# Research Papers — 2025-12 · bundle page 87 of 175

> Full-text markdown of papers #4301–4350 of 8703 published in 2025-12, newest first. AI-readable mirrors of the full 8703-paper 2025-12 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2025-12/page/87
[← Previous bundle](https://aipod.dev/papers/archive/2025-12/page/86.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-12/page/88.md)

### Next-generation reservoir computing validated by classification task

Authors: Ken-ichi Kitayama
Published: 2025-12-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2512.12903v1

An emerging computing paradigm, so-called next-generation reservoir computing (NG-RC) is investigated. True to its namesake, NG-RC requires no actual reservoirs for input data mixing but rather computing the polynomial terms directly from the time series inputs. However, benchmark tests so far reported have been one-sided, limited to prediction tasks of temporal waveforms such as Lorenz 63 attractor and Mackey-Glass chaotic signal. We will demonstrate for the first time that NG-RC can perform classification task as good as conventional RC. This validates the versatile computational capability of NG-RC in tasks of both prediction and classification.

---

### Predicted-occupancy grids for vehicle safety applications based on autoencoders and the Random Forest algorithm

Authors: Parthasarathy Nadarajan, Michael Botsch, Sebastian Sardina
Published: 2025-12-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2512.12901v1

In this paper, a probabilistic space-time representation of complex traffic scenarios is predicted using machine learning algorithms. Such a representation is significant for all active vehicle safety applications especially when performing dynamic maneuvers in a complex traffic scenario. As a first step, a hierarchical situation classifier is used to distinguish the different types of traffic scenarios. This classifier is responsible for identifying the type of the road infrastructure and the safety-relevant traffic participants of the driving environment. With each class representing similar traffic scenarios, a set of Random Forests (RFs) is individually trained to predict the probabilistic space-time representation, which depicts the future behavior of traffic participants. This representation is termed as a Predicted-Occupancy Grid (POG). The input to the RFs is an Augmented Occupancy Grid (AOG). In order to increase the learning accuracy of the RFs and to perform better predictions, the AOG is reduced to low-dimensional features using a Stacked Denoising Autoencoder (SDA). The excellent performance of the proposed machine learning approach consisting of SDAs and RFs is demonstrated in simulations and in experiments with real vehicles. An application of POGs to estimate the criticality of traffic scenarios and to determine safe trajectories is also presented.

---

### Towards High-Fidelity Gaussian Splatting with Queried-Convolution Neural Networks

Authors: Abhinav Kumar, Tristan Aumentado-Armstrong, Lazar Valkov, Gopal Sharma, Alex Levinshtein, Radek Grzeszczuk, Suren Kumar
Published: 2025-12-15
Categories: cs.CV, cs.GR, cs.LG
arXiv: https://arxiv.org/abs/2512.12898v2

Gaussian Splatting has revolutionized the field of Novel View Synthesis (NVS) with faster training and real-time rendering. However, its reconstruction fidelity still trails behind the powerful radiance models such as Zip-NeRF. Motivated by our theoretical result that both queries (such as coordinates) and neighborhood are important to learn high-fidelity signals, this paper proposes Queried-Convolutions (Qonvolutions), a simple yet powerful modification using the neighborhood properties of convolution. Qonvolutions convolve a low-fidelity signal with queries to output residual and achieve high-fidelity reconstruction. We empirically demonstrate that combining Gaussian splatting with Qonvolution neural networks (QNNs) results in state-of-the-art NVS on real-world scenes, even outperforming Zip-NeRF on image fidelity. QNNs also enhance performance of 1D regression, 2D regression and 2D super-resolution tasks.

---

### Probability Estimation for Predicted-Occupancy Grids in Vehicle Safety Applications Based on Machine Learning

Authors: Parthasarathy Nadarajan, Michael Botsch
Published: 2025-12-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2512.12896v1

This paper presents a method to predict the evolution of a complex traffic scenario with multiple objects. The current state of the scenario is assumed to be known from sensors and the prediction is taking into account various hypotheses about the behavior of traffic participants. This way, the uncertainties regarding the behavior of traffic participants can be modelled in detail. In the first part of this paper a model-based approach is presented to compute Predicted-Occupancy Grids (POG), which are introduced as a grid-based probabilistic representation of the future scenario hypotheses. However, due to the large number of possible trajectories for each traffic participant, the model-based approach comes with a very high computational load. Thus, a machine-learning approach is adopted for the computation of POGs. This work uses a novel grid-based representation of the current state of the traffic scenario and performs the mapping to POGs. This representation consists of augmented cells in an occupancy grid. The adopted machine-learning approach is based on the Random Forest algorithm. Simulations of traffic scenarios are performed to compare the machine-learning with the model-based approach. The results are promising and could enable the real-time computation of POGs for vehicle safety applications. With this detailed modelling of uncertainties, crucial components in vehicle safety systems like criticality estimation and trajectory planning can be improved.

---

### Wait, Wait, Wait... Why Do Reasoning Models Loop?

Authors: Charilaos Pipis, Shivam Garg, Vasilis Kontonis, Vaishnavi Shrivastava, Akshay Krishnamurthy, Dimitris Papailiopoulos
Published: 2025-12-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2512.12895v3

Reasoning models (e.g., DeepSeek-R1) generate long chains of thought to solve harder problems, but they often loop, repeating the same text at low temperatures or with greedy decoding. We study why this happens and what role temperature plays. With open reasoning models, we find that looping is common at low temperature. Larger models tend to loop less, and distilled students loop significantly even when their teachers rarely do. This points to mismatches between the training distribution and the learned model, which we refer to as errors in learning, as a key cause. To understand how such errors cause loops, we introduce a synthetic graph reasoning task and demonstrate two mechanisms. First, risk aversion caused by hardness of learning: when the correct progress-making action is hard to learn but an easy cyclic action is available, the model puts relatively more probability on the cyclic action and gets stuck. Second, even when there is no hardness, Transformers show an inductive bias toward temporally correlated errors, so the same few actions keep being chosen and loops appear. Higher temperature reduces looping by promoting exploration, but it does not fix the errors in learning, so generations remain much longer than necessary at high temperature; in this sense, temperature is a stopgap rather than a holistic solution. We end with a discussion of training-time interventions aimed at directly reducing errors in learning.

---

### Distillation of Discrete Diffusion by Exact Conditional Distribution Matching

Authors: Yansong Gao, Yu Sun
Published: 2025-12-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2512.12889v1

Discrete diffusion models (DDMs) are a powerful class of generative models for categorical data, but they typically require many function evaluations for a single sample, making inference expensive. Existing acceleration methods either rely on approximate simulators, such as $τ$-leaping, or on distillation schemes that train new student models and auxiliary networks with proxy objectives. We propose a simple and principled distillation alternative based on \emph{conditional distribution matching}. Our key observation is that the reverse conditional distribution of clean data given a noisy state, $p_{0\mid t}(x_0 \mid x_t)$, admits a Markov decomposition through intermediate times and can be recovered from marginal density ratios and the known forward CTMC kernel. We exploit this structure to define distillation objectives that directly match conditional distributions between a pre-trained teacher and a low-NFE student, both for one-step and few-step samplers.

---

### Meta-GPT: Decoding the Metasurface Genome with Generative Artificial Intelligence

Authors: David Dang, Stuart Love, Meena Salib, Quynh Dang, Samuel Rothfarb, Mysk Alnatour, Andrew Salij, Hou-Tong Chen, Ho Wai, Lee, Wilton J. M. Kort-Kamp
Published: 2025-12-15
Categories: physics.optics, cs.AI, cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2512.12888v1

Advancing artificial intelligence for physical sciences requires representations that are both interpretable and compatible with the underlying laws of nature. We introduce METASTRINGS, a symbolic language for photonics that expresses nanostructures as textual sequences encoding materials, geometries, and lattice configurations. Analogous to molecular textual representations in chemistry, METASTRINGS provides a framework connecting human interpretability with computational design by capturing the structural hierarchy of photonic metasurfaces. Building on this representation, we develop Meta-GPT, a foundation transformer model trained on METASTRINGS and finetuned with physics-informed supervised, reinforcement, and chain-of-thought learning. Across various design tasks, the model achieves <3% mean-squared spectral error and maintains >98% syntactic validity, generating diverse metasurface prototypes whose experimentally measured optical responses match their target spectra. These results demonstrate that Meta-GPT can learn the compositional rules of light-matter interactions through METASTRINGS, laying a rigorous foundation for AI-driven photonics and representing an important step toward a metasurface genome project.

---

### Revisiting 2D Foundation Models for Scalable 3D Medical Image Classification

Authors: Han Liu, Bogdan Georgescu, Yanbo Zhang, Youngjin Yoo, Michael Baumgartner, Riqiang Gao, Jianing Wang, Gengyan Zhao, Eli Gibson, Dorin Comaniciu, Sasa Grbic
Published: 2025-12-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.12887v3

3D medical image classification is essential for modern clinical workflows. Medical foundation models (FMs) have emerged as a promising approach for scaling to new tasks, yet current research suffers from three critical pitfalls: data-regime bias, suboptimal adaptation, and insufficient task coverage. In this paper, we address these pitfalls and introduce AnyMC3D, a scalable 3D classifier adapted from 2D FMs. Our method scales efficiently to new tasks by adding only lightweight plugins (about 1M parameters per task) on top of a single frozen backbone. This versatile framework also supports multi-view inputs, auxiliary pixel-level supervision, and interpretable heatmap generation. We establish a comprehensive benchmark of 12 tasks covering diverse pathologies, anatomies, and modalities, and systematically analyze state-of-the-art 3D classification techniques. Our analysis reveals key insights: (1) effective adaptation is essential to unlock FM potential, (2) general-purpose FMs can match medical-specific FMs if properly adapted, and (3) 2D-based methods surpass 3D architectures for 3D classification. For the first time, we demonstrate the feasibility of achieving state-of-the-art performance across diverse applications using a single scalable framework (including 1st place in the VLM3D challenge), eliminating the need for separate task-specific models.

---

### Comparative Analysis of Neural Retriever-Reranker Pipelines for Retrieval-Augmented Generation over Knowledge Graphs in E-commerce Applications

Authors: Teri Rumble, Zbyněk Gazdík, Javad Zarrin, Jagdeep Ahluwalia
Published: 2025-12-14
Categories: cs.IR, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2602.22219v1

Recent advancements in Large Language Models (LLMs) have transformed Natural Language Processing (NLP), enabling complex information retrieval and generation tasks. Retrieval-Augmented Generation (RAG) has emerged as a key innovation, enhancing factual accuracy and contextual grounding by integrating external knowledge sources with generative models. Although RAG demonstrates strong performance on unstructured text, its application to structured knowledge graphs presents challenges: scaling retrieval across connected graphs and preserving contextual relationships during response generation. Cross-encoders refine retrieval precision, yet their integration with structured data remains underexplored. Addressing these challenges is crucial for developing domain-specific assistants that operate in production environments. This study presents the design and comparative evaluation of multiple Retriever-Reranker pipelines for knowledge graph natural language queries in e-Commerce contexts. Using the STaRK Semi-structured Knowledge Base (SKB), a production-scale e-Commerce dataset, we evaluate multiple RAG pipeline configurations optimized for language queries. Experimental results demonstrate substantial improvements over published benchmarks, achieving 20.4% higher Hit@1 and 14.5% higher Mean Reciprocal Rank (MRR). These findings establish a practical framework for integrating domain-specific SKBs into generative systems. Our contributions provide actionable insights for the deployment of production-ready RAG systems, with implications that extend beyond e-Commerce to other domains that require information retrieval from structured knowledge bases.

---

### Social Determinants of Health Prediction for ICD-9 Code with Reasoning Models

Authors: Sharim Khan, Paul Landes, Adam Cross, Jimeng Sun
Published: 2025-12-14
Categories: cs.LG, cs.CL, cs.CY
arXiv: https://arxiv.org/abs/2601.09709v1

Social Determinants of Health correlate with patient outcomes but are rarely captured in structured data. Recent attention has been given to automatically extracting these markers from clinical text to supplement diagnostic systems with knowledge of patients' social circumstances. Large language models demonstrate strong performance in identifying Social Determinants of Health labels from sentences. However, prediction in large admissions or longitudinal notes is challenging given long distance dependencies. In this paper, we explore hospital admission multi-label Social Determinants of Health ICD-9 code classification on the MIMIC-III dataset using reasoning models and traditional large language models. We exploit existing ICD-9 codes for prediction on admissions, which achieved an 89% F1. Our contributions include our findings, missing SDoH codes in 139 admissions, and code to reproduce the results.

---

### DeliberationBench: When Do More Voices Hurt? A Controlled Study of Multi-LLM Deliberation Protocols

Authors: Vaarunay Kaushal, Taranveer Singh
Published: 2025-12-14
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2601.08835v1

Multi-agent systems where Large Language Models (LLMs) deliberate to form consensus have gained significant attention, yet their practical value over simpler methods remains under-scrutinized. We introduce DELIBERATIONBENCH, a controlled benchmark evaluating three deliberation protocols against a strong baseline of selecting the best response from a pool of model outputs. Across 270 questions and three independent seeds (810 total evaluations), we find a striking negative result: the best-single baseline achieves an 82.5% +- 3.3% win rate, dramatically outperforming the best deliberation protocol(13.8% +- 2.6%). This 6.0x performance gap is statistically significant (p < 0.01) and comes at 1.5-2.5x higher computational cost. Our findings challenge assumptions that complexity enhances quality in multi-LLM systems.

---

### AnimatedLLM: Explaining LLMs with Interactive Visualizations

Authors: Zdeněk Kasner, Ondřej Dušek
Published: 2025-12-14
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.04213v2

Large language models (LLMs) are becoming central to natural language processing education, yet materials showing their mechanics are sparse. We present AnimatedLLM, an interactive web application that provides step-by-step visualizations of a Transformer language model. AnimatedLLM runs entirely in the browser, using pre-computed traces of open LLMs applied on manually curated inputs. The application is available at https://animatedllm.github.io, both as a teaching aid and for self-educational purposes.

---

### Optimal Traffic Relief Road Design using Bilevel Programming and Greedy Seeded Simulated Annealing: A Case Study of Kinshasa

Authors: Yves Matanga, Chunling Du, Etienne van Wyk
Published: 2025-12-14
Categories: cs.NE
arXiv: https://arxiv.org/abs/2601.00804v1

Context: The city of Kinshasa faces severe traffic congestion, requiring strategic infrastructure capacity enhancements. Although a comprehensive master plan has been proposed, its implementation requires substantial financial investment, which remains constrained in the Democratic Republic of the Congo (DRC), an emerging economy. This research proposes a traffic flow based algorithm to support the development of priority road segments. The objective is to enable more effective prioritisation of road construction projects and facilitate the optimal allocation of limited infrastructure budgets. Methods: The study was conducted by formulating a standard transport network design problem (TNDP) that included estimated origin demand data specific to the city of Kinshasa. Given the high computational nature of the 30 node network design, TNDP relevant metaheuristics (GA, ACO, PSO, SA, TS, Greedy) were used selectively and hybridised to achieve high quality, stable solutions. A greedy search seeded simulated annealing and Tabu search were devised to achieve the design goals. Results: Greedy Simulated Annealing and Greedy Tabu search yielded the best travel time reduction and the most stable solutions compared to other solvers, also improving network edge betweenness centrality by nearly a scale of two and a half. Conclusions: Road priorities were proposed, including junctions connecting the Bandundu and Kongo Central entry point to main attraction centres (Limete Poids Lourd, Gombe, Airport) and additional inner city areas (Ngaliema, Selembao, Lemba, Masina, Kimwenza).

---

### EEG-to-Voice Decoding of Spoken and Imagined speech Using Non-Invasive EEG

Authors: Hanbeot Park, Yunjeong Cho, Hunhee Kim
Published: 2025-12-14
Categories: eess.SP, cs.LG, cs.SD
arXiv: https://arxiv.org/abs/2512.22146v1

Restoring speech communication from neural signals is a central goal of brain-computer interface research, yet EEG-based speech reconstruction remains challenging due to limited spatial resolution, susceptibility to noise, and the absence of temporally aligned acoustic targets in imagined speech. In this study, we propose an EEG-to-Voice paradigm that directly reconstructs speech from non-invasive EEG signals without dynamic time warping (DTW) or explicit temporal alignment. The proposed pipeline generates mel-spectrograms from EEG in an open-loop manner using a subject-specific generator, followed by pretrained vocoder and automatic speech recognition (ASR) modules to synthesize speech waveforms and decode text. Separate generators were trained for spoken speech and imagined speech, and transfer learning-based domain adaptation was applied by pretraining on spoken speech and adapting to imagined speech. A minimal language model-based correction module was optionally applied to correct limited ASR errors while preserving semantic structure. The framework was evaluated under 2 s and 4 s speech conditions using acoustic-level metrics (PCC, RMSE, MCD) and linguistic-level metrics (CER, WER). Stable acoustic reconstruction and comparable linguistic accuracy were observed for both spoken speech and imagined speech. While acoustic similarity decreased for longer utterances, text-level decoding performance was largely preserved, and word-position analysis revealed a mild increase in decoding errors toward later parts of sentences. The language model-based correction consistently reduced CER and WER without introducing semantic distortion. These results demonstrate the feasibility of direct, open-loop EEG-to-Voice reconstruction for spoken speech and imagined speech without explicit temporal alignment.

---

### Pre-review to Peer review: Pitfalls of Automating Reviews using Large Language Models

Authors: Akhil Pandey Akella, Harish Varma Siravuri, Shaurya Rohatgi
Published: 2025-12-14
Categories: cs.DL, cs.AI, cs.CY
arXiv: https://arxiv.org/abs/2512.22145v1

Large Language Models are versatile general-task solvers, and their capabilities can truly assist people with scholarly peer review as \textit{pre-review} agents, if not as fully autonomous \textit{peer-review} agents. While incredibly beneficial, automating academic peer-review, as a concept, raises concerns surrounding safety, research integrity, and the validity of the academic peer-review process. The majority of the studies performing a systematic evaluation of frontier LLMs generating reviews across science disciplines miss the mark on addressing the alignment/misalignment of reviews along with the utility of LLM generated reviews when compared against publication outcomes such as \textbf{Citations}, \textbf{Hit-papers}, \textbf{Novelty}, and \textbf{Disruption}. This paper presents an experimental study in which we gathered ground-truth reviewer ratings from OpenReview and used various frontier open-weight LLMs to generate reviews of papers to gauge the safety and reliability of incorporating LLMs into the scientific review pipeline. Our findings demonstrate the utility of frontier open-weight LLMs as pre-review screening agents despite highlighting fundamental misalignment risks when deployed as autonomous reviewers. Our results show that all models exhibit weak correlation with human peer reviewers (0.15), with systematic overestimation bias of 3-5 points and uniformly high confidence scores (8.0-9.0/10) despite prediction errors. However, we also observed that LLM reviews correlate more strongly with post-publication metrics than with human scores, suggesting potential utility as pre-review screening tools. Our findings highlight the potential and address the pitfalls of automating peer reviews with language models. We open-sourced our dataset $D_{LMRSD}$ to help the research community expand the safety framework of automating scientific reviews.

---

### The Complete Anatomy of the Madden-Julian Oscillation Revealed by Artificial Intelligence

Authors: Xiao Zhou, Yuze Sun, Jie Wu, Xiaomeng Huang
Published: 2025-12-14
Categories: physics.ao-ph, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2512.22144v1

Accurately defining the life cycle of the Madden-Julian Oscillation (MJO), the dominant mode of intraseasonal climate variability, remains a foundational challenge due to its propagating nature. The established linear-projection method (RMM index) often conflates mathematical artifacts with physical states, while direct clustering in raw data space is confounded by a "propagation penalty." Here, we introduce an "AI-for-theory" paradigm to objectively discover the MJO's intrinsic structure. We develop a deep learning model, PhysAnchor-MJO-AE, to learn a latent representation where vector distance corresponds to physical-feature similarity, enabling objective clustering of MJO dynamical states. Clustering these "MJO fingerprints" reveals the first complete, six-phase anatomical map of its life cycle. This taxonomy refines and critically completes the classical view by objectively isolating two long-hypothesized transitional phases: organizational growth over the Indian Ocean and the northward shift over the Philippine Sea. Derived from this anatomy, we construct a new physics-coherent monitoring framework that decouples location and intensity diagnostics. This framework reduces the rates of spurious propagation and convective misplacement by over an order of magnitude compared to the classical index. Our work transforms AI from a forecasting tool into a discovery microscope, establishing a reproducible template for extracting fundamental dynamical constructs from complex systems.

---

### FuseFi: Combining Irregularly Sampled CSI from Diverse Communication Packets and Frequency Bands for Wi-Fi Sensing

Authors: Gaofeng Dong, Kang Yang, Mani Srivastava
Published: 2025-12-14
Categories: eess.SP, cs.LG
arXiv: https://arxiv.org/abs/2512.22143v2

Existing Wi-Fi sensing systems rely on injecting high-rate probing packets to extract channel state information (CSI), leading to communication degradation and limited deployment flexibility. Although Integrated Sensing and Communication (ISAC) is a promising direction, existing solutions still rely on auxiliary packet injection because they exploit only uniform CSI from a single frame type, discarding approximately 70% of naturally available packets. We present FuseFi, a novel Wi-Fi-based ISAC framework that directly exploits irregularly sampled CSI from diverse communication packets across multiple frequency bands, eliminating intrusive packet injection and introducing no sensing-specific communication overhead. FuseFi integrates a CSI sanitization pipeline to harmonize heterogeneous packets and remove burst-induced redundancy, together with a time-aware attention model that learns directly from non-uniform CSI sequences without resampling. We further introduce CommCSI-HAR, a new dataset with irregularly sampled CSI from real-world dual-band communication traffic. Extensive evaluations on this dataset and, separately, on four public sensing tasks across three benchmark datasets show that FuseFi achieves state-of-the-art accuracy with a compact model size, while fully preserving communication throughput.

---

### EcoNet: Multiagent Planning and Control Of Household Energy Resources Using Active Inference

Authors: John C. Boik, Kobus Esterhuysen, Jacqueline B. Hynes, Axel Constant, Ines Hipolito, Mahault Albarracin, Alex B. Kiefer, Karl Friston
Published: 2025-12-14
Categories: eess.SY, cs.AI, cs.MA
arXiv: https://arxiv.org/abs/2512.21343v1

Advances in automated systems afford new opportunities for intelligent management of energy at household, local area, and utility scales. Home Energy Management Systems (HEMS) can play a role by optimizing the schedule and use of household energy devices and resources. One challenge is that the goals of a household can be complex and conflicting. For example, a household might wish to reduce energy costs and grid-associated greenhouse gas emissions, yet keep room temperatures comfortable. Another challenge is that an intelligent HEMS agent must make decisions under uncertainty. An agent must plan actions into the future, but weather and solar generation forecasts, for example, provide inherently uncertain estimates of future conditions. This paper introduces EcoNet, a Bayesian approach to household and neighborhood energy management that is based on active inference. The aim is to improve energy management and coordination, while accommodating uncertainties and taking into account potentially conditional and conflicting goals and preferences. Simulation results are presented and discussed.

---

### What's the Price of Monotonicity? A Multi-Dataset Benchmark of Monotone-Constrained Gradient Boosting for Credit PD

Authors: Petr Koklev
Published: 2025-12-14
Categories: cs.LG, q-fin.RM, q-fin.ST
arXiv: https://arxiv.org/abs/2512.17945v2

Financial institutions face a trade-off between predictive accuracy and interpretability when deploying machine learning models for credit risk. Monotonicity constraints align model behavior with domain knowledge, but their performance cost - the price of monotonicity - is not well quantified. This paper benchmarks monotone-constrained versus unconstrained gradient boosting models for credit probability of default across five public datasets and three libraries. We define the Price of Monotonicity (PoM) as the relative change in standard performance metrics when moving from unconstrained to constrained models, estimated via paired comparisons with bootstrap uncertainty. In our experiments, PoM in AUC ranges from essentially zero to about 2.9 percent: constraints are almost costless on large datasets (typically less than 0.2 percent, often indistinguishable from zero) and most costly on smaller datasets with extensive constraint coverage (around 2-3 percent). Thus, appropriately specified monotonicity constraints can often deliver interpretability with small accuracy losses, particularly in large-scale credit portfolios.

---

### Auto-Tuning Safety Guardrails for Black-Box Large Language Models

Authors: Perry Abdulkadir
Published: 2025-12-14
Categories: cs.CR, cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2512.15782v1

Large language models (LLMs) are increasingly deployed behind safety guardrails such as system prompts and content filters, especially in settings where product teams cannot modify model weights. In practice these guardrails are typically hand-tuned, brittle, and difficult to reproduce. This paper studies a simple but practical alternative: treat safety guardrail design itself as a hyperparameter optimization problem over a frozen base model. Concretely, I wrap Mistral-7B-Instruct with modular jailbreak and malware system prompts plus a ModernBERT-based harmfulness classifier, then evaluate candidate configurations on three public benchmarks covering malware generation, classic jailbreak prompts, and benign user queries. Each configuration is scored using malware and jailbreak attack success rate, benign harmful-response rate, and end-to-end latency. A 48-point grid search over prompt combinations and filter modes establishes a baseline. I then run a black-box Optuna study over the same space and show that it reliably rediscovers the best grid configurations while requiring an order of magnitude fewer evaluations and roughly 8x less wall-clock time. The results suggest that viewing safety guardrails as tunable hyperparameters is a feasible way to harden black-box LLM deployments under compute and time constraints.

---

### Adversarial Robustness in Financial Machine Learning: Defenses, Economic Impact, and Governance Evidence

Authors: Samruddhi Baviskar
Published: 2025-12-14
Categories: cs.LG, cs.AI, cs.CR
arXiv: https://arxiv.org/abs/2512.15780v1

We evaluate adversarial robustness in tabular machine learning models used in financial decision making. Using credit scoring and fraud detection data, we apply gradient based attacks and measure impacts on discrimination, calibration, and financial risk metrics. Results show notable performance degradation under small perturbations and partial recovery through adversarial training.

---

### Hyperparameter Tuning-Based Optimized Performance Analysis of Machine Learning Algorithms for Network Intrusion Detection

Authors: Sudhanshu Sekhar Tripathy, Bichitrananda Behera
Published: 2025-12-14
Categories: cs.CR, cs.LG, cs.NI
arXiv: https://arxiv.org/abs/2512.15779v1

Network Intrusion Detection Systems (NIDS) are essential for securing networks by identifying and mitigating unauthorized activities indicative of cyberattacks. As cyber threats grow increasingly sophisticated, NIDS must evolve to detect both emerging threats and deviations from normal behavior. This study explores the application of machine learning (ML) methods to improve the NIDS accuracy through analyzing intricate structures in deep-featured network traffic records. Leveraging the 1999 KDD CUP intrusion dataset as a benchmark, this research evaluates and optimizes several ML algorithms, including Support Vector Machines (SVM), Naïve Bayes variants (MNB, BNB), Random Forest (RF), k-Nearest Neighbors (k-NN), Decision Trees (DT), AdaBoost, XGBoost, Logistic Regression (LR), Ridge Classifier, Passive-Aggressive (PA) Classifier, Rocchio Classifier, Artificial Neural Networks (ANN), and Perceptron (PPN). Initial evaluations without hyper-parameter optimization demonstrated suboptimal performance, highlighting the importance of tuning to enhance classification accuracy. After hyper-parameter optimization using grid and random search techniques, the SVM classifier achieved 99.12% accuracy with a 0.0091 False Alarm Rate (FAR), outperforming its default configuration (98.08% accuracy, 0.0123 FAR) and all other classifiers. This result confirms that SVM accomplishes the highest accuracy among the evaluated classifiers. We validated the effectiveness of all classifiers using a tenfold cross-validation approach, incorporating Recursive Feature Elimination (RFE) for feature selection to enhance the classifiers accuracy and efficiency. Our outcomes indicate that ML classifiers are both adaptable and reliable, contributing to enhanced accuracy in systems for detecting network intrusions.

---

### COBRA: Catastrophic Bit-flip Reliability Analysis of State-Space Models

Authors: Sanjay Das, Swastik Bhattacharya, Shamik Kundu, Arnab Raha, Souvik Kundu, Kanad Basu
Published: 2025-12-14
Categories: cs.CR, cs.LG
arXiv: https://arxiv.org/abs/2512.15778v2

State-space models (SSMs), exemplified by the Mamba architecture, have recently emerged as state-of-the-art sequence-modeling frameworks, offering linear-time scalability together with strong performance in long-context settings. Owing to their unique combination of efficiency, scalability, and expressive capacity, SSMs have become compelling alternatives to transformer-based models, which suffer from the quadratic computational and memory costs of attention mechanisms. As SSMs are increasingly deployed in real-world applications, it is critical to assess their susceptibility to both software- and hardware-level threats to ensure secure and reliable operation. Among such threats, hardware-induced bit-flip attacks (BFAs) pose a particularly severe risk by corrupting model parameters through memory faults, thereby undermining model accuracy and functional integrity. To investigate this vulnerability, we introduce RAMBO, the first BFA framework specifically designed to target Mamba-based architectures. Through experiments on the Mamba-1.4b model with LAMBADA benchmark, a cloze-style word-prediction task, we demonstrate that flipping merely a single critical bit can catastrophically reduce accuracy from 74.64% to 0% and increase perplexity from 18.94 to 3.75 x 10^6. These results demonstrate the pronounced fragility of SSMs to adversarial perturbations.

---

### CODE ACROSTIC: Robust Watermarking for Code Generation

Authors: Li Lin, Siyuan Xin, Yang Cao, Xiaochun Cao
Published: 2025-12-14
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2512.14753v1

Watermarking large language models (LLMs) is vital for preventing their misuse, including the fabrication of fake news, plagiarism, and spam. It is especially important to watermark LLM-generated code, as it often contains intellectual property.However, we found that existing methods for watermarking LLM-generated code fail to address comment removal attack.In such cases, an attacker can simply remove the comments from the generated code without affecting its functionality, significantly reducing the effectiveness of current code-watermarking techniques.On the other hand, injecting a watermark into code is challenging because, as previous works have noted, most code represents a low-entropy scenario compared to natural language. Our approach to addressing this issue involves leveraging prior knowledge to distinguish between low-entropy and high-entropy parts of the code, as indicated by a Cue List of words.We then inject the watermark guided by this Cue List, achieving higher detectability and usability than existing methods.We evaluated our proposed method on HumanEvaland compared our method with three state-of-the-art code watermarking techniques. The results demonstrate the effectiveness of our approach.

---

### Cyberswarm: a novel swarm intelligence algorithm inspired by cyber community dynamics

Authors: Abdelsadeq Elfergany, Ammar Adl, Mohammed Kayed
Published: 2025-12-14
Categories: cs.SI, cs.AI
arXiv: https://arxiv.org/abs/2512.14752v1

Recommendation systems face challenges in dynamically adapting to evolving user preferences and interactions within complex social networks. Traditional approaches often fail to account for the intricate interactions within cyber-social systems and lack the flexibility to generalize across diverse domains, highlighting the need for more adaptive and versatile solutions. In this work, we introduce a general-purpose swarm intelligence algorithm for recommendation systems, designed to adapt seamlessly to varying applications. It was inspired by social psychology principles. The framework models user preferences and community influences within a dynamic hypergraph structure. It leverages centrality-based feature extraction and Node2Vec embeddings. Preference evolution is guided by message-passing mechanisms and hierarchical graph modeling, enabling real-time adaptation to changing behaviors. Experimental evaluations demonstrated the algorithm's superior performance in various recommendation tasks, including social networks and content discovery. Key metrics such as Hit Rate (HR), Mean Reciprocal Rank (MRR), and Normalized Discounted Cumulative Gain (NDCG) consistently outperformed baseline methods across multiple datasets. The model's adaptability to dynamic environments allowed for contextually relevant and precise recommendations. The proposed algorithm represents an advancement in recommendation systems by bridging individual preferences and community influences. Its general-purpose design enables applications in diverse domains, including social graphs, personalized learning, and medical graphs. This work highlights the potential of integrating swarm intelligence with network dynamics to address complex optimization challenges in recommendation systems.

---

### One Leak Away: How Pretrained Model Exposure Amplifies Jailbreak Risks in Finetuned LLMs

Authors: Yixin Tan, Zhe Yu, Rui Wen, Jun Sakuma
Published: 2025-12-14
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2512.14751v3

Finetuning pretrained large language models (LLMs) has become the standard paradigm for developing downstream applications. However, its security implications remain unclear, particularly regarding whether finetuned LLMs inherit jailbreak vulnerabilities from their pretrained sources. We investigate this question in a realistic pretrain-to-finetune threat model, where an attacker has full access to a released pretrained LLM but no access to its proprietary finetuned derivatives. Empirical analysis shows that adversarial prompts optimized on the pretrained model transfer most effectively to its finetuned variants, revealing inherited vulnerabilities from pretrained to finetuned LLMs. To further examine this inheritance, we conduct representation-level probing, which shows that transferable prompts are linearly separable within the pretrained hidden states, suggesting that transferability-relevant structure is already encoded in pretrained representations. Building on this insight, we propose the Probe-Guided Projection (PGP) attack, which steers optimization toward transferability-relevant directions. Experiments across multiple LLM families and diverse finetuned tasks confirm PGP's strong transfer success, underscoring the security risks inherent in the pretrain-to-finetune paradigm. Finally, we demonstrate that the same representation-level insights also enable a lightweight defense that mitigates pretrain-to-finetune jailbreak transfer while preserving downstream utility.

---

### DL$^3$M: A Vision-to-Language Framework for Expert-Level Medical Reasoning through Deep Learning and Large Language Models

Authors: Md. Najib Hasan, Imran Ahmad, Sourav Basak Shuvo, Md. Mahadi Hasan Ankon, Nazmul Siddique, Hui Wang
Published: 2025-12-14
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2512.13742v3

Medical image classifiers detect gastrointestinal diseases well, but they do not explain their decisions. Large language models can generate clinical text, yet they struggle with visual reasoning and often produce unstable or incorrect explanations. This leaves a gap between what a model sees and the type of reasoning a clinician expects. We introduce a framework that links image classification with structured clinical reasoning. A new hybrid model, MobileCoAtNet, is designed for endoscopic images and achieves high accuracy across eight stomach-related classes. Its outputs are then used to drive reasoning by several LLMs. To judge this reasoning, we build two expert-verified benchmarks covering causes, symptoms, treatment, lifestyle, and follow-up care. Thirty-two LLMs are evaluated against these gold standards. Strong classification improves the quality of their explanations, but none of the models reach human-level stability. Even the best LLMs change their reasoning when prompts vary. Our study shows that combining DL with LLMs can produce useful clinical narratives, but current LLMs remain unreliable for high-stakes medical decisions. The framework provides a clearer view of their limits and a path for building safer reasoning systems. The complete source code and datasets used in this study are available at https://github.com/souravbasakshuvo/DL3M.

---

### The Laminar Flow Hypothesis: Detecting Jailbreaks via Semantic Turbulence in Large Language Models

Authors: Md. Hasib Ur Rahman
Published: 2025-12-14
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2512.13741v1

As Large Language Models (LLMs) become ubiquitous, the challenge of securing them against adversarial "jailbreaking" attacks has intensified. Current defense strategies often rely on computationally expensive external classifiers or brittle lexical filters, overlooking the intrinsic dynamics of the model's reasoning process. In this work, the Laminar Flow Hypothesis is introduced, which posits that benign inputs induce smooth, gradual transitions in an LLM's high-dimensional latent space, whereas adversarial prompts trigger chaotic, high-variance trajectories - termed Semantic Turbulence - resulting from the internal conflict between safety alignment and instruction-following objectives. This phenomenon is formalized through a novel, zero-shot metric: the variance of layer-wise cosine velocity. Experimental evaluation across diverse small language models reveals a striking diagnostic capability. The RLHF-aligned Qwen2-1.5B exhibits a statistically significant 75.4% increase in turbulence under attack (p less than 0.001), validating the hypothesis of internal conflict. Conversely, Gemma-2B displays a 22.0% decrease in turbulence, characterizing a distinct, low-entropy "reflex-based" refusal mechanism. These findings demonstrate that Semantic Turbulence serves not only as a lightweight, real-time jailbreak detector but also as a non-invasive diagnostic tool for categorizing the underlying safety architecture of black-box models.

---

### Human-AI Collaboration Mechanism Study on AIGC Assisted Image Production for Special Coverage

Authors: Yajie Yang, Yuqing Zhao, Xiaochao Xi, Yinan Zhu
Published: 2025-12-14
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2512.13739v1

Artificial Intelligence Generated Content (AIGC) assisting image production triggers controversy in journalism while attracting attention from media agencies. Key issues involve misinformation, authenticity, semantic fidelity, and interpretability. Most AIGC tools are opaque "black boxes," hindering the dual demands of content accuracy and semantic alignment and creating ethical, sociotechnical, and trust dilemmas. This paper explores pathways for controllable image production in journalism's special coverage and conducts two experiments with projects from China's media agency: (1) Experiment 1 tests cross-platform adaptability via standardized prompts across three scenes, revealing disparities in semantic alignment, cultural specificity, and visual realism driven by training-corpus bias and platform-level filtering. (2) Experiment 2 builds a human-in-the-loop modular pipeline combining high-precision segmentation (SAM, GroundingDINO), semantic alignment (BrushNet), and style regulating (Style-LoRA, Prompt-to-Prompt), ensuring editorial fidelity through CLIP-based semantic scoring, NSFW/OCR/YOLO filtering, and verifiable content credentials. Traceable deployment preserves semantic representation. Consequently, we propose a human-AI collaboration mechanism for AIGC assisted image production in special coverage and recommend evaluating Character Identity Stability (CIS), Cultural Expression Accuracy (CEA), and User-Public Appropriateness (U-PA).

---

### Instilling Organisational Values in Firefighters through Simulation-Based Training

Authors: Nardine Osman, Manel Rodriguez-Soto, Jordi Sabater-Mir
Published: 2025-12-14
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2512.13737v1

In firefighting and other emergency operations, decisions made under pressure carry profound ethical weight and can significantly impact incident outcomes and firefighter safety. Traditional training methods, while foundational, often fall short in adequately preparing firefighters for the complex ethical dilemmas and value conflicts inherent in chaotic emergency environments. This paper proposes a conceptual framework for enhancing firefighter training by systematically integrating departmental values into simulation-based training. This approach fosters deeper value internalisation and improves value-driven decision-making under pressure. Furthermore, the underlying tools can also be leveraged to evaluate and refine departmental operational protocols for better alignment with preferred values.

---

### TF-MCL: Time-frequency Fusion and Multi-domain Cross-Loss for Self-supervised Depression Detection

Authors: Li-Xuan Zhao, Chen-Yang Xu, Wen-Qiang Li, Bo Wang, Rong-Xing Wei, Qing-Hao Menga
Published: 2025-12-14
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2512.13736v1

In recent years, there has been a notable increase in the use of supervised detection methods of major depressive disorder (MDD) based on electroencephalogram (EEG) signals. However, the process of labeling MDD remains challenging. As a self-supervised learning method, contrastive learning could address the shortcomings of supervised learning methods, which are unduly reliant on labels in the context of MDD detection. However, existing contrastive learning methods are not specifically designed to characterize the time-frequency distribution of EEG signals, and their capacity to acquire low-semantic data representations is still inadequate for MDD detection tasks. To address the problem of contrastive learning method, we propose a time-frequency fusion and multi-domain cross-loss (TF-MCL) model for MDD detection. TF-MCL generates time-frequency hybrid representations through the use of a fusion mapping head (FMH), which efficiently remaps time-frequency domain information to the fusion domain, and thus can effectively enhance the model's capacity to synthesize time-frequency information. Moreover, by optimizing the multi-domain cross-loss function, the distribution of the representations in the time-frequency domain and the fusion domain is reconstructed, thereby improving the model's capacity to acquire fusion representations. We evaluated the performance of our model on the publicly available datasets MODMA and PRED+CT and show a significant improvement in accuracy, outperforming the existing state-of-the-art (SOTA) method by 5.87% and 9.96%, respectively.

---

### DARTs: A Dual-Path Robust Framework for Anomaly Detection in High-Dimensional Multivariate Time Series

Authors: Xuechun Liu, Heli Sun, Xuecheng Wu, Ruichen Cao, Yunyun Shi, Dingkang Yang, Haoran Li
Published: 2025-12-14
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2512.13735v1

Multivariate time series anomaly detection (MTSAD) aims to accurately identify and localize complex abnormal patterns in the large-scale industrial control systems. While existing approaches excel in recognizing the distinct patterns under the low-dimensional scenarios, they often fail to robustly capture long-range spatiotemporal dependencies when learning representations from the high-dimensional noisy time series. To address these limitations, we propose DARTs, a robust long short-term dual-path framework with window-aware spatiotemporal soft fusion mechanism, which can be primarily decomposed into three complementary components. Specifically, in the short-term path, we introduce a Multi-View Sparse Graph Learner and a Diffusion Multi-Relation Graph Unit that collaborate to adaptively capture hierarchical discriminative short-term spatiotemporal patterns in the high-noise time series. While in the long-term path, we design a Multi-Scale Spatiotemporal Graph Constructor to model salient long-term dynamics within the high-dimensional representation space. Finally, a window-aware spatiotemporal soft-fusion mechanism is introduced to filter the residual noise while seamlessly integrating anomalous patterns. Extensive qualitative and quantitative experimental results across mainstream datasets demonstrate the superiority and robustness of our proposed DARTs. A series of ablation studies are also conducted to explore the crucial design factors of our proposed components. Our code and model will be made publicly open soon.

---

### Plug-and-Play Parameter-Efficient Tuning of Embeddings for Federated Recommendation

Authors: Haochen Yuan, Yang Zhang, Xiang He, Quan Z. Sheng, Zhongjie Wang
Published: 2025-12-14
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2512.13734v1

With the rise of cloud-edge collaboration, recommendation services are increasingly trained in distributed environments. Federated Recommendation (FR) enables such multi-end collaborative training while preserving privacy by sharing model parameters instead of raw data. However, the large number of parameters, primarily due to the massive item embeddings, significantly hampers communication efficiency. While existing studies mainly focus on improving the efficiency of FR models, they largely overlook the issue of embedding parameter overhead. To address this gap, we propose a FR training framework with Parameter-Efficient Fine-Tuning (PEFT) based embedding designed to reduce the volume of embedding parameters that need to be transmitted. Our approach offers a lightweight, plugin-style solution that can be seamlessly integrated into existing FR methods. In addition to incorporating common PEFT techniques such as LoRA and Hash-based encoding, we explore the use of Residual Quantized Variational Autoencoders (RQ-VAE) as a novel PEFT strategy within our framework. Extensive experiments across various FR model backbones and datasets demonstrate that our framework significantly reduces communication overhead while improving accuracy. The source code is available at https://github.com/young1010/FedPEFT.

---

### Low-Rank Compression of Language Models via Differentiable Rank Selection

Authors: Sidhant Sundrani, Francesco Tudisco, Pasquale Minervini
Published: 2025-12-14
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2512.13733v1

Approaches for compressing large-language models using low-rank decomposition have made strides, particularly with the introduction of activation and loss-aware SVD, which improves the trade-off between decomposition rank and downstream task performance. Despite these advancements, a persistent challenge remains--selecting the optimal ranks for each layer to jointly optimise compression rate and downstream task accuracy. Current methods either rely on heuristics that can yield sub-optimal results due to their limited discrete search space or are gradient-based but are not as performant as heuristic approaches without post-compression fine-tuning. To address these issues, we propose Learning to Low-Rank Compress (LLRC), a gradient-based approach which directly learns the weights of masks that select singular values in a fine-tuning-free setting. Using a calibration dataset, we train only the mask weights to select fewer and fewer singular values while minimising the divergence of intermediate activations from the original model. Our approach outperforms competing ranking selection methods that similarly require no post-compression fine-tuning across various compression rates on common-sense reasoning and open-domain question-answering tasks. For instance, with a compression rate of 20% on Llama-2-13B, LLRC outperforms the competitive Sensitivity-based Truncation Rank Searching (STRS) on MMLU, BoolQ, and OpenbookQA by 12%, 3.5%, and 4.4%, respectively. Compared to other compression techniques, our approach consistently outperforms fine-tuning-free variants of SVD-LLM and LLM-Pruner across datasets and compression rates. Our fine-tuning-free approach also performs competitively with the fine-tuning variant of LLM-Pruner.

---

### PIS: A Generalized Physical Inversion Solver for Arbitrary Sparse Observations via Set Conditioned Flow Matching

Authors: Weijie Yang, Xun Zhang, Simin Jiang, Yubao Zhou
Published: 2025-12-14
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2512.13732v2

The estimation of high-dimensional physical parameters constrained by partial differential equations (PDEs) from limited and indirect measurements is a highly ill-posed problem. Traditional methods face significant accuracy and efficiency bottlenecks, particularly when observations are sparse, irregularly sampled, and constrained by real-world sensor placement. We propose the Physical Inversion Solver (PIS), a unified framework that couples Set-Conditioned Flow Matching with a Cosine-Annealed Sparsity Curriculum (CASC) to enable stable inversion from arbitrary, off-grid sensors even under minimal guidance. By leveraging straight-path transport, PIS achieves instantaneous inference (50 NFEs), offering orders-of-magnitude speedup over iterative baselines. Extensive experiments demonstrate that PIS reduces error by up to 88.7% under extreme sparsity (<1%) across subsurface characterization, wave-based characterization, and structural health monitoring, while providing robust uncertainty quantification for optimal sensor placement.

---

### Complex Mathematical Expression Recognition: Benchmark, Large-Scale Dataset and Strong Baseline

Authors: Weikang Bai, Yongkun Du, Yuchen Su, Yazhen Xie, Zhineng Chen
Published: 2025-12-14
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2512.13731v1

Mathematical Expression Recognition (MER) has made significant progress in recognizing simple expressions, but the robust recognition of complex mathematical expressions with many tokens and multiple lines remains a formidable challenge. In this paper, we first introduce CMER-Bench, a carefully constructed benchmark that categorizes expressions into three difficulty levels: easy, moderate, and complex. Leveraging CMER-Bench, we conduct a comprehensive evaluation of existing MER models and general-purpose multimodal large language models (MLLMs). The results reveal that while current methods perform well on easy and moderate expressions, their performance degrades significantly when handling complex mathematical expressions, mainly because existing public training datasets are primarily composed of simple samples. In response, we propose MER-17M and CMER-3M that are large-scale datasets emphasizing the recognition of complex mathematical expressions. The datasets provide rich and diverse samples to support the development of accurate and robust complex MER models. Furthermore, to address the challenges posed by the complicated spatial layout of complex expressions, we introduce a novel expression tokenizer, and a new representation called Structured Mathematical Language, which explicitly models the hierarchical and spatial structure of expressions beyond LaTeX format. Based on these, we propose a specialized model named CMERNet, built upon an encoder-decoder architecture and trained on CMER-3M. Experimental results show that CMERNet, with only 125 million parameters, significantly outperforms existing MER models and MLLMs on CMER-Bench.

---

### Exploring the Modular Integration of "AI + Architecture" Pedagogy in Undergraduate Design Education: A Case Study of Architectural Design III/IV Courses at Zhejiang University

Authors: Wang Jiaqi, Lan Yi, Chen Xiang
Published: 2025-12-14
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2512.13730v1

This study investigates AI integration in architectural education through a teaching experiment in Zhejiang University's 2024-25 grade three undergraduate design studio. Adopting a dual-module framework (20-hour AI training + embedded ethics discussions), the course introduced deep learning models, LLMs, AIGC, LoRA, and ComfyUI while maintaining the original curriculum structure, supported by dedicated technical instructors. Findings demonstrate the effectiveness of phased guidance, balanced technical-ethical approaches, and institutional support. The model improved students' digital skills and strategic cognition while addressing AI ethics, providing a replicable approach combining technical and critical learning in design education.

---

### SignRAG: A Retrieval-Augmented System for Scalable Zero-Shot Road Sign Recognition

Authors: Minghao Zhu, Zhihao Zhang, Anmol Sidhu, Keith Redmill
Published: 2025-12-14
Categories: cs.CV, cs.AI, cs.CL, cs.IR, cs.RO
arXiv: https://arxiv.org/abs/2512.12885v1

Automated road sign recognition is a critical task for intelligent transportation systems, but traditional deep learning methods struggle with the sheer number of sign classes and the impracticality of creating exhaustive labeled datasets. This paper introduces a novel zero-shot recognition framework that adapts the Retrieval-Augmented Generation (RAG) paradigm to address this challenge. Our method first uses a Vision Language Model (VLM) to generate a textual description of a sign from an input image. This description is used to retrieve a small set of the most relevant sign candidates from a vector database of reference designs. Subsequently, a Large Language Model (LLM) reasons over the retrieved candidates to make a final, fine-grained recognition. We validate this approach on a comprehensive set of 303 regulatory signs from the Ohio MUTCD. Experimental results demonstrate the framework's effectiveness, achieving 95.58% accuracy on ideal reference images and 82.45% on challenging real-world road data. This work demonstrates the viability of RAG-based architectures for creating scalable and accurate systems for road sign recognition without task-specific training.

---

### Cross-Level Sensor Fusion with Object Lists via Transformer for 3D Object Detection

Authors: Xiangzhong Liu, Jiajie Zhang, Hao Shen
Published: 2025-12-14
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2512.12884v2

In automotive sensor fusion systems, smart sensors and Vehicle-to-Everything (V2X) modules are commonly utilized. Sensor data from these systems are typically available only as processed object lists rather than raw sensor data from traditional sensors. Instead of processing other raw data separately and then fusing them at the object level, we propose an end-to-end cross-level fusion concept with Transformer, which integrates highly abstract object list information with raw camera images for 3D object detection. Object lists are fed into a Transformer as denoising queries and propagated together with learnable queries through the latter feature aggregation process. Additionally, a deformable Gaussian mask, derived from the positional and size dimensional priors from the object lists, is explicitly integrated into the Transformer decoder. This directs attention toward the target area of interest and accelerates model training convergence. Furthermore, as there is no public dataset containing object lists as a standalone modality, we propose an approach to generate pseudo object lists from ground-truth bounding boxes by simulating state noise and false positives and negatives. As the first work to conduct cross-level fusion, our approach shows substantial performance improvements over the vision-based baseline on the nuScenes dataset. It demonstrates its generalization capability over diverse noise levels of simulated object lists and real detectors.

---

### Unsupervised learning of multiscale switching dynamical system models from multimodal neural data

Authors: DongKyu Kim, Han-Lin Hsieh, Maryam M. Shanechi
Published: 2025-12-14
Categories: cs.LG, q-bio.NC, stat.ML
arXiv: https://arxiv.org/abs/2512.12881v1

Neural population activity often exhibits regime-dependent non-stationarity in the form of switching dynamics. Learning accurate switching dynamical system models can reveal how behavior is encoded in neural activity. Existing switching approaches have primarily focused on learning models from a single neural modality, either continuous Gaussian signals or discrete Poisson signals. However, multiple neural modalities are often recorded simultaneously to measure different spatiotemporal scales of brain activity, and all these modalities can encode behavior. Moreover, regime labels are typically unavailable in training data, posing a significant challenge for learning models of regime-dependent switching dynamics. To address these challenges, we develop a novel unsupervised learning algorithm that learns the parameters of switching multiscale dynamical system models using only multiscale neural observations. We demonstrate our method using both simulations and two distinct experimental datasets with multimodal spike-LFP observations during different motor tasks. We find that our switching multiscale dynamical system models more accurately decode behavior than switching single-scale dynamical models, showing the success of multiscale neural fusion. Further, our models outperform stationary multiscale models, illustrating the importance of tracking regime-dependent non-stationarity in multimodal neural data. The developed unsupervised learning framework enables more accurate modeling of complex multiscale neural dynamics by leveraging information in multimodal recordings while incorporating regime switches. This approach holds promise for improving the performance and robustness of brain-computer interfaces over time and for advancing our understanding of the neural basis of behavior.

---

### Improving Recursive Transformers with Mixture of LoRAs

Authors: Mohammadmahdi Nouriborji, Morteza Rohanian, Omid Rohanian
Published: 2025-12-14
Categories: cs.LG
arXiv: https://arxiv.org/abs/2512.12880v2

Parameter sharing in recursive transformers reduces model size but collapses layer-wise expressivity. We propose Mixture of LoRAs (MoL), a lightweight conditional-computation mechanism that inserts Low-Rank Adaptation (LoRA) experts inside a shared feed-forward network (FFN). MoL enables token-conditional weight-space modulation of the shared FFN without untying backbone parameters, unlike prior approaches that add fixed or externally attached adapters. We pretrain a modernised recursive architecture, ModernALBERT, integrating rotary embeddings, GeGLU, FlashAttention, and a distillation-based initialisation. Across GLUE, SQuAD-v2, and BEIR, ModernALBERT (50M--120M) achieves state-of-the-art performance among compact models and surpasses larger fully parameterised baselines. We also propose an expert-merging procedure that compresses MoL into a single adapter at inference while preserving accuracy, enabling efficient deployment. Our results show that conditional weight-space modulation effectively restores the expressivity lost under aggressive parameter sharing in recursive transformers.

---

### Schrodinger Audio-Visual Editor: Object-Level Audiovisual Removal

Authors: Weihan Xu, Kan Jen Cheng, Koichi Saito, Muhammad Jehanzeb Mirza, Tingle Li, Yisi Liu, Alexander H. Liu, Liming Wang, Masato Ishii, Takashi Shibuya, Yuki Mitsufuji, Gopala Anumanchipalli et al.
Published: 2025-12-14
Categories: cs.CV, cs.MM, cs.SD
arXiv: https://arxiv.org/abs/2512.12875v1

Joint editing of audio and visual content is crucial for precise and controllable content creation. This new task poses challenges due to the limitations of paired audio-visual data before and after targeted edits, and the heterogeneity across modalities. To address the data and modeling challenges in joint audio-visual editing, we introduce SAVEBench, a paired audiovisual dataset with text and mask conditions to enable object-grounded source-to-target learning. With SAVEBench, we train the Schrodinger Audio-Visual Editor (SAVE), an end-to-end flow-matching model that edits audio and video in parallel while keeping them aligned throughout processing. SAVE incorporates a Schrodinger Bridge that learns a direct transport from source to target audiovisual mixtures. Our evaluation demonstrates that the proposed SAVE model is able to remove the target objects in audio and visual content while preserving the remaining content, with stronger temporal synchronization and audiovisual semantic correspondence compared with pairwise combinations of an audio editor and a video editor.

---

### Active Learning with Imperfect Labels: Optimal Labeler Assignment and Sample Selection

Authors: Pouya Ahadi, Blair Winograd, Camille Zaug, Karunesh Arora, Lijun Wang, Kamran Paynabar
Published: 2025-12-14
Categories: cs.LG, cs.AI, math.OC
arXiv: https://arxiv.org/abs/2512.12870v2

Active Learning (AL) is commonly used in applications where labeling data is expensive or time-consuming. In practice, however, labels are often noisy due to varying labeler expertise and annotation uncertainty, especially for complex or ambiguous samples. Learning from such imperfectly labeled data can degrade classifier performance. We propose an AL framework that explicitly accounts for label noise by optimally assigning labelers and selecting samples to minimize labeling error. Our approach, called OLAS (Optimal Labeler Assignment and Sampling), uses a noise model that depends on both labeler accuracy and model uncertainty to guide these decisions. We develop two tractable optimization formulations: one for assigning samples to labelers to minimize worst-case noise, and another for selecting samples while controlling overall label noise. Theoretical results provide closed-form solutions under mild conditions. Empirical evaluations on benchmark datasets and a real-world warranty claim classification problem show that OLAS achieves the highest or near-highest classification accuracy among existing AL strategies across most settings, using only a single label per sample.

---

### ERA-IT: Aligning Semantic Models with Revealed Economic Preference for Real-Time and Explainable Patent Valuation

Authors: Yongmin Yoo, Seungwoo Kim, Jingjiang Liu
Published: 2025-12-14
Categories: cs.CE, cs.CL
arXiv: https://arxiv.org/abs/2512.12869v2

Valuing intangible assets under uncertainty remains a critical challenge in the strategic management of technological innovation due to the information asymmetry inherent in high-dimensional technical specifications. Traditional bibliometric indicators, such as citation counts, fail to address this friction in a timely manner due to the systemic latency inherent in data accumulation. To bridge this gap, this study proposes the Economic Reasoning Alignment via Instruction Tuning (ERA-IT) framework. We theoretically conceptualize patent renewal history as a revealed economic preference and leverage it as an objective supervisory signal to align the generative reasoning of Large Language Models (LLMs) with market realities, a process we term Eco-Semantic Alignment. Using a randomly sampled dataset of 10,000 European Patent Office patents across diverse technological domains, we trained the model not only to predict value tiers but also to reverse-engineer the Economic Chain-of-Thought from unstructured text. Empirical results demonstrate that ERA-IT significantly outperforms both conventional econometric models and zero-shot LLMs in predictive accuracy. More importantly, by generating explicit, logically grounded rationales for valuation, the framework serves as a transparent cognitive scaffold for decision-makers, reducing the opacity of black-box AI in high-stakes intellectual property management.

---

### Counting Clues: A Lightweight Probabilistic Baseline Can Match an LLM

Authors: Furong Jia, Yuan Pu, Finn Guo, Monica Agrawal
Published: 2025-12-14
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2512.12868v1

Large language models (LLMs) excel on multiple-choice clinical diagnosis benchmarks, yet it is unclear how much of this performance reflects underlying probabilistic reasoning. We study this through questions from MedQA, where the task is to select the most likely diagnosis. We introduce the Frequency-Based Probabilistic Ranker (FBPR), a lightweight method that scores options with a smoothed Naive Bayes over concept-diagnosis co-occurrence statistics from a large corpus. When co-occurrence statistics were sourced from the pretraining corpora for OLMo and Llama, FBPR achieves comparable performance to the corresponding LLMs pretrained on that same corpus. Direct LLM inference and FBPR largely get different questions correct, with an overlap only slightly above random chance, indicating complementary strengths of each method. These findings highlight the continued value of explicit probabilistic baselines: they provide a meaningful performance reference point and a complementary signal for potential hybridization. While the performance of LLMs seems to be driven by a mechanism other than simple frequency aggregation, we show that an approach similar to the historically grounded, low-complexity expert systems still accounts for a substantial portion of benchmark performance.

---

### Information-Consistent Language Model Recommendations through Group Relative Policy Optimization

Authors: Sonal Prabhune, Balaji Padmanabhan, Kaushik Dutta
Published: 2025-12-14
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2512.12858v3

Large Language Models (LLMs) are increasingly deployed in business-critical domains such as finance, education, healthcare, and customer support, where users expect consistent and reliable recommendations. Yet LLMs often exhibit variability when prompts are phrased with minor differences, even when semantically equivalent. Such inconsistency undermines trust, complicates compliance, and disrupts user experience. While personalization is desirable in certain contexts, many enterprise scenarios, such as HR onboarding, customer support, or policy disclosure, require invariant information delivery regardless of phrasing or prior conversational history. Existing approaches, including retrieval-augmented generation (RAG) and temperature tuning, improve factuality or reduce stochasticity, but cannot guarantee stability across equivalent prompts. In this paper, we propose a reinforcement learning framework based on Group Relative Policy Optimization (GRPO) to directly optimize for consistency. Unlike prior applications of GRPO, which have been limited to reasoning and code generation, we adapt GRPO to enforce the stability of information content across groups of semantically equivalent prompts. We introduce entropy-based helpfulness and stability rewards, treating prompt variants as groups and resetting conversational context to isolate phrasing effects. Experiments on investment and job recommendation tasks show that our GRPO-fine-tuned model reduces variability compared to the baseline LLM model. To our knowledge, this is a novel application of GRPO for aligning LLMs toward information consistency, reframing variability not as an acceptable feature of generative diversity, but as a correctable flaw in enterprise deployments.

---

### Forgetful but Faithful: A Cognitive Memory Architecture and Benchmark for Privacy-Aware Generative Agents

Authors: Saad Alqithami
Published: 2025-12-14
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2512.12856v1

As generative agents become increasingly sophisticated and deployed in long-term interactive scenarios, their memory management capabilities emerge as a critical bottleneck for both performance and privacy. Current approaches either maintain unlimited memory stores, leading to computational intractability and privacy concerns, or employ simplistic forgetting mechanisms that compromise agent coherence and functionality. This paper introduces the Memory-Aware Retention Schema (MaRS), a novel framework for human-centered memory management in generative agents, coupled with six theoretically-grounded forgetting policies that balance performance, privacy, and computational efficiency. We present the Forgetful but Faithful Agent (FiFA) benchmark, a comprehensive evaluation framework that assesses agent performance across narrative coherence, goal completion, social recall accuracy, privacy preservation, and cost efficiency. Through extensive experimentation involving 300 evaluation runs across multiple memory budgets and agent configurations, we demonstrate that our hybrid forgetting policy achieves superior performance (composite score: 0.911) while maintaining computational tractability and privacy guarantees. Our work establishes new benchmarks for memory-budgeted agent evaluation and provides practical guidelines for deploying generative agents in resource-constrained, privacy-sensitive environments. The theoretical foundations, implementation framework, and empirical results contribute to the emerging field of human-centered AI by addressing fundamental challenges in agent memory management that directly impact user trust, system scalability, and regulatory compliance.

---

### KANELÉ: Kolmogorov-Arnold Networks for Efficient LUT-based Evaluation

Authors: Duc Hoang, Aarush Gupta, Philip Harris
Published: 2025-12-14
Categories: cs.AR, cs.LG, eess.SY, hep-ex
arXiv: https://arxiv.org/abs/2512.12850v3

Low-latency, resource-efficient neural network inference on FPGAs is essential for applications demanding real-time capability and low power. Lookup table (LUT)-based neural networks are a common solution, combining strong representational power with efficient FPGA implementation. In this work, we introduce KANELÉ, a framework that exploits the unique properties of Kolmogorov-Arnold Networks (KANs) for FPGA deployment. Unlike traditional multilayer perceptrons (MLPs), KANs employ learnable one-dimensional splines with fixed domains as edge activations, a structure naturally suited to discretization and efficient LUT mapping. We present the first systematic design flow for implementing KANs on FPGAs, co-optimizing training with quantization and pruning to enable compact, high-throughput, and low-latency KAN architectures. Our results demonstrate up to a 2700x speedup and orders of magnitude resource savings compared to prior KAN-on-FPGA approaches. Moreover, KANELÉ matches or surpasses other LUT-based architectures on widely used benchmarks, particularly for tasks involving symbolic or physical formulas, while balancing resource usage across FPGA hardware. Finally, we showcase the versatility of the framework by extending it to real-time, power-efficient control systems.

---

### HaShiFlex: A High-Throughput Hardened Shifter DNN Accelerator with Fine-Tuning Flexibility

Authors: Jonathan Herbst, Michael Pellauer, Sherief Reda
Published: 2025-12-14
Categories: cs.AR, cs.LG
arXiv: https://arxiv.org/abs/2512.12847v1

We introduce a high-throughput neural network accelerator that embeds most network layers directly in hardware, minimizing data transfer and memory usage while preserving a degree of flexibility via a small neural processing unit for the final classification layer. By leveraging power-of-two (Po2) quantization for weights, we replace multiplications with simple rewiring, effectively reducing each convolution to a series of additions. This streamlined approach offers high-throughput, energy-efficient processing, making it highly suitable for applications where model parameters remain stable, such as continuous sensing tasks at the edge or large-scale data center deployments. Furthermore, by including a strategically chosen reprogrammable final layer, our design achieves high throughput without sacrificing fine-tuning capabilities. We implement this accelerator in a 7nm ASIC flow using MobileNetV2 as a baseline and report throughput, area, accuracy, and sensitivity to quantization and pruning - demonstrating both the advantages and potential trade-offs of the proposed architecture. We find that for MobileNetV2, we can improve inference throughput by 20x over fully programmable GPUs, processing 1.21 million images per second through a full forward pass while retaining fine-tuning flexibility. If absolutely no post-deployment fine tuning is required, this advantage increases to 67x at 4 million images per second.

---

### Selective Conformal Risk Control

Authors: Yunpeng Xu, Wenge Guo, Zhi Wei
Published: 2025-12-14
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2512.12844v2

Reliable uncertainty quantification is essential for deploying machine learning systems in high-stakes domains. Conformal prediction provides distribution-free coverage guarantees but often produces overly large prediction sets, limiting its practical utility. To address this issue, we propose \textit{Selective Conformal Risk Control} (SCRC), a unified framework that integrates conformal prediction with selective classification. The framework formulates uncertainty control as a two-stage problem: the first stage selects confident samples for prediction, and the second stage applies conformal risk control on the selected subset to construct calibrated prediction sets. We develop two algorithms under this framework. The first, SCRC-T, preserves exchangeability by computing thresholds jointly over calibration and test samples, offering exact finite-sample guarantees. The second, SCRC-I, is a calibration-only variant that provides PAC-style probabilistic guarantees while being more computational efficient. Experiments on two public datasets show that both methods achieve the target coverage and risk levels, with nearly identical performance, while SCRC-I exhibits slightly more conservative risk control but superior computational practicality. Our results demonstrate that selective conformal risk control offers an effective and efficient path toward compact, reliable uncertainty quantification.

---