# Research Papers — 2026-01 · bundle page 58 of 189 > Full-text markdown of papers #2851–2900 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/58 [← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/57.md) [Next bundle →](https://aipod.dev/papers/archive/2026-01/page/59.md) ### Frequent Pattern Mining approach to Image Compression Authors: Avinash Kadimisetty, C. Oswald, B. Sivalselvan Published: 2026-01-24 Categories: eess.IV, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2602.00100v1 The paper focuses on Image Compression, explaining efficient approaches based on Frequent Pattern Mining(FPM). The proposed compression mechanism is based on clustering similar pixels in the image and thus using cluster identifiers in image compression. Redundant data in the image is effectively handled by replacing the DCT phase of conventional JPEG through a mixture of k-means Clustering and Closed Frequent Sequence Mining. To optimize the cardinality of pattern(s) in encoding, efficient pruning techniques have been used through the refinement of Conventional Generalized Sequential Pattern Mining(GSP) algorithm. We have proposed a mechanism for finding the frequency of a sequence which will yield significant reduction in the code table size. The algorithm is tested by compressing benchmark datasets yielding an improvement of 45% in compression ratios, often outperforming the existing alternatives. PSNR and SSIM, which are the image quality metrics, have been tested which show a negligible loss in visual quality. --- ### Gauss-Newton Natural Gradient Descent for Shape Learning Authors: James King, Arturs Berzins, Siddhartha Mishra, Marius Zeinhofer Published: 2026-01-24 Categories: cs.LG, math.OC arXiv: https://arxiv.org/abs/2602.00099v2 We explore the use of the Gauss-Newton method for optimization in shape learning, including implicit neural surfaces and geometry-informed neural networks. The method addresses key challenges in shape learning, such as the ill-conditioning of the underlying differential constraints and the mismatch between the optimization problem in parameter space and the function space where the problem is naturally posed. This leads to significantly faster and more stable convergence than standard first-order methods, while also requiring far fewer iterations. Experiments across benchmark shape optimization tasks demonstrate that the Gauss-Newton method consistently improves both training speed and final solution accuracy. --- ### MO-ELA: Rigorously Expanding Exploratory Landscape Features for Automated Algorithm Selection in Continuous Multi-Objective Optimisation Authors: Oliver Preuß, Jeroen Rook, Jakob Bossek, Heike Trautmann Published: 2026-01-24 Categories: cs.NE arXiv: https://arxiv.org/abs/2602.00098v1 Automated Algorithm Selection (AAS) is a popular meta-algorithmic approach and has demonstrated to work well for single-objective optimisation in combination with exploratory landscape features (ELA), i.e., (numerical) descriptive features derived from sampling the black-box (continuous) optimisation problem. In contrast to the abundance of features that describe single-objective optimisation problems, only a few features have been proposed for multi-objective optimisation so far. Building upon recent work on exploratory landscape features for box-constrained continuous multi-objective optimization problems, we propose a novel and complementary set of additional features (MO-ELA). These features are based on a random sample of points considering both the decision and objective space. The features are divided into 5 feature groups depending on how they are being calculated: non-dominated-sorting, descriptive statistics, principal component analysis, graph structures and gradient information. An AAS study conducted on well-established multi-objective benchmarks demonstrates that the proposed features contribute to successfully distinguishing between algorithm performance and thus adequately capture problem hardness resulting in models that come very close to the virtual best solver. After feature selection, the newly proposed features are frequently among the top contributors, underscoring their value in algorithm selection and problem characterisation. --- ### Mirage2Matter: A Physically Grounded Gaussian World Model from Video Authors: Zhengqing Gao, Ziwen Li, Xin Wang, Jiaxin Huang, Zhenyang Ren, Mingkai Shao, Hanlue Zhang, Tianyu Huang, Yongkang Cheng, Yandong Guo, Runqi Lin, Yuanyuan Wang et al. Published: 2026-01-24 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.00096v1 The scalability of embodied intelligence is fundamentally constrained by the scarcity of real-world interaction data. While simulation platforms provide a promising alternative, existing approaches often suffer from a substantial visual and physical gap to real environments and rely on expensive sensors, precise robot calibration, or depth measurements, limiting their practicality at scale. We present Simulate Anything, a graphics-driven world modeling and simulation framework that enables efficient generation of high-fidelity embodied training data using only multi-view environment videos and off-the-shelf assets. Our approach reconstructs real-world environments into a photorealistic scene representation using 3D Gaussian Splatting (3DGS), seamlessly capturing fine-grained geometry and appearance from video. We then leverage generative models to recover a physically realistic representation and integrate it into a simulation environment via a precision calibration target, enabling accurate scale alignment between the reconstructed scene and the real world. Together, these components provide a unified, editable, and physically grounded world model. Vision Language Action (VLA) models trained on our simulated data achieve strong zero-shot performance on downstream tasks, matching or even surpassing results obtained with real-world data, highlighting the potential of reconstruction-driven world modeling for scalable and practical embodied intelligence training. --- ### NCSAM Noise-Compensated Sharpness-Aware Minimization for Noisy Label Learning Authors: Jiayu Xu, Junbiao Pang Published: 2026-01-24 Categories: cs.LG, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2601.19947v3 Learning from Noisy Labels (LNL) remains a fundamental challenge in deep learning because real-world datasets often contain corrupted annotations. Most existing methods rely on label correction or sample selection mechanisms. In contrast, we study LNL from an optimization perspective by establishing a theoretical connection between label noise and the flatness-seeking behavior of Sharpness-Aware Minimization (SAM). Based on this analysis, we propose Noise-Compensated Sharpness-Aware Minimization (NCSAM), which uses a noise-compensated perturbation to counteract the optimization bias induced by noisy labels. By correcting distorted SAM perturbations, NCSAM mitigates the memorization of noisy labels during training while preserving the simplicity of optimization-based learning. Experiments on synthetic and real-world noisy-label benchmarks show that NCSAM consistently improves over SAM-based optimization baselines and remains competitive with representative noisy-label learning methods. --- ### MK-SGC-SC: Multiple Kernel Guided Sparse Graph Construction in Spectral Clustering for Unsupervised Speaker Diarization Authors: Nikhil Raghav, Avisek Gupta, Swagatam Das, Md Sahidullah Published: 2026-01-24 Categories: eess.AS, cs.LG, cs.SD arXiv: https://arxiv.org/abs/2601.19946v2 Speaker diarization aims to segment audio recordings into regions corresponding to individual speakers. Although unsupervised speaker diarization is inherently challenging, the prospect of identifying speaker regions without pretraining or weak supervision motivates research on clustering techniques. In this work, we share the notable observation that measuring multiple kernel similarities of speaker embeddings to thereafter craft a sparse graph for spectral clustering in a principled manner is sufficient to achieve state-of-the-art performances in a fully unsupervised setting. Specifically, we consider four polynomial kernels and a degree one arccosine kernel to measure similarities in speaker embeddings, using which sparse graphs are constructed in a principled manner to emphasize local similarities. Experiments show the proposed approach excels in unsupervised speaker diarization over a variety of challenging environments in the DIHARD-III, AMI, and VoxConverse corpora. To encourage further research, our implementations are available at https://github.com/nikhilraghav29/MK-SGC-SC. --- ### Lossy Image Compression -- A Frequent Sequence Mining perspective employing efficient Clustering Authors: Avinash Kadimisetty, Oswald C, Sivaselvan B, Alekhya Kadimisetty Published: 2026-01-24 Categories: eess.IV, cs.AI, cs.MM arXiv: https://arxiv.org/abs/2601.18821v1 This work explores the scope of Frequent Sequence Mining in the domain of Lossy Image Compression. The proposed work is based on the idea of clustering pixels and using the cluster identifiers in the compression. The DCT phase in JPEG is replaced with a combination of closed frequent sequence mining and k-means clustering to handle the redundant data effectively. This method focuses mainly on applying k-means clustering in parallel to all blocks of each component of the image to reduce the compression time. Conventional GSP algorithm is refined to optimize the cardinality of patterns through a novel pruning strategy, thus achieving a good reduction in the code table size. Simulations of the proposed algorithm indicate significant gains in compression ratio and quality in relation to the existing alternatives. --- ### BrainDistill: Implantable Motor Decoding with Task-Specific Knowledge Distillation Authors: Yuhan Xie, Jinhan Liu, Xiaoyong Ni, Fei Tan, Icare Sakr, Thibault Collin, Shiqi Sun, Alejandro Rodriguez Guajardo, Demon Fanny, Charles-francois Vincent Latchoumane, Henri Lorach, Jocelyne Bloch et al. Published: 2026-01-24 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.17625v1 Transformer-based neural decoders with large parameter counts, pre-trained on large-scale datasets, have recently outperformed classical machine learning models and small neural networks on brain-computer interface (BCI) tasks. However, their large parameter counts and high computational demands hinder deployment in power-constrained implantable systems. To address this challenge, we introduce BrainDistill, a novel implantable motor decoding pipeline that integrates an implantable neural decoder (IND) with a task-specific knowledge distillation (TSKD) framework. Unlike standard feature distillation methods that attempt to preserve teacher representations in full, TSKD explicitly prioritizes features critical for decoding through supervised projection. Across multiple neural datasets, IND consistently outperforms prior neural decoders on motor decoding tasks, while its TSKD-distilled variant further surpasses alternative distillation methods in few-shot calibration settings. Finally, we present a quantization-aware training scheme that enables integer-only inference with activation clipping ranges learned during training. The quantized IND enables deployment under the strict power constraints of implantable BCIs with minimal performance loss. --- ### Memento: Towards Proactive Visualization of Everyday Memories with Personal Wearable AR Assistant Authors: Yoonsang Kim, Yalong Yang, Arie E. Kaufman Published: 2026-01-24 Categories: cs.HC, cs.CL, cs.IR arXiv: https://arxiv.org/abs/2601.17622v4 We introduce Memento, a conversational AR assistant that permanently captures and memorizes user's verbal queries alongside their spatiotemporal and activity contexts. By storing these "memories," Memento discovers connections between users' recurring interests and the contexts that trigger them. Upon detection of similar or identical spatiotemporal activity, Memento proactively recalls user interests and delivers up-to-date responses through AR, seamlessly integrating AR experience into their daily routine. Unlike prior work, each interaction in Memento is not a transient event, but a connected series of interactions with coherent long--term perspective, tailored to the user's broader multimodal (visual, spatial, temporal, and embodied) context. We conduct a preliminary evaluation through user feedbacks with participants of diverse expertise in immersive apps, and explore the value of proactive context-aware AR assistant in everyday settings. We share our findings and challenges in designing a proactive, context-aware AR system. --- ### Agentic Search in the Wild: Intents and Trajectory Dynamics from 14M+ Real Search Requests Authors: Jingjie Ning, João Coelho, Yibo Kong, Yunfan Long, Bruno Martins, João Magalhães, Jamie Callan, Chenyan Xiong Published: 2026-01-24 Categories: cs.IR, cs.CL arXiv: https://arxiv.org/abs/2601.17617v3 LLM-powered search agents are increasingly being used for multi-step information seeking tasks, yet the IR community lacks empirical understanding of how agentic search sessions unfold and how retrieved evidence is reflected in later queries. This paper presents a large-scale log analysis of agentic search based on 14.44M search requests (3.97M sessions) collected from DeepResearchGym, i.e., an open-source search API accessed by external agentic clients. We sessionize the logs, assign session-level intents and step-wise query-reformulation labels using LLM-based annotation, and propose Context-driven Term Adoption Rate (CTAR) to quantify whether newly introduced query terms are lexically traceable to previously retrieved evidence. Our analyses reveal distinctive behavioral patterns. First, over 90\% of multi-turn sessions contain at most ten steps, and 89\% of inter-step intervals fall under one minute. Second, behavior varies by intent. Fact-seeking sessions exhibit high repetition that increases over time, while sessions requiring reasoning sustain broader exploration. Third, query reformulations are often traceable to retrieved evidence across steps. On average, 54\% of newly introduced query terms appear in the accumulated evidence context, with additional traceability to earlier steps beyond the most recent retrieval. These findings provide candidate signals for repetition-aware stopping, intent-adaptive retrieval budgeting, and explicit cross-step context tracking. We released the anonymized logs, making them available at a public HuggingFace~\chref{https://huggingface.co/datasets/cx-cmu/deepresearchgym-agentic-search-logs}{repository}. --- ### Split-on-Share: Mixture of Sparse Experts for Task-Agnostic Continual Learning Authors: Fatema Siddika, Md Anwar Hossen, Tanwi Mallick, Ali Jannesari Published: 2026-01-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.17616v2 Continual learning in Large Language Models (LLMs) is hindered by the plasticity-stability dilemma, where acquiring new capabilities often leads to catastrophic forgetting of previous knowledge. Existing methods typically treat parameters uniformly, failing to distinguish between specific task knowledge and shared capabilities. We introduce Mixture of Sparse Experts for Task-Agnostic Continual Learning, referred to as SETA, a framework that resolves the plasticity-stability conflict by decomposing the model into modular subspaces. Unlike standard updates, where tasks compete for the same parameters, SETA separates knowledge into unique experts, designed to isolate task-specific patterns, and shared experts, responsible for capturing common features. This structure is maintained through elastic weight anchoring, which protects critical shared knowledge and enables a unified gating network to automatically retrieve the correct expert combination for each task during inference. Extensive experiments across diverse domain-specific and general benchmarks demonstrate that SETA consistently outperforms state-of-the-art parameter-efficient fine-tuning-based continual learning methods. --- ### Athena: Synergizing Data Prefetching and Off-Chip Prediction via Online Reinforcement Learning Authors: Rahul Bera, Zhenrong Lang, Caroline Hengartner, Konstantinos Kanellopoulos, Rakesh Kumar, Mohammad Sadrosadati, Onur Mutlu Published: 2026-01-24 Categories: cs.AR, cs.DC, cs.LG arXiv: https://arxiv.org/abs/2601.17615v1 Prefetching and off-chip prediction are two techniques proposed to hide long memory access latencies in high-performance processors. In this work, we demonstrate that: (1) prefetching and off-chip prediction often provide complementary performance benefits, yet (2) naively combining them often fails to realize their full performance potential, and (3) existing prefetcher control policies leave significant room for performance improvement behind. Our goal is to design a holistic framework that can autonomously learn to coordinate an off-chip predictor with multiple prefetchers employed at various cache levels. To this end, we propose a new technique called Athena, which models the coordination between prefetchers and off-chip predictor (OCP) as a reinforcement learning (RL) problem. Athena acts as the RL agent that observes multiple system-level features (e.g., prefetcher/OCP accuracy, bandwidth usage) over an epoch of program execution, and uses them as state information to select a coordination action (i.e., enabling the prefetcher and/or OCP, and adjusting prefetcher aggressiveness). At the end of every epoch, Athena receives a numerical reward that measures the change in multiple system-level metrics (e.g., number of cycles taken to execute an epoch). Athena uses this reward to autonomously and continuously learn a policy to coordinate prefetchers with OCP. Our extensive evaluation using a diverse set of memory-intensive workloads shows that Athena consistently outperforms prior state-of-the-art coordination policies across a wide range of system configurations with various combinations of underlying prefetchers, OCPs, and main memory bandwidths, while incurring only modest storage overhead. Athena is freely available at https://github.com/CMU-SAFARI/Athena. --- ### ToS: A Team of Specialists ensemble framework for Stereo Sound Event Localization and Detection with distance estimation in Video Authors: Davide Berghi, Philip J. B. Jackson Published: 2026-01-24 Categories: eess.AS, cs.LG, cs.SD, eess.IV, eess.SP arXiv: https://arxiv.org/abs/2601.17611v1 Sound event localization and detection with distance estimation (3D SELD) in video involves identifying active sound events at each time frame while estimating their spatial coordinates. This multimodal task requires joint reasoning across semantic, spatial, and temporal dimensions, a challenge that single models often struggle to address effectively. To tackle this, we introduce the Team of Specialists (ToS) ensemble framework, which integrates three complementary sub-networks: a spatio-linguistic model, a spatio-temporal model, and a tempo-linguistic model. Each sub-network specializes in a unique pair of dimensions, contributing distinct insights to the final prediction, akin to a collaborative team with diverse expertise. ToS has been benchmarked against state-of-the-art audio-visual models for 3D SELD on the DCASE2025 Task 3 Stereo SELD development set, consistently outperforming existing methods across key metrics. Future work will extend this proof of concept by strengthening the specialists with appropriate tasks, training, and pre-training curricula. --- ### What Language Models Know But Don't Say: Non-Generative Prior Extraction for Generalization Authors: Sara Rezaeimanesh, Kundan Thind, Farzan Siddiqui, Mohammad M. Ghassemi Published: 2026-01-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.17609v3 In domains like medicine and finance, large-scale labeled data is costly and often unavailable, leading to models trained on small datasets that struggle to generalize to real-world populations. Large language models contain extensive knowledge from years of research across these domains. We propose LoID (Logit-Informed Distributions), a deterministic method for extracting informative prior distributions for Bayesian logistic regression by directly accessing their token-level predictions. Rather than relying on generated text, we probe the model's confidence in opposing semantic directions (positive vs. negative impact) through carefully constructed sentences. By measuring how consistently the LLM favors one direction across diverse phrasings, we extract the strength and reliability of the model's belief about each feature's influence. We evaluate LoID on ten real-world tabular datasets under synthetic out-of-distribution (OOD) settings characterized by covariate shift, where the training data represents only a subset of the population. We compare our approach against (1) standard uninformative priors, (2) AutoElicit, a recent method that prompts LLMs to generate priors via text completions, (3) LLMProcesses, a method that uses LLMs to generate numerical predictions through in-context learning and (4) an oracle-style upper bound derived from fitting logistic regression on the full dataset. We assess performance using Area Under the Curve (AUC). Across datasets, LoID significantly improves performance over logistic regression trained on OOD data, recovering up to \textbf{59\%} of the performance gap relative to the oracle model. LoID outperforms AutoElicit and LLMProcessesc on 8 out of 10 datasets, while providing a reproducible and computationally efficient mechanism for integrating LLM knowledge into Bayesian inference. --- ### A Thermodynamic Theory of Learning I: Irreversible Ensemble Transport and Epistemic Costs Authors: Daisuke Okanohara Published: 2026-01-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.17607v2 Learning systems acquire structured internal representations from data, yet classical information-theoretic results state that deterministic transformations do not increase information. This raises a fundamental question: how can learning produce abstraction and insight without violating information-theoretic limits? We argue that learning is inherently an irreversible process when performed over finite time, and that the realization of epistemic structure necessarily incurs entropy production. To formalize this perspective, we model learning as a transport process in the space of probability distributions over model configurations and introduce an epistemic free-energy framework. Within this framework, we define the free-energy reduction as a bookkeeping quantity that records the total reduction of epistemic free energy along a learning trajectory. This formulation highlights that realizing such a reduction over finite time necessarily incurs irreversible entropy production. We then derive the Epistemic Speed Limit (ESL), a finite-time inequality that lower-bounds the minimal entropy production required by any learning process to realize a given distributional transformation. This bound depends only on the Wasserstein distance between initial and final ensemble distributions and is independent of the specific learning algorithm. --- ### Human-Aligned Enhancement of Programming Answers with LLMs Guided by User Feedback Authors: Suborno Deb Bappon, Saikat Mondal, Chanchal K. Roy, Kevin Schneider Published: 2026-01-24 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2601.17604v1 Large Language Models (LLMs) are widely used to support software developers in tasks such as code generation, optimization, and documentation. However, their ability to improve existing programming answers in a human-like manner remains underexplored. On technical question-and-answer platforms such as Stack Overflow (SO), contributors often revise answers based on user comments that identify errors, inefficiencies, or missing explanations. Yet roughly one-third of this feedback is never addressed due to limited time, expertise, or visibility, leaving many answers incomplete or outdated. This study investigates whether LLMs can enhance programming answers by interpreting and incorporating comment-based feedback. We make four main contributions. First, we introduce ReSOlve, a benchmark consisting of 790 SO answers with associated comment threads, annotated for improvement-related and general feedback. Second, we evaluate four state-of-the-art LLMs on their ability to identify actionable concerns, finding that DeepSeek achieves the best balance between precision and recall. Third, we present AUTOCOMBAT, an LLM-powered tool that improves programming answers by jointly leveraging user comments and question context. Compared to human revised references, AUTOCOMBAT produces near-human quality improvements while preserving the original intent and significantly outperforming the baseline. Finally, a user study with 58 practitioners shows strong practical value, with 84.5 percent indicating they would adopt or recommend the tool. Overall, AUTOCOMBAT demonstrates the potential of scalable, feedback-driven answer refinement to improve the reliability and trustworthiness of technical knowledge platforms. --- ### Understanding Transformer Encoder-Decoder Representations through Bernoulli Dropout Authors: Xuanzhou Chen Published: 2026-01-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.17602v1 We study Transformer overparameterization through the lens of angular similarity in high-dimensional encoder-decoder embeddings. We apply Bernoulli dropout between the encoder and the decoder, varying the keep probability $p$ to identify a sparsity-dependent threshold above which the Top-1 prediction is preserved. Theoretically, we prove that, if the effective sparsity embeddings is sufficiently large, and thus decoder performance, remain stable under moderate coordinate dropout. Empirically, we implement the Bernoulli dropout by constructing a new Transformer model augmented with Binary Erasure Channel (BEC) and test its performance on an English-French translation task. Experimental results visualize the trends for validation accuracies and BLEU scores, both decline sharply at some threshold. --- ### Deep Intrinsic Surprise-Regularized Control (DISRC): A Biologically Inspired Mechanism for Efficient Deep Q-Learning in Sparse Environments Authors: Yash Kini, Shiv Davay, Shreya Polavarapu Published: 2026-01-24 Categories: cs.NE, cs.LG arXiv: https://arxiv.org/abs/2601.17598v1 Deep reinforcement learning (DRL) has driven major advances in autonomous control. Still, standard Deep Q-Network (DQN) agents tend to rely on fixed learning rates and uniform update scaling, even as updates are modulated by temporal-difference (TD) error. This rigidity destabilizes convergence, especially in sparse-reward settings where feedback is infrequent. We introduce Deep Intrinsic Surprise-Regularized Control (DISRC), a biologically inspired augmentation to DQN that dynamically scales Q-updates based on latent-space surprise. DISRC encodes states via a LayerNorm-based encoder and computes a deviation-based surprise score relative to a moving latent setpoint. Each update is then scaled in proportion to both TD error and surprise intensity, promoting plasticity during early exploration and stability as familiarity increases. We evaluate DISRC on two sparse-reward MiniGrid environments, which included MiniGrid-DoorKey-8x8 and MiniGrid-LavaCrossingS9N1, under identical settings as a vanilla DQN baseline. In DoorKey, DISRC reached the first successful episode (reward > 0.8) 33% faster than the vanilla DQN baseline (79 vs. 118 episodes), with lower reward standard deviation (0.25 vs. 0.34) and higher reward area under the curve (AUC: 596.42 vs. 534.90). These metrics reflect faster, more consistent learning - critical for sparse, delayed reward settings. In LavaCrossing, DISRC achieved a higher final reward (0.95 vs. 0.93) and the highest AUC of all agents (957.04), though it converged more gradually. These preliminary results establish DISRC as a novel mechanism for regulating learning intensity in off-policy agents, improving both efficiency and stability in sparse-reward domains. By treating surprise as an intrinsic learning signal, DISRC enables agents to modulate updates based on expectation violations, enhancing decision quality when conventional value-based methods fall short. --- ### Learning to Ideate for Machine Learning Engineering Agents Authors: Yunxiang Zhang, Kang Zhou, Zhichao Xu, Kiran Ramnath, Yun Zhou, Sangmin Woo, Haibo Ding, Lin Lee Cheong Published: 2026-01-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.17596v1 Existing machine learning engineering (MLE) agents struggle to iteratively optimize their implemented algorithms for effectiveness. To address this, we introduce MLE-Ideator, a dual-agent framework that separates ideation from implementation. In our system, an implementation agent can request strategic help from a dedicated Ideator. We show this approach is effective in two ways. First, in a training-free setup, our framework significantly outperforms implementation-only agent baselines on MLE-Bench. Second, we demonstrate that the Ideator can be trained with reinforcement learning (RL) to generate more effective ideas. With only 1K training samples from 10 MLE tasks, our RL-trained Qwen3-8B Ideator achieves an 11.5% relative improvement compared to its untrained counterpart and surpasses Claude Sonnet 3.5. These results highlights a promising path toward training strategic AI systems for scientific discovery. --- ### From Chains to DAGs: Probing the Graph Structure of Reasoning in LLMs Authors: Tianjun Zhong, Linyang He, Ziyang Li, Nima Mesgarani Published: 2026-01-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.17593v3 Recent progress in large language models has renewed interest in how multi-step reasoning is represented internally. While prior work often treats reasoning as a linear chain, many reasoning problems can be more naturally modeled as directed acyclic graphs (DAGs), where intermediate conclusions branch, merge, and are reused. Whether such graph structure is reflected in model internals remains unclear. We introduce Reasoning DAG Probing, a framework for testing whether LLM hidden states linearly encode properties of an underlying reasoning DAG. We formalize each premise, intermediate conclusion, and final answer as a DAG node, and train lightweight probes to predict node depth, pairwise distance, and adjacency from hidden states. Using these probes, we analyze the layerwise emergence of DAG structure, reconstruct approximate reasoning graphs, and evaluate controls that disrupt reasoning-relevant structure. Across logical, mathematical, and code reasoning benchmarks, we find that DAG structure is meaningfully encoded in LLM representations: recoverability peaks in intermediate layers; later layers preferentially recover deeper nodes and longer-range dependencies; larger models exhibit stronger DAG recoverability. During autoregressive generation, post-trained models exhibit clearer, progressively strengthening DAG structure, which broadly anticipates answer correctness trends. These findings suggest that LLM reasoning is not purely sequential, but exhibits measurable internal graph structure. --- ### Intelligence Requires Grounding But Not Embodiment Authors: Marcus Ma, Shrikanth Narayanan Published: 2026-01-24 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2601.17588v1 Recent advances in LLMs have reignited scientific debate over whether embodiment is necessary for intelligence. We present the argument that intelligence requires grounding, a phenomenon entailed by embodiment, but not embodiment itself. We define intelligence as the possession of four properties -- motivation, predictive ability, understanding of causality, and learning from experience -- and argue that each can be achieved by a non-embodied, grounded agent. We use this to conclude that grounding, not embodiment, is necessary for intelligence. We then present a thought experiment of an intelligent LLM agent in a digital environment and address potential counterarguments. --- ### Discovery of Feasible 3D Printing Configurations for Metal Alloys via AI-driven Adaptive Experimental Design Authors: Azza Fadhel, Nathaniel W. Zuckschwerdt, Aryan Deshwal, Susmita Bose, Amit Bandyopadhyay, Jana Doppa Published: 2026-01-24 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.17587v1 Configuring the parameters of additive manufacturing processes for metal alloys is a challenging problem due to complex relationships between input parameters (e.g., laser power, scan speed) and quality of printed outputs. The standard trial-and-error approach to find feasible parameter configurations is highly inefficient because validating each configuration is expensive in terms of resources (physical and human labor) and the configuration space is very large. This paper combines the general principles of AI-driven adaptive experimental design with domain knowledge to address the challenging problem of discovering feasible configurations. The key idea is to build a surrogate model from past experiments to intelligently select a small batch of input configurations for validation in each iteration. To demonstrate the effectiveness of this methodology, we deploy it for Directed Energy Deposition process to print GRCop--42, a high-performance copper--chromium--niobium alloy developed by NASA for aerospace applications. Within three months, our approach yielded multiple defect-free outputs across a range of laser powers dramatically reducing time to result and resource expenditure compared to several months of manual experimentation by domain scientists with no success. By enabling high-quality GRCop--42 fabrication on readily available infrared laser platforms for the first time, we democratize access to this critical alloy, paving the way for cost-effective, decentralized production for aerospace applications. --- ### Stylizing ViT: Anatomy-Preserving Instance Style Transfer for Domain Generalization Authors: Sebastian Doerrich, Francesco Di Salvo, Jonas Alle, Christian Ledig Published: 2026-01-24 Categories: cs.CV, cs.LG, eess.IV arXiv: https://arxiv.org/abs/2601.17586v1 Deep learning models in medical image analysis often struggle with generalizability across domains and demographic groups due to data heterogeneity and scarcity. Traditional augmentation improves robustness, but fails under substantial domain shifts. Recent advances in stylistic augmentation enhance domain generalization by varying image styles but fall short in terms of style diversity or by introducing artifacts into the generated images. To address these limitations, we propose Stylizing ViT, a novel Vision Transformer encoder that utilizes weight-shared attention blocks for both self- and cross-attention. This design allows the same attention block to maintain anatomical consistency through self-attention while performing style transfer via cross-attention. We assess the effectiveness of our method for domain generalization by employing it for data augmentation on three distinct image classification tasks in the context of histopathology and dermatology. Results demonstrate an improved robustness (up to +13% accuracy) over the state of the art while generating perceptually convincing images without artifacts. Additionally, we show that Stylizing ViT is effective beyond training, achieving a 17% performance improvement during inference when used for test-time augmentation. The source code is available at https://github.com/sdoerrich97/stylizing-vit . --- ### Sequence Repetition Enhances Token Embeddings and Improves Sequence Labeling with Decoder-only Language Models Authors: Matija Luka Kukić, Marko Čuljak, David Dukić, Martin Tutek, Jan Šnajder Published: 2026-01-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.17585v1 Modern language models (LMs) are trained in an autoregressive manner, conditioned only on the prefix. In contrast, sequence labeling (SL) tasks assign labels to each individual input token, naturally benefiting from bidirectional context. This discrepancy has historically led SL to rely on inherently bidirectional encoder-only models. However, the rapid development of decoder-only models has raised the question of whether they can be adapted to SL. While causal mask removal has emerged as a viable technique for adapting decoder-only models to leverage the full context for SL, it requires considerable changes to the base model functionality. In this work, we explore sequence repetition (SR) as a less invasive alternative for enabling bidirectionality in decoder-only models. Through fine-tuning experiments, we show that SR inherently makes decoders bidirectional, improving the quality of token-level embeddings and surpassing encoders and unmasked decoders. Contrary to earlier claims, we find that increasing the number of repetitions does not degrade SL performance. Finally, we demonstrate that embeddings from intermediate layers are highly effective for SR, comparable to those from final layers, while being significantly more efficient to compute. Our findings underscore that SR alleviates the structural limitations of decoders, enabling more efficient and adaptable LMs and broadening their applicability to other token-level tasks. --- ### Prompt Driven Development with Claude Code: Building a Complete TUI Framework for the Ring Programming Language Authors: Mahmoud Samir Fayed, Ahmed Samir Fayed Published: 2026-01-24 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2601.17584v1 Large language models are increasingly used in software development, yet their ability to generate and maintain large, multi module systems through natural language interaction remains insufficiently characterized. This study presents an empirical analysis of developing a 7420 line Terminal User Interface framework for the Ring programming language, completed in roughly ten hours of active work spread across three days using a purely prompt driven workflow with Claude Code, Opus 4.5. The system was produced through 107 prompts: 21 feature requests, 72 bug fix prompts, 9 prompts sharing information from Ring documentation, 4 prompts providing architectural guidance, and 1 prompt dedicated to generating documentation. Development progressed across five phases, with the Window Manager phase requiring the most interaction, followed by complex UI systems and controls expansion. Bug related prompts covered redraw issues, event handling faults, runtime errors, and layout inconsistencies, while feature requests focused primarily on new widgets, window manager capabilities, and advanced UI components. Most prompts were short, reflecting a highly iterative workflow in which the human role was limited to specifying requirements, validating behaviour, and issuing corrective prompts without writing any code manually. The resulting framework includes a complete windowing subsystem, event driven architecture, interactive widgets, hierarchical menus, grid and tree components, tab controls, and a multi window desktop environment. By combining quantitative prompt analysis with qualitative assessment of model behaviour, this study provides empirical evidence that modern LLMs can sustain architectural coherence and support the construction of production grade tooling for emerging programming languages, highlighting prompt driven development as a viable methodology within software engineering practice. --- ### GenAI-Net: A Generative AI Framework for Automated Biomolecular Network Design Authors: Maurice Filo, Nicolò Rossi, Zhou Fang, Mustafa Khammash Published: 2026-01-24 Categories: q-bio.QM, cs.AI, cs.LG, eess.SY, q-bio.MN arXiv: https://arxiv.org/abs/2601.17582v2 Biomolecular networks underpin emerging technologies in synthetic biology-from robust biomanufacturing and metabolic engineering to smart therapeutics and cell-based diagnostics-and also provide a mechanistic language for understanding complex dynamics in natural and ecological systems. Yet designing chemical reaction networks (CRNs) that implement a desired dynamical function remains largely manual: while a proposed network can be checked by simulation, the reverse problem of discovering a network from a behavioral specification is difficult, requiring substantial human insight to navigate a vast space of topologies and kinetic parameters with nonlinear and possibly stochastic dynamics. Here we introduce GenAI-Net, a generative AI framework that automates CRN design by coupling an agent that proposes reactions to simulation-based evaluation defined by a user-specified objective. GenAI-Net efficiently produces novel, topologically diverse solutions across multiple design tasks, including dose responses, complex logic gates, classifiers, oscillators, and robust perfect adaptation in deterministic and stochastic settings (including noise reduction). By turning specifications into families of circuit candidates and reusable motifs, GenAI-Net provides a general route to programmable biomolecular circuit design and accelerates the translation from desired function to implementable mechanisms. --- ### How AI Coding Agents Modify Code: A Large-Scale Study of GitHub Pull Requests Authors: Daniel Ogenrwot, John Businge Published: 2026-01-24 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2601.17581v3 AI coding agents are increasingly acting as autonomous contributors by generating and submitting pull requests (PRs). However, we lack empirical evidence on how these agent-generated PRs differ from human contributions, particularly in how they modify code and describe their changes. Understanding these differences is essential for assessing their reliability and impact on development workflows. Using the MSR 2026 Mining Challenge version of the AIDev dataset, we analyze 24,014 merged Agentic PRs (440,295 commits) and 5,081 merged Human PRs (23,242 commits). We examine additions, deletions, commits, and files touched, and evaluate the consistency between PR descriptions and their diffs using lexical and semantic similarity. Agentic PRs differ substantially from Human PRs in commit count (Cliff's $δ= 0.5429$) and show moderate differences in files touched and deleted lines. They also exhibit slightly higher description-to-diff similarity across all measures. These findings provide a large-scale empirical characterization of how AI coding agents contribute to open source development. --- ### Status Hierarchies in Language Models Authors: Emilio Barkett Published: 2026-01-24 Categories: cs.HC, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2601.17577v1 From school playgrounds to corporate boardrooms, status hierarchies -- rank orderings based on respect and perceived competence -- are universal features of human social organization. Language models trained on human-generated text inevitably encounter these hierarchical patterns embedded in language, raising the question of whether they might reproduce such dynamics in multi-agent settings. This thesis investigates when and how language models form status hierarchies by adapting Berger et al.'s (1972) expectation states framework. I create multi-agent scenarios where separate language model instances complete sentiment classification tasks, are introduced with varying status characteristics (e.g., credentials, expertise), then have opportunities to revise their initial judgments after observing their partner's responses. The dependent variable is deference, the rate at which models shift their ratings toward their partner's position based on status cues rather than task information. Results show that language models form significant status hierarchies when capability is equal (35 percentage point asymmetry, p < .001), but capability differences dominate status cues, with the most striking effect being that high-status assignments reduce higher-capability models' deference rather than increasing lower-capability models' deference. The implications for AI safety are significant: status-seeking behavior could introduce deceptive strategies, amplify discriminatory biases, and scale across distributed deployments far faster than human hierarchies form organically. This work identifies emergent social behaviors in AI systems and highlights a previously underexplored dimension of the alignment challenge. --- ### ME-WARD: A multimodal ergonomic analysis tool for musculoskeletal risk assessment from inertial and video data in working plac Authors: Javier González-Alonso, Paula Martín-Tapia, David González-Ortega, Míriam Antón-Rodríguez, Francisco Javier Díaz-Pernas, Mario Martínez-Zarzuela Published: 2026-01-24 Categories: eess.SP, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2601.17571v1 This study presents ME-WARD (Multimodal Ergonomic Workplace Assessment and Risk from Data), a novel system for ergonomic assessment and musculoskeletal risk evaluation that implements the Rapid Upper Limb Assessment (RULA) method. ME-WARD is designed to process joint angle data from motion capture systems, including inertial measurement unit (IMU)-based setups, and deep learning human body pose tracking models. The tool's flexibility enables ergonomic risk assessment using any system capable of reliably measuring joint angles, extending the applicability of RULA beyond proprietary setups. To validate its performance, the tool was tested in an industrial setting during the assembly of conveyor belts, which involved high-risk tasks such as inserting rods and pushing conveyor belt components. The experiments leveraged gold standard IMU systems alongside a state-of-the-art monocular 3D pose estimation system. The results confirmed that ME-WARD produces reliable RULA scores that closely align with IMU-derived metrics for flexion-dominated movements and comparable performance with the monocular system, despite limitations in tracking lateral and rotational motions. This work highlights the potential of integrating multiple motion capture technologies into a unified and accessible ergonomic assessment pipeline. By supporting diverse input sources, including low-cost video-based systems, the proposed multimodal approach offers a scalable, cost-effective solution for ergonomic assessments, paving the way for broader adoption in resource-constrained industrial environments. --- ### Quantum-Inspired Episode Selection for Monte Carlo Reinforcement Learning via QUBO Optimization Authors: Hadi Salloum, Ali Jnadi, Yaroslav Kholodov, Alexander Gasnikov Published: 2026-01-24 Categories: cs.LG, cs.RO arXiv: https://arxiv.org/abs/2601.17570v1 Monte Carlo (MC) reinforcement learning suffers from high sample complexity, especially in environments with sparse rewards, large state spaces, and correlated trajectories. We address these limitations by reformulating episode selection as a Quadratic Unconstrained Binary Optimization (QUBO) problem and solving it with quantum-inspired samplers. Our method, MC+QUBO, integrates a combinatorial filtering step into standard MC policy evaluation: from each batch of trajectories, we select a subset that maximizes cumulative reward while promoting state-space coverage. This selection is encoded as a QUBO, where linear terms favor high-reward episodes and quadratic terms penalize redundancy. We explore both Simulated Quantum Annealing (SQA) and Simulated Bifurcation (SB) as black-box solvers within this framework. Experiments in a finite-horizon GridWorld demonstrate that MC+QUBO outperforms vanilla MC in convergence speed and final policy quality, highlighting the potential of quantum-inspired optimization as a decision-making subroutine in reinforcement learning. --- ### Improving User Privacy in Personalized Generation: Client-Side Retrieval-Augmented Modification of Server-Side Generated Speculations Authors: Alireza Salemi, Hamed Zamani Published: 2026-01-24 Categories: cs.CL, cs.AI, cs.CR, cs.IR arXiv: https://arxiv.org/abs/2601.17569v1 Personalization is crucial for aligning Large Language Model (LLM) outputs with individual user preferences and background knowledge. State-of-the-art solutions are based on retrieval augmentation, where relevant context from a user profile is retrieved for LLM consumption. These methods deal with a trade-off between exposing retrieved private data to cloud providers and relying on less capable local models. We introduce $P^3$, an interactive framework for high-quality personalization without revealing private profiles to server-side LLMs. In $P^3$, a large server-side model generates a sequence of $k$ draft tokens based solely on the user query, while a small client-side model, with retrieval access to the user's private profile, evaluates and modifies these drafts to better reflect user preferences. This process repeats until an end token is generated. Experiments on LaMP-QA, a recent benchmark consisting of three personalized question answering datasets, show that $P^3$ consistently outperforms both non-personalized server-side and personalized client-side baselines, achieving statistically significant improvements of $7.4%$ to $9%$ on average. Importantly, $P^3$ recovers $90.3%$ to $95.7%$ of the utility of a ``leaky'' upper-bound scenario in which the full profile is exposed to the large server-side model. Privacy analyses, including linkability and attribute inference attacks, indicate that $P^3$ preserves the privacy of a non-personalized server-side model, introducing only marginal additional leakage ($1.5%$--$3.5%$) compared to submitting a query without any personal context. Additionally, the framework is efficient for edge deployment, with the client-side model generating only $9.2%$ of the total tokens. These results demonstrate that $P^3$ provides a practical, effective solution for personalized generation with improved privacy. --- ### Real-Time Trend Prediction via Continually-Aligned LLM Query Generation Authors: Zijing Hui, Wenhan Lyu, Shusen Wang, Li Chen, Chu Wang Published: 2026-01-24 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2601.17567v1 Trending news detection in low-traffic search environments faces a fundamental cold-start problem, where a lack of query volume prevents systems from identifying emerging or long-tail trends. Existing methods relying on keyword frequency or query spikes are inherently slow and ineffective in these sparse settings, lagging behind real-world shifts in attention. We introduce RTTP, a novel Real-Time Trending Prediction framework that generates search queries directly from news content instead of waiting for users to issue them. RTTP leverages a continual learning LLM (CL-LLM) that converts posts into search-style queries and scores them using engagement strength + creator authority, enabling early trend surfacing before search volume forms. To ensure adaptation without degrading reasoning, we propose Mix-Policy DPO, a new preference-based continual learning approach that combines on-policy stability with off-policy novelty to mitigate catastrophic forgetting during model upgrades. Deployed at production scale on Facebook and Meta AI products, RTTP delivers +91.4% improvement in tail-trend detection precision@500 and +19% query generation accuracy over industry baselines, while sustaining stable performance after multi-week online training. This work demonstrates that LLM-generated synthetic search signals, when aligned and continually updated, unlock timely trend understanding in low-traffic search environments. --- ### Sponge Tool Attack: Stealthy Denial-of-Efficiency against Tool-Augmented Agentic Reasoning Authors: Qi Li, Xinchao Wang Published: 2026-01-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.17566v1 Enabling large language models (LLMs) to solve complex reasoning tasks is a key step toward artificial general intelligence. Recent work augments LLMs with external tools to enable agentic reasoning, achieving high utility and efficiency in a plug-and-play manner. However, the inherent vulnerabilities of such methods to malicious manipulation of the tool-calling process remain largely unexplored. In this work, we identify a tool-specific attack surface and propose Sponge Tool Attack (STA), which disrupts agentic reasoning solely by rewriting the input prompt under a strict query-only access assumption. Without any modification on the underlying model or the external tools, STA converts originally concise and efficient reasoning trajectories into unnecessarily verbose and convoluted ones before arriving at the final answer. This results in substantial computational overhead while remaining stealthy by preserving the original task semantics and user intent. To achieve this, we design STA as an iterative, multi-agent collaborative framework with explicit rewritten policy control, and generates benign-looking prompt rewrites from the original one with high semantic fidelity. Extensive experiments across 6 models (including both open-source models and closed-source APIs), 12 tools, 4 agentic frameworks, and 13 datasets spanning 5 domains validate the effectiveness of STA. --- ### JaxARC: A High-Performance JAX-based Environment for Abstraction and Reasoning Research Authors: Aadam, Monu Verma, Mohamed Abdel-Mottaleb Published: 2026-01-24 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.17564v1 The Abstraction and Reasoning Corpus (ARC) tests AI systems' ability to perform human-like inductive reasoning from a few demonstration pairs. Existing Gymnasium-based RL environments severely limit experimental scale due to computational bottlenecks. We present JaxARC, an open-source, high-performance RL environment for ARC implemented in JAX. Its functional, stateless architecture enables massive parallelism, achieving 38-5,439x speedup over Gymnasium at matched batch sizes, with peak throughput of 790M steps/second. JaxARC supports multiple ARC datasets, flexible action spaces, composable wrappers, and configuration-driven reproducibility, enabling large-scale RL research previously computationally infeasible. JaxARC is available at https://github.com/aadimator/JaxARC. --- ### Towards Generalisable Imitation Learning Through Conditioned Transition Estimation and Online Behaviour Alignment Authors: Nathan Gavenski, Matteo Leonetti, Odinaldo Rodrigues Published: 2026-01-24 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.17563v1 State-of-the-art imitation learning from observation methods (ILfO) have recently made significant progress, but they still have some limitations: they need action-based supervised optimisation, assume that states have a single optimal action, and tend to apply teacher actions without full consideration of the actual environment state. While the truth may be out there in observed trajectories, existing methods struggle to extract it without supervision. In this work, we propose Unsupervised Imitation Learning from Observation (UfO) that addresses all of these limitations. UfO learns a policy through a two-stage process, in which the agent first obtains an approximation of the teacher's true actions in the observed state transitions, and then refines the learned policy further by adjusting agent trajectories to closely align them with the teacher's. Experiments we conducted in five widely used environments show that UfO not only outperforms the teacher and all other ILfO methods but also displays the smallest standard deviation. This reduction in standard deviation indicates better generalisation in unseen scenarios. --- ### Sparse RBF Networks for PDEs and nonlocal equations: function space theory, operator calculus, and training algorithms Authors: Zihan Shao, Konstantin Pieper, Xiaochuan Tian Published: 2026-01-24 Categories: math.NA, cs.LG arXiv: https://arxiv.org/abs/2601.17562v1 This work presents a systematic analysis and extension of the sparse radial basis function network (SparseRBFnet) previously introduced for solving nonlinear partial differential equations (PDEs). Based on its adaptive-width shallow kernel network formulation, we further investigate its function-space characterization, operator evaluation, and computational algorithm. We provide a unified description of the solution space for a broad class of radial basis functions (RBFs). Under mild assumptions, this space admits a characterization as a Besov space, independent of the specific kernel choice. We further demonstrate how the explicit kernel-based structure enables quasi-analytical evaluation of both differential and nonlocal operators, including fractional Laplacians. On the computational end, we study the adaptive-width network and related three-phase training strategy through a comparison with variants concerning the modeling and algorithmic details. In particular, we assess the roles of second-order optimization, inner-weight training, network adaptivity, and anisotropic kernel parameterizations. Numerical experiments on high-order, fractional, and anisotropic PDE benchmarks illustrate the empirical insensitivity to kernel choice, as well as the resulting trade-offs between accuracy, sparsity, and computational cost. Collectively, these results consolidate and generalize the theoretical and computational framework of SparseRBFnet, supporting accurate sparse representations with efficient operator evaluation and offering theory-grounded guidance for algorithmic and modeling choices. --- ### Saliency Driven Imagery Preprocessing for Efficient Compression -- Industrial Paper Authors: Justin Downes, Sam Saltwick, Anthony Chen Published: 2026-01-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.17555v1 The compression of satellite imagery remains an important research area as hundreds of terabytes of images are collected every day, which drives up storage and bandwidth costs. Although progress has been made in increasing the resolution of these satellite images, many downstream tasks are only interested in small regions of any given image. These areas of interest vary by task but, once known, can be used to optimize how information within the image is encoded. Whereas standard image encoding methods, even those optimized for remote sensing, work on the whole image equally, there are emerging methods that can be guided by saliency maps to focus on important areas. In this work we show how imagery preprocessing techniques driven by saliency maps can be used with traditional lossy compression coding standards to create variable rate image compression within a single large satellite image. Specifically, we use variable sized smoothing kernels that map to different quantized saliency levels to process imagery pixels in order to optimize downstream compression and encoding schemes. --- ### GreenServ: Energy-Efficient Context-Aware Dynamic Routing for Multi-Model LLM Inference Authors: Thomas Ziller, Shashikant Ilager, Alessandro Tundo, Ezio Bartocci, Leonardo Mariani, Ivona Brandic Published: 2026-01-24 Categories: cs.PF, cs.LG arXiv: https://arxiv.org/abs/2601.17551v2 Large language models (LLMs) demonstrate remarkable capabilities, but their broad deployment is limited by significant computational resource demands, particularly energy consumption during inference. Static, one-model-fits-all inference strategies are often inefficient, as they do not exploit the diverse range of available models or adapt to varying query requirements. This paper presents GreenServ, a dynamic, context-aware routing framework that optimizes the trade-off between inference accuracy and energy efficiency. GreenServ extracts lightweight contextual features from each query, including task type, semantic cluster, and text complexity, and routes queries to the most suitable model from a heterogeneous pool, based on observed accuracy and energy usage. We employ a multi-armed bandit approach to learn adaptive routing policies online. This approach operates under partial feedback, eliminates the need for extensive offline calibration, and streamlines the integration of new models into the inference pipeline. We evaluated GreenServ across five benchmark tasks and a pool of 16 contemporary open-access LLMs. Experimental results show that GreenServ consistently outperforms static (single-model) and random baselines. In particular, compared to random routing, GreenServ achieved a 22% increase in accuracy while reducing cumulative energy consumption by 31%. Finally, we evaluated GreenServ with RouterBench, achieving an average accuracy of 71.7% with a peak accuracy of 75.7%. All artifacts are open-source and available here: \href{https://github.com/TZData1/llm-inference-router}{GitHub} --- ### Breaking the Protocol: Security Analysis of the Model Context Protocol Specification and Prompt Injection Vulnerabilities in Tool-Integrated LLM Agents Authors: Narek Maloyan, Dmitry Namiot Published: 2026-01-24 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2601.17549v1 The Model Context Protocol (MCP) has emerged as a de facto standard for integrating Large Language Models with external tools, yet no formal security analysis of the protocol specification exists. We present the first rigorous security analysis of MCP's architectural design, identifying three fundamental protocol-level vulnerabilities: (1) absence of capability attestation allowing servers to claim arbitrary permissions, (2) bidirectional sampling without origin authentication enabling server-side prompt injection, and (3) implicit trust propagation in multi-server configurations. We implement \textsc{MCPBench}, a novel framework bridging existing agent security benchmarks to MCP-compliant infrastructure, enabling direct measurement of protocol-specific attack surfaces. Through controlled experiments on 847 attack scenarios across five MCP server implementations, we demonstrate that MCP's architectural choices amplify attack success rates by 23--41\% compared to equivalent non-MCP integrations. We propose \textsc{MCPSec}, a backward-compatible protocol extension adding capability attestation and message authentication, reducing attack success rates from 52.8\% to 12.4\% with median latency overhead of 8.3ms per message. Our findings establish that MCP's security weaknesses are architectural rather than implementation-specific, requiring protocol-level remediation. --- ### In-situ On-demand Digital Image Correlation: A New Data-rich Characterization Paradigm for Deformation and Damage Development in Solids Authors: Ravi Venkata Surya Sai Mogilisetti, Partha Pratim Das, Rassel Raihan, Shiyao Lin Published: 2026-01-24 Categories: eess.IV, cond-mat.mtrl-sci, cs.CV arXiv: https://arxiv.org/abs/2601.17545v1 Digital image correlation (DIC) has become one of the most popular methods for deformation characterization in experimental mechanics. DIC is based on optical images taken during experimentation and post-test image processing. Its advantages include the capability to capture full-field deformation in a non-contact manner, the robustness in characterizing excessive deformation induced by events such as yielding and cracking, and the versatility to integrate optical cameras with a variety of open-source and commercial codes. In this paper, we developed a new paradigm of DIC analysis by integrating camera control into the DIC process flow. The essential idea is to dynamically increase the camera imaging frame rate with excessive deformation or deformation rate, while maintaining a relatively low imaging frame rate with small and slow deformation. We refer to this new DIC paradigm as in-situ on-demand (ISOD) DIC. ISOD DIC enables real-time deformation analysis, visualization, and closed-loop camera control. ISOD DIC has captured approximately 178% more images than conventional DIC for samples undergoing crack growth due to its dynamically adjusted frame rate, with the potential to significantly enhance data richness for damage inspection without consuming excessive storage space and analysis time, thereby benefiting the characterization of intrinsic constitutive behaviors and damage mechanisms --- ### Cognitive Platform Engineering for Autonomous Cloud Operations Authors: Vinoth Punniyamoorthy, Nitin Saksena, Srivenkateswara Reddy Sankiti, Nachiappan Chockalingam, Aswathnarayan Muthukrishnan Kirubakaran, Shiva Kumar Reddy Carimireddy, Durgaraman Maruthavanan Published: 2026-01-24 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2601.17542v1 Modern DevOps practices have accelerated software delivery through automation, CI/CD pipelines, and observability tooling,but these approaches struggle to keep pace with the scale and dynamism of cloud-native systems. As telemetry volume grows and configuration drift increases, traditional, rule-driven automation often results in reactive operations, delayed remediation, and dependency on manual expertise. This paper introduces Cognitive Platform Engineering, a next-generation paradigm that integrates sensing, reasoning, and autonomous action directly into the platform lifecycle. This paper propose a four-plane reference architecture that unifies data collection, intelligent inference, policy-driven orchestration, and human experience layers within a continuous feedback loop. A prototype implementation built with Kubernetes, Terraform, Open Policy Agent, and ML-based anomaly detection demonstrates improvements in mean time to resolution, resource efficiency, and compliance. The results show that embedding intelligence into platform operations enables resilient, self-adjusting, and intent-aligned cloud environments. The paper concludes with research opportunities in reinforcement learning, explainable governance, and sustainable self-managing cloud ecosystems. --- ### OTI: A Model-free and Visually Interpretable Measure of Image Attackability Authors: Jiaming Liang, Haowei Liu, Chi-Man Pun Published: 2026-01-24 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2601.17536v1 Despite the tremendous success of neural networks, benign images can be corrupted by adversarial perturbations to deceive these models. Intriguingly, images differ in their attackability. Specifically, given an attack configuration, some images are easily corrupted, whereas others are more resistant. Evaluating image attackability has important applications in active learning, adversarial training, and attack enhancement. This prompts a growing interest in developing attackability measures. However, existing methods are scarce and suffer from two major limitations: (1) They rely on a model proxy to provide prior knowledge (e.g., gradients or minimal perturbation) to extract model-dependent image features. Unfortunately, in practice, many task-specific models are not readily accessible. (2) Extracted features characterizing image attackability lack visual interpretability, obscuring their direct relationship with the images. To address these, we propose a novel Object Texture Intensity (OTI), a model-free and visually interpretable measure of image attackability, which measures image attackability as the texture intensity of the image's semantic object. Theoretically, we describe the principles of OTI from the perspectives of decision boundaries as well as the mid- and high-frequency characteristics of adversarial perturbations. Comprehensive experiments demonstrate that OTI is effective and computationally efficient. In addition, our OTI provides the adversarial machine learning community with a visual understanding of attackability. --- ### Will It Zero-Shot?: Predicting Zero-Shot Classification Performance For Arbitrary Queries Authors: Kevin Robbins, Xiaotong Liu, Yu Wu, Le Sun, Grady McPeak, Abby Stylianou, Robert Pless Published: 2026-01-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.17535v2 Vision-Language Models like CLIP create aligned embedding spaces for text and images, making it possible for anyone to build a visual classifier by simply naming the classes they want to distinguish. However, a model that works well in one domain may fail in another, and non-expert users have no straightforward way to assess whether their chosen VLM will work on their problem. We build on prior work using text-only comparisons to evaluate how well a model works for a given natural language task, and explore approaches that also generate synthetic images relevant to that task to evaluate and refine the prediction of zero-shot accuracy. We show that generated imagery to the baseline text-only scores substantially improves the quality of these predictions. Additionally, it gives a user feedback on the kinds of images that were used to make the assessment. Experiments on standard CLIP benchmark datasets demonstrate that the image-based approach helps users predict, without any labeled examples, whether a VLM will be effective for their application. --- ### Reconstructing Training Data from Adapter-based Federated Large Language Models Authors: Silong Chen, Yuchuan Luo, Guilin Deng, Yi Liu, Min Xu, Shaojing Fu, Xiaohua Jia Published: 2026-01-24 Categories: cs.CR, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.17533v1 Adapter-based Federated Large Language Models (FedLLMs) are widely adopted to reduce the computational, storage, and communication overhead of full-parameter fine-tuning for web-scale applications while preserving user privacy. By freezing the backbone and training only compact low-rank adapters, these methods appear to limit gradient leakage and thwart existing Gradient Inversion Attacks (GIAs). Contrary to this assumption, we show that low-rank adapters create new, exploitable leakage channels. We propose the Unordered-word-bag-based Text Reconstruction (UTR) attack, a novel GIA tailored to the unique structure of adapter-based FedLLMs. UTR overcomes three core challenges: low-dimensional gradients, frozen backbones, and combinatorially large reconstruction spaces by: (i) inferring token presence from attention patterns in frozen layers, (ii) performing sentence-level inversion within the low-rank subspace of adapter gradients, and (iii) enforcing semantic coherence through constrained greedy decoding guided by language priors. Extensive experiments across diverse models (GPT2-Large, BERT, Qwen2.5-7B) and datasets (CoLA, SST-2, Rotten Tomatoes) demonstrate that UTR achieves near-perfect reconstruction accuracy (ROUGE-1/2 > 99), even with large batch size settings where prior GIAs fail completely. Our results reveal a fundamental tension between parameter efficiency and privacy in FedLLMs, challenging the prevailing belief that lightweight adaptation inherently enhances security. Our code and data are available at https://github.com/shwksnshwowk-wq/GIA. --- ### Less is More for RAG: Information Gain Pruning for Generator-Aligned Reranking and Evidence Selection Authors: Zhipeng Song, Yizhi Zhou, Xiangyu Kong, Jiulong Jiao, Xinrui Bao, Xu You, Xueqing Shi, Yuhang Zhou, Heng Qi Published: 2026-01-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.17532v1 Retrieval-augmented generation (RAG) grounds large language models with external evidence, but under a limited context budget, the key challenge is deciding which retrieved passages should be injected. We show that retrieval relevance metrics (e.g., NDCG) correlate weakly with end-to-end QA quality and can even become negatively correlated under multi-passage injection, where redundancy and mild conflicts destabilize generation. We propose \textbf{Information Gain Pruning (IGP)}, a deployment-friendly reranking-and-pruning module that selects evidence using a generator-aligned utility signal and filters weak or harmful passages before truncation, without changing existing budget interfaces. Across five open-domain QA benchmarks and multiple retrievers and generators, IGP consistently improves the quality--cost trade-off. In a representative multi-evidence setting, IGP delivers about +12--20% relative improvement in average F1 while reducing final-stage input tokens by roughly 76--79% compared to retriever-only baselines. --- ### Revealing the Truth with ConLLM for Detecting Multi-Modal Deepfakes Authors: Gautam Siddharth Kashyap, Harsh Joshi, Niharika Jain, Ebad Shabbir, Jiechao Gao, Nipun Joshi, Usman Naseem Published: 2026-01-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.17530v1 The rapid rise of deepfake technology poses a severe threat to social and political stability by enabling hyper-realistic synthetic media capable of manipulating public perception. However, existing detection methods struggle with two core limitations: (1) modality fragmentation, which leads to poor generalization across diverse and adversarial deepfake modalities; and (2) shallow inter-modal reasoning, resulting in limited detection of fine-grained semantic inconsistencies. To address these, we propose ConLLM (Contrastive Learning with Large Language Models), a hybrid framework for robust multimodal deepfake detection. ConLLM employs a two-stage architecture: stage 1 uses Pre-Trained Models (PTMs) to extract modality-specific embeddings; stage 2 aligns these embeddings via contrastive learning to mitigate modality fragmentation, and refines them using LLM-based reasoning to address shallow inter-modal reasoning by capturing semantic inconsistencies. ConLLM demonstrates strong performance across audio, video, and audio-visual modalities. It reduces audio deepfake EER by up to 50%, improves video accuracy by up to 8%, and achieves approximately 9% accuracy gains in audio-visual tasks. Ablation studies confirm that PTM-based embeddings contribute 9%-10% consistent improvements across modalities. --- ### FMIR, a foundation model-based Image Registration Framework for Robust Image Registration Authors: Fengting Zhang, Yue He, Qinghao Liu, Yaonan Wang, Xiang Chen, Hang Zhang Published: 2026-01-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.17529v2 Deep learning has revolutionized medical image registration by achieving unprecedented speeds, yet its clinical application is hindered by a limited ability to generalize beyond the training domain, a critical weakness given the typically small scale of medical datasets. In this paper, we introduce FMIR, a foundation model-based registration framework that overcomes this limitation.Combining a foundation model-based feature encoder for extracting anatomical structures with a general registration head, and trained with a channel regularization strategy on just a single dataset, FMIR achieves state-of-the-art(SOTA) in-domain performance while maintaining robust registration on out-of-domain images.Our approach demonstrates a viable path toward building generalizable medical imaging foundation models with limited resources. The code is available at https://github.com/Monday0328/FMIR.git. --- ### Bridging Expectation Signals: LLM-Based Experiments and a Behavioral Kalman Filter Framework Authors: Yu Wang, Xiangchen Liu Published: 2026-01-24 Categories: econ.GN, cs.AI arXiv: https://arxiv.org/abs/2601.17527v1 As LLMs increasingly function as economic agents, the specific mechanisms LLMs use to update their belief with heterogeneous signals remain opaque. We design experiments and develop a Behavioral Kalman Filter framework to quantify how LLM-based agents update expectations, acting as households or firm CEOs, update expectations when presented with individual and aggregate signals. The results from experiments and model estimation reveal four consistent patterns: (1) agents' weighting of priors and signals deviates from unity; (2) both household and firm CEO agents place substantially larger weights on individual signals compared to aggregate signals; (3) we identify a significant and negative interaction between concurrent signals, implying that the presence of multiple information sources diminishes the marginal weight assigned to each individual signal; and (4) expectation formation patterns differ significantly between household and firm CEO agents. Finally, we demonstrate that LoRA fine-tuning mitigates, but does not fully eliminate, behavioral biases in LLM expectation formation. --- ### EuleroDec: A Complex-Valued RVQ-VAE for Efficient and Robust Audio Coding Authors: Luca Cerovaz, Michele Mancusi, Emanuele Rodolà Published: 2026-01-24 Categories: cs.SD, cs.LG, eess.AS arXiv: https://arxiv.org/abs/2601.17517v2 Audio codecs power discrete music generative modelling, music streaming and immersive media by shrinking PCM audio to bandwidth-friendly bit-rates. Recent works have gravitated towards processing in the spectral domain; however, spectrogram-domains typically struggle with phase modeling which is naturally complex-valued. Most frequency-domain neural codecs either disregard phase information or encode it as two separate real-valued channels, limiting spatial fidelity. This entails the need to introduce adversarial discriminators at the expense of convergence speed and training stability to compensate for the inadequate representation power of the audio signal. In this work we introduce an end-to-end complex-valued RVQ-VAE audio codec that preserves magnitude-phase coupling across the entire analysis-quantization-synthesis pipeline and removes adversarial discriminators and diffusion post-filters. Without GANs or diffusion we match or surpass much longer-trained baselines in-domain and reach SOTA out-of-domain performance. Compared to standard baselines that train for hundreds of thousands of steps, our model reducing training budget by an order of magnitude is markedly more compute-efficient while preserving high perceptual quality. --- ### Constrained Multi-Objective Genetic Algorithm Variants for Design and Optimization of Tri-Band Microstrip Patch Antenna loaded CSRR for IoT Applications: A Comparative Case Study Authors: Moahmed Hamza Boulaich, Said Ohamouddou, Mohammed Ali Ennasar, Abdelatif El Afia Published: 2026-01-24 Categories: cs.NE, cs.NI arXiv: https://arxiv.org/abs/2601.17513v1 This paper presents an automated antenna design and optimization framework employing multi-objective genetic algorithms (MOGAs) to investigate various evolutionary optimization approaches, with a primary emphasis on multi-band frequency optimization. Five MOGA variants were implemented and compared: the Pareto genetic algorithm (PGA), non-dominated sorting genetic algorithm with niching (NSGA-I), non-dominated sorting genetic algorithm with elitism (NSGA-II), non-dominated sorting genetic algorithm using reference points (NSGA-III), and strength Pareto evolutionary algorithm (SPEA). These algorithms are employed to design and optimize microstrip patch antennas loaded with complementary split-ring resonators (CSRRs). A weighted-sum scalarization approach was adopted within a single-objective genetic algorithm framework enhanced with domain-specific constraint handling mechanisms. The optimization addresses the conflicting objectives of minimizing the return loss ($S_{11} < -10$~dB) and achieving multi-band resonance at 2.4~GHz, 3.6~GHz, and 5.2~GHz. The proposed method delivers a superior overall performance by aggregating these objectives into a unified fitness function encompassing $S_{11}$(2.4~GHz), $S_{11}$(3.6~GHz), and $S_{11}$(5.2~GHz). This approach effectively balances all three frequency bands simultaneously, rather than exploring trade-off solutions typical of traditional multi-objective approaches. The antenna was printed on a Rogers RT5880 substrate with a dielectric constant of 2.2 , loss tangent of 0.0009 , and thickness of 1.57~mm . Scalarization approach achieved return loss values of $-21.56$~dB, $-16.60$~dB, and $-27.69$~dB, with corresponding gains of 1.96~dBi, 2.6~dB, and 3.99~dBi at 2.4~GHz, 3.6~GHz, and 5.2~GHz, respectively. ---