# Research Papers — 2026-02 · bundle page 209 of 210

> Full-text markdown of papers #10401–10450 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/209
[← Previous bundle](https://aipod.dev/papers/archive/2026-02/page/208.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-02/page/210.md)

### On the Expressive Power of Permutation-Equivariant Weight-Space Networks

Authors: Adir Dayan, Yam Eitan, Haggai Maron
Published: 2026-02-01
Categories: cs.LG
arXiv: https://arxiv.org/abs/2602.01083v2

Weight-space learning studies neural architectures that operate directly on the parameters of other neural networks. Motivated by the growing availability of pretrained models, recent work has demonstrated the effectiveness of weight-space networks across a wide range of tasks. SOTA weight-space networks rely on permutation-equivariant designs to improve generalization. However, this may negatively affect expressive power, warranting theoretical investigation. Importantly, unlike other structured domains, weight-space learning targets maps operating on both weight and function spaces, making expressivity analysis particularly subtle. While a few prior works provide partial expressivity results, a comprehensive characterization is still missing. In this work, we address this gap by developing a systematic theory for expressivity of weight-space networks. We first prove that all prominent permutation-equivariant networks are equivalent in expressive power. We then establish universality in both weight- and function-space settings under mild, natural assumptions on the input weights, and characterize the edge-case regimes where universality no longer holds. Guided by our theoretical results, we show that slight modifications to existing weight-space models yield a 34% improvement over prior SOTA, demonstrating the practical relevance of our framework.

---

### EvoOpt-LLM: Evolving industrial optimization models with large language models

Authors: Yiliu He, Tianle Li, Binghao Ji, Zhiyuan Liu, Di Huang
Published: 2026-02-01
Categories: cs.AI
arXiv: https://arxiv.org/abs/2602.01082v2

Optimization modeling via mixed-integer linear programming (MILP) is fundamental to industrial planning and scheduling, yet translating natural-language requirements into solver-executable models and maintaining them under evolving business rules remains highly expertise-intensive. While large language models (LLMs) offer promising avenues for automation, existing methods often suffer from low data efficiency, limited solver-level validity, and poor scalability to industrial-scale problems. To address these challenges, we present EvoOpt-LLM, a unified LLM-based framework supporting the full lifecycle of industrial optimization modeling, including automated model construction, dynamic business-constraint injection, and end-to-end variable pruning. Built on a 7B-parameter LLM and adapted via parameter-efficient LoRA fine-tuning, EvoOpt-LLM achieves a generation rate of 91% and an executability rate of 65.9% with only 3,000 training samples, with critical performance gains emerging under 1,500 samples. The constraint injection module reliably augments existing MILP models while preserving original objectives, and the variable pruning module enhances computational efficiency, achieving an F1 score of ~0.56 on medium-sized LP models with only 400 samples. EvoOpt-LLM demonstrates a practical, data-efficient approach to industrial optimization modeling, reducing reliance on expert intervention while improving adaptability and solver efficiency.

---

### MedAD-R1: Eliciting Consistent Reasoning in Interpretible Medical Anomaly Detection via Consistency-Reinforced Policy Optimization

Authors: Haitao Zhang, Yingying Wang, Jiaxiang Wang, Haote Xu, Hongyang Zhang, Yirong Chen, Yue Huang, Xinghao Ding
Published: 2026-02-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2602.01081v1

Medical Anomaly Detection (MedAD) presents a significant opportunity to enhance diagnostic accuracy using Large Multimodal Models (LMMs) to interpret and answer questions based on medical images. However, the reliance on Supervised Fine-Tuning (SFT) on simplistic and fragmented datasets has hindered the development of models capable of plausible reasoning and robust multimodal generalization. To overcome this, we introduce MedAD-38K, the first large-scale, multi-modal, and multi-center benchmark for MedAD featuring diagnostic Chain-of-Thought (CoT) annotations alongside structured Visual Question-Answering (VQA) pairs. On this foundation, we propose a two-stage training framework. The first stage, Cognitive Injection, uses SFT to instill foundational medical knowledge and align the model with a structured think-then-answer paradigm. Given that standard policy optimization can produce reasoning that is disconnected from the final answer, the second stage incorporates Consistency Group Relative Policy Optimization (Con-GRPO). This novel algorithm incorporates a crucial consistency reward to ensure the generated reasoning process is relevant and logically coherent with the final diagnosis. Our proposed model, MedAD-R1, achieves state-of-the-art (SOTA) performance on the MedAD-38K benchmark, outperforming strong baselines by more than 10\%. This superior performance stems from its ability to generate transparent and logically consistent reasoning pathways, offering a promising approach to enhancing the trustworthiness and interpretability of AI for clinical decision support.

---

### AutoHealth: An Uncertainty-Aware Multi-Agent System for Autonomous Health Data Modeling

Authors: Tong Xia, Weibin Li, Gang Liu, Yong Li
Published: 2026-02-01
Categories: cs.AI
arXiv: https://arxiv.org/abs/2602.01078v1

LLM-based agents have demonstrated strong potential for autonomous machine learning, yet their applicability to health data remains limited. Existing systems often struggle to generalize across heterogeneous health data modalities, rely heavily on predefined solution templates with insufficient adaptation to task-specific objectives, and largely overlook uncertainty estimation, which is essential for reliable decision-making in healthcare. To address these challenges, we propose \textit{AutoHealth}, a novel uncertainty-aware multi-agent system that autonomously models health data and assesses model reliability. \textit{AutoHealth} employs closed-loop coordination among five specialized agents to perform data exploration, task-conditioned model construction, training, and optimization, while jointly prioritizing predictive performance and uncertainty quantification. Beyond producing ready-to-use models, the system generates comprehensive reports to support trustworthy interpretation and risk-aware decision-making. To rigorously evaluate its effectiveness, we curate a challenging real-world benchmark comprising 17 tasks across diverse data modalities and learning settings. \textit{AutoHealth} completes all tasks and outperforms state-of-the-art baselines by 29.2\% in prediction performance and 50.2\% in uncertainty estimation.

---

### PISA: Piecewise Sparse Attention Is Wiser for Efficient Diffusion Transformers

Authors: Haopeng Li, Shitong Shao, Wenliang Zhong, Zikai Zhou, Lichen Bai, Hui Xiong, Zeke Xie
Published: 2026-02-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2602.01077v2

Diffusion Transformers are fundamental for video and image generation, but their efficiency is bottlenecked by the quadratic complexity of attention. While block sparse attention accelerates computation by attending only critical key-value blocks, it suffers from degradation at high sparsity by discarding context. In this work, we discover that attention scores of non-critical blocks exhibit distributional stability, allowing them to be approximated accurately and efficiently rather than discarded, which is essentially important for sparse attention design. Motivated by this key insight, we propose PISA, a training-free Piecewise Sparse Attention that covers the full attention span with sub-quadratic complexity. Unlike the conventional keep-or-drop paradigm that directly drop the non-critical block information, PISA introduces a novel exact-or-approximate strategy: it maintains exact computation for critical blocks while efficiently approximating the remainder through block-wise Taylor expansion. This design allows PISA to serve as a faithful proxy to full attention, effectively bridging the gap between speed and quality. Experimental results demonstrate that PISA achieves 1.91 times and 2.57 times speedups on Wan2.1-14B and Hunyuan-Video, respectively, while consistently maintaining the highest quality among sparse attention methods. Notably, even for image generation on FLUX, PISA achieves a 1.2 times acceleration without compromising visual quality. Code is available at: https://github.com/xie-lab-ml/piecewise-sparse-attention.

---

### ConvexBench: Can LLMs Recognize Convex Functions?

Authors: Yepeng Liu, Yu Huang, Yu-Xiang Wang, Yingbin Liang, Yuheng Bu
Published: 2026-02-01
Categories: cs.AI
arXiv: https://arxiv.org/abs/2602.01075v2

Convex analysis is a modern branch of mathematics with many applications. As Large Language Models (LLMs) start to automate research-level math and sciences, it is important for LLMs to demonstrate the ability to understand and reason with convexity. We introduce \cb, a scalable and mechanically verifiable benchmark for testing \textit{whether LLMs can identify the convexity of a symbolic objective under deep functional composition.} Experiments on frontier LLMs reveal a sharp compositional reasoning gap: performance degrades rapidly with increasing depth, dropping from an F1-score of $1.0$ at depth $2$ to approximately $0.2$ at depth $100$. Inspection of models' reasoning traces indicates two failure modes: \textit{parsing failure} and \textit{lazy reasoning}. To address these limitations, we propose an agentic divide-and-conquer framework that (i) offloads parsing to an external tool to construct an abstract syntax tree (AST) and (ii) enforces recursive reasoning over each intermediate sub-expression with focused context. This framework reliably mitigates deep-composition failures, achieving substantial performance improvement at large depths (e.g., F1-Score $= 1.0$ at depth $100$).

---

### Vortex Stretching in the Navier-Stokes Equations and Information Dissipation in Diffusion Models: A Reformulation from a Partial Differential Equation Viewpoint

Authors: Tsuyoshi Yoneda
Published: 2026-02-01
Categories: math.AP, cs.AI, physics.flu-dyn
arXiv: https://arxiv.org/abs/2602.01071v1

We present a new inverse-time formulation of vortex stretching in the Navier-Stokes equations, based on a PDE framework inspired by score-based diffusion models. By absorbing the ill-posed backward Laplacian arising from time reversal into a drift term expressed through a score function, the inverse-time dynamics are formulated in a Lagrangian manner. Using a discrete Lagrangian flow of an axisymmetric vortex-stretching field, the score function is learned with a neural network and employed to construct backward-time particle trajectories. Numerical results demonstrate that information about initial positions is rapidly lost in the compressive direction, whereas it is relatively well preserved in the stretching direction.

---

### What If We Allocate Test-Time Compute Adaptively?

Authors: Ahsan Bilal, Ahmed Mohsin, Muhammad Umer, Ali Subhan, Hassan Rizwan, Ayesha Mohsin, Dean Hougen
Published: 2026-02-01
Categories: cs.CL
arXiv: https://arxiv.org/abs/2602.01070v5

Test-time compute scaling allocates inference computation uniformly, uses fixed sampling strategies, and applies verification only for reranking. In contrast, we propose a verifier-guided adaptive framework treating reasoning as iterative trajectory generation and selection. For each problem, the agent runs multiple inference iterations. In each iteration, it optionally produces a high-level plan, selects a set of reasoning tools and a compute strategy together with an exploration parameter, and then generates a candidate reasoning trajectory. A process reward model (PRM) serves as a unified control signal: within each iteration, step-level PRM scores are aggregated to guide pruning and expansion during generation, and across iterations, aggregated trajectory rewards are used to select the final response. Across datasets, our dynamic, PRM-guided approach consistently outperforms direct test-time scaling, yielding large gains on MATH-500 and several-fold improvements on harder benchmarks such as AIME24 and AMO-Bench. We characterize efficiency using theoretical FLOPs and a compute intensity metric penalizing wasted generation and tool overhead, demonstrating that verification-guided allocation concentrates computation on high-utility reasoning paths.

---

### PDE-Constrained Optimization for Neural Image Segmentation with Physics Priors

Authors: Seema K. Poudel, Sunny K. Khadka
Published: 2026-02-01
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2602.01069v1

Segmentation of microscopy images constitutes an ill-posed inverse problem due to measurement noise, weak object boundaries, and limited labeled data. Although deep neural networks provide flexible nonparametric estimators, unconstrained empirical risk minimization often leads to unstable solutions and poor generalization. In this work, image segmentation is formulated as a PDE-constrained optimization problem that integrates physically motivated priors into deep learning models through variational regularization. The proposed framework minimizes a composite objective function consisting of a data fidelity term and penalty terms derived from reaction-diffusion equations and phase-field interface energies, all implemented as differentiable residual losses. Experiments are conducted on the LIVECell dataset, a high-quality, manually annotated collection of phase-contrast microscopy images. Training is performed on two cell types, while evaluation is carried out on a distinct, unseen cell type to assess generalization. A UNet architecture is used as the unconstrained baseline model. Experimental results demonstrate consistent improvements in segmentation accuracy and boundary fidelity compared to unconstrained deep learning baselines. Moreover, the PDE-regularized models exhibit enhanced stability and improved generalization in low-sample regimes, highlighting the advantages of incorporating structured priors. The proposed approach illustrates how PDE-constrained optimization can strengthen data-driven learning frameworks, providing a principled bridge between variational methods, statistical learning, and scientific machine learning.

---

### From Utterance to Vividity: Training Expressive Subtitle Translation LLM via Adaptive Local Preference Optimization

Authors: Chaoqun Cui, Shijing Wang, Liangbin Huang, Qingqing Gu, Zhaolong Huang, Xiao Zeng, Wenji Mao
Published: 2026-02-01
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2602.01068v1

The rapid development of Large Language Models (LLMs) has significantly enhanced the general capabilities of machine translation. However, as application scenarios become more complex, the limitations of LLMs in vertical domain translations are gradually becoming apparent. In this study, we focus on how to construct translation LLMs that meet the needs of domain customization. We take visual media subtitle translation as our topic and explore how to train expressive and vivid translation LLMs. We investigated the situations of subtitle translation and other domains of literal and liberal translation, verifying the reliability of LLM as reward model and evaluator for translation. Additionally, to train an expressive translation LLM, we constructed and released a multidirectional subtitle parallel corpus dataset and proposed the Adaptive Local Preference Optimization (ALPO) method to address fine-grained preference alignment. Experimental results demonstrate that ALPO achieves outstanding performance in multidimensional evaluation of translation quality.

---

### Exploring Knowledge Purification in Multi-Teacher Knowledge Distillation for LLMs

Authors: Ruihan Jin, Pengpeng Shao, Zhengqi Wen, Jinyang Wu, Mingkuan Feng, Shuo Yang, Chu Yuan Zhang, Jianhua Tao
Published: 2026-02-01
Categories: cs.CL
arXiv: https://arxiv.org/abs/2602.01064v2

Knowledge distillation has emerged as a pivotal technique for transferring knowledge from stronger large language models (LLMs) to smaller, more efficient models. However, traditional distillation approaches face challenges related to knowledge conflicts and high resource demands, particularly when leveraging multiple teacher models. In this paper, we introduce the concept of \textbf{Knowledge Purification}, which consolidates the rationales from multiple teacher LLMs into a single rationale, thereby mitigating conflicts and enhancing efficiency. To investigate the effectiveness of knowledge purification, we further propose five purification methods from various perspectives. Our experiments demonstrate that these methods not only improve the performance of the distilled model but also effectively alleviate knowledge conflicts. Moreover, router-based methods exhibit robust generalization capabilities, underscoring the potential of innovative purification techniques in optimizing multi-teacher distillation and facilitating the practical deployment of powerful yet lightweight models.

---

### Personality Expression Across Contexts: Linguistic and Behavioral Variation in LLM Agents

Authors: Bin Han, Deuksin Kwon, Jonathan Gratch
Published: 2026-02-01
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2602.01063v1

Large Language Models (LLMs) can be conditioned with explicit personality prompts, yet their behavioral realization often varies depending on context. This study examines how identical personality prompts lead to distinct linguistic, behavioral, and emotional outcomes across four conversational settings: ice-breaking, negotiation, group decision, and empathy tasks. Results show that contextual cues systematically influence both personality expression and emotional tone, suggesting that the same traits are expressed differently depending on social and affective demands. This raises an important question for LLM-based dialogue agents: whether such variations reflect inconsistency or context-sensitive adaptation akin to human behavior. Viewed through the lens of Whole Trait Theory, these findings highlight that LLMs exhibit context-sensitive rather than fixed personality expression, adapting flexibly to social interaction goals and affective conditions.

---

### SetPO: Set-Level Policy Optimization for Diversity-Preserving LLM Reasoning

Authors: Chenyi Li, Yuan Zhang, Bo Wang, Guoqing Ma, Wei Tang, Haoyang Huang, Nan Duan
Published: 2026-02-01
Categories: cs.AI
arXiv: https://arxiv.org/abs/2602.01062v1

Reinforcement learning with verifiable rewards has shown notable effectiveness in enhancing large language models (LLMs) reasoning performance, especially in mathematics tasks. However, such improvements often come with reduced outcome diversity, where the model concentrates probability mass on a narrow set of solutions. Motivated by diminishing-returns principles, we introduce a set level diversity objective defined over sampled trajectories using kernelized similarity. Our approach derives a leave-one-out marginal contribution for each sampled trajectory and integrates this objective as a plug-in advantage shaping term for policy optimization. We further investigate the contribution of a single trajectory to language model diversity within a distribution perturbation framework. This analysis theoretically confirms a monotonicity property, proving that rarer trajectories yield consistently higher marginal contributions to the global diversity. Extensive experiments across a range of model scales demonstrate the effectiveness of our proposed algorithm, consistently outperforming strong baselines in both Pass@1 and Pass@K across various benchmarks.

---

### TLDiffGAN: A Latent Diffusion-GAN Framework with Temporal Information Fusion for Anomalous Sound Detection

Authors: Chengyuan Ma, Peng Jia, Hongyue Guo, Wenming Yang
Published: 2026-02-01
Categories: cs.SD, cs.AI, eess.AS
arXiv: https://arxiv.org/abs/2602.01060v1

Existing generative models for unsupervised anomalous sound detection are limited by their inability to fully capture the complex feature distribution of normal sounds, while the potential of powerful diffusion models in this domain remains largely unexplored. To address this challenge, we propose a novel framework, TLDiffGAN, which consists of two complementary branches. One branch incorporates a latent diffusion model into the GAN generator for adversarial training, thereby making the discriminator's task more challenging and improving the quality of generated samples. The other branch leverages pretrained audio model encoders to extract features directly from raw audio waveforms for auxiliary discrimination. This framework effectively captures feature representations of normal sounds from both raw audio and Mel spectrograms. Moreover, we introduce a TMixup spectrogram augmentation technique to enhance sensitivity to subtle and localized temporal patterns that are often overlooked. Extensive experiments on the DCASE 2020 Challenge Task 2 dataset demonstrate the superior detection performance of TLDiffGAN, as well as its strong capability in anomalous time-frequency localization.

---

### DRFormer: A Dual-Regularized Bidirectional Transformer for Person Re-identification

Authors: Ying Shu, Pujian Zhan, Huiqi Yang, Hehe Fan, Youfang Lin, Kai Lv
Published: 2026-02-01
Categories: cs.CV, cs.MM
arXiv: https://arxiv.org/abs/2602.01059v1

Both fine-grained discriminative details and global semantic features can contribute to solving person re-identification challenges, such as occlusion and pose variations. Vision foundation models (\textit{e.g.}, DINO) excel at mining local textures, and vision-language models (\textit{e.g.}, CLIP) capture strong global semantic difference. Existing methods predominantly rely on a single paradigm, neglecting the potential benefits of their integration. In this paper, we analyze the complementary roles of these two architectures and propose a framework to synergize their strengths by a \textbf{D}ual-\textbf{R}egularized Bidirectional \textbf{Transformer} (\textbf{DRFormer}). The dual-regularization mechanism ensures diverse feature extraction and achieves a better balance in the contributions of the two models. Extensive experiments on five benchmarks show that our method effectively harmonizes local and global representations, achieving competitive performance against state-of-the-art methods.

---

### Good SFT Optimizes for SFT, Better SFT Prepares for Reinforcement Learning

Authors: Dylan Zhang, Yufeng Xu, Haojin Wang, Qingzhi Chen, Hao Peng
Published: 2026-02-01
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2602.01058v2

Post-training of reasoning LLMs is a holistic process that typically consists of an offline SFT stage followed by an online reinforcement learning (RL) stage. However, SFT is often optimized in isolation to maximize SFT performance alone. We show that, after identical RL training, models initialized from stronger SFT checkpoints can significantly underperform those initialized from weaker ones. We attribute this to a mismatch typical in current SFT-RL pipelines: the distribution that generates the offline SFT data can differ substantially from the policy optimized during online RL, which learns from its own rollouts. We propose PEAR (Policy Evaluation-inspired Algorithm for Offline Learning Loss Re-weighting), an SFT-stage method that corrects this mismatch and better prepares the model for RL. PEAR uses importance sampling to reweight the SFT loss, with three variants operating at the token, block, and sequence levels. It can be used to augment standard SFT objectives and incurs little additional training overhead once probabilities for the offline data are collected. We conduct controlled experiments on verifiable reasoning games and mathematical reasoning tasks on Qwen 2.5 and 3 and DeepSeek-distilled models. PEAR consistently improves post-RL performance over canonical SFT, with pass at 8 gains up to a 14.6 percent on AIME2025. Our results suggest that PEAR is an effective step toward more holistic LLM post-training by designing and evaluating SFT with downstream RL in mind rather than in isolation.

---

### Radioactive 3D Gaussian Ray Tracing for Tomographic Reconstruction

Authors: Ling Chen, Bao Yang
Published: 2026-02-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2602.01057v1

3D Gaussian Splatting (3DGS) has recently emerged in computer vision as a promising rendering technique. By adapting the principles of Elliptical Weighted Average (EWA) splatting to a modern differentiable pipeline, 3DGS enables real-time, high-quality novel view synthesis. Building upon this, R2-Gaussian extended the 3DGS paradigm to tomographic reconstruction by rectifying integration bias, achieving state-of-the-art performance in computed tomography (CT). To enable differentiability, R2-Gaussian adopts a local affine approximation: each 3D Gaussian is locally mapped to a 2D Gaussian on the detector and composed via alpha blending to form projections. However, the affine approximation can degrade reconstruction quantitative accuracy and complicate the incorporation of nonlinear geometric corrections. To address these limitations, we propose a tomographic reconstruction framework based on 3D Gaussian ray tracing. Our approach provides two key advantages over splatting-based models: (i) it computes the line integral through 3D Gaussian primitives analytically, avoiding the local affine collapse and thus yielding a more physically consistent forward projection model; and (ii) the ray-tracing formulation gives explicit control over ray origins and directions, which facilitates the precise application of nonlinear geometric corrections, e.g., arc-correction used in positron emission tomography (PET). These properties extend the applicability of Gaussian-based reconstruction to a wider range of realistic tomography systems while improving projection accuracy.

---

### Baseline Method of the Foundation Model Challenge for Ultrasound Image Analysis

Authors: Bo Deng, Yitong Tang, Jiake Li, Yuxin Huang, Li Wang, Yu Zhang, Yufei Zhan, Hua Lu, Xiaoshen Zhang, Jieyun Bai
Published: 2026-02-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2602.01055v1

Ultrasound (US) imaging exhibits substantial heterogeneity across anatomical structures and acquisition protocols, posing significant challenges to the development of generalizable analysis models. Most existing methods are task-specific, limiting their suitability as clinically deployable foundation models. To address this limitation, the Foundation Model Challenge for Ultrasound Image Analysis (FM\_UIA~2026) introduces a large-scale multi-task benchmark comprising 27 subtasks across segmentation, classification, detection, and regression. In this paper, we present the official baseline for FM\_UIA~2026 based on a unified Multi-Head Multi-Task Learning (MH-MTL) framework that supports all tasks within a single shared network. The model employs an ImageNet-pretrained EfficientNet--B4 backbone for robust feature extraction, combined with a Feature Pyramid Network (FPN) to capture multi-scale contextual information. A task-specific routing strategy enables global tasks to leverage high-level semantic features, while dense prediction tasks exploit spatially detailed FPN representations. Training incorporates a composite loss with task-adaptive learning rate scaling and a cosine annealing schedule. Validation results demonstrate the feasibility and robustness of this unified design, establishing a strong and extensible baseline for ultrasound foundation model research. The code and dataset are publicly available at \href{https://github.com/lijiake2408/Foundation-Model-Challenge-for-Ultrasound-Image-Analysis}{GitHub}.

---

### The Quantum Learning Menagerie (A survey on Quantum learning for Classical concepts)

Authors: Sagnik Chatterjee
Published: 2026-02-01
Categories: quant-ph, cs.LG
arXiv: https://arxiv.org/abs/2602.01054v1

This paper surveys various results in the field of Quantum Learning theory, specifically focusing on learning quantum-encoded classical concepts in the Probably Approximately Correct (PAC) framework. The cornerstone of this work is the emphasis on query, sample, and time complexity separations between classical and quantum learning that emerge under learning with query access to different labeling oracles. This paper aims to consolidate all known results in the area under the above umbrella and underscore the limits of our understanding by leaving the reader with 23 open problems.

---

### LRAgent: Efficient KV Cache Sharing for Multi-LoRA LLM Agents

Authors: Hyesung Jeon, Hyeongju Ha, Jae-Joon Kim
Published: 2026-02-01
Categories: cs.LG
arXiv: https://arxiv.org/abs/2602.01053v2

Role specialization in multi-LLM agent systems is often realized via multi-LoRA, where agents share a pretrained backbone and differ only by lightweight adapters. Despite sharing base model weights, each agent independently builds and stores its own KV cache for the same long, tool-augmented trajectories, incurring substantial memory and compute overhead. Existing KV cache sharing methods largely overlook this multi-LoRA setting. We observe that, cache differences across agents are dominated by adapter outputs, while activations from the shared pretrained backbone remain highly similar. Based on this observation, we propose LRAgent, a KV cache sharing framework for multi-LoRA agents. It decomposes the cache into two components, a shared base component derived from pretrained weights and an adapter-dependent component derived from LoRA weights. LRAgent reduces memory overhead by sharing the base component across agents and storing the adapter component in its inherent low-rank form. It also reduces computational overhead by sharing the low-rank cache, enabled by a shared-A multi-LoRA architecture. This avoids redundant computations for contexts that have already been processed by other agents. To efficiently reconstruct adapter contributions at runtime, we introduce Flash-LoRA-Attention, a kernel that reorders attention computation to avoid materializing the low-rank cache to full dimension. LRAgent achieves throughput and time-to-first-token latency close to fully shared caching, while preserving accuracy near the non-shared caching baseline across agentic question-answering benchmarks.

---

### SwiftRepertoire: Few-Shot Immune-Signature Synthesis via Dynamic Kernel Codes

Authors: Rong Fu, Yang Li, Yabin Jin, Jiekai Wu, Chunlei Meng, Simon Fong
Published: 2026-02-01
Categories: cs.LG
arXiv: https://arxiv.org/abs/2602.01051v7

Repertoire-level analysis of T cell receptors offers a biologically grounded signal for disease detection and immune monitoring, yet practical deployment is impeded by label sparsity, cohort heterogeneity, and the computational burden of adapting large encoders to new tasks. We introduce a framework that synthesizes compact task-specific parameterizations from a learned dictionary of prototypes conditioned on lightweight task descriptors derived from repertoire probes and pooled embedding statistics. This synthesis produces small adapter modules applied to a frozen pretrained backbone, enabling immediate adaptation to novel tasks with only a handful of support examples and without full model fine-tuning. The architecture preserves interpretability through motif-aware probes and a calibrated motif discovery pipeline that links predictive decisions to sequence-level signals. Together, these components yield a practical, sample-efficient, and interpretable pathway for translating repertoire-informed models into diverse clinical and research settings where labeled data are scarce and computational resources are constrained.

---

### SwiftRepertoire: Few-Shot Immune-Signature Synthesis via Dynamic Kernel Codes

Authors: Rong Fu, Yang Li, Yabin Jin, Jiekai Wu, Chunlei Meng, Youjin Wang
Published: 2026-02-01
Categories: cs.LG
arXiv: https://arxiv.org/abs/2602.01051v6

Repertoire-level analysis of T cell receptors offers a biologically grounded signal for disease detection and immune monitoring, yet practical deployment is impeded by label sparsity, cohort heterogeneity, and the computational burden of adapting large encoders to new tasks. We introduce a framework that synthesizes compact task-specific parameterizations from a learned dictionary of prototypes conditioned on lightweight task descriptors derived from repertoire probes and pooled embedding statistics. This synthesis produces small adapter modules applied to a frozen pretrained backbone, enabling immediate adaptation to novel tasks with only a handful of support examples and without full model fine-tuning. The architecture preserves interpretability through motif-aware probes and a calibrated motif discovery pipeline that links predictive decisions to sequence-level signals. Together, these components yield a practical, sample-efficient, and interpretable pathway for translating repertoire-informed models into diverse clinical and research settings where labeled data are scarce and computational resources are constrained.

---

### Residual Decoding: Mitigating Hallucinations in Large Vision-Language Models via History-Aware Residual Guidance

Authors: Xinrong Chen, Xu Chu, Yingmin Qiu, Hengyuan Zhang, Jing Xiong, Shiyu Tang, Shuai Liu, Shaokang Yang, Cheng Yang, Hayden Kwok-Hay So, Ngai Wong
Published: 2026-02-01
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2602.01047v3

Large Vision-Language Models (LVLMs) can reason from image-text inputs and perform well in various multimodal tasks. Despite this success, they are affected by language priors and often produce hallucinations. Hallucinations denote generated content that is grammatically and syntactically coherent, yet bears no match or direct relevance to visual input. To address this problem, we propose Residual Decoding (ResDec). It is a novel training-free method that uses historical information to aid decoding. The method relies on the internal implicit reasoning mechanism and token logits evolution mechanism of LVLMs to correct biases. Extensive experiments demonstrate that ResDec effectively suppresses hallucinations induced by language priors, significantly improves visual grounding, and reduces object hallucinations. In addition to mitigating hallucinations, ResDec also performs exceptionally well on comprehensive LVLM benchmarks, highlighting its broad applicability.

---

### ReLayout: Versatile and Structure-Preserving Design Layout Editing via Relation-Aware Design Reconstruction

Authors: Jiawei Lin, Shizhao Sun, Danqing Huang, Ting Liu, Ji Li, Jiang Bian
Published: 2026-02-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2602.01046v1

Automated redesign without manual adjustments marks a key step forward in the design workflow. In this work, we focus on a foundational redesign task termed design layout editing, which seeks to autonomously modify the geometric composition of a design based on user intents. To overcome the ambiguity of user needs expressed in natural language, we introduce four basic and important editing actions and standardize the format of editing operations. The underexplored task presents a unique challenge: satisfying specified editing operations while simultaneously preserving the layout structure of unedited elements. Besides, the scarcity of triplet (original design, editing operation, edited design) samples poses another formidable challenge. To this end, we present ReLayout, a novel framework for versatile and structure-preserving design layout editing that operates without triplet data. Specifically, ReLayout first introduces the relation graph, which contains the position and size relationships among unedited elements, as the constraint for layout structure preservation. Then, relation-aware design reconstruction (RADR) is proposed to bypass the data challenge. By learning to reconstruct a design from its elements, a relation graph, and a synthesized editing operation, RADR effectively emulates the editing process in a self-supervised manner. A multi-modal large language model serves as the backbone for RADR, unifying multiple editing actions within a single model and thus achieving versatile editing after fine-tuning. Qualitative, quantitative results and user studies show that ReLayout significantly outperforms the baseline models in terms of editing quality, accuracy, and layout structure preservation.

---

### Superposition unifies power-law training dynamics

Authors: Zixin Jessie Chen, Hao Chen, Yizhou Liu, Jeff Gore
Published: 2026-02-01
Categories: cs.LG, cs.AI, physics.data-an, stat.ML
arXiv: https://arxiv.org/abs/2602.01045v1

We investigate the role of feature superposition in the emergence of power-law training dynamics using a teacher-student framework. We first derive an analytic theory for training without superposition, establishing that the power-law training exponent depends on both the input data statistics and channel importance. Remarkably, we discover that a superposition bottleneck induces a transition to a universal power-law exponent of $\sim 1$, independent of data and channel statistics. This one over time training with superposition represents an up to tenfold acceleration compared to the purely sequential learning that takes place in the absence of superposition. Our finding that superposition leads to rapid training with a data-independent power law exponent may have important implications for a wide range of neural networks that employ superposition, including production-scale large language models.

---

### Adaptive Dual-Weighting Framework for Federated Learning via Out-of-Distribution Detection

Authors: Zhiwei Ling, Hailiang Zhao, Chao Zhang, Xiang Ao, Ziqi Wang, Cheng Zhang, Zhen Qin, Xinkui Zhao, Kingsum Chow, Yuanqing Wu, MengChu Zhou
Published: 2026-02-01
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2602.01039v1

Federated Learning (FL) enables collaborative model training across large-scale distributed service nodes while preserving data privacy, making it a cornerstone of intelligent service systems in edge-cloud environments. However, in real-world service-oriented deployments, data generated by heterogeneous users, devices, and application scenarios are inherently non-IID. This severe data heterogeneity critically undermines the convergence stability, generalization ability, and ultimately the quality of service delivered by the global model. To address this challenge, we propose FLood, a novel FL framework inspired by out-of-distribution (OOD) detection. FLood dynamically counteracts the adverse effects of heterogeneity through a dual-weighting mechanism that jointly governs local training and global aggregation. At the client level, it adaptively reweights the supervised loss by upweighting pseudo-OOD samples, thereby encouraging more robust learning from distributionally misaligned or challenging data. At the server level, it refines model aggregation by weighting client contributions according to their OOD confidence scores, prioritizing updates from clients with higher in-distribution consistency and enhancing the global model's robustness and convergence stability. Extensive experiments across multiple benchmarks under diverse non-IID settings demonstrate that FLood consistently outperforms state-of-the-art FL methods in both accuracy and generalization. Furthermore, FLood functions as an orthogonal plug-in module: it seamlessly integrates with existing FL algorithms to boost their performance under heterogeneity without modifying their core optimization logic. These properties make FLood a practical and scalable solution for deploying reliable intelligent services in real-world federated environments.

---

### From Videos to Conversations: Egocentric Instructions for Task Assistance

Authors: Lavisha Aggarwal, Vikas Bahirwani, Andrea Colaco
Published: 2026-02-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2602.01038v1

Many everyday tasks, ranging from appliance repair and cooking to car maintenance, require expert knowledge, particularly for complex, multi-step procedures. Despite growing interest in AI agents for augmented reality (AR) assistance, progress remains limited by the scarcity of large-scale multimodal conversational datasets grounded in real-world task execution, in part due to the cost and logistical complexity of human-assisted data collection. In this paper, we present a framework to automatically transform single person instructional videos into two-person multimodal task-guidance conversations. Our fully automatic pipeline, based on large language models, provides a scalable and cost efficient alternative to traditional data collection approaches. Using this framework, we introduce HowToDIV, a multimodal dataset comprising 507 conversations, 6,636 question answer pairs, and 24 hours of video spanning multiple domains. Each session consists of a multi-turn expert-novice interaction. Finally, we report baseline results using Gemma 3 and Qwen 2.5 on HowToDIV, providing an initial benchmark for multimodal procedural task assistance.

---

### VEQ: Modality-Adaptive Quantization for MoE Vision-Language Models

Authors: Guangshuo Qin, Zhiteng Li, Zheng Chen, Weihang Zhang, Linghe Kong, Yulun Zhang
Published: 2026-02-01
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2602.01037v1

Mixture-of-Experts(MoE) Vision-Language Models (VLMs) offer remarkable performance but incur prohibitive memory and computational costs, making compression essential. Post-Training Quantization (PTQ) is an effective training-free technique to address the massive memory and computation overhead. Existing quantization paradigms fall short as they are oblivious to two critical forms of heterogeneity: the inherent discrepancy between vision and language tokens, and the non-uniform contribution of different experts. To bridge this gap, we propose Visual Expert Quantization (VEQ), a dual-aware quantization framework designed to simultaneously accommodate cross-modal differences and heterogeneity between experts. Specifically, VEQ incorporates 1)Modality-expert-aware Quantization, which utilizes expert activation frequency to prioritize error minimization for pivotal experts, and 2)Modality-affinity-aware Quantization, which constructs an enhanced Hessian matrix by integrating token-expert affinity with modality information to guide the calibration process. Extensive experiments across diverse benchmarks verify that VEQ consistently outperforms state-of-the-art baselines. Specifically, under the W3A16 configuration, our method achieves significant average accuracy gains of 2.04\% on Kimi-VL and 3.09\% on Qwen3-VL compared to the previous SOTA quantization methods, demonstrating superior robustness across various multimodal tasks. Our code will be available at https://github.com/guangshuoqin/VEQ.

---

### FUSE-Flow: Scalable Real-Time Multi-View Point Cloud Reconstruction Using Confidence

Authors: Chentian Sun
Published: 2026-02-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2602.01035v2

Real-time multi-view point cloud reconstruction is a core problem in 3D vision and immersive perception, with wide applications in VR, AR, robotic navigation, digital twins, and computer interaction. Despite advances in multi-camera systems and high-resolution depth sensors, fusing large-scale multi-view depth observations into high-quality point clouds under strict real-time constraints remains challenging. Existing methods relying on voxel-based fusion, temporal accumulation, or global optimization suffer from high computational complexity, excessive memory usage, and limited scalability, failing to simultaneously achieve real-time performance, reconstruction quality, and multi-camera extensibility. We propose FUSE-Flow, a frame-wise, stateless, and linearly scalable point cloud streaming reconstruction framework. Each frame independently generates point cloud fragments, fused via two weights, measurement confidence and 3D distance consistency to suppress noise while preserving geometric details. For large-scale multi-camera efficiency, we introduce an adaptive spatial hashing-based weighted aggregation method: 3D space is adaptively partitioned by local point cloud density, representative points are selected per cell, and weighted fusion is performed to handle both sparse and dense regions. With GPU parallelization, FUSE-Flow achieves high-throughput, low-latency point cloud generation and fusion with linear complexity. Experiments demonstrate that the framework improves reconstruction stability and geometric fidelity in overlapping, depth-discontinuous, and dynamic scenes, while maintaining real-time frame rates on modern GPUs, verifying its effectiveness, robustness, and scalability.

---

### Stepwise Intrinsic Rewards for Reasoning in Large Language Models

Authors: Xiangwei Wang, Wei Wang, Ken Chen, Nanduni Nimalsiri, Sachith Seneviratne, Saman Halgamuge
Published: 2026-02-01
Categories: cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2602.01034v2

Reinforcement learning (RL) has become a widely used paradigm for improving the reasoning abilities of large language models (LLMs) and Vision-language models (VLMs). Sparse binary outcome rewards, however, score only final correctness and cannot identify which intermediate steps contributed to it; in multimodal tasks, they may also reward answers driven by linguistic priors rather than visual evidence. Process reward models (PRMs) densify supervision but usually require process annotations, auxiliary models, or inference-time search. In this paper, we introduce Stepwise Marginal Information Gain (MIG), an intrinsic process reward computed from the policy itself. MIG measures how each structured reasoning prefix changes the length-normalized, teacher-forced log-likelihood of the reference answer. A monotonic historical watermark rewards only new likelihood maxima, avoiding duplicate credit after sub-record detours. We combine this signal with outcome and format rewards and a gated self-distillation objective that retains only structurally valid and correct trajectories. For VLMs, a real-versus-blank likelihood gate down-weights rewards when answers remain predictable without the image. Across eight task-specific benchmarks, the full method exceeds outcome-only GRPO in every single-run comparison. In broad-data transfer, it improves average accuracy by up to 4.8 points over binary-reward training and gains 12.6 points on MathVerse. At 7B, it exceeds an external PRM-BoN@16 baseline by 12.9 points on vision-language transfer without inference-time reranking. These results support policy-derived stepwise credit as an annotation-free alternative to explicit process reward modeling.

---

### Discovering Process-Outcome Credit in Multi-Step LLM Reasoning

Authors: Xiangwei Wang, Wei Wang, Ken Chen, Nanduni Nimalsiri, Saman Halgamuge
Published: 2026-02-01
Categories: cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2602.01034v1

Reinforcement Learning (RL) serves as a potent paradigm for enhancing reasoning capabilities in Large Language Models (LLMs), yet standard outcome-based approaches often suffer from reward sparsity and inefficient credit assignment. In this paper, we propose a novel framework designed to provide continuous reward signals, which introduces a Step-wise Marginal Information Gain (MIG) mechanism that quantifies the intrinsic value of reasoning steps against a Monotonic Historical Watermark, effectively filtering out training noise. To ensure disentangled credit distribution, we implement a Decoupled Masking Strategy, applying process-oriented rewards specifically to the chain-of-thought (CoT) and outcome-oriented rewards to the full completion. Additionally, we incorporate a Dual-Gated SFT objective to stabilize training with high-quality structural and factual signals. Extensive experiments across textual and multi-modal benchmarks (e.g., MATH, Super-CLEVR) demonstrate that our approach consistently outperforms baselines such as GRPO in both sample efficiency and final accuracy. Furthermore, our model exhibits superior out-of-distribution robustness, demonstrating promising zero-shot transfer capabilities to unseen and challenging reasoning tasks.

---

### GMAC: Global Multi-View Constraint for Automatic Multi-Camera Extrinsic Calibration

Authors: Chentian Sun
Published: 2026-02-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2602.01033v2

Automatic calibration of multi-camera systems, namely the accurate estimation of spatial extrinsic parameters, is fundamental for 3D reconstruction, panoramic perception, and multi-view data fusion. Existing methods typically rely on calibration targets, explicit geometric modeling, or task-specific neural networks. Such approaches often exhibit limited robustness and applicability in complex dynamic environments or online scenarios, making them difficult to deploy in practical applications. To address this, this paper proposes GMAC, a multi-camera extrinsic estimation framework based on the implicit geometric representations learned by multi-view reconstruction networks. GMAC models extrinsics as global variables constrained by the latent multi-view geometric structure and prunes and structurally reconfigures existing networks so that their latent features can directly support extrinsic prediction through a lightweight regression head, without requiring a completely new network design. Furthermore, GMAC jointly optimizes cross-view reprojection consistency and multi-view cycle consistency, ensuring geometric coherence across cameras while improving prediction accuracy and optimization stability. Experiments on both synthetic and real-world multi-camera datasets demonstrate that GMAC achieves accurate and stable extrinsic estimation without explicit 3D reconstruction or manual calibration, providing a new solution for efficient deployment and online calibration of multi-camera systems.

---

### HierCon: Hierarchical Contrastive Attention for Audio Deepfake Detection

Authors: Zhili Nicholas Liang, Soyeon Caren Han, Qizhou Wang, Christopher Leckie
Published: 2026-02-01
Categories: cs.SD, cs.AI, eess.AS
arXiv: https://arxiv.org/abs/2602.01032v1

Audio deepfakes generated by modern TTS and voice conversion systems are increasingly difficult to distinguish from real speech, raising serious risks for security and online trust. While state-of-the-art self-supervised models provide rich multi-layer representations, existing detectors treat layers independently and overlook temporal and hierarchical dependencies critical for identifying synthetic artefacts. We propose HierCon, a hierarchical layer attention framework combined with margin-based contrastive learning that models dependencies across temporal frames, neighbouring layers, and layer groups, while encouraging domain-invariant embeddings. Evaluated on ASVspoof 2021 DF and In-the-Wild datasets, our method achieves state-of-the-art performance (1.93% and 6.87% EER), improving over independent layer weighting by 36.6% and 22.5% respectively. The results and attention visualisations confirm that hierarchical modelling enhances generalisation to cross-domain generation techniques and recording conditions.

---

### HalluHard: A Hard Multi-Turn Hallucination Benchmark

Authors: Dongyang Fan, Sebastien Delsad, Nicolas Flammarion, Maksym Andriushchenko
Published: 2026-02-01
Categories: cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2602.01031v1

Large language models (LLMs) still produce plausible-sounding but ungrounded factual claims, a problem that worsens in multi-turn dialogue as context grows and early errors cascade. We introduce $\textbf{HalluHard}$, a challenging multi-turn hallucination benchmark with 950 seed questions spanning four high-stakes domains: legal cases, research questions, medical guidelines, and coding. We operationalize groundedness by requiring inline citations for factual assertions. To support reliable evaluation in open-ended settings, we propose a judging pipeline that iteratively retrieves evidence via web search. It can fetch, filter, and parse full-text sources (including PDFs) to assess whether cited material actually supports the generated content. Across a diverse set of frontier proprietary and open-weight models, hallucinations remain substantial even with web search ($\approx 30\%$ for the strongest configuration, Opus-4.5 with web search), with content-grounding errors persisting at high rates. Finally, we show that hallucination behavior is shaped by model capacity, turn position, effective reasoning, and the type of knowledge required.

---

### Bias in the Ear of the Listener: Assessing Sensitivity in Audio Language Models Across Linguistic, Demographic, and Positional Variations

Authors: Sheng-Lun Wei, Yu-Ling Liao, Yen-Hua Chang, Hen-Hsen Huang, Hsin-Hsi Chen
Published: 2026-02-01
Categories: cs.CL, cs.SD, eess.AS
arXiv: https://arxiv.org/abs/2602.01030v1

This work presents the first systematic investigation of speech bias in multilingual MLLMs. We construct and release the BiasInEar dataset, a speech-augmented benchmark based on Global MMLU Lite, spanning English, Chinese, and Korean, balanced by gender and accent, and totaling 70.8 hours ($\approx$4,249 minutes) of speech with 11,200 questions. Using four complementary metrics (accuracy, entropy, APES, and Fleiss' $κ$), we evaluate nine representative models under linguistic (language and accent), demographic (gender), and structural (option order) perturbations. Our findings reveal that MLLMs are relatively robust to demographic factors but highly sensitive to language and option order, suggesting that speech can amplify existing structural biases. Moreover, architectural design and reasoning strategy substantially affect robustness across languages. Overall, this study establishes a unified framework for assessing fairness and robustness in speech-integrated LLMs, bridging the gap between text- and speech-based evaluation. The resources can be found at https://github.com/ntunlplab/BiasInEar.

---

### SFMP: Fine-Grained, Hardware-Friendly and Search-Free Mixed-Precision Quantization for Large Language Models

Authors: Xin Nie, Haicheng Zhang, Liang Dong, Beining Feng, Jinhong Weng, Guiling Sun
Published: 2026-02-01
Categories: cs.LG
arXiv: https://arxiv.org/abs/2602.01027v2

Mixed-precision quantization is a promising approach for compressing large language models under tight memory budgets. However, existing mixed-precision methods typically suffer from one of two limitations: they either rely on expensive discrete optimization to determine precision allocation, or introduce hardware inefficiencies due to irregular memory layouts. We propose SFMP, a search-free and hardware-friendly mixed-precision quantization framework for large language models. The framework is built upon four novel ideas: Fractional bit-width, which extends integer bit-width for weight matrix to fractional value and transforms discrete precision allocation as a continuous problem; 2)Block-wise mixed-precision, enabling fine-grained precision within weight matrices while remaining hardware-friendly; 3)Row-column weight reordering, which aggregates salient weights via row and column reordering, incurring only a small activation reordering overhead during inference; 4)Unified GEMM kernel, which supports mixed-precision GEMM at arbitrary average bit-width. Extensive experiments demonstrate that SFMP outperforms state-of-the-art layer-wise mixed-precision methods under the same memory constraints, while significantly reducing quantization cost and improving inference efficiency. Code is available at https://github.com/Nkniexin/SFMP

---

### The Stacked Autoencoder Evolution Hypothesis

Authors: Hiroyuki Iizuka
Published: 2026-02-01
Categories: cs.NE, cs.LG
arXiv: https://arxiv.org/abs/2602.01026v1

This study introduces a novel theoretical framework, the Stacked Autoencoder Evolution Hypothesis, which proposes that biological evolutionary systems operate through multi-layered self-encoding and decoding processes, analogous to stacked autoencoders in deep learning. Rather than viewing evolution solely as gradual changes driven by mutation and selection, this hypothesis suggests that self-replication inherently compresses and reconstructs genetic information across hierarchical layers of abstraction. This layered structure enables evolutionary systems to explore diverse possibilities not only at the sequence level but also across progressively more abstract layers of representation, making it possible for even simple mutations to navigate these higher-order spaces.Such a mechanism may explain punctuated evolutionary patterns and changes that can appear as if they are goal-directed in natural evolution, by allowing mutations at deeper latent layers to trigger sudden, large-scale phenotypic shifts. To illustrate the plausibility of this mechanism, artificial chemistry simulations were conducted, demonstrating the spontaneous emergence of hierarchical autoencoder structures. This framework offers a new perspective on the informational dynamics underlying both continuous and discontinuous evolutionary change.

---

### Toward Universal and Transferable Jailbreak Attacks on Vision-Language Models

Authors: Kaiyuan Cui, Yige Li, Yutao Wu, Xingjun Ma, Sarah Erfani, Christopher Leckie, Hanxun Huang
Published: 2026-02-01
Categories: cs.LG, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2602.01025v1

Vision-language models (VLMs) extend large language models (LLMs) with vision encoders, enabling text generation conditioned on both images and text. However, this multimodal integration expands the attack surface by exposing the model to image-based jailbreaks crafted to induce harmful responses. Existing gradient-based jailbreak methods transfer poorly, as adversarial patterns overfit to a single white-box surrogate and fail to generalise to black-box models. In this work, we propose Universal and transferable jailbreak (UltraBreak), a framework that constrains adversarial patterns through transformations and regularisation in the vision space, while relaxing textual targets through semantic-based objectives. By defining its loss in the textual embedding space of the target LLM, UltraBreak discovers universal adversarial patterns that generalise across diverse jailbreak objectives. This combination of vision-level regularisation and semantically guided textual supervision mitigates surrogate overfitting and enables strong transferability across both models and attack targets. Extensive experiments show that UltraBreak consistently outperforms prior jailbreak methods. Further analysis reveals why earlier approaches fail to transfer, highlighting that smoothing the loss landscape via semantic objectives is crucial for enabling universal and transferable jailbreaks. The code is publicly available in our \href{https://github.com/kaiyuanCui/UltraBreak}{GitHub repository}.

---

### Unifying Ranking and Generation in Query Auto-Completion via Retrieval-Augmented Generation and Multi-Objective Alignment

Authors: Kai Yuan, Anthony Zheng, Jia Hu, Divyanshu Sheth, Hemanth Velaga, Kylee Kim, Matteo Guarrera, Besim Avci, Jianhua Li, Xuetao Yin, Rajyashree Mukherjee, Sean Suchter
Published: 2026-02-01
Categories: cs.IR, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2602.01023v4

Query Auto-Completion (QAC) suggests query completions as users type, helping them articulate intent and reach results more efficiently. Existing approaches face fundamental challenges: traditional retrieve-and-rank pipelines have limited long-tail coverage and require extensive feature engineering, while recent generative methods suffer from hallucination and safety risks. We present a unified framework that reformulates QAC as end-to-end list generation through Retrieval-Augmented Generation (RAG) and multi-objective Direct Preference Optimization (DPO). Our approach combines three key innovations: (1) reformulating QAC as end-to-end list generation with multi-objective optimization; (2) defining and deploying a suite of rule-based, model-based, and LLM-as-judge verifiers for QAC, and using them in a comprehensive methodology that combines RAG, multi-objective DPO, and iterative critique-revision for high-quality synthetic data; (3) a hybrid serving architecture enabling efficient production deployment under strict latency constraints. Evaluation on a large-scale commercial search platform demonstrates substantial improvements: offline metrics show gains across all dimensions, human evaluation yields +0.40 to +0.69 preference scores, and a controlled online experiment achieves 5.44\% reduction in keystrokes and 3.46\% increase in suggestion adoption, validating that unified generation with RAG and multi-objective alignment provides an effective solution for production QAC. This work represents a paradigm shift to end-to-end generation powered by large language models, RAG, and multi-objective alignment, establishing a production-validated framework that can benefit the broader search and recommendation industry.

---

### Calibrating Behavioral Parameters with Large Language Models

Authors: Brandon Yee, Pairie Koh
Published: 2026-02-01
Categories: econ.GN, cs.AI
arXiv: https://arxiv.org/abs/2602.01022v3

Behavioral parameters such as loss aversion, herding, and extrapolation are central to asset pricing models but remain difficult to measure reliably. We develop a framework that treats large language models (LLMs) as calibrated measurement instruments for behavioral parameters. Using four models and 24{,}000 agent--scenario pairs, we document systematic rationality bias in baseline LLM behavior, including attenuated loss aversion, weak herding, and near-zero disposition effects relative to human benchmarks. Profile-based calibration induces large, stable, and theoretically coherent shifts in several parameters, with calibrated loss aversion, herding, extrapolation, and anchoring reaching or exceeding benchmark magnitudes. To assess external validity, we embed calibrated parameters in an agent-based asset pricing model, where calibrated extrapolation generates short-horizon momentum and long-horizon reversal patterns consistent with empirical evidence. Our results establish measurement ranges, calibration functions, and explicit boundaries for eight canonical behavioral biases.

---

### Effectiveness of Automatically Curated Dataset in Thyroid Nodules Classification Algorithms Using Deep Learning

Authors: Jichen Yang, Jikai Zhang, Benjamin Wildman-Tobriner, Maciej A. Mazurowski
Published: 2026-02-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2602.01020v1

The diagnosis of thyroid nodule cancers commonly utilizes ultrasound images. Several studies showed that deep learning algorithms designed to classify benign and malignant thyroid nodules could match radiologists' performance. However, data availability for training deep learning models is often limited due to the significant effort required to curate such datasets. The previous study proposed a method to curate thyroid nodule datasets automatically. It was tested to have a 63% yield rate and 83% accuracy. However, the usefulness of the generated data for training deep learning models remains unknown. In this study, we conducted experiments to determine whether using a automatically-curated dataset improves deep learning algorithms' performance. We trained deep learning models on the manually annotated and automatically-curated datasets. We also trained with a smaller subset of the automatically-curated dataset that has higher accuracy to explore the optimum usage of such dataset. As a result, the deep learning model trained on the manually selected dataset has an AUC of 0.643 (95% confidence interval (CI): 0.62, 0.66). It is significantly lower than the AUC of the 6automatically-curated dataset trained deep learning model, 0.694 (95% confidence interval (CI): 0.67, 0.73, P < .001). The AUC of the accurate subset trained deep learning model is 0.689 (95% confidence interval (CI): 0.66, 0.72, P > .43), which is insignificantly worse than the AUC of the full automatically-curated dataset. In conclusion, we showed that using a automatically-curated dataset can substantially increase the performance of deep learning algorithms, and it is suggested to use all the data rather than only using the accurate subset.

---

### Inter- and Intra-Subject Variability in EEG: A Systematic Survey

Authors: Xuan-The Tran, Thien-Nhan Vo, Son-Tung Vu, Thoa-Thi Tran, Manh-Dat Nguyen, Thomas Do, Chin-Teng Lin
Published: 2026-02-01
Categories: q-bio.NC, cs.AI, cs.HC
arXiv: https://arxiv.org/abs/2602.01019v1

Electroencephalography (EEG) underpins neuroscience, clinical neurophysiology, and brain-computer interfaces (BCIs), yet pronounced inter- and intra-subject variability limits reliability, reproducibility, and translation. This systematic review studies that quantified or modeled EEG variability across resting-state, event-related potentials (ERPs), and task-related/BCI paradigms (including motor imagery and SSVEP) in healthy and clinical cohorts. Across paradigms, inter-subject differences are typically larger than within-subject fluctuations, but both affect inference and model generalization. Stability is feature-dependent: alpha-band measures and individual alpha peak frequency are often relatively reliable, whereas higher-frequency and many connectivity-derived metrics show more heterogeneous reliability; ERP reliability varies by component, with P300 measures frequently showing moderate-to-good stability. We summarize major sources of variability (biological, state-related, technical, and analytical), review common quantification and modeling approaches (e.g., ICC, CV, SNR, generalizability theory, and multivariate/learning-based methods), and provide recommendations for study design, reporting, and harmonization. Overall, EEG variability should be treated as both a practical constraint to manage and a meaningful signal to leverage for precision neuroscience and robust neurotechnology.

---

### Offline Discovery of Interpretable Skills from Multi-Task Trajectories

Authors: Chongyu Zhu, Mithun Vanniasinghe, Jiayu Chen, Chi-Guhn Lee
Published: 2026-02-01
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2602.01018v1

Hierarchical Imitation Learning is a powerful paradigm for acquiring complex robot behaviors from demonstrations. A central challenge, however, lies in discovering reusable skills from long-horizon, multi-task offline data, especially when the data lacks explicit rewards or subtask annotations. In this work, we introduce LOKI, a three-stage end-to-end learning framework designed for offline skill discovery and hierarchical imitation. The framework commences with a two-stage, weakly supervised skill discovery process: Stage one performs coarse, task-aware macro-segmentation by employing an alignment-enforced Vector Quantized VAE guided by weak task labels. Stage two then refines these segments at a micro-level using a self-supervised sequential model, followed by an iterative clustering process to consolidate skill boundaries. The third stage then leverages these precise boundaries to construct a hierarchical policy within an option-based framework-complete with a learned termination condition beta for explicit skill switching. LOKI achieves high success rates on the challenging D4RL Kitchen benchmark and outperforms standard HIL baselines. Furthermore, we demonstrate that the discovered skills are semantically meaningful, aligning with human intuition, and exhibit compositionality by successfully sequencing them to solve a novel, unseen task.

---

### How Does Unfaithful Reasoning Emerge from Autoregressive Training? A Study of Synthetic Experiments

Authors: Fuxin Wang, Amr Alazali, Yiqiao Zhong
Published: 2026-02-01
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2602.01017v1

Chain-of-thought (CoT) reasoning generated by large language models (LLMs) is often unfaithful: intermediate steps can be logically inconsistent or fail to reflect the causal relationship leading to the final answer. Despite extensive empirical observations, a fundamental understanding of CoT is lacking--what constitutes faithful CoT reasoning, and how unfaithfulness emerges from autoregressive training. We study these questions using well-controlled synthetic experiments, training small transformers on noisy data to solve modular arithmetic expressions step by step, a task we term Arithmetic Expression Reasoning. We find that models can learn faithful reasoning that causally follows the underlying arithmetic rules, but only when the training noise is below a critical threshold, a phenomenon attributable to simplicity bias. At higher noise levels, training dynamics exhibit a transition from faithful stepwise reasoning to unfaithful skip-step reasoning via an intermediate mixed mode characterized by a transient increase in prediction entropy. Mechanistic analysis reveals that models learn to encode internal uncertainty by resolving inconsistent reasoning steps, which suggests the emergence of implicit self-verification from autoregressive training.

---

### Large Language Models as Students Who Think Aloud: Overly Coherent, Verbose, and Confident

Authors: Conrad Borchers, Jill-Jênn Vie, Roger Azevedo
Published: 2026-02-01
Categories: cs.CL, cs.CY
arXiv: https://arxiv.org/abs/2602.01015v2

Large language models (LLMs) are increasingly embedded in AI-based tutoring systems. Can they faithfully model novice reasoning and metacognitive judgments? Existing evaluations emphasize problem-solving accuracy, overlooking the fragmented and imperfect reasoning that characterizes human learning. We evaluate LLMs as novices using 630 think-aloud utterances from multi-step chemistry tutoring problems with problem-solving logs of student hint use, attempts, and problem context. We compare LLM-generated reasoning to human learner utterances under minimal and extended contextual prompting, and assess the models' ability to predict step-level learner success. Although GPT-4.1 generates fluent and contextually appropriate continuations, its reasoning is systematically over-coherent, verbose, and less variable than human think-alouds. These effects intensify with a richer problem-solving context during prompting. Learner performance was consistently overestimated. These findings highlight epistemic limitations of simulating learning with LLMs. We attribute these limitations to LLM training data, including expert-like solutions devoid of expressions of affect and working memory constraints during problem solving. Our evaluation framework can guide future design of adaptive systems that more faithfully support novice learning and self-regulation using generative artificial intelligence.

---

### LocalScore: Local Density-Aware Similarity Scoring for Biometrics

Authors: Yiyang Su, Minchul Kim, Jie Zhu, Christopher Perry, Feng Liu, Anil Jain, Xiaoming Liu
Published: 2026-02-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2602.01012v1

Open-set biometrics faces challenges with probe subjects who may not be enrolled in the gallery, as traditional biometric systems struggle to detect these non-mated probes. Despite the growing prevalence of multi-sample galleries in real-world deployments, most existing methods collapse intra-subject variability into a single global representation, leading to suboptimal decision boundaries and poor open-set robustness. To address this issue, we propose LocalScore, a simple yet effective scoring algorithm that explicitly incorporates the local density of the gallery feature distribution using the k-th nearest neighbors. LocalScore is architecture-agnostic, loss-independent, and incurs negligible computational overhead, making it a plug-and-play solution for existing biometric systems. Extensive experiments across multiple modalities demonstrate that LocalScore consistently achieves substantial gains in open-set retrieval (FNIR@FPIR reduced from 53% to 40%) and verification (TAR@FAR improved from 51% to 74%). We further provide theoretical analysis and empirical validation explaining when and why the method achieves the most significant gains based on dataset characteristics.

---

### Multi-Agent Teams Hold Experts Back

Authors: Aneesh Pappu, Batu El, Hancheng Cao, Carmelo di Nolfo, Yanchao Sun, Meng Cao, James Zou
Published: 2026-02-01
Categories: cs.MA, cs.AI
arXiv: https://arxiv.org/abs/2602.01011v4

Multi-agent LLM systems are increasingly deployed as autonomous collaborators, where agents interact freely rather than execute fixed, pre-specified workflows. In such settings, effective coordination cannot be fully designed in advance and must instead emerge through interaction. However, most prior work enforces coordination through fixed roles, workflows, or aggregation rules, leaving open the question of how well self-organizing teams perform when coordination is unconstrained. Drawing on organizational psychology, we study whether self-organizing LLM teams achieve strong synergy, where team performance matches or exceeds the best individual member. Across human-inspired and frontier ML benchmarks, we find that -- unlike human teams -- LLM teams consistently fail to match their expert agent's performance, even when explicitly told who the expert is, incurring performance losses of up to 41.1% on ML benchmarks. Decomposing this failure, we show that expert leveraging, rather than identification, is the primary bottleneck. Conversational analysis reveals a tendency toward integrative compromise -- averaging expert and non-expert views rather than appropriately weighting expertise -- which increases with team size and correlates negatively with performance. Interestingly, this consensus-seeking behavior improves robustness to adversarial agents, suggesting a trade-off between alignment and effective expertise utilization. Our findings reveal a significant gap in the ability of self-organizing multi-agent teams to harness the collective expertise of their members.

---

### LASS-ODE: Scaling ODE Computations to Connect Foundation Models with Dynamical Physical Systems

Authors: Haoran Li, Chenhan Xiao, Lihao Mai, Yang Weng, Erik Blasch
Published: 2026-02-01
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2602.01009v2

Foundation models have transformed language, vision, and time series data analysis, yet progress on dynamic predictions for physical systems remains limited. Given the complexity of physical constraints, two challenges stand out. $(i)$ Physics-computation scalability: physics-informed learning can enforce physical regularization, but its computation (e.g., ODE integration) does not scale to extensive systems. $(ii)$ Knowledge-sharing efficiency: the attention mechanism is primarily computed within each system, which limits the extraction of shared ODE structures across systems. We show that enforcing ODE consistency does not require expensive nonlinear integration: a token-wise locally linear ODE representation preserves physical fidelity while scaling to foundation-model regimes. Thus, we propose novel token representations that respect locally linear ODE evolution. Such linearity substantially accelerates integration while accurately approximating the local data manifold. Second, we introduce a simple yet effective inter-system attention that augments attention with a common structure hub (CSH) that stores shared tokens and aggregates knowledge across systems. The resulting model, termed LASS-ODE (\underline{LA}rge-\underline{S}cale \underline{S}mall \underline{ODE}), is pretrained on our $40$GB ODE trajectory collections to enable strong in-domain performance, zero-shot generalization across diverse ODE systems, and additional improvements through fine-tuning.

---

### Adapting Where It Matters: Depth-Aware Adaptation for Efficient Multilingual Speech Recognition in Low-Resource Languages

Authors: Yang Xiao, Eun-Jung Holden, Ting Dang
Published: 2026-02-01
Categories: eess.AS, cs.CL, cs.SD
arXiv: https://arxiv.org/abs/2602.01008v1

Recent speech foundation models excel at multilingual automatic speech recognition (ASR) for high-resource languages, but adapting them to low-resource languages remains challenging due to data scarcity and efficiency constraints. Full-model fine-tuning is computationally expensive and prone to overfitting, while parameter-efficient methods like LoRA apply adaptation uniformly across layers, overlooking internal representations thus compromising effectiveness and efficiency. We analyze multilingual ASR models and reveal a U-shaped adaptability pattern: early and late layers are language-specific and require more adaptation, while intermediate layers retain shared semantics and need less. Building on this observation, we propose DAMA, a Depth-Aware Model Adaptation framework that allocates adaptation capacity according to each layer's role. DAMA also introduces Singular Value Decomposition (SVD)-based initialization to constrain adaptation and preserve the U-shaped pattern, as well as a frozen middle-layer basis for further efficiency. Evaluated on 18 low-resource languages across two benchmark datasets, DAMA matches or surpasses state-of-the-art accuracy with 80% fewer trainable parameters, achieves a 29% error reduction under extreme data scarcity, and significantly improves memory, training time, and computational efficiency over baselines. These results highlight the benefits of structure-aware adaptation for efficient, scalable multilingual ASR.

---

### Distilling Token-Trained Models into Byte-Level Models

Authors: Zishuo Bao, Jiaqi Leng, Junxiong Wang, Bowen Peng, Yucheng Lu
Published: 2026-02-01
Categories: cs.CL
arXiv: https://arxiv.org/abs/2602.01007v1

Byte Language Models (BLMs) have emerged as a promising direction for scaling language models beyond tokenization. However, existing BLMs typically require training from scratch on trillions of bytes, making them prohibitively expensive. In this paper, we propose an efficient distillation recipe that converts existing token-trained LLMs into BLMs while retaining comparable capabilities. Our recipe follows a two-stage curriculum: (1) Progressive Knowledge Distillation, which aligns byte-level representations with the embeddings of the token-trained teacher model; and (2) Byte-Level Supervised Fine-Tuning, which enables end-to-end generation entirely in the byte space. We validate our approach across multiple model families, including Llama, Qwen, and OLMo, and demonstrate that the distilled BLMs retain most of the teacher models' performance using only approximately 125B bytes.

---