directions and languages unseen in post-training. We further study the case where some in-domain data is available, using low-resource language ASR as a case study, since recognition for underserved languages is central to globalizing generative AI. Here, using MetaSICL as a warmup for in-domain reinforcement learning yields the strongest results, outperforming direct fine-tuning across five typologically diverse languages. Overall, MetaSICL offers a practical route toward globalizing auditory LLMs by building inference-time adaptation into the model. --- ### Flatter Tokens are More Valuable for Speculative Draft Model Training Authors: Jiaming Fan, Daming Cao, Xiangzhong Luo, Jiale Fu, Chonghan Liu, Xu Yang Published: 2026-01-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.18902v2 Speculative Decoding (SD) is a key technique for accelerating Large Language Model (LLM) inference, but it typically requires training a draft model on a large dataset. We approach this problem from a data-centric perspective, finding that not all training samples contribute equally to the SD acceptance rate. Specifically, our theoretical analysis and empirical validation reveals that tokens inducing flatter predictive distributions from the target model are more valuable than those yielding sharply peaked distributions. Based on this insight, we propose flatness, a new metric to quantify this property, and develop the Sample-level-flatness-based Dataset Distillation (SFDD) approach, which filters the training data to retain only the most valuable samples. Experiments on the EAGLE framework demonstrate that SFDD can achieve over 2$\times$ training speedup using only 50% of the data, while keeping the final model's inference speedup within 4% of the full-dataset baseline. This work introduces an effective, data-centric approach that substantially improves the training efficiency for Speculative Decoding. Our code is available at https://github.com/fjm9933/Flatness. --- ### Self-Aware Knowledge Probing: Evaluating Language Models' Relational Knowledge through Confidence Calibration Authors: Christopher Kissling, Elena Merdjanovska, Alan Akbik Published: 2026-01-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.18901v1 Knowledge probing quantifies how much relational knowledge a language model (LM) has acquired during pre-training. Existing knowledge probes evaluate model capabilities through metrics like prediction accuracy and precision. Such evaluations fail to account for the model's reliability, reflected in the calibration of its confidence scores. In this paper, we propose a novel calibration probing framework for relational knowledge, covering three modalities of model confidence: (1) intrinsic confidence, (2) structural consistency and (3) semantic grounding. Our extensive analysis of ten causal and six masked language models reveals that most models, especially those pre-trained with the masking objective, are overconfident. The best-calibrated scores come from confidence estimates that account for inconsistencies due to statement rephrasing. Moreover, even the largest pre-trained models fail to encode the semantics of linguistic confidence expressions accurately. --- ### RealStats: A Rigorous Real-Only Statistical Framework for Fake Image Detection Authors: Haim Zisman, Uri Shaham Published: 2026-01-26 Categories: cs.CV, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2601.18900v1 As generative models continue to evolve, detecting AI-generated images remains a critical challenge. While effective detection methods exist, they often lack formal interpretability and may rely on implicit assumptions about fake content, potentially limiting robustness to distributional shifts. In this work, we introduce a rigorous, statistically grounded framework for fake image detection that focuses on producing a probability score interpretable with respect to the real-image population. Our method leverages the strengths of multiple existing detectors by combining training-free statistics. We compute p-values over a range of test statistics and aggregate them using classical statistical ensembling to assess alignment with the unified real-image distribution. This framework is generic, flexible, and training-free, making it well-suited for robust fake image detection across diverse and evolving settings. --- ### Language Family Matters: Evaluating LLM-Based ASR Across Linguistic Boundaries Authors: Yuchen Zhang, Ravi Shekhar, Haralambos Mouratidis Published: 2026-01-26 Categories: cs.CL, cs.AI, cs.SD arXiv: https://arxiv.org/abs/2601.18899v3 Large Language Model (LLM)-powered Automatic Speech Recognition (ASR) systems achieve strong performance with limited resources by linking a frozen speech encoder to a pretrained LLM via a lightweight connector. Prior work trains a separate connector per language, overlooking linguistic relatedness. We propose an efficient and novel connector-sharing strategy based on linguistic family membership, enabling one connector per family, and empirically validate its effectiveness across two multilingual LLMs and two real-world corpora spanning curated and crowd-sourced speech. Our results show that family-based connectors reduce parameter count while improving generalization across domains, offering a practical and scalable strategy for multilingual ASR deployment. --- ### Explainable Uncertainty Quantification for Wastewater Treatment Energy Prediction via Interval Type-2 Neuro-Fuzzy System Authors: Qusai Khaled, Bahjat Mallak, Uzay Kaymak, Laura Genga Published: 2026-01-26 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.18897v1 Wastewater treatment plants consume 1-3% of global electricity, making accurate energy forecasting critical for operational optimization and sustainability. While machine learning models provide point predictions, they lack explainable uncertainty quantification essential for risk-aware decision-making in safety-critical infrastructure. This study develops an Interval Type-2 Adaptive Neuro-Fuzzy Inference System (IT2-ANFIS) that generates interpretable prediction intervals through fuzzy rule structures. Unlike black-box probabilistic methods, the proposed framework decomposes uncertainty across three levels: feature-level, footprint of uncertainty identify which variables introduce ambiguity, rule-level analysis reveals confidence in local models, and instance-level intervals quantify overall prediction uncertainty. Validated on Melbourne Water's Eastern Treatment Plant dataset, IT2-ANFIS achieves comparable predictive performance to first order ANFIS with substantially reduced variance across training runs, while providing explainable uncertainty estimates that link prediction confidence directly to operational conditions and input variables. --- ### Weakly supervised framework for wildlife detection and counting in challenging Arctic environments: a case study on caribou (Rangifer tarandus) Authors: Ghazaleh Serati, Samuel Foucher, Jerome Theau Published: 2026-01-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.18891v3 Caribou across the Arctic has declined in recent decades, motivating scalable and accurate monitoring approaches to guide evidence-based conservation actions and policy decisions. Manual interpretation from this imagery is labor-intensive and error-prone, underscoring the need for automatic and reliable detection across varying scenes. Yet, such automatic detection is challenging due to severe background heterogeneity, dominant empty terrain (class imbalance), small or occluded targets, and wide variation in density and scale. To make the detection model (HerdNet) more robust to these challenges, a weakly supervised patch-level pretraining based on a detection network's architecture is proposed. The detection dataset includes five caribou herds distributed across Alaska. By learning from empty vs. non-empty labels in this dataset, the approach produces early weakly supervised knowledge for enhanced detection compared to HerdNet, which is initialized from generic weights. Accordingly, the patch-based pretrain network attained high accuracy on multi-herd imagery (2017) and on an independent year's (2019) test sets (F1: 93.7%/92.6%, respectively), enabling reliable mapping of regions containing animals to facilitate manual counting on large aerial imagery. Transferred to detection, initialization from weakly supervised pretraining yielded consistent gains over ImageNet weights on both positive patches (F1: 92.6%/93.5% vs. 89.3%/88.6%), and full-image counting (F1: 95.5%/93.3% vs. 91.5%/90.4%). Remaining limitations are false positives from animal-like background clutter and false negatives related to low animal density occlusions. Overall, pretraining on coarse labels prior to detection makes it possible to rely on weakly-supervised pretrained weights even when labeled data are limited, achieving results comparable to generic-weight initialization. --- ### XProvence: Zero-Cost Multilingual Context Pruning for Retrieval-Augmented Generation Authors: Youssef Mohamed, Mohamed Elhoseiny, Thibault Formal, Nadezhda Chirkova Published: 2026-01-26 Categories: cs.IR, cs.CL arXiv: https://arxiv.org/abs/2601.18886v1 This paper introduces XProvence, a multilingual zero-cost context pruning model for retrieval-augmented generation (RAG), trained on 16 languages and supporting 100+ languages through effective cross-lingual transfer. Motivated by the growing use of RAG systems across diverse languages, we explore several strategies to generalize the Provence framework-which first integrated efficient zero-cost context pruning directly into the re-ranking model-beyond English. Across four multilingual question answering benchmarks, we show how XProvence can prune RAG contexts with minimal-to-no performance degradation and outperforms strong baselines. Our model is available at https://huggingface.co/naver/xprovence-reranker-bgem3-v2. --- ### Representational Homomorphism Predicts and Improves Compositional Generalization In Transformer Language Model Authors: Zhiyu An, Wan Du Published: 2026-01-26 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.18858v2 Compositional generalization-the ability to interpret novel combinations of familiar components-remains a persistent challenge for neural networks. Behavioral evaluations reveal \emph{when} models fail but offer limited insight into \emph{why} failures arise at the representational level. We introduce \textit{Homomorphism Error} (HE), a structural metric that measures the inconsistency between a set of established rules for which words combine to form new meaning (linguistic syntax) and model's learned rules for which hidden states combine to form new states (semantic syntax). We formulate this inconsistency as deviations from approximate homomorphisms between the linguistic expression algebra and a model's hidden-state space. We designed experiments to test if i) HE predicts compositional generalization performance, and ii) will regularizing for low HE during training improve such performance. To avoid the effect of data spoilage, we train small decoder-only Transformers from scratch using an adapted version of established dataset, SCAN, for testing compositional generalization. Across controlled experiments, HE predicts out-of-distribution (OOD) compositional generalization under noise injection, achieving $R^2=0.73$ correlation between HE and OOD accuracy. Ablations show that model depth has minimal effect on either HE or OOD accuracy, training data coverage exhibits threshold effects, and randomly inserted noise tokens increase HE. Intervention experiment shows that HE-regularized training significantly reduces HE ($p=1.1\times10^{-4}$) and yields a statistically significant improvement in OOD accuracy ($p=0.023$). Together, these results indicate the potential of HE to be both a diagnostic and an actionable training signal for improving compositional generalization. --- ### Statistical Inference for Explainable Boosting Machines Authors: Haimo Fang, Kevin Tan, Jonathan Pipping-Gamon, Giles Hooker Published: 2026-01-26 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2601.18857v2 Explainable boosting machines (EBMs) are popular "glass-box" models that learn a set of univariate functions using boosting trees. These achieve explainability through visualizations of each feature's effect. However, unlike linear model coefficients, uncertainty quantification for the learned univariate functions requires computationally intensive bootstrapping, making it hard to know which features truly matter. We provide an alternative using recent advances in statistical inference for gradient boosting, deriving methods for statistical inference as well as end-to-end theoretical guarantees. Using a moving average instead of a sum of trees (Boulevard regularization) allows the boosting process to converge to a feature-wise kernel ridge regression. This produces asymptotically normal predictions that achieve the minimax-optimal MSE for fitting Lipschitz GAMs with $p$ features of $O(p n^{-2/3})$, successfully avoiding the curse of dimensionality. We then construct prediction intervals for the response and confidence intervals for each learned univariate function with a runtime independent of the number of datapoints, enabling further explainability within EBMs. Code is available at https://github.com/hetankevin/ebm-inference. --- ### SelfieAvatar: Real-time Head Avatar reenactment from a Selfie Video Authors: Wei Liang, Hui Yu, Derui Ding, Rachael E. Jack, Philippe G. Schyns Published: 2026-01-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.18851v1 Head avatar reenactment focuses on creating animatable personal avatars from monocular videos, serving as a foundational element for applications like social signal understanding, gaming, human-machine interaction, and computer vision. Recent advances in 3D Morphable Model (3DMM)-based facial reconstruction methods have achieved remarkable high-fidelity face estimation. However, on the one hand, they struggle to capture the entire head, including non-facial regions and background details in real time, which is an essential aspect for producing realistic, high-fidelity head avatars. On the other hand, recent approaches leveraging generative adversarial networks (GANs) for head avatar generation from videos can achieve high-quality reenactments but encounter limitations in reproducing fine-grained head details, such as wrinkles and hair textures. In addition, existing methods generally rely on a large amount of training data, and rarely focus on using only a simple selfie video to achieve avatar reenactment. To address these challenges, this study introduces a method for detailed head avatar reenactment using a selfie video. The approach combines 3DMMs with a StyleGAN-based generator. A detailed reconstruction model is proposed, incorporating mixed loss functions for foreground reconstruction and avatar image generation during adversarial training to recover high-frequency details. Qualitative and quantitative evaluations on self-reenactment and cross-reenactment tasks demonstrate that the proposed method achieves superior head avatar reconstruction with rich and intricate textures compared to existing approaches. --- ### Audio-Driven Talking Face Generation with Blink Embedding and Hash Grid Landmarks Encoding Authors: Yuhui Zhang, Hui Yu, Wei Liang, Sunjie Zhang Published: 2026-01-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.18849v1 Dynamic Neural Radiance Fields (NeRF) have demonstrated considerable success in generating high-fidelity 3D models of talking portraits. Despite significant advancements in the rendering speed and generation quality, challenges persist in accurately and efficiently capturing mouth movements in talking portraits. To tackle this challenge, we propose an automatic method based on blink embedding and hash grid landmarks encoding in this study, which can substantially enhance the fidelity of talking faces. Specifically, we leverage facial features encoded as conditional features and integrate audio features as residual terms into our model through a Dynamic Landmark Transformer. Furthermore, we employ neural radiance fields to model the entire face, resulting in a lifelike face representation. Experimental evaluations have validated the superiority of our approach to existing methods. --- ### MulVul: Retrieval-augmented Multi-Agent Code Vulnerability Detection via Cross-Model Prompt Evolution Authors: Zihan Wu, Jie Xu, Yun Peng, Chun Yong Chong, Xiaohua Jia Published: 2026-01-26 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2601.18847v1 Large Language Models (LLMs) struggle to automate real-world vulnerability detection due to two key limitations: the heterogeneity of vulnerability patterns undermines the effectiveness of a single unified model, and manual prompt engineering for massive weakness categories is unscalable. To address these challenges, we propose \textbf{MulVul}, a retrieval-augmented multi-agent framework designed for precise and broad-coverage vulnerability detection. MulVul adopts a coarse-to-fine strategy: a \emph{Router} agent first predicts the top-$k$ coarse categories and then forwards the input to specialized \emph{Detector} agents, which identify the exact vulnerability types. Both agents are equipped with retrieval tools to actively source evidence from vulnerability knowledge bases to mitigate hallucinations. Crucially, to automate the generation of specialized prompts, we design \emph{Cross-Model Prompt Evolution}, a prompt optimization mechanism where a generator LLM iteratively refines candidate prompts while a distinct executor LLM validates their effectiveness. This decoupling mitigates the self-correction bias inherent in single-model optimization. Evaluated on 130 CWE types, MulVul achieves 34.79\% Macro-F1, outperforming the best baseline by 41.5\%. Ablation studies validate cross-model prompt evolution, which boosts performance by 51.6\% over manual prompts by effectively handling diverse vulnerability patterns. --- ### LLM Driven Design of Continuous Optimization Problems with Controllable High-level Properties Authors: Urban Skvorc, Niki van Stein, Moritz Seiler, Britta Grimme, Thomas Bäck, Heike Trautmann Published: 2026-01-26 Categories: cs.AI, cs.NE arXiv: https://arxiv.org/abs/2601.18846v1 Benchmarking in continuous black-box optimisation is hindered by the limited structural diversity of existing test suites such as BBOB. We explore whether large language models embedded in an evolutionary loop can be used to design optimisation problems with clearly defined high-level landscape characteristics. Using the LLaMEA framework, we guide an LLM to generate problem code from natural-language descriptions of target properties, including multimodality, separability, basin-size homogeneity, search-space homogeneity and globallocal optima contrast. Inside the loop we score candidates through ELA-based property predictors. We introduce an ELA-space fitness-sharing mechanism that increases population diversity and steers the generator away from redundant landscapes. A complementary basin-of-attraction analysis, statistical testing and visual inspection, verifies that many of the generated functions indeed exhibit the intended structural traits. In addition, a t-SNE embedding shows that they expand the BBOB instance space rather than forming an unrelated cluster. The resulting library provides a broad, interpretable, and reproducible set of benchmark problems for landscape analysis and downstream tasks such as automated algorithm selection. --- ### Dynamic Mask-Based Backdoor Attack Against Vision AI Models: A Case Study on Mushroom Detection Authors: Zeineb Dridi, Jihen Bennaceur, Amine Ben Hassouna Published: 2026-01-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.18845v1 Deep learning has revolutionized numerous tasks within the computer vision field, including image classification, image segmentation, and object detection. However, the increasing deployment of deep learning models has exposed them to various adversarial attacks, including backdoor attacks. This paper presents a novel dynamic mask-based backdoor attack method, specifically designed for object detection models. We exploit a dataset poisoning technique to embed a malicious trigger, rendering any models trained on this compromised dataset vulnerable to our backdoor attack. We particularly focus on a mushroom detection dataset to demonstrate the practical risks posed by such attacks on critical real-life domains. Our work also emphasizes the importance of creating a detailed backdoor attack scenario to illustrate the significant risks associated with the outsourcing practice. Our approach leverages SAM, a recent and powerful image segmentation AI model, to create masks for dynamic trigger placement, introducing a new and stealthy attack method. Through extensive experimentation, we show that our sophisticated attack scenario maintains high accuracy on clean data with the YOLOv7 object detection model while achieving high attack success rates on poisoned samples. Our approach surpasses traditional methods for backdoor injection, which are based on static and consistent patterns. Our findings underscore the urgent need for robust countermeasures to protect deep learning models from these evolving adversarial threats. --- ### Reducing False Positives in Static Bug Detection with LLMs: An Empirical Study in Industry Authors: Xueying Du, Jiayi Feng, Yi Zou, Wei Xu, Jie Ma, Wei Zhang, Sisi Liu, Xin Peng, Yiling Lou Published: 2026-01-26 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2601.18844v1 Static analysis tools (SATs) are widely adopted in both academia and industry for improving software quality, yet their practical use is often hindered by high false positive rates, especially in large-scale enterprise systems. These false alarms demand substantial manual inspection, creating severe inefficiencies in industrial code review. While recent work has demonstrated the potential of large language models (LLMs) for false alarm reduction on open-source benchmarks, their effectiveness in real-world enterprise settings remains unclear. To bridge this gap, we conduct the first comprehensive empirical study of diverse LLM-based false alarm reduction techniques in an industrial context at Tencent, one of the largest IT companies in China. Using data from Tencent's enterprise-customized SAT on its large-scale Advertising and Marketing Services software, we construct a dataset of 433 alarms (328 false positives, 105 true positives) covering three common bug types. Through interviewing developers and analyzing the data, our results highlight the prevalence of false positives, which wastes substantial manual effort (e.g., 10-20 minutes of manual inspection per alarm). Meanwhile, our results show the huge potential of LLMs for reducing false alarms in industrial settings (e.g., hybrid techniques of LLM and static analysis eliminate 94-98% of false positives with high recall). Furthermore, LLM-based techniques are cost-effective, with per-alarm costs as low as 2.1-109.5 seconds and $0.0011-$0.12, representing orders-of-magnitude savings compared to manual review. Finally, our case analysis further identifies key limitations of LLM-based false alarm reduction in industrial settings. ---