# Research Papers — 2026-01 · bundle page 170 of 189 > Full-text markdown of papers #8451–8500 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/170 [← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/169.md) [Next bundle →](https://aipod.dev/papers/archive/2026-01/page/171.md) ### Hierarchical temporal receptive windows and zero-shot timescale generalization in biologically constrained scale-invariant deep networks Authors: Aakash Sarkar, Marc W. Howard Published: 2026-01-06 Categories: q-bio.NC, cs.AI, cs.CL, cs.LG, cs.NE arXiv: https://arxiv.org/abs/2601.02618v1 Human cognition integrates information across nested timescales. While the cortex exhibits hierarchical Temporal Receptive Windows (TRWs), local circuits often display heterogeneous time constants. To reconcile this, we trained biologically constrained deep networks, based on scale-invariant hippocampal time cells, on a language classification task mimicking the hierarchical structure of language (e.g., 'letters' forming 'words'). First, using a feedforward model (SITHCon), we found that a hierarchy of TRWs emerged naturally across layers, despite the network having an identical spectrum of time constants within layers. We then distilled these inductive priors into a biologically plausible recurrent architecture, SITH-RNN. Training a sequence of architectures ranging from generic RNNs to this restricted subset showed that the scale-invariant SITH-RNN learned faster with orders-of-magnitude fewer parameters, and generalized zero-shot to out-of-distribution timescales. These results suggest the brain employs scale-invariant, sequential priors - coding "what" happened "when" - making recurrent networks with such priors particularly well-suited to describe human cognition. --- ### Reliable conformal novelty detection at the decision boundary Authors: Zijun Gao, Etienne Roquain, Daniel Xiang Published: 2026-01-06 Categories: stat.ME, stat.ML arXiv: https://arxiv.org/abs/2601.02610v3 Novelty detection via conformal $p$-values and BH procedure provides distribution-free global false discovery rate (FDR) control. We present here fundamental limits of this approach by showing that it does not produce reliable detection at the decision boundary. We study boundary false discovery rate (bFDR), the probability that the least extreme reported novelty is in fact a null observation. We first show that the support line (SL) procedure, controlling the bFDR in the continuous independent framework, fails to control the bFDR in the conformal case. We then present several modifications of the SL procedure that restore reliability at the decision boundary, by controlling the bFDR, with specific improvements in situations where many novelties are expected (adaptive procedures) and when the calibration sample is too small with respect to the test sample (subsampled procedures). Numerical experiments with both synthetic and real data support our findings and show the relevance of the new proposed approach. --- ### Chronicals: A High-Performance Framework for LLM Fine-Tuning with 3.51x Speedup over Unsloth Authors: Arjun S. Nair Published: 2026-01-06 Categories: cs.LG, cs.AI, cs.CL, cs.DC, stat.ML arXiv: https://arxiv.org/abs/2601.02609v1 Large language model fine-tuning is bottlenecked by memory: a 7B parameter model requires 84GB--14GB for weights, 14GB for gradients, and 56GB for FP32 optimizer states--exceeding even A100-40GB capacity. We present Chronicals, an open-source training framework achieving 3.51x speedup over Unsloth through four synergistic optimizations: (1) fused Triton kernels eliminating 75% of memory traffic via RMSNorm (7x), SwiGLU (5x), and QK-RoPE (2.3x) fusion; (2) Cut Cross-Entropy reducing logit memory from 5GB to 135MB through online softmax computation; (3) LoRA+ with theoretically-derived 16x differential learning rates between adapter matrices; and (4) Best-Fit Decreasing sequence packing recovering 60-75% of compute wasted on padding. On Qwen2.5-0.5B with A100-40GB, Chronicals achieves 41,184 tokens/second for full fine-tuning versus Unsloth's 11,736 tokens/second (3.51x). For LoRA at rank 32, we reach 11,699 tokens/second versus Unsloth MAX's 2,857 tokens/second (4.10x). Critically, we discovered that Unsloth's reported 46,000 tokens/second benchmark exhibited zero gradient norms--the model was not training. We provide complete mathematical foundations: online softmax correctness proofs, FlashAttention IO complexity bounds O(N^2 d^2 M^{-1}), LoRA+ learning rate derivations from gradient magnitude analysis, and bin-packing approximation guarantees. All implementations, benchmarks, and proofs are available at https://github.com/Ajwebdevs/Chronicals with pip installation via https://pypi.org/project/chronicals/. --- ### ELF: A Family of Encoder-Free ECG-Language Models Authors: William Han, Tony Chen, Chaojing Duan, Xiaoyu Song, Yihang Yao, Yuzhe Yang, Michael A. Rosenberg, Emerson Liu, Ding Zhao Published: 2026-01-05 Categories: cs.MM, cs.AI arXiv: https://arxiv.org/abs/2601.18798v3 ECG-Language Models (ELMs) extend recent advances in Multimodal Large Language Models (MLLMs) to automated ECG interpretation. However, most existing ELMs inherit Vision-Language Model (VLM) design choices and rely on pretrained ECG encoders, introducing substantial architectural and training complexity. Inspired by encoder-free VLMs, we introduce ELF, a family of three encoder-free ELMs that remain competitive with, and often outperform, prior state-of-the-art ELMs across two datasets despite substantially simpler architectures and training pipelines. All code and data are available at github.com/ELM-Research/ECG-Language-Models. --- ### Evaluating Reward Model Generalization via Pairwise Maximum Discrepancy Competitions Authors: Shunyang Luo, Peibei Cao, Zhihui Zhu, Kehua Feng, Zhihua Wang, Keyan Ding Published: 2026-01-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.16987v1 Reward models (RMs) are central to aligning large language models, yet their practical effectiveness hinges on generalization to unseen prompts and shifting distributions. Most existing RM evaluations rely on static, pre-annotated preference datasets, which provide limited coverage and often fail to faithfully assess generalization in open-world settings. We introduce Pairwise Maximum Discrepancy Competition (PMDC), a dynamic and annotation-efficient framework for evaluating RM generalization using a large, unlabeled, open-domain prompt pool. PMDC actively selects prompt--response pairs that maximize disagreement between two RMs, yielding a compact set of highly contentious test cases. These cases are adjudicated by an oracle, and the resulting outcomes are aggregated via a Bradley--Terry model to produce a global ranking and pairwise win-rate landscape of RMs. We apply PMDC to re-evaluate 10 representative RMs and observe substantial rank reshuffling compared with conventional benchmarks. Qualitative analyses further uncover systematic generalization failures, providing valuable insights for improving reward modeling. --- ### Crystal-KV: Efficient KV Cache Management for Chain-of-Thought LLMs via Answer-First Principle Authors: Zihan Wang, Cheng Tang, Lei Gong, Cheng Li, Chao Wang, teng wang, Wenqi Lou, Xuehai Zhou Published: 2026-01-05 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.16986v1 Chain-of-Thought (CoT) reasoning in large language models (LLMs) significantly improves accuracy on complex tasks, yet incurs excessive memory overhead due to the long think-stage sequences stored in the Key-Value (KV) cache. Unlike traditional generation tasks where all tokens are uniformly important, CoT emphasizes the final answer, rendering conventional KV compression strategies ineffective. In this paper, we present Crystal-KV, an efficient KV cache management framework tailored for CoT reasoning. Our key insight is the answer-first principle. By mapping answer preferences into think-stage attention map, we distinguish between SlipKV, which mainly maintains the reasoning flow but may occasionally introduce misleading context, and CrystalKV, which truly contributes to the correctness of the final answer. Next, we propose an attention-based Least Recently Frequently Used algorithm. It precisely identifies when a SlipKV entry's utility expires and evicts it, retaining CrystalKV without disrupting reasoning flow. Finally, we introduce an adaptive cache budget allocation algorithm. Based on the dynamic proportion of CrystalKV, it estimates the importance of each layer/head and adjusts the KV cache budget during inference, amplifying critical components to improve budget utilization. Results show that Crystal-KV achieves state-of-the-art KV cache compression, significantly improves throughput, and enables faster response time, while maintaining, or even improving, answer accuracy for CoT reasoning. --- ### Domain Specific Specialization in Low-Resource Settings: The Efficacy of Offline Response-Based Knowledge Distillation in Large Language Models Authors: Erdem Aslan, Pakize Erdoğmuş Published: 2026-01-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.16219v1 Large Language Models (LLMs) excel in general tasks but often struggle with hallucinations when handling domain-specific or institutional knowledge absent from their pre-training. We present an offline response-based knowledge distillation method that develops high-accuracy specialized assistants under constrained hardware resources. We evaluate three distinct data strategies: general domain adaptation (15,000 lines), unstructured knowledge injection (2,000 lines), and a context-aware synthetic dataset (500 lines) generated by a teacher model. To minimize computational costs, we utilize the Unsloth library to optimize the Qwen-2.5-7B student model, reducing NVIDIA A100 GPU memory requirements from 40 GB to 16 GB. Experimental results demonstrate that while larger unstructured datasets suffer from persistent hallucinations, the 500-line context-aware dataset achieves a 96.7% accuracy rate and robust rejection capability. These findings validate the LIMA hypothesis, showing that data quality and structural alignment are more critical than quantity for domain adaptation in low-resource settings. --- ### EPD-Serve: A Flexible Multimodal EPD Disaggregation Inference Serving System On Ascend Authors: Fan Bai, Pai Peng, Zhengzhi Tang, Zhe Wang, Gong Chen, Xiang Lu, Yinuo Li, Huan Lin, Weizhe Lin, Yaoyuan Wang, Xiaosong Li Published: 2026-01-05 Categories: cs.DC, cs.AI arXiv: https://arxiv.org/abs/2601.11590v1 With the widespread adoption of large multimodal models, efficient inference across text, image, audio, and video modalities has become critical. However, existing multimodal inference systems typically employ monolithic architectures that tightly couple the Encode, Prefill, and Decode stages on homogeneous hardware, neglecting the heterogeneous computational characteristics of each stage. This design leads to inefficient resource utilization and limited system throughput. To address these issues, we propose EPD-Serve, a stage-level disaggregated inference serving system for multimodal models. EPD-Serve decouples the inference pipeline into independent Encode, Prefill, and Decode stages, enabling logical isolation and flexible co-located deployment through dynamic orchestration. Leveraging the Ascend interconnect topology, EPD-Serve introduces asynchronous feature prefetching between Encode and Prefill stages and a hierarchical grouped KV cache transmission mechanism between Prefill and Decode stages to improve cross-node communication efficiency. In addition, EPD-Serve incorporates multi-route scheduling, instance-level load balancing, and multi-stage hardware co-location with spatial multiplexing to better support diverse multimodal workloads. Comprehensive experiments on multimodal understanding models demonstrate that, under high-concurrency scenarios, EPD-Serve improves end-to-end throughput by 57.37-69.48% compared to PD-disaggregated deployment, while satisfying strict SLO constraints, including TTFT below 2000 ms and TPOT below 50 ms. These results highlight the effectiveness of stage-level disaggregation for optimizing multimodal large model inference systems. --- ### OrbitFlow: SLO-Aware Long-Context LLM Serving with Fine-Grained KV Cache Reconfiguration Authors: Xinyue Ma, Heelim Hong, Taegeon Um, Jongseop Lee, Seoyeong Choy, Woo-Yeon Lee, Myeongjae Jeon Published: 2026-01-05 Categories: cs.AI, cs.LG, cs.PF arXiv: https://arxiv.org/abs/2601.10729v2 Serving long-context LLMs is challenging because request lengths and batch composition vary during token generation, causing the memory footprint to fluctuate significantly at runtime. Offloading KV caches to host memory limits effective memory usage, but existing static and predetermined offloading strategies cannot adapt to the rapidly shifting memory demands of long-context serving. This often leads to excessive CPU-to-GPU KV transfers that translate into latency spikes and frequent SLO violations. To address these challenges, we introduce OrbitFlow, a fine-grained and adaptive KV cache management system that meets latency SLOs in long-context LLM serving. OrbitFlow employs a lightweight ILP solver to decide which layers' KV caches to retain on the GPU for each request, within memory capacity constraints. It continuously refines KV placements based on runtime feedback when the active plan becomes suboptimal during token generation. Under heavy load, OrbitFlow invokes a fallback mechanism to temporarily defer in-flight requests with large memory footprints, preserving overall SLO attainment. Our experiments demonstrate that OrbitFlow improves SLO attainment for TPOT and TBT by up to 66% and 48%, respectively, while reducing the 95th percentile latency by 38% and achieving up to 3.3x higher throughput compared to existing offloading methods. --- ### A Foundation Model Approach for Fetal Stress Prediction During Labor From cardiotocography (CTG) recordings Authors: Naomi Fridman, Berta Ben Shachar Published: 2026-01-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.06149v1 Intrapartum cardiotocography (CTG) is widely used for fetal monitoring during labor, yet its interpretation suffers from high inter-observer variability and limited predictive accuracy. Deep learning approaches have been constrained by the scarcity of CTG recordings with clinical outcome labels. We present the first application of self-supervised pre-training to intrapartum CTG analysis, leveraging 2,444 hours of unlabeled recordings for masked pre-training followed by fine-tuning on the 552-recording CTU-UHB benchmark. Using a PatchTST transformer architecture with a channel-asymmetric masking scheme designed for fetal heart rate reconstruction, we achieve an area under the receiver operating characteristic curve of 0.83 on the full test set and 0.853 on uncomplicated vaginal deliveries, exceeding previously reported results on this benchmark (0.68-0.75). Error analysis reveals that false-positive alerts typically correspond to CTG patterns judged concerning on retrospective clinical review, suggesting clinically meaningful predictions even when umbilical pH is normal. We release standardized dataset splits and model weights to enable reproducible benchmarking. Our results demonstrate that self-supervised pre-training can address data scarcity in fetal monitoring, offering a path toward reliable decision support in the labor room. --- ### LLM Flow Processes for Text-Conditioned Regression Authors: Felix Biggs, Samuel Willis Published: 2026-01-05 Categories: cs.LG, cs.CL, stat.ML arXiv: https://arxiv.org/abs/2601.06147v2 Recent work has demonstrated surprisingly good performance of pre-trained LLMs on regression tasks (for example, time-series prediction), with the ability to incorporate expert prior knowledge and the information contained in textual metadata. However we observe major error cascades even in short sequences < ~100 points; these models are also computationally intensive and difficult to parallelise. Marginal LLM predictions do not suffer this issue and are trivially parallelised, but can predict over-broad densities. To address this, we propose combining these densities with a lightweight (diffusion-based) neural process. We show that this combination leads to better-calibrated predictions overall, outputs locally consistent trajectories, and leads to text-conditioned function space selection in the meta-learner. As part of this work we propose a gradient-free (and non-Monte Carlo) method for sampling from a product-of-experts of a score model and an 'expert' (here the LLM predictive densities). We believe this general method is of independent interest as it is applicable whenever an expert can be convolved with a Gaussian in closed form. --- ### Bridging the AI divide in sub-Saharan Africa: Challenges and opportunities for inclusivity Authors: Masike Malatji Published: 2026-01-05 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2601.06145v1 The artificial intelligence (AI) digital divide in sub-Saharan Africa (SSA) presents significant disparities in AI access, adoption, and development due to varying levels of infrastructure, education, and policy support. This study investigates the extent of AI readiness among the top SSA countries using the 2024 Government AI Readiness Index, alongside an analysis of AI initiatives to foster inclusivity. A comparative analysis of AI readiness scores highlights disparities across nations, with Mauritius (53.94) and South Africa (52.91) leading, while Zambia (42.58) and Uganda (43.32) lag. Quartile analysis reveals a concentration of AI preparedness among a few nations, suggesting uneven AI development. The study further examines the relationship between AI readiness and economic indicators, identifying instances where AI progress does not strictly correlate with Gross Domestic Product per capita, as seen in Rwanda and Uganda. Using case studies of AI initiatives across SSA, this research contextualises quantitative findings, identifying key strategies contributing to AI inclusivity, including talent development programs, research networks, and policy interventions. The study concludes with recommendations to bridge the AI digital divide, emphasising investments in AI education, localised AI solutions, and cross-country collaborations to accelerate AI adoption in SSA. --- ### The Patient/Industry Trade-off in Medical Artificial Intelligence Authors: Rina Khan, Annabelle Sauve, Imaan Bayoumi, Amber L. Simpson, Catherine Stinson Published: 2026-01-05 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2601.06144v1 Artificial intelligence (AI) in healthcare has led to many promising developments; however, increasingly, AI research is funded by the private sector leading to potential trade-offs between benefits to patients and benefits to industry. Health AI practitioners should prioritize successful adaptation into clinical practice in order to provide meaningful benefits to patients, but translation usually requires collaboration with industry. We discuss three features of AI studies that hamper the integration of AI into clinical practice from the perspective of researchers and clinicians. These include lack of clinically relevant metrics, lack of clinical trials and longitudinal studies to validate results, and lack of patient and physician involvement in the development process. For partnerships between industry and health research to be sustainable, a balance must be established between patient and industry benefit. We propose three approaches for addressing this gap: improved transparency and explainability of AI models, fostering relationships with industry partners that have a reputation for centering patient benefit in their practices, and prioritization of overall healthcare benefits. With these priorities, we can sooner realize meaningful AI technologies used by clinicians where mutua --- ### Emergent Complexity in Nuclear Reaction Networks: A Study of Stellar Nucleosynthesis through Chemical Organization Theory Authors: Pedro Maldonado-Lang, Clément Vidal Published: 2026-01-05 Categories: astro-ph.SR, cs.AI arXiv: https://arxiv.org/abs/2601.06143v1 We explore the emergence of complex structures within reaction networks, focusing on nuclear reaction networks relevant to stellar nucleosynthesis. The work presents a theoretical framework rooted in Chemical Organization Theory (COT) to characterize how stable, self-sustaining structures arise from the interactions of basic components. Key theoretical contributions include the formalization of atom sets as fundamental reactive units and the concept of synergy to describe the emergence of new reactions and species from the interaction of these units. The property of separability is defined to distinguish dynamically coupled systems from those that can be decomposed. This framework is then applied to the STARLIB nuclear reaction network database, analyzing how network structure, particularly the formation and properties of atom sets and semi-self-maintaining sets, changes as a function of temperature. Results indicate that increasing temperature generally enhances network cohesion, leading to fewer, larger atom sets. Critical temperatures are identified where significant structural reorganizations occur, such as the merging of distinct clusters of atom sets and the disappearance of small, isolated reactive units. The analysis reveals core clusters - large (containing more that 1000 reactions), semi-self-maintaining structures that appear to form the core of all potentially stable nucleosynthetic configurations at various temperatures. Overall, the paper provides insights into the structural underpinnings of stability and emergence in complex reaction networks, with specific implications for understanding stellar evolution and nucleosynthesis. --- ### Is Sanskrit the most token-efficient language? A quantitative study using GPT, Gemini, and SentencePiece Authors: Anshul Kumar Published: 2026-01-05 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.06142v1 Tokens are the basic units of Large Language Models (LLMs). LLMs rely on tokenizers to segment text into these tokens, and tokenization is the primary determinant of computational and inference cost. Sanskrit, one of the oldest languages, is hypothesized to express more meaning per token due to its morphology and grammar rules; however, no prior work has quantified this. We use a dataset of 701 parallel verses of the Bhagavad Gita, which comprises three languages-Sanskrit, English, and Hindi along with transliteration of Sanskrit into English. We test tokenizers including SentencePiece (SPM), older GPT models, and the latest generation tokenizers from Gemini and GPT. We use metrics of token count, characters per token (token efficiency), and tokens per character (token cost). Results show a ~2x difference in token counts between Sanskrit and English/Hindi under the unbiased SPM baseline. English/Hindi translations of Sanskrit commentary resulted in an approximately 20x increase in token count. GPT o200k base (latest, used by GPT-4o) and Gemini (latest) reduce bias by a significant degree compared to GPT cl100k base (used until GPT-4), but still fail to fully capture Sanskrit's compactness. This matters because there might be a penalty bias for non-English users, which inflates the token count. This research provides a foundation for improving future tokenizer design and shows the potential of Sanskrit for highly compact encoding, saving on cost while speeding up training and inference. The code and dataset are available at https://github.com/anshulkr713/sanskrit-token-efficiency --- ### Causal and Federated Multimodal Learning for Cardiovascular Risk Prediction under Heterogeneous Populations Authors: Rohit Kaushik, Eva Kaushik Published: 2026-01-05 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2601.06140v1 Cardiovascular disease (CVD) continues to be the major cause of death globally, calling for predictive models that not only handle diverse and high-dimensional biomedical signals but also maintain interpretability and privacy. We create a single multimodal learning framework that integrates cross modal transformers with graph neural networks and causal representation learning to measure personalized CVD risk. The model combines genomic variation, cardiac MRI, ECG waveforms, wearable streams, and structured EHR data to predict risk while also implementing causal invariance constraints across different clinical subpopulations. To maintain transparency, we employ SHAP based feature attribution, counterfactual explanations and causal latent alignment for understandable risk factors. Besides, we position the design in a federated, privacy, preserving optimization protocol and establish rules for convergence, calibration and uncertainty quantification under distributional shift. Experimental studies based on large-scale biobank and multi institutional datasets reveal state discrimination and robustness, exhibiting fair performance across demographic strata and clinically distinct cohorts. This study paves the way for a principled approach to clinically trustworthy, interpretable and privacy respecting CVD prediction at the population level. --- ### Low-Back Pain Physical Rehabilitation by Movement Analysis in Clinical Trial Authors: Sao Mai Nguyen Published: 2026-01-05 Categories: cs.CV, cs.HC, cs.RO arXiv: https://arxiv.org/abs/2601.06138v1 To allow the development and assessment of physical rehabilitation by an intelligent tutoring system, we propose a medical dataset of clinical patients carrying out low back-pain rehabilitation exercises and benchmark on state of the art human movement analysis algorithms. This dataset is valuable because it includes rehabilitation motions in a clinical setting with patients in their rehabilitation program. This paper introduces the Keraal dataset, a clinically collected dataset to enable intelligent tutoring systems (ITS) for rehabilitation. It addresses four challenges in exercise monitoring: motion assessment, error recognition, spatial localization, temporal localization --- ### RainBalance: Alleviating Dual Imbalance in GNSS-based Precipitation Nowcasting via Continuous Probability Modeling Authors: Yifang Zhang, Shengwu Xiong, Henan Wang, Wenjie Yin, Jiawang Peng, Duan Zhou, Yuqiang Zhang, Chen Zhou, Hua Chen, Qile Zhao, Pengfei Duan Published: 2026-01-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.06137v1 Global navigation satellite systems (GNSS) station-based Precipitation Nowcasting aims to predict rainfall within the next 0-6 hours by leveraging a GNSS station's historical observations of precipitation, GNSS-PWV, and related meteorological variables, which is crucial for disaster mitigation and real-time decision-making. In recent years, time-series forecasting approaches have been extensively applied to GNSS station-based precipitation nowcasting. However, the highly imbalanced temporal distribution of precipitation, marked not only by the dominance of non-rainfall events but also by the scarcity of extreme precipitation samples, significantly limits model performance in practical applications. To address the dual imbalance problem in precipitation nowcasting, we propose a continuous probability modeling-based framework, RainBalance. This plug-and-play module performs clustering for each input sample to obtain its cluster probability distribution, which is further mapped into a continuous latent space via a variational autoencoder (VAE). By learning in this continuous probabilistic space, the task is reformulated from fitting single and imbalance-prone precipitation labels to modeling continuous probabilistic label distributions, thereby alleviating the imbalance issue. We integrate this module into multiple state-of-the-art models and observe consistent performance gains. Comprehensive statistical analysis and ablation studies further validate the effectiveness of our approach. --- ### Attention in Geometry: Scalable Spatial Modeling via Adaptive Density Fields and FAISS-Accelerated Kernels Authors: Zhaowen Fan, Yunxiang Han Published: 2026-01-05 Categories: cs.LG, cs.CV, cs.GR arXiv: https://arxiv.org/abs/2601.06135v3 Spatial computation in geographic systems increasingly requires query-conditioned, local, interpretable aggregation under metric constraints. Many classical approaches rely on global summation and treat approximation as an implementation concern, limiting interpretability and scalability at large scales. We propose the Adaptive Density Field (ADF), a geometric attention framework that formulates spatial aggregation as a query-conditioned, metric-induced attention operator in continuous space. Given a set of labelled spatial points with associated scalar scores, ADF defines a continuous intensity field over space. For a given query location, the field value is obtained via a local adaptive Gaussian kernel mixture centered on the query's nearest neighbors, where kernel bandwidths are modulated by point-specific scores to evaluate local aggregated influence. Additionally, approximate nearest-neighbor search is introduced, enabling scalable execution while preserving locality. The proposed ADF bridges concepts from adaptive kernel methods, classical GIS methods, and attention mechanisms by reinterpreting spatial influence as geometry-embedded attention, grounded in physical distance rather than learned latent projections. The proposed framework is formulation-level rather than algorithm-specific, allowing flexible kernel choices, score-to-bandwidth mappings, and approximation parameters. This approach provides a unifying perspective on spatial influence modeling that emphasizes structure, scalability, and geometric interpretability, with relevance to geographic information systems and spatial machine learning. --- ### DeeperBrain: A Neuro-Grounded EEG Foundation Model Towards Universal BCI Authors: Jiquan Wang, Sha Zhao, Yangxuan Zhou, Yiming Kang, Shijian Li, Gang Pan Published: 2026-01-05 Categories: cs.LG, eess.SP, q-bio.NC arXiv: https://arxiv.org/abs/2601.06134v2 Electroencephalography (EEG) foundation models hold significant promise for universal Brain-Computer Interfaces (BCIs). However, existing approaches often rely on end-to-end fine-tuning and exhibit limited efficacy under frozen-probing protocols, lacking the intrinsic universality required for broad generalization. This limitation stems from adapting general-purpose sequence architectures that overlook the biophysical and dynamical principles of neural activity. To bridge this gap, we propose DeeperBrain, a neuro-grounded foundation model integrating domain-specific inductive biases into its model design and learning objectives. Architecturally, DeeperBrain incorporates a volume conduction-aware channel encoding to model spatial mixing via 3D geometry, and a neurodynamics-aware temporal encoding capturing slow adaptations using oscillatory and exponential bases. For pretraining, we introduce a dual-objective strategy combining Masked EEG Reconstruction (MER) for local fidelity and Neurodynamics Statistics Prediction (NSP). NSP enforces alignment with macroscopic brain states by predicting interpretable order parameters, including spectral power, functional connectivity, cross-frequency coupling, and dynamic complexity. Extensive experiments demonstrate that DeeperBrain achieves state-of-the-art or highly competitive performance under end-to-end fine-tuning. Crucially, it maintains superior efficacy under a rigorous frozen-probing protocol, verifying that embedding neuroscientific first principles endows learned representations with the intrinsic universality essential for universal BCI. The code will be publicly available. --- ### A Review of Online Diffusion Policy RL Algorithms for Scalable Robotic Control Authors: Wonhyeok Choi, Shutong Ding, Minwoo Choi, Jungwan Woo, Kyumin Hwang, Jaeyeul Kim, Ye Shi, Sunghoon Im Published: 2026-01-05 Categories: cs.LG, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2601.06133v2 Diffusion policies have emerged as a powerful approach for robotic control, demonstrating superior expressiveness in modeling multimodal action distributions compared to conventional policy networks. However, their integration with online reinforcement learning remains challenging due to fundamental incompatibilities between diffusion model training objectives and standard RL policy improvement mechanisms. This paper presents the first comprehensive review and empirical analysis of current Online Diffusion Policy Reinforcement Learning (Online DPRL) algorithms for scalable robotic control systems. We propose a novel taxonomy that categorizes existing approaches into four distinct families--Action-Gradient, Q-Weighting, Proximity-Based, and Backpropagation Through Time (BPTT) methods--based on their policy improvement mechanisms. Through extensive experiments on a unified NVIDIA Isaac Lab benchmark encompassing 12 diverse robotic tasks, we systematically evaluate representative algorithms across five critical dimensions: task diversity, parallelization capability, diffusion step scalability, cross-embodiment generalization, and environmental robustness. Our analysis identifies key findings regarding the fundamental trade-offs inherent in each algorithmic family, particularly concerning sample efficiency and scalability. Furthermore, we reveal critical computational and algorithmic bottlenecks that currently limit the practical deployment of online DPRL. Based on these findings, we provide concrete guidelines for algorithm selection tailored to specific operational constraints and outline promising future research directions to advance the field toward more general and scalable robotic learning systems. --- ### An evaluation of LLMs for political bias in Western media: Israel-Hamas and Ukraine-Russia wars Authors: Rohitash Chandra, Haoyan Chen, Yaqing Zhang, Jiacheng Chen, Yuting Wu Published: 2026-01-05 Categories: cs.CY, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2601.06132v1 Political bias in media plays a critical role in shaping public opinion, voter behaviour, and broader democratic discourse. Subjective opinions and political bias can be found in media sources, such as newspapers, depending on their funding mechanisms and alliances with political parties. Automating the detection of political biases in media content can limit biases in elections. The impact of large language models (LLMs) in politics and media studies is becoming prominent. In this study, we utilise LLMs to compare the left-wing, right-wing, and neutral political opinions expressed in the Guardian and BBC. We review newspaper reporting that includes significant events such as the Russia-Ukraine war and the Hamas-Israel conflict. We analyse the proportion for each opinion to find the bias under different LLMs, including BERT, Gemini, and DeepSeek. Our results show that after the outbreak of the wars, the political bias of Western media shifts towards the left-wing and each LLM gives a different result. DeepSeek consistently showed a stable Left-leaning tendency, while BERT and Gemini remained closer to the Centre. The BBC and The Guardian showed distinct reporting behaviours across the two conflicts. In the Russia-Ukraine war, both outlets maintained relatively stable positions; however, in the Israel-Hamas conflict, we identified larger political bias shifts, particularly in Guardian coverage, suggesting a more event-driven pattern of reporting bias. These variations suggest that LLMs are shaped not only by their training data and architecture, but also by underlying worldviews with associated political biases. --- ### Large Language Models are Shannon Lossy Compressors Not Solomonoff Induction Estimators: Self-improvement and Singularity Are Not Near Without Symbolic Model Synthesis Authors: Hector Zenil, Abicumaran Uthamacumaran, Luan Ozelim Published: 2026-01-05 Categories: cs.IT, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.05280v6 We connect two questions in Algorithmic Information Theory (AIT), Machine Learning (ML) and Artificial General Intelligence (AGI): whether LLMs estimate Solomonoff induction, and whether they can self-improve towards an AI Singularity. We provide theoretical, methodological and empirical answers in the negative but show how limits can be circumvented. Cross-entropy, negative log-likelihood and related next-token objectives cannot alone implement Solomonoff induction: they fit supplied conditionals rather than a program-weighted universal mixture. More computation can improve fit within a fixed objective but cannot change its inductive principle without external hyperparameter or architectural tuning; they alone do not deliver Solomonoff-Levin optimal prediction. For finite learners and observers, theoretical boundaries become less decisive and approaches diverge. Resource-bounded estimators are finite mechanism-search tools whose divergence does not violate algorithmic information conservation. All 26 served language-model checkpoints across five pre-training families, 0.8-35 billion parameters and 1.9-8.5 bits per weight, evaluated at their commitments over a closed alphabet, violate the dominance guarantee defining a universal mixture. Against a 3.32-bit bound attained by a genuine mixture, the best model trails a Krichevsky-Trofimov code by 4.5 bits, the median by 36 and the worst by 128; excess grows to every stream's end rather than settling to a constant. Served conditionals fail to form a mixture over the declared class in 79 of 91 checkpoint-designs; neither scale nor post-training closes the gap. Frontier developers adopt neurosymbolic approaches, including Fable and Astra, incorporating model synthesis via neurosymbolic computation. They are no longer purely statistical LLMs, making them better, though still limited, candidates for higher forms of induction & model synthesis. --- ### Solving Cyclic Antibandwidth Problem by SAT Authors: Hieu Truong Xuan, Khanh To Van Published: 2026-01-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.04239v1 The Cyclic Antibandwidth Problem (CABP), a variant of the Antibandwidth Problem, is an NP-hard graph labeling problem with numerous applications. Despite significant research efforts, existing state-of-the-art approaches for CABP are exclusively heuristic or metaheuristic in nature, and exact methods have been limited to restricted graph classes. In this paper, we present the first exact approach for the CABP on general graphs, based on SAT solving, called SAT-CAB. The proposed method is able to systematically explore the solution space and guarantee global optimality, overcoming the limitations of previously reported heuristic algorithms. This approach relies on a novel and efficient SAT encoding of CABP, in which the problem is transformed into a sequence of At-Most-One constraints. In particular, we introduce a compact representation of the At-Most-One constraints inherent to CABP, which significantly reduces the size of the resulting formulas and enables modern SAT solvers to effectively explore the solution space and to certify global optimality. Extensive computational experiments on standard benchmark instances show that the proposed method efficiently solves CABP instances of practical relevance, while identifying several previously unknown optimal solutions. Moreover, global optimal cyclic antibandwidth values are proven for a number of benchmark instances for the first time. Comparative results indicate that SAT-CAB consistently matches or surpasses the best-known solutions obtained by state-of-the-art heuristic algorithms such as MS-GVNS, HABC-CAB, and MACAB, as well as strong commercial Constraint Programming and Mixed Integer Programming solvers like CPLEX and Gurobi, particularly on general graphs, while also providing optimality guarantees. These results advance the state of the art for CABP and provide a new baseline for exact and hybrid methods on general graphs. --- ### MetagenBERT: a Transformer-based Architecture using Foundational genomic Large Language Models for novel Metagenome Representation Authors: Gaspar Roy, Eugeni Belda, Baptiste Hennecart, Yann Chevaleyre, Edi Prifti, Jean-Daniel Zucker Published: 2026-01-05 Categories: q-bio.GN, cs.LG arXiv: https://arxiv.org/abs/2601.03295v1 Metagenomic disease prediction commonly relies on species abundance tables derived from large, incomplete reference catalogs, constraining resolution and discarding valuable information contained in DNA reads. To overcome these limitations, we introduce MetagenBERT, a Transformer based framework that produces end to end metagenome embeddings directly from raw DNA sequences, without taxonomic or functional annotations. Reads are embedded using foundational genomic language models (DNABERT2 and the microbiome specialized DNABERTMS), then aggregated through a scalable clustering strategy based on FAISS accelerated KMeans. Each metagenome is represented as a cluster abundance vector summarizing the distribution of its embedded reads. We evaluate this approach on five benchmark gut microbiome datasets (Cirrhosis, T2D, Obesity, IBD, CRC). MetagenBERT achieves competitive or superior AUC performance relative to species abundance baselines across most tasks. Concatenating both representations further improves prediction, demonstrating complementarity between taxonomic and embedding derived signals. Clustering remains robust when applied to as little as 10% of reads, highlighting substantial redundancy in metagenomes and enabling major computational gains. We additionally introduce MetagenBERT Glob Mcardis, a cross cohort variant trained on the large, phenotypically diverse MetaCardis cohort and transferred to other datasets, retaining predictive signal including for unseen phenotypes, indicating the feasibility of a foundation model for metagenome representation. Robustness analyses (PERMANOVA, PERMDISP, entropy) show consistent separation of different states across subsamples. Overall, MetagenBERT provides a scalable, annotation free representation of metagenomes pointing toward future phenotype aware generalization across heterogeneous cohorts and sequencing technologies. --- ### AgentMark: Utility-Preserving Behavioral Watermarking for Agents Authors: Kaibo Huang, Jin Tan, Yukun Wei, Wanling Li, Zipei Zhang, Hui Tian, Zhongliang Yang, Linna Zhou Published: 2026-01-05 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2601.03294v2 LLM-based agents are increasingly deployed to autonomously solve complex tasks, raising urgent needs for IP protection and regulatory provenance. While content watermarking effectively attributes LLM-generated outputs, it fails to directly identify the high-level planning behaviors (e.g., tool and subgoal choices) that govern multi-step execution. Critically, watermarking at the planning-behavior layer faces unique challenges: minor distributional deviations in decision-making can compound during long-term agent operation, degrading utility, and many agents operate as black boxes that are difficult to intervene in directly. To bridge this gap, we propose AgentMark, a behavioral watermarking framework that embeds multi-bit identifiers into planning decisions while preserving utility. It operates by eliciting an explicit behavior distribution from the agent and applying distribution-preserving conditional sampling, enabling deployment under black-box APIs while remaining compatible with action-layer content watermarking. Experiments across embodied, tool-use, and social environments demonstrate practical multi-bit capacity, robust recovery from partial logs, and utility preservation. The code is available at https://github.com/Tooooa/AgentMark. --- ### Lightweight Transformer Architectures for Edge Devices in Real-Time Applications Authors: Hema Hariharan Samson Published: 2026-01-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.03290v1 The deployment of transformer-based models on resource-constrained edge devices represents a critical challenge in enabling real-time artificial intelligence applications. This comprehensive survey examines lightweight transformer architectures specifically designed for edge deployment, analyzing recent advances in model compression, quantization, pruning, and knowledge distillation techniques. We systematically review prominent lightweight variants including MobileBERT, TinyBERT, DistilBERT, EfficientFormer, EdgeFormer, and MobileViT, providing detailed performance benchmarks on standard datasets such as GLUE, SQuAD, ImageNet-1K, and COCO. Our analysis encompasses current industry adoption patterns across major hardware platforms (NVIDIA Jetson, Qualcomm Snapdragon, Apple Neural Engine, ARM architectures), deployment frameworks (TensorFlow Lite, ONNX Runtime, PyTorch Mobile, CoreML), and optimization strategies. Experimental results demonstrate that modern lightweight transformers can achieve 75-96% of full-model accuracy while reducing model size by 4-10x and inference latency by 3-9x, enabling deployment on devices with as little as 2-5W power consumption. We identify sparse attention mechanisms, mixed-precision quantization (INT8/FP16), and hardware-aware neural architecture search as the most effective optimization strategies. Novel findings include memory-bandwidth bottleneck analysis revealing 15-40M parameter models achieve optimal hardware utilization (60-75% efficiency), quantization sweet spots for different model types, and comprehensive energy efficiency profiling across edge platforms. We establish real-time performance boundaries and provide a practical 6-step deployment pipeline achieving 8-12x size reduction with less than 2% accuracy degradation. --- ### Scalable Construction of a Lung Cancer Knowledge Base: Profiling Semantic Reasoning in LLMs Authors: Cesar Felipe Martínez Cisneros, Jesús Ulises Quiroz Bautista, Claudia Anahí Guzmán Solano, Bogdan Kaleb García Rivera, Iván García Pacheco, Yalbi Itzel Balderas Martínez, Kolawole John Adebayoc, Ignacio Arroyo Fernández Published: 2026-01-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.02604v1 The integration of Large Language Models (LLMs) into biomedical research offers new opportunities for domainspecific reasoning and knowledge representation. However, their performance depends heavily on the semantic quality of training data. In oncology, where precision and interpretability are vital, scalable methods for constructing structured knowledge bases are essential for effective fine-tuning. This study presents a pipeline for developing a lung cancer knowledge base using Open Information Extraction (OpenIE). The process includes: (1) identifying medical concepts with the MeSH thesaurus; (2) filtering open-access PubMed literature with permissive licenses (CC0); (3) extracting (subject, relation, object) triplets using OpenIE method; and (4) enriching triplet sets with Named Entity Recognition (NER) to ensure biomedical relevance. The resulting triplet sets provide a domain-specific, large-scale, and noise-aware resource for fine-tuning LLMs. We evaluated T5 models finetuned on this dataset through Supervised Semantic Fine-Tuning. Comparative assessments with ROUGE and BERTScore show significantly improved performance and semantic coherence, demonstrating the potential of OpenIE-derived resources as scalable, low-cost solutions for enhancing biomedical NLP. --- ### SWaRL: Safeguard Code Watermarking via Reinforcement Learning Authors: Neusha Javidnia, Ruisi Zhang, Ashish Kundu, Farinaz Koushanfar Published: 2026-01-05 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2601.02602v2 We present SWaRL, a robust and fidelity-preserving watermarking framework designed to protect the intellectual property of code LLMs by embedding unique and verifiable signatures in the generated program. Existing watermarking approaches either rely on handcrafted code transformations or manipulate token generation probabilities at inference time, making them vulnerable to removal attacks or prone to breaking functional correctness. To address these challenges, SWaRL employs a reinforcement learning-based co-training framework that uses compiler feedback for functional correctness and a jointly trained confidential verifier as a reward signal to maintain watermark detectability. Furthermore, SWaRL employs low-rank adaptation (LoRA) during fine-tuning, enabling efficient integration of watermarking behavior and transferability across model updates. Extensive experiments show that SWaRL achieves strong watermark detection accuracy compared to prior methods while fully maintaining watermarked code functionality. Moreover, SWaRL exhibits strong resilience against refactoring and adversarial transformation attacks, which maintains reliable attribution without substantial computational overhead. --- ### LongDA: Benchmarking LLM Agents for Long-Document Data Analysis Authors: Yiyang Li, Zheyuan Zhang, Tianyi Ma, Zehong Wang, Keerthiram Murugesan, Chuxu Zhang, Yanfang Ye Published: 2026-01-05 Categories: cs.DL, cs.AI arXiv: https://arxiv.org/abs/2601.02598v2 We introduce LongDA, a data analysis benchmark for evaluating LLM-based agents under documentation-intensive analytical workflows. In contrast to existing benchmarks that assume well-specified schemas and inputs, LongDA targets real-world settings in which navigating long documentation and complex data is the primary bottleneck. To this end, we manually curate raw data files, long and heterogeneous documentation, and expert-written publications from 17 publicly available U.S. national surveys, from which we extract 505 analytical queries grounded in real analytical practice. Solving these queries requires agents to first retrieve and integrate key information from multiple unstructured documents, before performing multi-step computations and writing executable code, which remains challenging for existing data analysis agents. To support the systematic evaluation under this setting, we develop LongTA, a tool-augmented agent framework that enables document access, retrieval, and code execution, and evaluate a range of proprietary and open-source models. Our experiments reveal substantial performance gaps even among state-of-the-art models, highlighting the challenges researchers should consider before applying LLM agents for decision support in real-world, high-stakes analytical settings. --- ### Annealed Langevin Posterior Sampling (ALPS): A Rapid Algorithm for Image Restoration with Multiscale Energy Models Authors: Jyothi Rikhab Chand, Mathews Jacob Published: 2026-01-05 Categories: eess.IV, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2601.02594v1 Solving inverse problems in imaging requires models that support efficient inference, uncertainty quantification, and principled probabilistic reasoning. Energy-Based Models (EBMs), with their interpretable energy landscapes and compositional structure, are well-suited for this task but have historically suffered from high computational costs and training instability. To overcome the historical shortcomings of EBMs, we introduce a fast distillation strategy to transfer the strengths of pre-trained diffusion models into multi-scale EBMs. These distilled EBMs enable efficient sampling and preserve the interpretability and compositionality inherent to potential-based frameworks. Leveraging EBM compositionality, we propose Annealed Langevin Posterior Sampling (ALPS) algorithm for Maximum-A-Posteriori (MAP), Minimum Mean Square Error (MMSE), and uncertainty estimates for inverse problems in imaging. Unlike diffusion models that use complex guidance strategies for latent variables, we perform annealing on static posterior distributions that are well-defined and composable. Experiments on image inpainting and MRI reconstruction demonstrate that our method matches or surpasses diffusion-based baselines in both accuracy and efficiency, while also supporting MAP recovery. Overall, our framework offers a scalable and principled solution for inverse problems in imaging, with potential for practical deployment in scientific and clinical settings. ALPS code is available at the GitHub repository \href{https://github.com/JyoChand/ALPS}{ALPS}. --- ### FlowPlan-G2P: A Structured Generation Framework for Transforming Scientific Papers into Patent Descriptions Authors: Kris W Pan, Yongmin Yoo Published: 2026-01-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.02589v4 Generating patent descriptions from scientific papers is challenging due to fundamental rhetorical and structural disparities between the two genres. Existing approaches treat this as surface-level rewriting, failing to capture the hierarchical reasoning and statutory constraints inherent in patent drafting. We propose FlowPlan-G2P, a graph-mediated generation framework that decomposes this transformation into three stages: (1) Concept Graph Induction, extracting technical entities and functional dependencies into a directed graph; (2) Section-level Planning, partitioning the graph into coherent subgraphs aligned with canonical patent sections; and (3) Graph-Conditioned Generation, synthesizing legally compliant paragraphs conditioned on section-specific subgraphs. Experiments on expert-validated benchmarks reveal that standard NLG metrics systematically favor legally non-compliant outputs over valid patent descriptions, motivating our domain-specific evaluation. Under this evaluation, FlowPlan-G2P with an open-weight backbone consistently outperforms vanilla proprietary models, demonstrating that structured decomposition is a stronger determinant of quality than model scale. --- ### Threat Detection in Social Media Networks Using Machine Learning Based Network Analysis Authors: Aditi Sanjay Agrawal Published: 2026-01-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.02581v1 The accelerated development of social media websites has posed intricate security issues in cyberspace, where these sites have increasingly become victims of criminal activities including attempts to intrude into them, abnormal traffic patterns, and organized attacks. The conventional rule-based security systems are not always scalable and dynamic to meet such a threat. This paper introduces a threat detection framework based on machine learning that can be used to classify malicious behavior in the social media network environment based on the nature of network traffic. Exploiting a rich network traffic dataset, the massive preprocessing and exploratory data analysis is conducted to overcome the problem of data imbalance, feature inconsistency, and noise. A model of artificial neural network (ANN) is then created to acquire intricate, non-linear tendencies of malicious actions. The proposed model is tested on conventional performance metrics, such as accuracy, accuracy, recall, F1-score, and ROC-AUC, and shows good detection and high levels of strength. The findings suggest that neural network-based solutions have the potential to be used effectively to identify the latent threat dynamics within the context of a large-scale social media network and that they can be employed to complement the existing intrusion detection system and better to conduct proactive cybersecurity operations. --- ### Reconstructing Item Characteristic Curves using Fine-Tuned Large Language Models Authors: Christopher Ormerod Published: 2026-01-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.02580v2 Traditional methods for determining assessment item parameters, such as difficulty and discrimination, rely heavily on expensive field testing to collect student performance data for Item Response Theory (IRT) calibration. This study introduces a novel approach that implicitly models these psychometric properties by fine-tuning Large Language Models (LLMs) to simulate student responses across a spectrum of latent abilities. Leveraging the Qwen-3 dense model series and Low-Rank Adaptation (LoRA), we train models to generate responses to multiple choice questions conditioned on discrete ability descriptors. We reconstruct the probability of a correct response as a function of student ability, effectively generating synthetic Item Characteristic Curves (ICCs) to estimate IRT parameters. Evaluation on a dataset of Grade 6 English Language Arts (ELA) items and the BEA 2024 Shared Task dataset demonstrates that this method competes with or outperforms baseline approaches. This simulation-based technique seems particularly effective at modeling item discrimination. --- ### DataParasite Enables Scalable and Repurposable Online Data Curation Authors: Mengyi Sun Published: 2026-01-05 Categories: cs.CL, cs.IR arXiv: https://arxiv.org/abs/2601.02578v1 Many questions in computational social science rely on datasets assembled from heterogeneous online sources, a process that is often labor-intensive, costly, and difficult to reproduce. Recent advances in large language models enable agentic search and structured extraction from the web, but existing systems are frequently opaque, inflexible, or poorly suited to scientific data curation. Here we introduce DataParasite, an open-source, modular pipeline for scalable online data collection. DataParasite decomposes tabular curation tasks into independent, entity-level searches defined through lightweight configuration files and executed through a shared, task-agnostic python script. Crucially, the same pipeline can be repurposed to new tasks, including those without predefined entity lists, using only natural-language instructions. We evaluate the pipeline on multiple canonical tasks in computational social science, including faculty hiring histories, elite death events, and political career trajectories. Across tasks, DataParasite achieves high accuracy while reducing data-collection costs by an order of magnitude relative to manual curation. By lowering the technical and labor barriers to online data assembly, DataParasite provides a practical foundation for scalable, transparent, and reusable data curation in computational social science and beyond. --- ### Orchestral AI: A Framework for Agent Orchestration Authors: Alexander Roman, Jacob Roman Published: 2026-01-05 Categories: cs.AI, astro-ph.IM, hep-ph arXiv: https://arxiv.org/abs/2601.02577v1 The rapid proliferation of LLM agent frameworks has forced developers to choose between vendor lock-in through provider-specific SDKs and complex multi-package ecosystems that obscure control flow and hinder reproducibility. Integrating tool calling across multiple LLM providers remains a core engineering challenge due to fragmented APIs, incompatible message formats, and inconsistent streaming and tool-calling behavior, making it difficult to build portable, reliable agent systems. We introduce Orchestral, a lightweight Python framework that provides a unified, type-safe interface for building LLM agents across major providers while preserving the simplicity required for scientific computing and production deployment. Orchestral defines a single universal representation for messages, tools, and LLM usage that operates seamlessly across providers, eliminating manual format translation and reducing framework-induced complexity. Automatic tool schema generation from Python type hints removes the need for handwritten descriptors while maintaining type safety across provider boundaries. A synchronous execution model with streaming support enables deterministic behavior, straightforward debugging, and real-time interaction without introducing server dependencies. The framework's modular architecture cleanly separates provider integration, tool execution, conversation orchestration, and user-facing interfaces, enabling extensibility without architectural entanglement. Orchestral supports advanced agent capabilities found in larger frameworks, including rich tool calling, context compaction, workspace sandboxing, user approval workflows, sub-agents, memory management, and MCP integration. --- ### Fact-Checking with Large Language Models via Probabilistic Certainty and Consistency Authors: Haoran Wang, Maryam Khalid, Qiong Wu, Jian Gao, Cheng Cao Published: 2026-01-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.02574v1 Large language models (LLMs) are increasingly used in applications requiring factual accuracy, yet their outputs often contain hallucinated responses. While fact-checking can mitigate these errors, existing methods typically retrieve external evidence indiscriminately, overlooking the model's internal knowledge and potentially introducing irrelevant noise. Moreover, current systems lack targeted mechanisms to resolve specific uncertainties in the model's reasoning. Inspired by how humans fact-check, we argue that LLMs should adaptively decide whether to rely on internal knowledge or initiate retrieval based on their confidence in a given claim. We introduce Probabilistic Certainty and Consistency (PCC), a framework that estimates factual confidence by jointly modeling an LLM's probabilistic certainty and reasoning consistency. These confidence signals enable an adaptive verification strategy: the model answers directly when confident, triggers targeted retrieval when uncertain or inconsistent, and escalates to deep search when ambiguity is high. Our confidence-guided routing mechanism ensures that retrieval is invoked only when necessary, improving both efficiency and reliability. Extensive experiments across three challenging benchmarks show that PCC achieves better uncertainty quantification than verbalized confidence and consistently outperforms strong LLM-based fact-checking baselines. Furthermore, we demonstrate that PCC generalizes well across various LLMs. --- ### LendNova: Towards Automated Credit Risk Assessment with Language Models Authors: Kiarash Shamsi, Danijel Novokmet, Joshua Peters, Mao Lin Liu, Paul K Edwards, Vahab Khoshdel Published: 2026-01-05 Categories: cs.LG, cs.AI, cs.CE arXiv: https://arxiv.org/abs/2601.02573v1 Credit risk assessment is essential in the financial sector, but has traditionally depended on costly feature-based models that often fail to utilize all available information in raw credit records. This paper introduces LendNova, the first practical automated end-to-end pipeline for credit risk assessment, designed to utilize all available information in raw credit records by leveraging advanced NLP techniques and language models. LendNova transforms risk modeling by operating directly on raw, jargon-heavy credit bureau text using a language model that learns task-relevant representations without manual feature engineering. By automatically capturing patterns and risk signals embedded in the text, it replaces manual preprocessing steps, reducing costs and improving scalability. Evaluation on real-world data further demonstrates its strong potential in accurate and efficient risk assessment. LendNova establishes a baseline for intelligent credit risk agents, demonstrating the feasibility of language models in this domain. It lays the groundwork for future research toward foundation systems that enable more accurate, adaptable, and automated financial decision-making. --- ### LoRA-Drop: Temporal LoRA Decoding for Efficient LLM Inference Authors: Hossein Rajabzadeh, Maryam Dialameh, Chul B. Park, Il-Min Kim, Hyock Ju Kwon Published: 2026-01-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.02569v1 Autoregressive large language models (LLMs) are bottlenecked by sequential decoding, where each new token typically requires executing all transformer layers. Existing dynamic-depth and layer-skipping methods reduce this cost, but often rely on auxiliary routing mechanisms or incur accuracy degradation when bypassed layers are left uncompensated. We present \textbf{LoRA-Drop}, a plug-and-play inference framework that accelerates decoding by applying a \emph{temporal compute schedule} to a fixed subset of intermediate layers: on most decoding steps, selected layers reuse the previous-token hidden state and apply a low-rank LoRA correction, while periodic \emph{refresh} steps execute the full model to prevent drift. LoRA-Drop requires no routing network, is compatible with standard KV caching, and can reduce KV-cache footprint by skipping KV updates in droppable layers during LoRA steps and refreshing periodically. Across \textbf{LLaMA2-7B}, \textbf{LLaMA3-8B}, \textbf{Qwen2.5-7B}, and \textbf{Qwen2.5-14B}, LoRA-Drop achieves up to \textbf{2.6$\times$ faster decoding} and \textbf{45--55\% KV-cache reduction} while staying within \textbf{0.5 percentage points (pp)} of baseline accuracy. Evaluations on reasoning (GSM8K, MATH, BBH), code generation (HumanEval, MBPP), and long-context/multilingual benchmarks (LongBench, XNLI, XCOPA) identify a consistent \emph{safe zone} of scheduling configurations that preserves quality while delivering substantial efficiency gains, providing a simple path toward adaptive-capacity inference in LLMs. Codes are available at https://github.com/hosseinbv/LoRA-Drop.git. --- ### Shallow- and Deep-fake Image Manipulation Localization Using Vision Mamba and Guided Graph Neural Network Authors: Junbin Zhang, Hamid Reza Tohidypour, Yixiao Wang, Panos Nasiopoulos Published: 2026-01-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.02566v1 Image manipulation localization is a critical research task, given that forged images may have a significant societal impact of various aspects. Such image manipulations can be produced using traditional image editing tools (known as "shallowfakes") or advanced artificial intelligence techniques ("deepfakes"). While numerous studies have focused on image manipulation localization on either shallowfake images or deepfake videos, few approaches address both cases. In this paper, we explore the feasibility of using a deep learning network to localize manipulations in both shallow- and deep-fake images, and proposed a solution for such purpose. To precisely differentiate between authentic and manipulated pixels, we leverage the Vision Mamba network to extract feature maps that clearly describe the boundaries between tampered and untouched regions. To further enhance this separation, we propose a novel Guided Graph Neural Network (G-GNN) module that amplifies the distinction between manipulated and authentic pixels. Our evaluation results show that our proposed method achieved higher inference accuracy compared to other state-of-the-art methods. --- ### Comparative Analysis of Binarization Methods For Medical Image Hashing On Odir Dataset Authors: Nedim Muzoglu Published: 2026-01-05 Categories: eess.IV, cs.CV, cs.IR arXiv: https://arxiv.org/abs/2601.02564v2 In this study, we evaluated four binarization methods. Locality-Sensitive Hashing (LSH), Iterative Quantization (ITQ), Kernel-based Supervised Hashing (KSH), and Supervised Discrete Hashing (SDH) on the ODIR dataset using deep feature embeddings. Experimental results show that SDH achieved the best performance, with an mAP@100 of 0.9184 using only 32-bit codes, outperforming LSH, ITQ, and KSH. Compared with prior studies, our method proved highly competitive: Fang et al. reported 0.7528 (Fundus-iSee, 48 bits) and 0.8856 (ASOCT-Cataract, 48 bits), while Wijesinghe et al. achieved 94.01 (KVASIR, 256 bits). Despite using significantly fewer bits, our SDH-based framework reached retrieval accuracy close to the state-of-the-art. These findings demonstrate that SDH is the most effective approach among those tested, offering a practical balance of accuracy, storage, and efficiency for medical image retrieval and device inventory management. --- ### Compressed code: the hidden effects of quantization and distillation on programming tokens Authors: Viacheslav Siniaev, Iaroslav Chelombitko, Aleksey Komissarov Published: 2026-01-05 Categories: cs.SE, cs.CL, cs.LG, cs.PL arXiv: https://arxiv.org/abs/2601.02563v3 Large Language Models (LLMs) have demonstrated exceptional code generation capabilities, yet their token-level mechanisms remain underexplored, particularly in compressed models. Through systematic analysis of programming language token representations, we characterize how programming languages are encoded in LLM tokenizers by analyzing their vocabulary distribution and keyword coverage patterns. We introduce a novel cold-start probability analysis method that provides insights into model behavior without requiring explicit prompts. Additionally, we present a comprehensive evaluation of how different model optimization techniques - including quantization, distillation, model scaling, and task-specific fine-tuning - affect token-level representations and code generation quality. Our experiments, supported by comprehensive probability distribution analysis and evaluation metrics, reveal critical insights into token-level behavior and provide empirically-validated guidelines for maintaining code generation quality under various optimization constraints. These findings advance both theoretical understanding of LLM code generation and practical implementation of optimized models in production environments. --- ### CutisAI: Deep Learning Framework for Automated Dermatology and Cancer Screening Authors: Rohit Kaushik, Eva Kaushik Published: 2026-01-05 Categories: cs.LG, eess.IV arXiv: https://arxiv.org/abs/2601.02562v1 The rapid growth of dermatological imaging and mobile diagnostic tools calls for systems that not only demonstrate empirical performance but also provide strong theoretical guarantees. Deep learning models have shown high predictive accuracy; however, they are often criticized for lacking well, calibrated uncertainty estimates without which these models are hardly deployable in a clinical setting. To this end, we present the Conformal Bayesian Dermatological Classifier (CBDC), a well, founded framework that combines Statistical Learning Theory, Topological Data Analysis (TDA), and Bayesian Conformal Inference. CBDC offers distribution, dependent generalization bounds that reflect dermatological variability, proves a topological stability theorem that guarantees the invariance of convolutional neural network embeddings under photometric and morphological perturbations and provides finite conformal coverage guarantees for trustworthy uncertainty quantification. Through exhaustive experiments on the HAM10000, PH2, and ISIC 2020 datasets, we show that CBDC not only attains classification accuracy but also generates calibrated predictions that are interpretable from a clinical perspective. This research constitutes a theoretical and practical leap for deep dermatological diagnostics, thereby opening the machine learning theory clinical applicability interface. --- ### AI-exposed jobs deteriorated before ChatGPT Authors: Morgan R. Frank, Alireza Javadian Sabet, Lisa Simon, Sarah H. Bana, Renzhe Yu Published: 2026-01-05 Categories: econ.GN, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2601.02554v1 Public debate links worsening job prospects for AI-exposed occupations to the release of ChatGPT in late 2022. Using monthly U.S. unemployment insurance records, we measure occupation- and location-specific unemployment risk and find that risk rose in AI-exposed occupations beginning in early 2022, months before ChatGPT. Analyzing millions of LinkedIn profiles, we show that graduate cohorts from 2021 onward entered AI-exposed jobs at lower rates than earlier cohorts, with gaps opening before late 2022. Finally, from millions of university syllabi, we find that graduates taking more AI-exposed curricula had higher first-job pay and shorter job searches after ChatGPT. Together, these results point to forces pre-dating generative AI and to the ongoing value of LLM-relevant education. --- ### SimpleMem: Efficient Lifelong Memory for LLM Agents Authors: Jiaqi Liu, Yaofeng Su, Peng Xia, Siwei Han, Zeyu Zheng, Cihang Xie, Mingyu Ding, Huaxiu Yao Published: 2026-01-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.02553v3 To support long-term interaction in complex environments, LLM agents require memory systems that manage historical experiences. Existing approaches either retain full interaction histories via passive context extension, leading to substantial redundancy, or rely on iterative reasoning to filter noise, incurring high token costs. To address this challenge, we introduce SimpleMem, an efficient memory framework based on semantic lossless compression. We propose a three-stage pipeline designed to maximize information density and token utilization: (1) Semantic Structured Compression, which distills unstructured interactions into compact, multi-view indexed memory units; (2) Online Semantic Synthesis, an intra-session process that instantly integrates related context into unified abstract representations to eliminate redundancy; and (3) Intent-Aware Retrieval Planning, which infers search intent to dynamically determine retrieval scope and construct precise context efficiently. Experiments on benchmark datasets show that our method consistently outperforms baseline approaches in accuracy, retrieval efficiency, and inference cost, achieving an average F1 improvement of 26.4% in LoCoMo while reducing inference-time token consumption by up to 30-fold, demonstrating a superior balance between performance and efficiency. Code is available at https://github.com/aiming-lab/SimpleMem. --- ### Normalized Conditional Mutual Information Surrogate Loss for Deep Neural Classifiers Authors: Linfeng Ye, Zhixiang Chi, Konstantinos N. Plataniotis, En-hui Yang Published: 2026-01-05 Categories: cs.LG, cs.AI, cs.CV, cs.IT arXiv: https://arxiv.org/abs/2601.02543v3 In this paper, we propose a novel information theoretic surrogate loss; normalized conditional mutual information (NCMI); as a drop in alternative to the de facto cross-entropy (CE) for training deep neural network (DNN) based classifiers. We first observe that the model's NCMI is inversely proportional to its accuracy. Building on this insight, we introduce an alternating algorithm to efficiently minimize the NCMI. Across image recognition and whole-slide imaging (WSI) subtyping benchmarks, NCMI-trained models surpass state of the art losses by substantial margins at a computational cost comparable to that of CE. Notably, on ImageNet, NCMI yields a 2.77% top-1 accuracy improvement with ResNet-50 comparing to the CE; on CAMELYON-17, replacing CE with NCMI improves the macro-F1 by 8.6% over the strongest baseline. Gains are consistent across various architectures and batch sizes, suggesting that NCMI is a practical and competitive alternative to CE. --- ### A Green Solution for Breast Region Segmentation Using Deep Active Learning Authors: Sam Narimani, Solveig Roth Hoff, Kathinka Dæhli Kurz, Kjell-Inge Gjesdal, Jürgen Geisler, Endre Grøvik Published: 2026-01-05 Categories: physics.med-ph, cs.CV, eess.IV arXiv: https://arxiv.org/abs/2601.02538v1 Purpose: Annotation of medical breast images is an essential step toward better diagnostic but a time consuming task. This research aims to focus on different selecting sample strategies within deep active learning on Breast Region Segmentation (BRS) to lessen computational cost of training and effective use of resources. Methods: The Stavanger breast MRI dataset containing 59 patients was used in this study, with FCN-ResNet50 adopted as a sustainable deep learning (DL) model. A novel sample selection approach based on Breast Anatomy Geometry (BAG) analysis was introduced to group data with similar informative features for DL. Patient positioning and Breast Size were considered the key selection criteria in this process. Four selection strategies including Random Selection, Nearest Point, Breast Size, and a hybrid of all three strategies were evaluated using an active learning framework. Four training data proportions of 10%, 20%, 30%, and 40% were used for model training, with the remaining data reserved for testing. Model performance was assessed using Dice score, Intersection over Union, precision, and recall, along with 5-fold cross-validation to enhance generalizability. Results: Increasing the training data proportion from 10% to 40% improved segmentation performance for nearly all strategies, except for Random Selection. The Nearest Point strategy consistently achieved the lowest carbon footprint at 30% and 40% data proportions. Overall, combining the Nearest Point strategy with 30% of the training data provided the best balance between segmentation performance, efficiency, and environmental sustainability. Keywords: Deep Active Learning, Breast Region Segmentation, Human-center analysis --- ### MovieRecapsQA: A Multimodal Open-Ended Video Question-Answering Benchmark Authors: Shaden Shaar, Bradon Thymes, Sirawut Chaixanien, Claire Cardie, Bharath Hariharan Published: 2026-01-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.02536v2 Understanding real-world videos such as movies requires integrating visual and dialogue cues. Yet existing VideoQA benchmarks struggle to capture this multimodal reasoning and, given the difficulty of evaluating free-form answers, largely resort to simple multiple choice questions. We introduce a novel open-ended multimodal VideoQA benchmark, MovieRecapsQA, created using movie recap videos -- a distinctive type of YouTube content that summarizes a film via a voiceover description of key clips from the movie (recap video). From the transcribed voiceover (recap summary) of 60 recap videos, we generate $\approx$8.2K questions along with the necessary ``facts'' expected in each answer; the former facilitates the creation of questions that require mutimodal reasoning and the latter allow the construction of a reference-free evaluation metric that can be applied to open-ended responses. To our knowledge, this is the first reference-free open-ended VideoQA benchmark. The benchmark allows each question to be evaluated in different input video settings: given (a) the full-length movie, (b) the full ($\approx$11 min) recap video (visual only), (c) $\approx$14 min of aligned movie scenes, i.e, movie scenes relevant to the question, and (d) $\approx$1.2 min of aligned recap video scenes. In all cases, the text of any associated movie dialogue is provided. Each question is categorized by the modality required to answer it -- visual, dialogue, or both -- enabling fine-grained evaluation of multimodal capabilities. We benchmark (setting (d)) seven state-of-the-art MLLMs and find that (i) only our reference-free metric produces meaningful human-aligned model separation; (ii) vision-centric questions yield the lowest scores across all models; (iii) removing visual input often \textit{improves} model factuality; and (iv) the primary bottleneck is visual perception, not visual reasoning. --- ### ModeX: Evaluator-Free Best-of-N Selection for Open-Ended Generation Authors: Hyeong Kyu Choi, Sharon Li Published: 2026-01-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.02535v2 Selecting a single high-quality output from multiple stochastic generations remains a fundamental challenge for large language models (LLMs), particularly in open-ended tasks where no canonical answer exists. While Best-of-N and self-consistency methods show that aggregating multiple generations can improve performance, existing approaches typically rely on external evaluators, reward models, or exact string-match voting, limiting their applicability and efficiency. We propose Mode Extraction (ModeX), an evaluator-free Best-of-N selection framework that generalizes majority voting to open-ended text generation by identifying the modal output representing the dominant semantic consensus among generated texts. ModeX constructs a similarity graph over candidate generations and recursively applies spectral clustering to select a representative centroid, without requiring additional inference or auxiliary models. We further instantiate this selection principle as ModeX-Lite, an improved version of ModeX with early pruning for efficiency. Across open-ended tasks -- including text summarization, code generation, and mathematical reasoning -- our approaches consistently outperform standard single- and multi-path baselines, providing a computationally efficient solution for robust open-ended text generation. Code is released in https://github.com/deeplearning-wisc/ModeX. --- ### Losses that Cook: Topological Optimal Transport for Structured Recipe Generation Authors: Mattia Ottoborgo, Daniele Rege Cambrin, Paolo Garza Published: 2026-01-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.02531v2 Cooking recipes are complex procedures that require not only a fluent and factual text, but also accurate timing, temperature, and procedural coherence, as well as the correct composition of ingredients. Standard training procedures are primarily based on cross-entropy and focus solely on fluency. Building on RECIPE-NLG, we investigate the use of several composite objectives and present a new topological loss that represents ingredient lists as point clouds in embedding space, minimizing the divergence between predicted and gold ingredients. Using both standard NLG metrics and recipe-specific metrics, we find that our loss significantly improves ingredient- and action-level metrics. Meanwhile, the Dice loss excels in time/temperature precision, and the mixed loss yields competitive trade-offs with synergistic gains in quantity and time. A human preference analysis supports our finding, showing our model is preferred in 62% of the cases. ---