# Research Papers — 2026-01 · bundle page 150 of 189 > Full-text markdown of papers #7451–7500 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/150 [← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/149.md) [Next bundle →](https://aipod.dev/papers/archive/2026-01/page/151.md) ### Prototypicality Bias Reveals Blindspots in Multimodal Evaluation Metrics Authors: Subhadeep Roy, Gagan Bhatia, Steffen Eger Published: 2026-01-08 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2601.04946v3 Automatic metrics are widely used to evaluate text-to-image models, often replacing human judgment in benchmarking, model selection, and large-scale data filtering. Yet they may reward images that look plausible or prototypical rather than images that faithfully satisfy the prompt. We identify prototypicality bias as a systematic blindspot in multimodal evaluation: metrics can prefer a semantically incorrect but visually or socially prototypical image over a correct but less prototypical one. We introduce PROTOBIAS, a controlled diagnostic benchmark across Animals, Objects, and Demography, where semantically correct images are contrasted with plausible prototypical adversaries containing a single controlled semantic violation. Grounded in prototype theory and social-category prototypicality, PROTOBIAS is constructed with multiple prompt generators, image generators, and independent VLM filters, and validated through prompt-quality, human-annotation, and image-quality controls. Using PROTOBIAS, we show that widely used embedding, reward, VQA-based, and VLM-as-judge metrics frequently fail these contrasts, while human judgments remain more faithful to semantic correctness. We further introduce PROTOSCORE, a lightweight contrastively trained evaluator, as an initial mitigation baseline. PROTOBIAS provides a focused benchmark for measuring prototypicality-driven metric failures and developing more semantically faithful T2I evaluators. --- ### Prototypicality Bias Reveals Blindspots in Multimodal Evaluation Metrics Authors: Subhadeep Roy, Gagan Bhatia, Steffen Eger Published: 2026-01-08 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2601.04946v1 Automatic metrics are now central to evaluating text-to-image models, often substituting for human judgment in benchmarking and large-scale filtering. However, it remains unclear whether these metrics truly prioritize semantic correctness or instead favor visually and socially prototypical images learned from biased data distributions. We identify and study \emph{prototypicality bias} as a systematic failure mode in multimodal evaluation. We introduce a controlled contrastive benchmark \textsc{\textbf{ProtoBias}} (\textit{\textbf{Proto}typical \textbf{Bias}}), spanning Animals, Objects, and Demography images, where semantically correct but non-prototypical images are paired with subtly incorrect yet prototypical adversarial counterparts. This setup enables a directional evaluation of whether metrics follow textual semantics or default to prototypes. Our results show that widely used metrics, including CLIPScore, PickScore, and VQA-based scores, frequently misrank these pairs, while even LLM-as-Judge systems exhibit uneven robustness in socially grounded cases. Human evaluations consistently favour semantic correctness with larger decision margins. Motivated by these findings, we propose \textbf{\textsc{ProtoScore}}, a robust 7B-parameter metric that substantially reduces failure rates and suppresses misranking, while running at orders of magnitude faster than the inference time of GPT-5, approaching the robustness of much larger closed-source judges. --- ### T-Retriever: Tree-based Hierarchical Retrieval Augmented Generation for Textual Graphs Authors: Chunyu Wei, Huaiyu Qin, Siyuan He, Yunhai Wang, Yueguo Chen Published: 2026-01-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04945v1 Retrieval-Augmented Generation (RAG) has significantly enhanced Large Language Models' ability to access external knowledge, yet current graph-based RAG approaches face two critical limitations in managing hierarchical information: they impose rigid layer-specific compression quotas that damage local graph structures, and they prioritize topological structure while neglecting semantic content. We introduce T-Retriever, a novel framework that reformulates attributed graph retrieval as tree-based retrieval using a semantic and structure-guided encoding tree. Our approach features two key innovations: (1) Adaptive Compression Encoding, which replaces artificial compression quotas with a global optimization strategy that preserves the graph's natural hierarchical organization, and (2) Semantic-Structural Entropy ($S^2$-Entropy), which jointly optimizes for both structural cohesion and semantic consistency when creating hierarchical partitions. Experiments across diverse graph reasoning benchmarks demonstrate that T-Retriever significantly outperforms state-of-the-art RAG methods, providing more coherent and contextually relevant responses to complex queries. --- ### Cardinality augmented loss functions Authors: Miguel O'Malley Published: 2026-01-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.04941v1 Class imbalance is a common and pernicious issue for the training of neural networks. Often, an imbalanced majority class can dominate training to skew classifier performance towards the majority outcome. To address this problem we introduce cardinality augmented loss functions, derived from cardinality-like invariants in modern mathematics literature such as magnitude and the spread. These invariants enrich the concept of cardinality by evaluating the `effective diversity' of a metric space, and as such represent a natural solution to overly homogeneous training data. In this work, we establish a methodology for applying cardinality augmented loss functions in the training of neural networks and report results on both artificially imbalanced datasets as well as a real-world imbalanced material science dataset. We observe significant performance improvement among minority classes, as well as improvement in overall performance metrics. --- ### CurricuLLM: Designing Personalized and Workforce-Aligned Cybersecurity Curricula Using Fine-Tuned LLMs Authors: Arthur Nijdam, Harri Kähkönen, Valtteri Niemi, Paul Stankovski Wagner, Sara Ramezanian Published: 2026-01-08 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2601.04940v1 The cybersecurity landscape is constantly evolving, driven by increased digitalization and new cybersecurity threats. Cybersecurity programs often fail to equip graduates with skills demanded by the workforce, particularly concerning recent developments in cybersecurity, as curriculum design is costly and labor-intensive. To address this misalignment, we present a novel Large Language Model (LLM)-based framework for automated design and analysis of cybersecurity curricula, called CurricuLLM. Our approach provides three key contributions: (1) automation of personalized curriculum design, (2) a data-driven pipeline aligned with industry demands, and (3) a comprehensive methodology for leveraging fine-tuned LLMs in curriculum development. CurricuLLM utilizes a two-tier approach consisting of PreprocessLM, which standardizes input data, and ClassifyLM, which assigns course content to nine Knowledge Areas in cybersecurity. We systematically evaluated multiple Natural Language Processing (NLP) architectures and fine-tuning strategies, ultimately selecting the Bidirectional Encoder Representations from Transformers (BERT) model as ClassifyLM, fine-tuned on foundational cybersecurity concepts and workforce competencies. We are the first to validate our method with human experts who analyzed real-world cybersecurity curricula and frameworks, motivating that CurricuLLM is an efficient solution to replace labor-intensive curriculum analysis. Moreover, once course content has been classified, it can be integrated with established cybersecurity role-based weights, enabling alignment of the educational program with specific job roles, workforce categories, or general market needs. This lays the foundation for personalized, workforce-aligned cybersecurity curricula that prepare students for the evolving demands in cybersecurity. --- ### GenProve: Learning to Generate Text with Fine-Grained Provenance Authors: Jingxuan Wei, Xingyue Wang, Yanghaoyu Liao, Jie Dong, Yuchen Liu, Caijun Jia, Bihui Yu, Junnan Zhu Published: 2026-01-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.04932v2 Large language models (LLM) often hallucinate, and while adding citations is a common solution, it is frequently insufficient for accountability as users struggle to verify how a cited source supports a generated claim. Existing methods are typically coarse-grained and fail to distinguish between direct quotes and complex reasoning. In this paper, we introduce Generation-time Fine-grained Provenance, a task where models must generate fluent answers while simultaneously producing structured, sentence-level provenance triples. To enable this, we present ReFInE (Relation-aware Fine-grained Interpretability & Evidence), a dataset featuring expert verified annotations that distinguish between Quotation, Compression, and Inference. Building on ReFInE, we propose GenProve, a framework that combines Supervised Fine-Tuning (SFT) with Group Relative Policy Optimization (GRPO). By optimizing a composite reward for answer fidelity and provenance correctness, GenProve significantly outperforms 14 strong LLMs in joint evaluation. Crucially, our analysis uncovers a reasoning gap where models excel at surface-level quotation but struggle significantly with inference-based provenance, suggesting that verifiable reasoning remains a frontier challenge distinct from surface-level citation. --- ### GenProve: Learning to Generate Text with Fine-Grained Provenance Authors: Jingxuan Wei, Xingyue Wang, Yanghaoyu Liao, Jie Dong, Yuchen Liu, Caijun Jia, Bihui Yu, Junnan Zhu Published: 2026-01-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.04932v1 Large language models (LLM) often hallucinate, and while adding citations is a common solution, it is frequently insufficient for accountability as users struggle to verify how a cited source supports a generated claim. Existing methods are typically coarse-grained and fail to distinguish between direct quotes and complex reasoning. In this paper, we introduce Generation-time Fine-grained Provenance, a task where models must generate fluent answers while simultaneously producing structured, sentence-level provenance triples. To enable this, we present ReFInE (Relation-aware Fine-grained Interpretability & Evidence), a dataset featuring expert verified annotations that distinguish between Quotation, Compression, and Inference. Building on ReFInE, we propose GenProve, a framework that combines Supervised Fine-Tuning (SFT) with Group Relative Policy Optimization (GRPO). By optimizing a composite reward for answer fidelity and provenance correctness, GenProve significantly outperforms 14 strong LLMs in joint evaluation. Crucially, our analysis uncovers a reasoning gap where models excel at surface-level quotation but struggle significantly with inference-based provenance, suggesting that verifiable reasoning remains a frontier challenge distinct from surface-level citation. --- ### Can AI-Generated Persuasion Be Detected? Persuaficial Benchmark and AI vs. Human Linguistic Differences Authors: Arkadiusz Modzelewski, Paweł Golik, Anna Kołos, Giovanni Da San Martino Published: 2026-01-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.04925v2 Large Language Models (LLMs) can generate highly persuasive text, raising concerns about their misuse for propaganda, manipulation, and other harmful purposes. This leads us to our central question: Is LLM-generated persuasion more difficult to automatically detect than human-written persuasion? To address this, we categorize controllable generation approaches for producing persuasive content with LLMs and introduce Persuaficial, a high-quality multilingual benchmark covering six languages: English, German, Polish, Italian, French and Russian. Using this benchmark, we conduct extensive empirical evaluations comparing human-authored and LLM-generated persuasive texts. We find that although overtly persuasive LLM-generated texts can be easier to detect than human-written ones, subtle LLM-generated persuasion consistently degrades automatic detection performance. Beyond detection performance, we provide the first comprehensive linguistic analysis contrasting human and LLM-generated persuasive texts, offering insights that may guide the development of more interpretable and robust detection tools. --- ### Can AI-Generated Persuasion Be Detected? Persuaficial Benchmark and AI vs. Human Linguistic Differences Authors: Arkadiusz Modzelewski, Paweł Golik, Anna Kołos, Giovanni Da San Martino Published: 2026-01-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.04925v1 Large Language Models (LLMs) can generate highly persuasive text, raising concerns about their misuse for propaganda, manipulation, and other harmful purposes. This leads us to our central question: Is LLM-generated persuasion more difficult to automatically detect than human-written persuasion? To address this, we categorize controllable generation approaches for producing persuasive content with LLMs and introduce Persuaficial, a high-quality multilingual benchmark covering six languages: English, German, Polish, Italian, French and Russian. Using this benchmark, we conduct extensive empirical evaluations comparing human-authored and LLM-generated persuasive texts. We find that although overtly persuasive LLM-generated texts can be easier to detect than human-written ones, subtle LLM-generated persuasion consistently degrades automatic detection performance. Beyond detection performance, we provide the first comprehensive linguistic analysis contrasting human and LLM-generated persuasive texts, offering insights that may guide the development of more interpretable and robust detection tools. --- ### Conversational AI for Rapid Scientific Prototyping: A Case Study on ESA's ELOPE Competition Authors: Nils Einecke Published: 2026-01-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04920v2 Large language models (LLMs) are increasingly used as coding partners, yet their role in accelerating scientific discovery remains underexplored. This paper presents a case study of using ChatGPT for rapid prototyping in ESA's ELOPE (Event-based Lunar OPtical flow Egomotion estimation) competition. The competition required participants to process event camera data to estimate lunar lander trajectories. Despite joining late, we achieved second place with a score of 0.01282, highlighting the potential of human-AI collaboration in competitive scientific settings. ChatGPT contributed not only executable code but also algorithmic reasoning, data handling routines, and methodological suggestions, such as using fixed number of events instead of fixed time spans for windowing. At the same time, we observed limitations: the model often introduced unnecessary structural changes, gets confused by intermediate discussions about alternative ideas, occasionally produced critical errors and forgets important aspects in longer scientific discussions. By analyzing these strengths and shortcomings, we show how conversational AI can both accelerate development and support conceptual insight in scientific research. We argue that structured integration of LLMs into the scientific workflow can enhance rapid prototyping by proposing best practices for AI-assisted scientific work. --- ### Conversational AI for Rapid Scientific Prototyping: A Case Study on ESA's ELOPE Competition Authors: Nils Einecke Published: 2026-01-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04920v1 Large language models (LLMs) are increasingly used as coding partners, yet their role in accelerating scientific discovery remains underexplored. This paper presents a case study of using ChatGPT for rapid prototyping in ESA's ELOPE (Event-based Lunar OPtical flow Egomotion estimation) competition. The competition required participants to process event camera data to estimate lunar lander trajectories. Despite joining late, we achieved second place with a score of 0.01282, highlighting the potential of human-AI collaboration in competitive scientific settings. ChatGPT contributed not only executable code but also algorithmic reasoning, data handling routines, and methodological suggestions, such as using fixed number of events instead of fixed time spans for windowing. At the same time, we observed limitations: the model often introduced unnecessary structural changes, gets confused by intermediate discussions about alternative ideas, occasionally produced critical errors and forgets important aspects in longer scientific discussions. By analyzing these strengths and shortcomings, we show how conversational AI can both accelerate development and support conceptual insight in scientific research. We argue that structured integration of LLMs into the scientific workflow can enhance rapid prototyping by proposing best practices for AI-assisted scientific work. --- ### What Students Ask, How a Generative AI Assistant Responds: Exploring Higher Education Students' Dialogues on Learning Analytics Feedback Authors: Yildiz Uzun, Andrea Gauthier, Mutlu Cukurova Published: 2026-01-08 Categories: cs.AI, cs.HC arXiv: https://arxiv.org/abs/2601.04919v1 Learning analytics dashboards (LADs) aim to support students' regulation of learning by translating complex data into feedback. Yet students, especially those with lower self-regulated learning (SRL) competence, often struggle to engage with and interpret analytics feedback. Conversational generative artificial intelligence (GenAI) assistants have shown potential to scaffold this process through real-time, personalised, dialogue-based support. Further advancing this potential, we explored authentic dialogues between students and GenAI assistant integrated into LAD during a 10-week semester. The analysis focused on questions students with different SRL levels posed, the relevance and quality of the assistant's answers, and how students perceived the assistant's role in their learning. Findings revealed distinct query patterns. While low SRL students sought clarification and reassurance, high SRL students queried technical aspects and requested personalised strategies. The assistant provided clear and reliable explanations but limited in personalisation, handling emotionally charged queries, and integrating multiple data points for tailored responses. Findings further extend that GenAI interventions can be especially valuable for low SRL students, offering scaffolding that supports engagement with feedback and narrows gaps with their higher SRL peers. At the same time, students' reflections underscored the importance of trust, need for greater adaptivity, context-awareness, and technical refinement in future systems. --- ### Breaking Robustness Barriers in Cognitive Diagnosis: A One-Shot Neural Architecture Search Perspective Authors: Ziwen Wang, Shangshang Yang, Xiaoshan Yu, Haiping Ma, Xingyi Zhang Published: 2026-01-08 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2601.04918v1 With the advancement of network technologies, intelligent tutoring systems (ITS) have emerged to deliver increasingly precise and tailored personalized learning services. Cognitive diagnosis (CD) has emerged as a core research task in ITS, aiming to infer learners' mastery of specific knowledge concepts by modeling the mapping between learning behavior data and knowledge states. However, existing research prioritizes model performance enhancement while neglecting the pervasive noise contamination in observed response data, significantly hindering practical deployment. Furthermore, current cognitive diagnosis models (CDMs) rely heavily on researchers' domain expertise for structural design, which fails to exhaustively explore architectural possibilities, thus leaving model architectures' full potential untapped. To address this issue, we propose OSCD, an evolutionary multi-objective One-Shot neural architecture search method for Cognitive Diagnosis, designed to efficiently and robustly improve the model's capability in assessing learner proficiency. Specifically, OSCD operates through two distinct stages: training and searching. During the training stage, we construct a search space encompassing diverse architectural combinations and train a weight-sharing supernet represented via the complete binary tree topology, enabling comprehensive exploration of potential architectures beyond manual design priors. In the searching stage, we formulate the optimal architecture search under heterogeneous noise scenarios as a multi-objective optimization problem (MOP), and develop an optimization framework integrating a Pareto-optimal solution search strategy with cross-scenario performance evaluation for resolution. Extensive experiments on real-world educational datasets validate the effectiveness and robustness of the optimal architectures discovered by our OSCD model for CD tasks. --- ### Decentralized Privacy-Preserving Federal Learning of Computer Vision Models on Edge Devices Authors: Damian Harenčák, Lukáš Gajdošech, Martin Madaras Published: 2026-01-08 Categories: cs.CR, cs.CV arXiv: https://arxiv.org/abs/2601.04912v1 Collaborative training of a machine learning model comes with a risk of sharing sensitive or private data. Federated learning offers a way of collectively training a single global model without the need to share client data, by sharing only the updated parameters from each client's local model. A central server is then used to aggregate parameters from all clients and redistribute the aggregated model back to the clients. Recent findings have shown that even in this scenario, private data can be reconstructed only using information about model parameters. Current efforts to mitigate this are mainly focused on reducing privacy risks on the server side, assuming that other clients will not act maliciously. In this work, we analyzed various methods for improving the privacy of client data concerning both the server and other clients for neural networks. Some of these methods include homomorphic encryption, gradient compression, gradient noising, and discussion on possible usage of modified federated learning systems such as split learning, swarm learning or fully encrypted models. We have analyzed the negative effects of gradient compression and gradient noising on the accuracy of convolutional neural networks used for classification. We have shown the difficulty of data reconstruction in the case of segmentation networks. We have also implemented a proof of concept on the NVIDIA Jetson TX2 module used in edge devices and simulated a federated learning process. --- ### From Stories to Cities to Games: A Qualitative Evaluation of Behaviour Planning Authors: Mustafa F. Abdelwahed, Joan Espasa, Alice Toniolo, Ian P. Gent Published: 2026-01-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04911v2 The primary objective of a diverse planning approach is to generate a set of plans that are distinct from one another. Such an approach is applied in a variety of real-world domains, including risk management, automated stream data analysis, and malware detection. More recently, a novel diverse planning paradigm, referred to as behaviour planning, has been proposed. This approach extends earlier methods by explicitly incorporating a diversity model into the planning process and supporting multiple planning categories. In this paper, we demonstrate the usefulness of behaviour planning in real-world settings by presenting three case studies. The first case study focuses on storytelling, the second addresses urban planning, and the third examines game evaluation. --- ### Distributed Online Convex Optimization with Efficient Communication: Improved Algorithm and Lower bounds Authors: Sifan Yang, Wenhao Yang, Wei Jiang, Lijun Zhang Published: 2026-01-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.04907v2 We investigate distributed online convex optimization with compressed communication, where $n$ learners connected by a network collaboratively minimize a sequence of global loss functions using only local information and compressed data from neighbors. Prior work has established regret bounds of $O(\max\{ω^{-2}ρ^{-4}n^{1/2},ω^{-4}ρ^{-8}\}n\sqrt{T})$ and $O(\max\{ω^{-2}ρ^{-4}n^{1/2},ω^{-4}ρ^{-8}\}n\ln{T})$ for convex and strongly convex functions, respectively, where $ω\in(0,1)$ is the compression quality factor ($ω=1$ means no compression) and $ρ<1$ is the spectral gap of the communication matrix. However, these regret bounds suffer from a quadratic or even quartic dependence on $ω^{-1}$. Moreover, the super-linear dependence on $n$ is also undesirable. To overcome these limitations, we propose a novel algorithm that achieves improved regret bounds of $\tilde{O}(ω^{-1/2}ρ^{-1}n\sqrt{T})$ and $\tilde{O}(ω^{-1}ρ^{-2}n\ln{T})$ for convex and strongly convex functions, respectively. The primary idea is to design a two-level blocking update framework incorporating two novel ingredients: an online gossip strategy and an error compensation scheme, which collaborate to achieve a better consensus among learners. Furthermore, we establish the first lower bounds for this problem, justifying the optimality of our results with respect to both $ω$ and $T$. Additionally, we consider the bandit feedback scenario, and extend our method with the classic gradient estimators to enhance existing regret bounds. --- ### Rotation-Robust Regression with Convolutional Model Trees Authors: Hongyi Li, William Ward Armstrong, Jun Xu Published: 2026-01-08 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2601.04899v1 We study rotation-robust learning for image inputs using Convolutional Model Trees (CMTs) (1), whose split and leaf coefficients can be structured on the image grid and transformed geometrically at deployment time. In a controlled MNIST setting with a rotation-invariant regression target, we introduce three geometry-aware inductive biases for split directions -- convolutional smoothing, a tilt dominance constraint, and importance-based pruning -- and quantify their impact on robustness under in-plane rotations. We further evaluate a deployment-time orientation search that selects a discrete rotation maximizing a forest-level confidence proxy without updating model parameters. Orientation search improves robustness under severe rotations but can be harmful near the canonical orientation when confidence is misaligned with correctness. Finally, we observe consistent trends on MNIST digit recognition implemented as one-vs-rest regression, highlighting both the promise and limitations of confidence-based orientation selection for model-tree ensembles. --- ### V-FAT: Benchmarking Visual Fidelity Against Text-bias Authors: Ziteng Wang, Yujie He, Guanliang Li, Siqi Yang, Jiaqi Xiong, Songxiang Liu Published: 2026-01-08 Categories: cs.CL, cs.CV, cs.LG, cs.MM arXiv: https://arxiv.org/abs/2601.04897v1 Recent advancements in Multimodal Large Language Models (MLLMs) have demonstrated impressive performance on standard visual reasoning benchmarks. However, there is growing concern that these models rely excessively on linguistic shortcuts rather than genuine visual grounding, a phenomenon we term Text Bias. In this paper, we investigate the fundamental tension between visual perception and linguistic priors. We decouple the sources of this bias into two dimensions: Internal Corpus Bias, stemming from statistical correlations in pretraining, and External Instruction Bias, arising from the alignment-induced tendency toward sycophancy. To quantify this effect, we introduce V-FAT (Visual Fidelity Against Text-bias), a diagnostic benchmark comprising 4,026 VQA instances across six semantic domains. V-FAT employs a Three-Level Evaluation Framework that systematically increases the conflict between visual evidence and textual information: (L1) internal bias from atypical images, (L2) external bias from misleading instructions, and (L3) synergistic bias where both coincide. We introduce the Visual Robustness Score (VRS), a metric designed to penalize "lucky" linguistic guesses and reward true visual fidelity. Our evaluation of 12 frontier MLLMs reveals that while models excel in existing benchmarks, they experience significant visual collapse under high linguistic dominance. --- ### DVD: A Robust Method for Detecting Variant Contamination in Large Language Model Evaluation Authors: Renzhao Liang, Jingru Chen, Bo Jia, Bo Deng, Chenggang Xie, Yidong Wang, Ke Jin, Xin Wang, Linfeng Zhang, Cunxiang Wang Published: 2026-01-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04895v1 Evaluating large language models (LLMs) is increasingly confounded by \emph{variant contamination}: the training corpus contains semantically equivalent yet lexically or syntactically altered versions of test items. Unlike verbatim leakage, these paraphrased or structurally transformed variants evade existing detectors based on sampling consistency or perplexity, thereby inflating benchmark scores via memorization rather than genuine reasoning. We formalize this problem and introduce \textbf{DVD} (\textbf{D}etection via \textbf{V}ariance of generation \textbf{D}istribution), a single-sample detector that models the local output distribution induced by temperature sampling. Our key insight is that contaminated items trigger alternation between a \emph{memory-adherence} state and a \emph{perturbation-drift} state, yielding abnormally high variance in the synthetic difficulty of low-probability tokens; uncontaminated items remain in drift with comparatively smooth variance. We construct the first benchmark for variant contamination across two domains Omni-MATH and SuperGPQA by generating and filtering semantically equivalent variants, and simulate contamination via fine-tuning models of different scales and architectures (Qwen2.5 and Llama3.1). Across datasets and models, \textbf{DVD} consistently outperforms perplexity-based, Min-$k$\%++, edit-distance (CDD), and embedding-similarity baselines, while exhibiting strong robustness to hyperparameters. Our results establish variance of the generation distribution as a principled and practical fingerprint for detecting variant contamination in LLM evaluation. --- ### Scaling Vision Language Models for Pharmaceutical Long Form Video Reasoning on Industrial GenAI Platform Authors: Suyash Mishra, Qiang Li, Srikanth Patil, Satyanarayan Pati, Baddu Narendra Published: 2026-01-08 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2601.04891v1 Vision Language Models (VLMs) have shown strong performance on multimodal reasoning tasks, yet most evaluations focus on short videos and assume unconstrained computational resources. In industrial settings such as pharmaceutical content understanding, practitioners must process long-form videos under strict GPU, latency, and cost constraints, where many existing approaches fail to scale. In this work, we present an industrial GenAI framework that processes over 200,000 PDFs, 25,326 videos across eight formats (e.g., MP4, M4V, etc.), and 888 multilingual audio files in more than 20 languages. Our study makes three contributions: (i) an industrial large-scale architecture for multimodal reasoning in pharmaceutical domains; (ii) empirical analysis of over 40 VLMs on two leading benchmarks (Video-MME and MMBench) and proprietary dataset of 25,326 videos across 14 disease areas; and (iii) four findings relevant to long-form video reasoning: the role of multimodality, attention mechanism trade-offs, temporal reasoning limits, and challenges of video splitting under GPU constraints. Results show 3-8 times efficiency gains with SDPA attention on commodity GPUs, multimodality improving up to 8/12 task domains (especially length-dependent tasks), and clear bottlenecks in temporal alignment and keyframe detection across open- and closed-source VLMs. Rather than proposing a new "A+B" model, this paper characterizes practical limits, trade-offs, and failure patterns of current VLMs under realistic deployment constraints, and provide actionable guidance for both researchers and practitioners designing scalable multimodal systems for long-form video understanding in industrial domains. --- ### Learnable Multipliers: Freeing the Scale of Language Model Matrix Layers Authors: Maksim Velikanov, Ilyas Chahed, Jingwei Zuo, Dhia Eddine Rhaiem, Younes Belkada, Hakim Hacid Published: 2026-01-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.04890v1 Applying weight decay (WD) to matrix layers is standard practice in large-language-model pretraining. Prior work suggests that stochastic gradient noise induces a Brownian-like expansion of the weight matrices W, whose growth is counteracted by WD, leading to a WD-noise equilibrium with a certain weight norm ||W||. In this work, we view the equilibrium norm as a harmful artifact of the training procedure, and address it by introducing learnable multipliers to learn the optimal scale. First, we attach a learnable scalar multiplier to W and confirm that the WD-noise equilibrium norm is suboptimal: the learned scale adapts to data and improves performance. We then argue that individual row and column norms are similarly constrained, and free their scale by introducing learnable per-row and per-column multipliers. Our method can be viewed as a learnable, more expressive generalization of muP multipliers. It outperforms a well-tuned muP baseline, reduces the computational overhead of multiplier tuning, and surfaces practical questions such as forward-pass symmetries and the width-scaling of the learned multipliers. Finally, we validate learnable multipliers with both Adam and Muon optimizers, where it shows improvement in downstream evaluations matching the improvement of the switching from Adam to Muon. --- ### Faithful Summarisation under Disagreement via Belief-Level Aggregation Authors: Favour Yahdii Aghaebe, Tanefa Apekey, Elizabeth Williams, Nafise Sadat Moosavi Published: 2026-01-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.04889v1 Opinion and multi-document summarisation often involve genuinely conflicting viewpoints, yet many existing approaches, particularly LLM-based systems, implicitly smooth disagreement and over-represent majority opinions. This limits the faithfulness of generated summaries in opinion-heavy settings. We introduce a disagreement-aware synthesis pipeline that separates belief-level aggregation from language generation. Documents are first represented as structured belief sets and aggregated using distance-based belief merging operators that explicitly model conflict. Large language models are then used only to realise the aggregated beliefs as natural language summaries. We evaluate the approach across multiple model families and scales, comparing it to methods that perform explicit aggregation during generation. Our results show that while sufficiently large models can match belief-level aggregation when aggregation is handled at generation time, this behaviour is not stable across architectures or capacities. In contrast, belief-level aggregation combined with simple prompting yields consistently strong disagreement-aware performance across models, while maintaining fluent and grounded summaries. --- ### SmartSearch: Process Reward-Guided Query Refinement for Search Agents Authors: Tongyu Wen, Guanting Dong, Zhicheng Dou Published: 2026-01-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04888v1 Large language model (LLM)-based search agents have proven promising for addressing knowledge-intensive problems by incorporating information retrieval capabilities. Existing works largely focus on optimizing the reasoning paradigms of search agents, yet the quality of intermediate search queries during reasoning remains overlooked. As a result, the generated queries often remain inaccurate, leading to unexpected retrieval results and ultimately limiting search agents' overall effectiveness. To mitigate this issue, we introduce SmartSearch, a framework built upon two key mechanisms: (1) Process rewards, which provide fine-grained supervision for the quality of each intermediate search query through Dual-Level Credit Assessment. (2) Query refinement, which promotes the optimization of query generation by selectively refining low-quality search queries and regenerating subsequent search rounds based on these refinements. To enable the search agent to progressively internalize the ability to improve query quality under the guidance of process rewards, we design a three-stage curriculum learning framework. This framework guides the agent through a progression from imitation, to alignment, and ultimately to generalization. Experimental results show that SmartSearch consistently surpasses existing baselines, and additional quantitative analyses further confirm its significant gains in both search efficiency and query quality. The code is available at https://github.com/MYVAE/SmartSearch. --- ### Flexible Manufacturing Systems Intralogistics: Dynamic Optimization of AGVs and Tool Sharing Using Coloured-Timed Petri Nets and Actor-Critic RL with Actions Masking Authors: Sofiene Lassoued, Laxmikant Shrikant Bahetic, Nathalie Weiß-Borkowskib, Stefan Lierc, Andreas Schwunga Published: 2026-01-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04887v1 Flexible Manufacturing Systems (FMS) are pivotal in optimizing production processes in today's rapidly evolving manufacturing landscape. This paper advances the traditional job shop scheduling problem by incorporating additional complexities through the simultaneous integration of automated guided vehicles (AGVs) and tool-sharing systems. We propose a novel approach that combines Colored-Timed Petri Nets (CTPNs) with actor-critic model-based reinforcement learning (MBRL), effectively addressing the multifaceted challenges associated with FMS. CTPNs provide a formal modeling structure and dynamic action masking, significantly reducing the action search space, while MBRL ensures adaptability to changing environments through the learned policy. Leveraging the advantages of MBRL, we incorporate a lookahead strategy for optimal positioning of AGVs, improving operational efficiency. Our approach was evaluated on small-sized public benchmarks and a newly developed large-scale benchmark inspired by the Taillard benchmark. The results show that our approach matches traditional methods on smaller instances and outperforms them on larger ones in terms of makespan while achieving a tenfold reduction in computation time. To ensure reproducibility, we propose a gym-compatible environment and an instance generator. Additionally, an ablation study evaluates the contribution of each framework component to its overall performance. --- ### Analyzing Message-Code Inconsistency in AI Coding Agent-Authored Pull Requests Authors: Jingzhi Gong, Giovanni Pinna, Yixin Bian, Jie M. Zhang Published: 2026-01-08 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2601.04886v2 Pull request (PR) descriptions generated by AI coding agents are the primary channel for communicating code changes to human reviewers. However, the alignment between these messages and the actual changes remains unexplored, raising concerns about the trustworthiness of AI agents. To fill this gap, we analyzed 23,247 agentic PRs across five agents using PR message-code inconsistency (PR-MCI). We contributed 974 manually annotated PRs, found 406 PRs (1.7%) exhibited high PR-MCI, and identified eight PR-MCI types, revealing that "descriptions claim unimplemented changes" was the most common issue (45.4%). Statistical tests confirmed that high-MCI PRs had 51.7% lower acceptance rates (28.3% vs. 80.0%) and took 3.5 times longer to merge (55.8 vs. 16.0 hours). Our findings suggest that unreliable PR descriptions undermine trust in AI agents, highlighting the need for PR-MCI verification mechanisms and improved PR generation to enable trustworthy human-AI collaboration. --- ### CuMA: Aligning LLMs with Sparse Cultural Values via Demographic-Aware Mixture of Adapters Authors: Ao Sun, Xiaoyu Wang, Zhe Tan, Yu Li, Jiachen Zhu, Yuheng Jia, Shu Su Published: 2026-01-08 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.04885v3 As Large Language Models (LLMs) serve a global audience, alignment must transition from enforcing universal consensus to respecting cultural pluralism. We demonstrate that dense models, when forced to fit conflicting value distributions, suffer from \textbf{Mean Collapse}, converging to a generic average that fails to represent diverse groups. We attribute this to \textbf{Cultural Sparsity}, where gradient interference prevents dense parameters from spanning distinct cultural modes. To resolve this, we propose \textbf{\textsc{CuMA}} (\textbf{Cu}ltural \textbf{M}ixture of \textbf{A}dapters), a framework that frames alignment as a \textbf{conditional capacity separation} problem. By incorporating demographic-aware routing, \textsc{CuMA} internalizes a \textit{Latent Cultural Topology} to explicitly disentangle conflicting gradients into specialized expert subspaces. Extensive evaluations on WorldValuesBench, Community Alignment, and PRISM demonstrate that \textsc{CuMA} achieves state-of-the-art performance, significantly outperforming both dense baselines and semantic-only MoEs. Crucially, our analysis confirms that \textsc{CuMA} effectively mitigates mean collapse, preserving cultural diversity. Our code is available at https://github.com/Throll/CuMA. --- ### Precomputing Multi-Agent Path Replanning Using Temporal Flexibility Authors: Issa Hanou, Eric Kemmeren, Devin Wild Thomas, Mathijs de Weerdt Published: 2026-01-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04884v3 Executing a multi-agent plan can be challenging when an agent is delayed, because this typically creates conflicts with other agents. So, we need to quickly find a new safe plan. Replanning only the delayed agent often does not yield an efficient plan, and sometimes cannot even yield a feasible one. On the other hand, replanning other agents may lead to a cascade of changes and delays, and it is computationally expensive. We show how to efficiently replan a single delayed agent by tracking and using the temporal flexibility of other agents while avoiding cascading delays. This flexibility is the maximum delay that the agent can take without changing the order with agents other than the initially delayed agent, or further delaying other agents. Our algorithm, FlexSIPP, precomputes all possible plans for the delayed agent and returns the changes to the other agents within the given scenario. We demonstrate our method in a real-world case study of replanning trains in the densely-used Dutch railway network and in the MovingAI MAPF benchmark set. Our experiments show that FlexSIPP provides effective solutions relevant to real-world adjustments, and within a reasonable timeframe. --- ### Mind2Report: Expert-Level Commercial Report Synthesis via Cognitive Deep Research Agent Authors: Mingyue Cheng, Daoyu Wang, Qi Liu, Shuo Yu, Xiaoyu Tao, Yuqian Wang, Chengzhong Chu, Yu Duan, Mingkang Long, Enhong Chen Published: 2026-01-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.04879v2 Synthesizing informative commercial reports from massive and noisy web sources is critical for high-stakes business decisions. Although recent deep research agents (DRAs) achieve notable progress, their reports remain limited in quality, reliability, and coverage. These mainly stem from ambiguous intents that cause search drift, retrieved web content that rapidly saturates the context window, and single-pass synthesis that limits report comprehensiveness. In this work, we propose Mind2Report, a cognitive deep research agent that emulates commercial analysts to synthesize expert-level reports. Mind2Report first probes fine-grained commercial intent to establish a structured outline, then recursively explores web sources and distills validated evidence into research memory to preserve context efficiency. Meanwhile, the research memory and outline continuously co-evolve, refining the report structure to avoid rigid initial planning. Finally, Mind2Report iteratively synthesizes the report based on the evolving outline and accumulated evidence. Together, these designs enable reliable and context-efficient long-horizon commercial deep research. To rigorously evaluate commercial DRAs, we further construct QRC-Eval, comprising 200 real-world commercial tasks and a holistic evaluation framework covering report quality, reliability, and coverage. Extensive experiments demonstrate that Mind2Report consistently outperforms leading proprietary and open-source DRAs, while ablations verify the effectiveness of each module and further analyze the challenges they address. We expect this work to advance the development of commercial deep research agents. --- ### Higher-Order Knowledge Representations for Agentic Scientific Reasoning Authors: Isabella A. Stewart, Markus J. Buehler Published: 2026-01-08 Categories: cs.AI, cond-mat.mtrl-sci, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2601.04878v1 Scientific inquiry requires systems-level reasoning that integrates heterogeneous experimental data, cross-domain knowledge, and mechanistic evidence into coherent explanations. While Large Language Models (LLMs) offer inferential capabilities, they often depend on retrieval-augmented contexts that lack structural depth. Traditional Knowledge Graphs (KGs) attempt to bridge this gap, yet their pairwise constraints fail to capture the irreducible higher-order interactions that govern emergent physical behavior. To address this, we introduce a methodology for constructing hypergraph-based knowledge representations that faithfully encode multi-entity relationships. Applied to a corpus of ~1,100 manuscripts on biocomposite scaffolds, our framework constructs a global hypergraph of 161,172 nodes and 320,201 hyperedges, revealing a scale-free topology (power law exponent ~1.23) organized around highly connected conceptual hubs. This representation prevents the combinatorial explosion typical of pairwise expansions and explicitly preserves the co-occurrence context of scientific formulations. We further demonstrate that equipping agentic systems with hypergraph traversal tools, specifically using node-intersection constraints, enables them to bridge semantically distant concepts. By exploiting these higher-order pathways, the system successfully generates grounded mechanistic hypotheses for novel composite materials, such as linking cerium oxide to PCL scaffolds via chitosan intermediates. This work establishes a "teacherless" agentic reasoning system where hypergraph topology acts as a verifiable guardrail, accelerating scientific discovery by uncovering relationships obscured by traditional graph methods. --- ### EvolSQL: Structure-Aware Evolution for Scalable Text-to-SQL Data Synthesis Authors: Xuanguang Pan, Chongyang Tao, Jiayuan Bai, Jianling Gao, Zhengwei Tao, Xiansheng Zhou, Gavin Cheung, Shuai Ma Published: 2026-01-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.04875v1 Training effective Text-to-SQL models remains challenging due to the scarcity of high-quality, diverse, and structurally complex datasets. Existing methods either rely on limited human-annotated corpora, or synthesize datasets directly by simply prompting LLMs without explicit control over SQL structures, often resulting in limited structural diversity and complexity. To address this, we introduce EvolSQL, a structure-aware data synthesis framework that evolves SQL queries from seed data into richer and more semantically diverse forms. EvolSQL starts with an exploratory Query-SQL expansion to broaden question diversity and improve schema coverage, and then applies an adaptive directional evolution strategy using six atomic transformation operators derived from the SQL Abstract Syntax Tree to progressively increase query complexity across relational, predicate, aggregation, and nesting dimensions. An execution-grounded SQL refinement module and schema-aware deduplication further ensure the creation of high-quality, structurally diverse mapping pairs. Experimental results show that a 7B model fine-tuned on our data outperforms one trained on the much larger SynSQL dataset using only 1/18 of the data. --- ### FibreCastML: An Open Web Platform for Predicting Electrospun Nanofibre Diameter Distributions Authors: Elisa Roldan, Kirstie Andrews, Stephen M. Richardson, Reyhaneh Fatahian, Glen Cooper, Rasool Erfani, Tasneem Sabir, Neil D. Reeves Published: 2026-01-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.04873v2 Electrospinning is a scalable technique for producing fibrous scaffolds with tunable micro- and nanoscale architectures for applications in tissue engineering, drug delivery, and wound care. While machine learning (ML) has been used to support electrospinning process optimisation, most existing approaches predict only mean fibre diameters, neglecting the full diameter distribution that governs scaffold performance. This work presents FibreCastML, an open, distribution-aware ML framework that predicts complete fibre diameter spectra from routinely reported electrospinning parameters and provides interpretable insights into process structure relationships. A meta-dataset comprising 68538 individual fibre diameter measurements extracted from 1778 studies across 16 biomedical polymers was curated. Six standard processing parameters, namely solution concentration, applied voltage, flow rate, tip to collector distance, needle diameter, and collector rotation speed, were used to train seven ML models using nested cross validation with leave one study out external folds. Model interpretability was achieved using variable importance analysis, SHapley Additive exPlanations, correlation matrices, and three dimensional parameter maps. Non linear models consistently outperformed linear baselines, achieving coefficients of determination above 0.91 for several widely used polymers. Solution concentration emerged as the dominant global driver of fibre diameter distributions. Experimental validation across different electrospinning systems demonstrated close agreement between predicted and measured distributions. FibreCastML enables more reproducible and data driven optimisation of electrospun scaffold architectures. --- ### Gradient-based Optimisation of Modulation Effects Authors: Alistair Carson, Alec Wright, Stefan Bilbao Published: 2026-01-08 Categories: eess.AS, cs.LG, cs.SD arXiv: https://arxiv.org/abs/2601.04867v3 Modulation effects such as phasers, flangers and chorus effects are heavily used in conjunction with the electric guitar. Machine learning based emulation of analog modulation units has been investigated in recent years, but most methods have either been limited to one class of effect or suffer from a high computational cost or latency compared to canonical digital implementations. Here, we build on previous work and present a framework for modelling flanger, chorus and phaser effects based on differentiable digital signal processing. The model is trained in the time-frequency domain, but at inference operates in the time-domain, requiring zero latency. We investigate the challenges associated with gradient-based optimisation of such effects, and show that low-frequency weighting of loss functions avoids convergence to local minima when learning delay times. We show that when trained against analog effects units, sound output from the model is in some cases perceptually indistinguishable from the reference, but challenges still remain for effects with long delay times and feedback. --- ### Key-Value Pair-Free Continual Learner via Task-Specific Prompt-Prototype Authors: Haihua Luo, Xuming Ran, Zhengji Li, Huiyan Xue, Tingting Jiang, Jiangrong Shen, Tommi Kärkkäinen, Qi Xu, Fengyu Cong Published: 2026-01-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04864v2 Continual learning aims to enable models to acquire new knowledge while retaining previously learned information. Prompt-based methods have shown remarkable performance in this domain; however, they typically rely on key-value pairing, which can introduce inter-task interference and hinder scalability. To overcome these limitations, we propose a novel approach employing task-specific Prompt-Prototype (ProP), thereby eliminating the need for key-value pairs. In our method, task-specific prompts facilitate more effective feature learning for the current task, while corresponding prototypes capture the representative features of the input. During inference, predictions are generated by binding each task-specific prompt with its associated prototype. Additionally, we introduce regularization constraints during prompt initialization to penalize excessively large values, thereby enhancing stability. Experiments on several widely used datasets demonstrate the effectiveness of the proposed method. In contrast to mainstream prompt-based approaches, our framework removes the dependency on key-value pairs, offering a fresh perspective for future continual learning research. --- ### Orchestrating Intelligence: Confidence-Aware Routing for Efficient Multi-Agent Collaboration across Multi-Scale Models Authors: Jingbo Wang, Sendong Zhao, Jiatong Liu, Haochun Wang, Wanting Li, Bing Qin, Ting Liu Published: 2026-01-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04861v2 While multi-agent systems (MAS) have demonstrated superior performance over single-agent approaches in complex reasoning tasks, they often suffer from significant computational inefficiencies. Existing frameworks typically deploy large language models (LLMs) uniformly across all agent roles, failing to account for the varying cognitive demands of different reasoning stages. We address this inefficiency by proposing OI-MAS framework, a novel multi-agent framework that implements an adaptive model-selection policy across a heterogeneous pool of multi-scale LLMs. Specifically, OI-MAS introduces a state-dependent routing mechanism that dynamically selects agent roles and model scales throughout the reasoning process. In addition, we introduce a confidence-aware mechanism that selects appropriate model scales conditioned on task complexity, thus reducing unnecessary reliance on large-scale models. Experimental results show that OI-MAS consistently outperforms baseline multi-agent systems, improving accuracy by up to 12.88\% while reducing cost by up to 79.78\%. --- ### DivAS: Interactive 3D Segmentation by Depth-Weighted Voxel Aggregation Authors: Ayush Pande, Mayank Vatsa Published: 2026-01-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.04860v2 Interactive 3D segmentation of a reconstructed scene should not require a representation-specific optimization loop. We observe that the recipe for lifting 2D foundation-model masks into 3D, namely prompting a few views, refining the resulting masks with rendered depth, and fusing the multi-view evidence into a voxel grid, is shared across scene representations. What remains representation-specific is only the depth signal returned by the renderer and the occupancy prior that gates fusion. We present **DivAS** (Depth-interactive Voxel Aggregation Segmentation), an optimization-free, training-free framework that realizes this recipe as a single interaction-and-fusion skeleton with lightweight, representation-specific adapters, instantiated on both Gaussian Splatting (GS) and NeRF backbones. On standard forward-facing and unbounded benchmarks, the GS instantiation attains segmentation quality competitive with state-of-the-art optimization-based methods, and the best on LLFF, while being the only one to reach this quality within the consumer-hardware memory envelope at standard resolution. Both instantiations run end-to-end around $2$x faster than feature-field baselines, with a per-update fusion-kernel cost below $70$ ms. Because segmentation evidence is gathered from a small, bounded set of anchor views, user effort and computation remain independent of the training-set size. The same skeleton applied to a NeRF backbone matches or exceeds the performance of optimization-based NeRF baselines, confirming that the recipe transfers across fundamentally different 3D representations. --- ### A Navigational Approach for Comprehensive RAG via Traversal over Proposition Graphs Authors: Maxime Delmas, Lei Xu, André Freitas Published: 2026-01-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.04859v1 Standard RAG pipelines based on chunking excel at simple factual retrieval but fail on complex multi-hop queries due to a lack of structural connectivity. Conversely, initial strategies that interleave retrieval with reasoning often lack global corpus awareness, while Knowledge Graph (KG)-based RAG performs strongly on complex multi-hop tasks but suffers on fact-oriented single-hop queries. To bridge this gap, we propose a novel RAG framework: ToPG (Traversal over Proposition Graphs). ToPG models its knowledge base as a heterogeneous graph of propositions, entities, and passages, effectively combining the granular fact density of propositions with graph connectivity. We leverage this structure using iterative Suggestion-Selection cycles, where the Suggestion phase enables a query-aware traversal of the graph, and the Selection phase provides LLM feedback to prune irrelevant propositions and seed the next iteration. Evaluated on three distinct QA tasks (Simple, Complex, and Abstract QA), ToPG demonstrates strong performance across both accuracy- and quality-based metrics. Overall, ToPG shows that query-aware graph traversal combined with factual granularity is a critical component for efficient structured RAG systems. ToPG is available at https://github.com/idiap/ToPG. --- ### MisSpans: Fine-Grained False Span Identification in Cross-Domain Fake News Authors: Zhiwei Liu, Paul Thompson, Jiaqi Rong, Baojie Qu, Runteng Guo, Min Peng, Qianqian Xie, Sophia Ananiadou Published: 2026-01-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.04857v1 Online misinformation is increasingly pervasive, yet most existing benchmarks and methods evaluate veracity at the level of whole claims or paragraphs using coarse binary labels, obscuring how true and false details often co-exist within single sentences. These simplifications also limit interpretability: global explanations cannot identify which specific segments are misleading or differentiate how a detail is false (e.g., distorted vs. fabricated). To address these gaps, we introduce MisSpans, the first multi-domain, human-annotated benchmark for span-level misinformation detection and analysis, consisting of paired real and fake news stories. MisSpans defines three complementary tasks: MisSpansIdentity for pinpointing false spans within sentences, MisSpansType for categorising false spans by misinformation type, and MisSpansExplanation for providing rationales grounded in identified spans. Together, these tasks enable fine-grained localisation, nuanced characterisation beyond true/false and actionable explanations. Expert annotators were guided by standardised guidelines and consistency checks, leading to high inter-annotator agreement. We evaluate 15 representative LLMs, including reasoning-enhanced and non-reasoning variants, under zero-shot and one-shot settings. Results reveal the challenging nature of fine-grained misinformation identification and analysis, and highlight the need for a deeper understanding of how performance may be influenced by multiple interacting factors, including model size and reasoning capabilities, along with domain-specific textual features. This project will be available at https://github.com/lzw108/MisSpans. --- ### Rethinking GNNs and Missing Features: Challenges, Evaluation and a Robust Solution Authors: Francesco Ferrini, Veronica Lachi, Antonio Longa, Bruno Lepri, Matono Akiyoshi, Andrea Passerini, Xin Liu, Manfred Jaeger Published: 2026-01-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.04855v2 Handling missing node features is a key challenge for deploying Graph Neural Networks (GNNs) in real-world domains such as healthcare and sensor networks. Existing studies mostly address relatively benign scenarios, namely benchmark datasets with (a) high-dimensional but sparse node features and (b) incomplete data generated under Missing Completely At Random (MCAR) mechanisms. For (a), we theoretically prove that high sparsity substantially limits the information loss caused by missingness, making all models appear robust and preventing a meaningful comparison of their performance. To overcome this limitation, we introduce one synthetic and three real-world datasets with dense, semantically meaningful features. For (b), we move beyond MCAR and design evaluation protocols with more realistic missingness mechanisms. Moreover, we provide a theoretical background to state explicit assumptions on the missingness process and analyze their implications for different methods. Building on this analysis, we propose GNNmim, a simple yet effective baseline for node classification with incomplete feature data. Experiments show that GNNmim is competitive with respect to specialized architectures across diverse datasets and missingness regimes. --- ### Projected Autoregression: Autoregressive Language Generation in Continuous State Space Authors: Oshri Naparstek Published: 2026-01-08 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.04854v3 Standard autoregressive language models generate text by repeatedly selecting a discrete next token, coupling prediction with irreversible commitment at every step. We show that token selection is not the only viable autoregressive interface. \textbf{Projected Autoregression} replaces token selection with continuous prediction in embedding space followed by discrete projection at commitment time. The model predicts next-token vectors via regression and contrastive objectives, while discrete tokens arise only by nearest-neighbor projection. An optional mutable suffix (``liquid tail'') enables iterative refinement before commitment, but the central change is more basic: next-step prediction is continuous, and discrete tokens are produced only as a downstream interface. Projected Autoregression establishes a concrete alternative to token-selection autoregression: language generation can be organized around continuous-state prediction with delayed discrete commitment. Refinement remains local to a short causal suffix within a left-to-right causal process, rather than a sequence-wide denoising process. This separation has two consequences. First, it induces a \emph{distinct generation regime}: even with immediate projection ($K{=}1$), continuous prediction yields text structure and dynamics that differ from tested token-space AR baselines, including a compute-matched best-of-16 reranking baseline. Second, it exposes a \emph{continuous control surface} inside autoregressive generation: direction rate, history noise, delayed commitment, state-space guidance, and embedding geometry act directly on the evolving generative state before token commitment. Taken together, these results place repeated token selection within a larger family of autoregressive interfaces and expose continuous state space as a broader algorithmic design space for language generation. --- ### RAAR: Retrieval Augmented Agentic Reasoning for Cross-Domain Misinformation Detection Authors: Zhiwei Liu, Runteng Guo, Baojie Qu, Yuechen Jiang, Min Peng, Qianqian Xie, Sophia Ananiadou Published: 2026-01-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.04853v1 Cross-domain misinformation detection is challenging, as misinformation arises across domains with substantial differences in knowledge and discourse. Existing methods often rely on single-perspective cues and struggle to generalize to challenging or underrepresented domains, while reasoning large language models (LLMs), though effective on complex tasks, are limited to same-distribution data. To address these gaps, we introduce RAAR, the first retrieval-augmented agentic reasoning framework for cross-domain misinformation detection. To enable cross-domain transfer beyond same-distribution assumptions, RAAR retrieves multi-perspective source-domain evidence aligned with each target sample's semantics, sentiment, and writing style. To overcome single-perspective modeling and missing systematic reasoning, RAAR constructs verifiable multi-step reasoning paths through specialized multi-agent collaboration, where perspective-specific agents produce complementary analyses and a summary agent integrates them under verifier guidance. RAAR further applies supervised fine-tuning and reinforcement learning to train a single multi-task verifier to enhance verification and reasoning capabilities. Based on RAAR, we trained the RAAR-8b and RAAR-14b models. Evaluation on three cross-domain misinformation detection tasks shows that RAAR substantially enhances the capabilities of the base models and outperforms other cross-domain methods, advanced LLMs, and LLM-based adaptation approaches. The project will be released at https://github.com/lzw108/RAAR. --- ### Character Detection using YOLO for Writer Identification in multiple Medieval books Authors: Alessandra Scotto di Freca, Tiziana D Alessandro, Francesco Fontanella, Filippo Sarria, Claudio De Stefano Published: 2026-01-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.04834v1 Paleography is the study of ancient and historical handwriting, its key objectives include the dating of manuscripts and understanding the evolution of writing. Estimating when a document was written and tracing the development of scripts and writing styles can be aided by identifying the individual scribes who contributed to a medieval manuscript. Although digital technologies have made significant progress in this field, the general problem remains unsolved and continues to pose open challenges. ... We previously proposed an approach focused on identifying specific letters or abbreviations that characterize each writer. In that study, we considered the letter "a", as it was widely present on all pages of text and highly distinctive, according to the suggestions of expert paleographers. We used template matching techniques to detect the occurrences of the character "a" on each page and the convolutional neural network (CNN) to attribute each instance to the correct scribe. Moving from the interesting results achieved from this previous system and being aware of the limitations of the template matching technique, which requires an appropriate threshold to work, we decided to experiment in the same framework with the use of the YOLO object detection model to identify the scribe who contributed to the writing of different medieval books. We considered the fifth version of YOLO to implement the YOLO object detection model, which completely substituted the template matching and CNN used in the previous work. The experimental results demonstrate that YOLO effectively extracts a greater number of letters considered, leading to a more accurate second-stage classification. Furthermore, the YOLO confidence score provides a foundation for developing a system that applies a rejection threshold, enabling reliable writer identification even in unseen manuscripts. --- ### When AI Settles Down: Late-Stage Stability as a Signature of AI-Generated Text Detection Authors: Ke Sun, Guangsheng Bao, Han Cui, Yue Zhang Published: 2026-01-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.04833v1 Zero-shot detection methods for AI-generated text typically aggregate token-level statistics across entire sequences, overlooking the temporal dynamics inherent to autoregressive generation. We analyze over 120k text samples and reveal Late-Stage Volatility Decay: AI-generated text exhibits rapidly stabilizing log probability fluctuations as generation progresses, while human writing maintains higher variability throughout. This divergence peaks in the second half of sequences, where AI-generated text shows 24--32\% lower volatility. Based on this finding, we propose two simple features: Derivative Dispersion and Local Volatility, which computed exclusively from late-stage statistics. Without perturbation sampling or additional model access, our method achieves state-of-the-art performance on EvoBench and MAGE benchmarks and demonstrates strong complementarity with existing global methods. --- ### Illumination Angular Spectrum Encoding for Controlling the Functionality of Diffractive Networks Authors: Matan Kleiner, Lior Michaeli, Tomer Michaeli Published: 2026-01-08 Categories: physics.optics, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2601.04825v1 Diffractive neural networks have recently emerged as a promising framework for all-optical computing. However, these networks are typically trained for a single task, limiting their potential adoption in systems requiring multiple functionalities. Existing approaches to achieving multi-task functionality either modify the mechanical configuration of the network per task or use a different illumination wavelength or polarization state for each task. In this work, we propose a new control mechanism, which is based on the illumination's angular spectrum. Specifically, we shape the illumination using an amplitude mask that selectively controls its angular spectrum. We employ different illumination masks for achieving different network functionalities, so that the mask serves as a unique task encoder. Interestingly, we show that effective control can be achieved over a very narrow angular range, within the paraxial regime. We numerically illustrate the proposed approach by training a single diffractive network to perform multiple image-to-image translation tasks. In particular, we demonstrate translating handwritten digits into typeset digits of different values, and translating handwritten English letters into typeset numbers and typeset Greek letters, where the type of the output is determined by the illumination's angular components. As we show, the proposed framework can work under different coherence conditions, and can be combined with existing control strategies, such as different wavelengths. Our results establish the illumination angular spectrum as a powerful degree of freedom for controlling diffractive networks, enabling a scalable and versatile framework for multi-task all-optical computing. --- ### SOVABench: A Vehicle Surveillance Action Retrieval Benchmark for Multimodal Large Language Models Authors: Oriol Rabasseda, Zenjie Li, Kamal Nasrollahi, Sergio Escalera Published: 2026-01-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.04824v2 Automatic identification of events and recurrent behavior analysis are critical for video surveillance. However, most existing content-based video retrieval benchmarks focus on scene-level similarity and do not evaluate the action discrimination required in surveillance. To address this gap, we introduce SOVABench (Surveillance Opposite Vehicle Actions Benchmark), a real-world retrieval benchmark built from surveillance footage and centered on vehicle-related actions. SOVABench defines two evaluation protocols (inter-pair and intra-pair) to assess cross-action discrimination and temporal direction understanding. Although action distinctions are generally intuitive for human observers, our experiments show that they remain challenging for state-of-the-art vision and multimodal models. Leveraging the visual reasoning and instruction-following capabilities of Multimodal Large Language Models (MLLMs), we present a training-free framework for producing interpretable embeddings from MLLM-generated descriptions for both images and videos. The framework achieves strong performance on SOVABench as well as on several spatial and counting benchmarks where contrastive Vision-Language Models often fail. The code, annotations, and instructions to construct the benchmark are publicly available. --- ### DR-LoRA: Dynamic Rank LoRA for Fine-Tuning Mixture-of-Experts Models Authors: Guanzhi Deng, Bo Li, Ronghao Chen, Xiujin Liu, Zhuo Han, Huacan Wang, Lijie Wen, Linqi Song Published: 2026-01-08 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2601.04823v6 Mixture-of-Experts (MoE) has become a prominent paradigm for scaling Large Language Models (LLMs). Parameter-efficient fine-tuning methods, such as LoRA, are widely adopted to adapt pretrained MoE LLMs to downstream tasks. However, existing approaches typically assign identical LoRA ranks to all expert modules, ignoring the heterogeneous specialization of pretrained experts. This uniform allocation leads to a resource mismatch: task-relevant experts are under-provisioned, while less relevant ones receive redundant parameters. To address this, we propose DR-LoRA, a Dynamic Rank LoRA framework for fine-tuning pretrained MoE models. Specifically, DR-LoRA initializes all expert LoRA modules with a small active rank and uses an expert saliency score, which combines routing frequency and gradient-based rank importance, to identify which experts would benefit most from additional capacity. It then periodically expands the active ranks of the task-critical expert LoRA, progressively constructing a heterogeneous rank distribution tailored to the target task. Experiments on three MoE models across six tasks show that DR-LoRA consistently outperforms LoRA and other strong baselines, demonstrating that task-adaptive heterogeneous rank allocation is an effective strategy to improve active capacity utilization in MoE fine-tuning. --- ### AECV-Bench: Benchmarking Multimodal Models on Architectural and Engineering Drawings Understanding Authors: Aleksei Kondratenko, Mussie Birhane, Houssame E. Hsain, Guido Maciocci Published: 2026-01-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04819v1 AEC drawings encode geometry and semantics through symbols, layout conventions, and dense annotation, yet it remains unclear whether modern multimodal and vision-language models can reliably interpret this graphical language. We present AECV-Bench, a benchmark for evaluating multimodal and vision-language models on realistic AEC artefacts via two complementary use cases: (i) object counting on 120 high-quality floor plans (doors, windows, bedrooms, toilets), and (ii) drawing-grounded document QA spanning 192 question-answer pairs that test text extraction (OCR), instance counting, spatial reasoning, and comparative reasoning over common drawing regions. Object-counting performance is reported using per-field exact-match accuracy and MAPE results, while document-QA performance is reported using overall accuracy and per-category breakdowns with an LLM-as-a-judge scoring pipeline and targeted human adjudication for edge cases. Evaluating a broad set of state-of-the-art models under a unified protocol, we observe a stable capability gradient; OCR and text-centric document QA are strongest (up to 0.95 accuracy), spatial reasoning is moderate, and symbol-centric drawing understanding - especially reliable counting of doors and windows - remains unsolved (often 0.40-0.55 accuracy) with substantial proportional errors. These results suggest that current systems function well as document assistants but lack robust drawing literacy, motivating domain-specific representations and tool-augmented, human-in-the-loop workflows for an efficient AEC automation. --- ### SCALER:Synthetic Scalable Adaptive Learning Environment for Reasoning Authors: Caijun Xu, Changyi Xiao, Zhongyuan Peng, Xinrun Wang, Yixin Cao Published: 2026-01-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04809v5 Reinforcement learning (RL) offers a principled way to enhance the reasoning capabilities of large language models, yet its effectiveness hinges on training signals that remain informative as models evolve. In practice, RL progress often slows when task difficulty becomes poorly aligned with model capability, or when training is dominated by a narrow set of recurring problem patterns. To jointly address these issues, we propose SCALER (Synthetic sCalable Adaptive Learning Environment for Reasoning), a framework that sustains effective learning signals through adaptive environment design. SCALER introduces a scalable synthesis pipeline that converts real-world programming problems into verifiable reasoning environments with controllable difficulty and unbounded instance generation, enabling RL training beyond finite datasets while preserving strong correctness guarantees. Building on this, SCALER further employs an adaptive multi-environment RL strategy that dynamically adjusts instance difficulty and curates the active set of environments to track the model's capability frontier and maintain distributional diversity. This co-adaptation prevents reward sparsity, mitigates overfitting to narrow task patterns, and supports sustained improvement throughout training. Extensive experiments show that SCALER consistently outperforms dataset-based RL baselines across diverse reasoning benchmarks and exhibits more stable, long-horizon training dynamics. --- ### Comparison of Maximum Likelihood Classification Before and After Applying Weierstrass Transform Authors: Muhammad Shoaib, Zaka Ur Rehman, Muhammad Qasim Published: 2026-01-08 Categories: stat.AP, cs.LG, math.PR arXiv: https://arxiv.org/abs/2601.04808v1 The aim of this paper is to use Maximum Likelihood (ML) Classification on multispectral data by means of qualitative and quantitative approaches. Maximum Likelihood is a supervised classification algorithm which is based on the Classical Bayes theorem. It makes use of a discriminant function to assign pixel to the class with the highest likelihood. Class means vector and covariance matrix are the key inputs to the function and can be estimated from training pixels of a particular class. As Maximum Likelihood need some assumptions before it has to be applied on the data. In this paper we will compare the results of Maximum Likelihood Classification (ML) before apply the Weierstrass Transform and apply Weierstrass Transform and will see the difference between the accuracy on training pixels of high resolution Quickbird satellite image. Principle Component analysis (PCA) is also used for dimension reduction and also used to check the variation in bands. The results shows that the separation between mean of the classes in the decision space is to be the main factor that leads to the high classification accuracy of Maximum Likelihood (ML) after using Weierstrass Transform than without using it. --- ### Parallelizing Node-Level Explainability in Graph Neural Networks Authors: Oscar Llorente, Jaime Boal, Eugenio F. Sánchez-Úbeda, Antonio Diaz-Cano, Miguel Familiar Published: 2026-01-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.04807v1 Graph Neural Networks (GNNs) have demonstrated remarkable performance in a wide range of tasks, such as node classification, link prediction, and graph classification, by exploiting the structural information in graph-structured data. However, in node classification, computing node-level explainability becomes extremely time-consuming as the size of the graph increases, while batching strategies often degrade explanation quality. This paper introduces a novel approach to parallelizing node-level explainability in GNNs through graph partitioning. By decomposing the graph into disjoint subgraphs, we enable parallel computation of explainability for node neighbors, significantly improving the scalability and efficiency without affecting the correctness of the results, provided sufficient memory is available. For scenarios where memory is limited, we further propose a dropout-based reconstruction mechanism that offers a controllable trade-off between memory usage and explanation fidelity. Experimental results on real-world datasets demonstrate substantial speedups, enabling scalable and transparent explainability for large-scale GNN models. --- ### Thinking-Based Non-Thinking: Solving the Reward Hacking Problem in Training Hybrid Reasoning Models via Reinforcement Learning Authors: Siyuan Gan, Jiaheng Liu, Boyan Wang, Tianpei Yang, Runqing Miao, Yuyao Zhang, Fanyu Meng, Junlan Feng, Linjian Meng, Jing Huo, Yang Gao Published: 2026-01-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04805v2 Large reasoning models (LRMs) have attracted much attention due to their exceptional performance. However, their performance mainly stems from thinking, a long Chain of Thought (CoT), which significantly increase computational overhead. To address this overthinking problem, existing work focuses on using reinforcement learning (RL) to train hybrid reasoning models that automatically decide whether to engage in thinking or not based on the complexity of the query. Unfortunately, using RL will suffer the the reward hacking problem, e.g., the model engages in thinking but is judged as not doing so, resulting in incorrect rewards. To mitigate this problem, existing works either employ supervised fine-tuning (SFT), which incurs high computational costs, or enforce uniform token limits on non-thinking responses, which yields limited mitigation of the problem. In this paper, we propose Thinking-Based Non-Thinking (TNT). It does not employ SFT, and sets different maximum token usage for responses not using thinking across various queries by leveraging information from the solution component of the responses using thinking. Experiments on five mathematical benchmarks demonstrate that TNT reduces token usage by around 50% compared to DeepSeek-R1-Distill-Qwen-1.5B/7B and DeepScaleR-1.5B, while significantly improving accuracy. In fact, TNT achieves the optimal trade-off between accuracy and efficiency among all tested methods. Additionally, the probability of reward hacking problem in TNT's responses, which are classified as not using thinking, remains below 10% across all tested datasets. ---