# Research Papers — 2025-10 · bundle page 9 of 217

> Full-text markdown of papers #401–450 of 10843 published in 2025-10, newest first. AI-readable mirrors of the full 10843-paper 2025-10 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2025-10/page/9
[← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/8.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-10/page/10.md)

### Residual Distribution Predictive Systems

Authors: Sam Allen, Enrico Pescara, Johanna Ziegel
Published: 2025-10-30
Categories: stat.ME, stat.ML
arXiv: https://arxiv.org/abs/2510.26914v1

Conformal predictive systems are sets of predictive distributions with theoretical out-of-sample calibration guarantees. The calibration guarantees are typically that the set of predictions contains a forecast distribution whose prediction intervals exhibit the correct marginal coverage at all levels. Conformal predictive systems are constructed using conformity measures that quantify how well possible outcomes conform with historical data. However, alternative methods have been proposed to construct predictive systems with more appealing theoretical properties. We study an approach to construct predictive systems that we term Residual Distribution Predictive Systems. In the split conformal setting, this approach nests conformal predictive systems with a popular class of conformity measures, providing an alternative perspective on the classical approach. In the full conformal setting, the two approaches differ, and the new approach has the advantage that it does not rely on a conformity measure satisfying fairly stringent requirements to ensure that the predictive system is well-defined; it can readily be implemented alongside any point-valued regression method to yield predictive systems with out-of-sample calibration guarantees. The empirical performance of this approach is assessed using simulated data, where it is found to perform competitively with conformal predictive systems. However, the new approach offers considerable scope for implementation with alternative regression methods.

---

### Understanding and Enhancing Mamba-Transformer Hybrids for Memory Recall and Language Modeling

Authors: Hyunji Lee, Wenhao Yu, Hongming Zhang, Kaixin Ma, Jiyeon Kim, Dong Yu, Minjoon Seo
Published: 2025-10-30
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.26912v1

Hybrid models that combine state space models (SSMs) with attention mechanisms have shown strong performance by leveraging the efficiency of SSMs and the high recall ability of attention. However, the architectural design choices behind these hybrid models remain insufficiently understood. In this work, we analyze hybrid architectures through the lens of memory utilization and overall performance, and propose a complementary method to further enhance their effectiveness. We first examine the distinction between sequential and parallel integration of SSM and attention layers. Our analysis reveals several interesting findings, including that sequential hybrids perform better on shorter contexts, whereas parallel hybrids are more effective for longer contexts. We also introduce a data-centric approach of continually training on datasets augmented with paraphrases, which further enhances recall while preserving other capabilities. It generalizes well across different base models and outperforms architectural modifications aimed at enhancing recall. Our findings provide a deeper understanding of hybrid SSM-attention models and offer practical guidance for designing architectures tailored to various use cases. Our findings provide a deeper understanding of hybrid SSM-attention models and offer practical guidance for designing architectures tailored to various use cases.

---

### Discovering EV Charging Site Archetypes Through Few Shot Forecasting: The First U.S.-Wide Study

Authors: Kshitij Nikhal, Lucas Ackerknecht, Benjamin S. Riggan, Phillip Stahlfeld
Published: 2025-10-30
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.26910v2

The decarbonization of transportation relies on the widespread adoption of electric vehicles (EVs), which requires an accurate understanding of charging behavior to ensure cost-effective, grid-resilient infrastructure. Existing work is constrained by small-scale datasets, simple proximity-based modeling of temporal dependencies, and weak generalization to sites with limited operational history. To overcome these limitations, this work proposes a framework that integrates clustering with few-shot forecasting to uncover site archetypes using a novel large-scale dataset of charging demand. The results demonstrate that archetype-specific expert models outperform global baselines in forecasting demand at unseen sites. By establishing forecast performance as a basis for infrastructure segmentation, we generate actionable insights that enable operators to lower costs, optimize energy and pricing strategies, and support grid resilience critical to climate goals.

---

### Generative diffusion modeling protocols for improving the Kikuchi pattern indexing in electron back-scatter diffraction

Authors: Meghraj Prajapat, Alankar Alankar
Published: 2025-10-30
Categories: cond-mat.mtrl-sci, cs.CV
arXiv: https://arxiv.org/abs/2510.26907v2

Electron back-scatter diffraction (EBSD) has traditionally relied upon methods such as the Hough transform and dictionary Indexing to interpret diffraction patterns and extract crystallographic orientation. However, these methods encounter significant limitations, particularly when operating at high scanning speeds, where the exposure time per pattern is decreased beyond the operating sensitivity of CCD camera. Hence the signal to noise ratio decreases for the observed pattern which makes the pattern noisy, leading to reduced indexing accuracy. This research work aims to develop generative machine learning models for the post-processing or on-the-fly processing of Kikuchi patterns which are capable of restoring noisy EBSD patterns obtained at high scan speeds. These restored patterns can be used for the determination of crystal orientations to provide reliable indexing results. We compare the performance of such generative models in enhancing the quality of patterns captured at short exposure times (high scan speeds). An interesting observation is that the methodology is not data-hungry as typical machine learning methods.

---

### Cognition Envelopes for Bounded Decision Making in Autonomous UAS Operations

Authors: Pedro Antonio Alarcon Granadeno, Arturo Miguel Bernal Russell, Sofia Nelson, Demetrius Hernandez, Maureen Petterson, Michael Murphy, Walter J. Scheirer, Jane Cleland-Huang
Published: 2025-10-30
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.26905v3

Cyber-physical systems increasingly rely on foundational models, such as Large Language Models (LLMs) and Vision-Language Models (VLMs) to increase autonomy through enhanced perception, inference, and planning. However, these models also introduce new types of errors, such as hallucinations, over-generalizations, and context misalignments, resulting in incorrect and flawed decisions. To address this, we introduce the concept of Cognition Envelopes, designed to establish reasoning boundaries that constrain AI-generated decisions while complementing the use of meta-cognition and traditional safety envelopes. As with safety envelopes, Cognition Envelopes require practical guidelines and systematic processes for their definition, validation, and assurance. In this paper we describe an LLM/VLM-supported pipeline for dynamic clue analysis within the domain of small autonomous Uncrewed Aerial Systems deployed on Search and Rescue (SAR) missions, and a Cognition Envelope based on probabilistic reasoning and resource analysis. We evaluate the approach through assessing decisions made by our Clue Analysis Pipeline in a series of SAR missions. Finally, we identify key software engineering challenges for systematically designing, implementing, and validating Cognition Envelopes for AI-supported decisions in cyber-physical systems.

---

### PF-DAformer: Proximal Femur Segmentation via Domain Adaptive Transformer for Dual-Center QCT

Authors: Rochak Dhakal, Chen Zhao, Zixin Shi, Joyce H. Keyak, Tadashi S. Kaneko, Kuan-Jui Su, Hui Shen, Hong-Wen Deng, Weihua Zhou
Published: 2025-10-30
Categories: cs.CV, physics.med-ph
arXiv: https://arxiv.org/abs/2510.26903v1

Quantitative computed tomography (QCT) plays a crucial role in assessing bone strength and fracture risk by enabling volumetric analysis of bone density distribution in the proximal femur. However, deploying automated segmentation models in practice remains difficult because deep networks trained on one dataset often fail when applied to another. This failure stems from domain shift, where scanners, reconstruction settings, and patient demographics vary across institutions, leading to unstable predictions and unreliable quantitative metrics. Overcoming this barrier is essential for multi-center osteoporosis research and for ensuring that radiomics and structural finite element analysis results remain reproducible across sites. In this work, we developed a domain-adaptive transformer segmentation framework tailored for multi-institutional QCT. Our model is trained and validated on one of the largest hip fracture related research cohorts to date, comprising 1,024 QCT images scans from Tulane University and 384 scans from Rochester, Minnesota for proximal femur segmentation. To address domain shift, we integrate two complementary strategies within a 3D TransUNet backbone: adversarial alignment via Gradient Reversal Layer (GRL), which discourages the network from encoding site-specific cues, and statistical alignment via Maximum Mean Discrepancy (MMD), which explicitly reduces distributional mismatches between institutions. This dual mechanism balances invariance and fine-grained alignment, enabling scanner-agnostic feature learning while preserving anatomical detail.

---

### How Similar Are Grokipedia and Wikipedia? A Multi-Dimensional Textual and Structural Comparison

Authors: Taha Yasseri, Saeedeh Mohammadi
Published: 2025-10-30
Categories: cs.CY, cs.AI, cs.SI
arXiv: https://arxiv.org/abs/2510.26899v5

The launch of Grokipedia, an AI-generated encyclopedia developed by Elon Musk's xAI, was presented as a response to perceived ideological and structural biases in Wikipedia, aiming to produce "truthful" entries using the Grok large language model. Yet whether an AI-driven alternative can escape the biases and limitations of human-edited platforms remains unclear. This study conducts a large-scale computational comparison of 17,790 matched article pairs from the 20,000 most-edited English Wikipedia pages. Using metrics spanning lexical richness, readability, reference density, structural features, and semantic similarity, we assess how closely the two platforms align in form and substance. We find that Grokipedia articles are substantially longer and contain significantly fewer references per word. Moreover, Grokipedia's content divides into two distinct groups: one that remains semantically and stylistically aligned with Wikipedia, and another that diverges sharply. Among the dissimilar articles, we observe a systematic rightward shift in the political bias of frequently cited news media sources, concentrated primarily in entries related to history and religion, and literature and art. More broadly, the findings indicate that AI-generated encyclopedic content departs from established editorial norms, favoring narrative expansion over citation-based verification, raising questions about transparency, provenance, and the governance of knowledge in automated information systems.

---

### Integrating Ontologies with Large Language Models for Enhanced Control Systems in Chemical Engineering

Authors: Crystal Su, Kuai Yu, Jingrui Zhang, Mingyuan Shao, Daniel Bauer
Published: 2025-10-30
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.26898v2

This work presents an ontology-integrated large language model (LLM) framework for chemical engineering that unites structured domain knowledge with generative reasoning. The proposed pipeline aligns model training and inference with the COPE ontology through a sequence of data acquisition, semantic preprocessing, information extraction, and ontology mapping steps, producing templated question-answer pairs that guide fine-tuning. A control-focused decoding stage and citation gate enforce syntactic and factual grounding by constraining outputs to ontology-linked terms, while evaluation metrics quantify both linguistic quality and ontological accuracy. Feedback and future extensions, including semantic retrieval and iterative validation, further enhance the system's interpretability and reliability. This integration of symbolic structure and neural generation provides a transparent, auditable approach for applying LLMs to process control, safety analysis, and other critical engineering contexts.

---

### BI-DCGAN: A Theoretically Grounded Bayesian Framework for Efficient and Diverse GANs

Authors: Mahsa Valizadeh, Rui Tuo, James Caverlee
Published: 2025-10-30
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.26892v1

Generative Adversarial Networks (GANs) are proficient at generating synthetic data but continue to suffer from mode collapse, where the generator produces a narrow range of outputs that fool the discriminator but fail to capture the full data distribution. This limitation is particularly problematic, as generative models are increasingly deployed in real-world applications that demand both diversity and uncertainty awareness. In response, we introduce BI-DCGAN, a Bayesian extension of DCGAN that incorporates model uncertainty into the generative process while maintaining computational efficiency. BI-DCGAN integrates Bayes by Backprop to learn a distribution over network weights and employs mean-field variational inference to efficiently approximate the posterior distribution during GAN training. We establishes the first theoretical proof, based on covariance matrix analysis, that Bayesian modeling enhances sample diversity in GANs. We validate this theoretical result through extensive experiments on standard generative benchmarks, demonstrating that BI-DCGAN produces more diverse and robust outputs than conventional DCGANs, while maintaining training efficiency. These findings position BI-DCGAN as a scalable and timely solution for applications where both diversity and uncertainty are critical, and where modern alternatives like diffusion models remain too resource-intensive.

---

### The Denario project: Deep knowledge AI agents for scientific discovery

Authors: Francisco Villaescusa-Navarro, Boris Bolliet, Pablo Villanueva-Domingo, Adrian E. Bayer, Aidan Acquah, Chetana Amancharla, Almog Barzilay-Siegal, Pablo Bermejo, Camille Bilodeau, Pablo Cárdenas Ramírez, Miles Cranmer, Urbano L. França et al.
Published: 2025-10-30
Categories: cs.AI, cs.CL, cs.LG, cs.MA
arXiv: https://arxiv.org/abs/2510.26887v1

We present Denario, an AI multi-agent system designed to serve as a scientific research assistant. Denario can perform many different tasks, such as generating ideas, checking the literature, developing research plans, writing and executing code, making plots, and drafting and reviewing a scientific paper. The system has a modular architecture, allowing it to handle specific tasks, such as generating an idea, or carrying out end-to-end scientific analysis using Cmbagent as a deep-research backend. In this work, we describe in detail Denario and its modules, and illustrate its capabilities by presenting multiple AI-generated papers generated by it in many different scientific disciplines such as astrophysics, biology, biophysics, biomedical informatics, chemistry, material science, mathematical physics, medicine, neuroscience and planetary science. Denario also excels at combining ideas from different disciplines, and we illustrate this by showing a paper that applies methods from quantum physics and machine learning to astrophysical data. We report the evaluations performed on these papers by domain experts, who provided both numerical scores and review-like feedback. We then highlight the strengths, weaknesses, and limitations of the current system. Finally, we discuss the ethical implications of AI-driven research and reflect on how such technology relates to the philosophy of science. We publicly release the code at https://github.com/AstroPilot-AI/Denario. A Denario demo can also be run directly on the web at https://huggingface.co/spaces/astropilot-ai/Denario, and the full app will be deployed on the cloud.

---

### Do Vision-Language Models Measure Up? Benchmarking Visual Measurement Reading with MeasureBench

Authors: Fenfen Lin, Yesheng Liu, Haiyu Xu, Chen Yue, Zheqi He, Mingxuan Zhao, Miguel Hu Chen, Jiakang Liu, JG Yao, Xi Yang
Published: 2025-10-30
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.26865v2

Reading measurement instruments is effortless for humans and requires relatively little domain expertise, yet it remains surprisingly challenging for current vision-language models (VLMs) as we find in preliminary evaluation. In this work, we introduce MeasureBench, a benchmark on visual measurement reading covering both real-world and synthesized images of various types of measurements, along with an extensible pipeline for data synthesis. Our pipeline procedurally generates a specified type of gauge with controllable visual appearance, enabling scalable variation in key details such as pointers, scales, fonts, lighting, and clutter. Evaluation on popular proprietary and open-weight VLMs shows that even the strongest frontier VLMs struggle with measurement reading in general. We have also conducted preliminary experiments with reinforcement finetuning (RFT) over synthetic data, and find a significant improvement on both in-domain synthetic subset and real-world images. Our analysis highlights a fundamental limitation of current VLMs in fine-grained spatial grounding. We hope this resource and our code releases can help future advances on visually grounded numeracy and precise spatial perception of VLMs, bridging the gap between recognizing numbers and measuring the world.

---

### Evaluating Perspectival Biases in Cross-Modal Retrieval

Authors: Teerapol Saengsukhiran, Peerawat Chomphooyod, Narabodee Rodjananant, Chompakorn Chaksangchaichot, Patawee Prakrankamanant, Witthawin Sripheanpol, Pak Lovichit, Sarana Nutanong, Ekapol Chuangsuwanich
Published: 2025-10-30
Categories: cs.IR, cs.CL
arXiv: https://arxiv.org/abs/2510.26861v4

Multimodal retrieval systems are expected to operate in a semantic space, agnostic to the language or cultural origin of the query. In practice, however, retrieval outcomes systematically reflect perspectival biases: deviations shaped by linguistic prevalence and cultural associations. We introduce the Cross-Cultural, Cross-Modal, Cross-lingual Multimodal (3XCM) benchmark to isolate these effects. Results from our studies indicate that, for image-to-text retrieval, models tend to favor entries from prevalent languages over those that are semantically faithful. For text-to-image retrieval, we observe a consistent "tugging effect" in the joint embedding space between semantic alignment and language-conditioned cultural association. When semantic representations are insufficiently resolved, particularly in low-resource languages, similarity is increasingly governed by culturally familiar visual patterns, leading to systematic association bias in retrieval. Our findings suggest that achieving equitable multimodal retrieval necessitates targeted strategies that explicitly decouple language from culture, rather than relying solely on broader data exposure. This work highlights the need to treat linguistic and cultural biases as distinct, measurable challenges in multimodal representation learning.

---

### Leveraging Foundation Models for Enhancing Robot Perception and Action

Authors: Reihaneh Mirjalili
Published: 2025-10-30
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2510.26855v1

This thesis investigates how foundation models can be systematically leveraged to enhance robotic capabilities, enabling more effective localization, interaction, and manipulation in unstructured environments. The work is structured around four core lines of inquiry, each addressing a fundamental challenge in robotics while collectively contributing to a cohesive framework for semantics-aware robotic intelligence.

---

### Inverse Knowledge Search over Verifiable Reasoning: Synthesizing a Scientific Encyclopedia from a Long Chains-of-Thought Knowledge Base

Authors: Yu Li, Yuan Huang, Tao Wang, Caiyu Fan, Xiansheng Cai, Sihan Hu, Xinzijian Liu, Cheng Shi, Mingjun Xu, Zhen Wang, Yan Wang, Xiangqi Jin et al.
Published: 2025-10-30
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.26854v3

Most scientific materials compress reasoning, presenting conclusions while omitting the derivational chains that justify them. This compression hinders verification by lacking explicit, step-wise justifications and inhibits cross-domain links by collapsing the very pathways that establish the logical and causal connections between concepts. We introduce a scalable framework that decompresses scientific reasoning, constructing a verifiable Long Chain-of-Thought (LCoT) knowledge base and projecting it into an emergent encyclopedia, SciencePedia. Our pipeline operationalizes an endpoint-driven, reductionist strategy: a Socratic agent, guided by a curriculum of around 200 courses, generates approximately 3 million first-principles questions. To ensure high fidelity, multiple independent solver models generate LCoTs, which are then rigorously filtered by prompt sanitization and cross-model answer consensus, retaining only those with verifiable endpoints. This verified corpus powers the Brainstorm Search Engine, which performs inverse knowledge search -- retrieving diverse, first-principles derivations that culminate in a target concept. This engine, in turn, feeds the Plato synthesizer, which narrates these verified chains into coherent articles. The initial SciencePedia comprises approximately 200,000 fine-grained entries spanning mathematics, physics, chemistry, biology, engineering, and computation. In evaluations across six disciplines, Plato-synthesized articles (conditioned on retrieved LCoTs) exhibit substantially higher knowledge-point density and significantly lower factual error rates than an equally-prompted baseline without retrieval (as judged by an external LLM). Built on this verifiable LCoT knowledge base, this reasoning-centric approach enables trustworthy, cross-domain scientific synthesis at scale and establishes the foundation for an ever-expanding encyclopedia.

---

### CATArena: Evaluating Evolutionary Capabilities of Code Agents via Iterative Tournaments

Authors: Lingyue Fu, Xin Ding, Linyue Pan, Yaoming Zhu, Shao Zhang, Lin Qiu, Weiwen Liu, Weinan Zhang, Xuezhi Cao, Xunliang Cai, Jiaxin Ding, Yong Yu
Published: 2025-10-30
Categories: cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2510.26852v2

Current evaluation for Large Language Model (LLM) code agents predominantly focus on generating functional code in single-turn scenarios, which fails to evaluate the agent's capability for continuous code optimization and multi-turn iterative development. To bridge this gap, we introduce CATArena, a framework designed to evaluate the evolutionary capabilities of code agents via iterative tournaments. Agents engage in multi-turn tournaments and continuously refine their code through self-reflection and peer-learning based on comprehensive execution feedback. For evaluation, we propose a dual-metric system to decouple static generation proficiency from evolutionary potential. Extensive experiments reveal that an agent's evolutionary potential is not strictly correlated with its initial proficiency. Our analysis further reveals that current agents struggle to concurrently leverage both peer-learning and self-reflection for effective performance gains. Furthermore, the results validate CATArena's high extensibility and resistance to variance tasks, establishing it as a continuous and reliable standard for assessing the evolutionary capability of LLM code agents.

---

### Broken-Token: Filtering Obfuscated Prompts by Counting Characters-Per-Token

Authors: Shaked Zychlinski, Yuval Kainan
Published: 2025-10-30
Categories: cs.CR, cs.AI, cs.CL, cs.IT
arXiv: https://arxiv.org/abs/2510.26847v1

Large Language Models (LLMs) are susceptible to jailbreak attacks where malicious prompts are disguised using ciphers and character-level encodings to bypass safety guardrails. While these guardrails often fail to interpret the encoded content, the underlying models can still process the harmful instructions. We introduce CPT-Filtering, a novel, model-agnostic with negligible-costs and near-perfect accuracy guardrail technique that aims to mitigate these attacks by leveraging the intrinsic behavior of Byte-Pair Encoding (BPE) tokenizers. Our method is based on the principle that tokenizers, trained on natural language, represent out-of-distribution text, such as ciphers, using a significantly higher number of shorter tokens. Our technique uses a simple yet powerful artifact of using language models: the average number of Characters Per Token (CPT) in the text. This approach is motivated by the high compute cost of modern methods - relying on added modules such as dedicated LLMs or perplexity models. We validate our approach across a large dataset of over 100,000 prompts, testing numerous encoding schemes with several popular tokenizers. Our experiments demonstrate that a simple CPT threshold robustly identifies encoded text with high accuracy, even for very short inputs. CPT-Filtering provides a practical defense layer that can be immediately deployed for real-time text filtering and offline data curation.

---

### CAS-Spec: Cascade Adaptive Self-Speculative Decoding for On-the-Fly Lossless Inference Acceleration of LLMs

Authors: Zhiyuan Ning, Jiawei Shao, Ruge Xu, Xinfei Guo, Jun Zhang, Chi Zhang, Xuelong Li
Published: 2025-10-30
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.26843v1

Speculative decoding has become a widely adopted as an effective technique for lossless inference acceleration when deploying large language models (LLMs). While on-the-fly self-speculative methods offer seamless integration and broad utility, they often fall short of the speed gains achieved by methods relying on specialized training. Cascading a hierarchy of draft models promises further acceleration and flexibility, but the high cost of training multiple models has limited its practical application. In this paper, we propose a novel Cascade Adaptive Self-Speculative Decoding (CAS-Spec) method which constructs speculative draft models by leveraging dynamically switchable inference acceleration (DSIA) strategies, including layer sparsity and activation quantization. Furthermore, traditional vertical and horizontal cascade algorithms are inefficient when applied to self-speculative decoding methods. We introduce a Dynamic Tree Cascade (DyTC) algorithm that adaptively routes the multi-level draft models and assigns the draft lengths, based on the heuristics of acceptance rates and latency prediction. Our CAS-Spec method achieves state-of-the-art acceleration compared to existing on-the-fly speculative decoding methods, with an average speedup from $1.1\times$ to $2.3\times$ over autoregressive decoding across various LLMs and datasets. DyTC improves the average speedup by $47$\% and $48$\% over cascade-based baseline and tree-based baseline algorithms, respectively. CAS-Spec can be easily integrated into most existing LLMs and holds promising potential for further acceleration as self-speculative decoding techniques continue to evolve.

---

### FedPF: Accurate Target Privacy Preserving Federated Learning Balancing Fairness and Utility

Authors: Kangkang Sun, Jun Wu, Minyi Guo, Jianhua Li, Jianwei Huang
Published: 2025-10-30
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.26841v2

Federated Learning (FL) enables collaborative model training without data sharing, yet participants face a fundamental challenge, e.g., simultaneously ensuring fairness across demographic groups while protecting sensitive client data. We introduce a differentially private fair FL algorithm (FedPF) that transforms this multi-objective optimization into a zero-sum game where fairness and privacy constraints compete against model utility. Our theoretical analysis reveals an inverse relationship: privacy mechanisms that protect sensitive attributes can reduce the statistical power available for detecting and correcting demographic biases under finite samples in federated settings. We further show that our theoretical bounds are consistent with a non-monotonic fairness-utility relationship, which is empirically validated by experiments where moderate fairness constraints improve generalization before excessive enforcement degrades performance. Compared with mainstream algorithms, even under strict privacy constraints, FedPF still maintains the lowest discrimination level among all tested algorithms while retaining high utility. Experimental validation demonstrates up to 42.9 % discrimination reduction across three datasets while maintaining competitive accuracy, but more importantly, reveals that achieving strong privacy and fairness simultaneously requires carefully balanced tradeoffs rather than optimizing either objective in isolation. Furthermore, hardware-level simulations demonstrate that FedPF maintains a low computational footprint, making it suitable for resource-constrained edge devices. The source code for our proposed algorithm is publicly accessible at https://github.com/szpsunkk/FedPF.

---

### SpotIt: Evaluating Text-to-SQL Evaluation with Formal Verification

Authors: Rocky Klopfenstein, Yang He, Andrew Tremante, Yuepeng Wang, Nina Narodytska, Haoze Wu
Published: 2025-10-30
Categories: cs.DB, cs.AI, cs.FL, cs.LO
arXiv: https://arxiv.org/abs/2510.26840v2

Community-driven Text-to-SQL evaluation platforms play a pivotal role in tracking the state of the art of Text-to-SQL performance. The reliability of the evaluation process is critical for driving progress in the field. Current evaluation methods are largely test-based, which involves comparing the execution results of a generated SQL query and a human-labeled ground-truth on a static test database. Such an evaluation is optimistic, as two queries can coincidentally produce the same output on the test database while actually being different. In this work, we propose a new alternative evaluation pipeline, called SpotIt, where a formal bounded equivalence verification engine actively searches for a database that differentiates the generated and ground-truth SQL queries. We develop techniques to extend existing verifiers to support a richer SQL subset relevant to Text-to-SQL. A performance evaluation of ten Text-to-SQL methods on the high-profile BIRD dataset suggests that test-based methods can often overlook differences between the generated query and the ground-truth. Further analysis of the verification results reveals a more complex picture of the current Text-to-SQL evaluation.

---

### Are Video Models Ready as Zero-Shot Reasoners? An Empirical Study with the MME-CoF Benchmark

Authors: Ziyu Guo, Xinyan Chen, Renrui Zhang, Ruichuan An, Yu Qi, Dongzhi Jiang, Xiangtai Li, Manyuan Zhang, Hongsheng Li, Pheng-Ann Heng
Published: 2025-10-30
Categories: cs.CV, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2510.26802v1

Recent video generation models can produce high-fidelity, temporally coherent videos, indicating that they may encode substantial world knowledge. Beyond realistic synthesis, they also exhibit emerging behaviors indicative of visual perception, modeling, and manipulation. Yet, an important question still remains: Are video models ready to serve as zero-shot reasoners in challenging visual reasoning scenarios? In this work, we conduct an empirical study to comprehensively investigate this question, focusing on the leading and popular Veo-3. We evaluate its reasoning behavior across 12 dimensions, including spatial, geometric, physical, temporal, and embodied logic, systematically characterizing both its strengths and failure modes. To standardize this study, we curate the evaluation data into MME-CoF, a compact benchmark that enables in-depth and thorough assessment of Chain-of-Frame (CoF) reasoning. Our findings reveal that while current video models demonstrate promising reasoning patterns on short-horizon spatial coherence, fine-grained grounding, and locally consistent dynamics, they remain limited in long-horizon causal reasoning, strict geometric constraints, and abstract logic. Overall, they are not yet reliable as standalone zero-shot reasoners, but exhibit encouraging signs as complementary visual engines alongside dedicated reasoning models. Project page: https://video-cof.github.io

---

### OmniX: From Unified Panoramic Generation and Perception to Graphics-Ready 3D Scenes

Authors: Yukun Huang, Jiwen Yu, Yanning Zhou, Jianan Wang, Xintao Wang, Pengfei Wan, Xihui Liu
Published: 2025-10-30
Categories: cs.CV, cs.GR, cs.LG
arXiv: https://arxiv.org/abs/2510.26800v2

There are two prevalent ways for automatic 3D scene construction: procedural generation and 2D lifting. Among these, panorama-based 2D lifting has emerged as a promising technique, leveraging powerful 2D generative priors to produce immersive, realistic, and diverse 3D environments. In this work, we advance this technique to generate graphics-ready 3D scenes suitable for physically based rendering (PBR), relighting, and simulation. Our key insight is to repurpose 2D generative models for panorama perception of geometry, textures, and PBR materials. Unlike existing 2D lifting approaches that emphasize appearance generation and neglect the perception of intrinsic properties, we present OmniX, a versatile and unified framework for panorama generation, perception, and completion. Built upon cross-modal adapter structure and cyclic spatial operators, OmniX effectively repurposes pre-trained 2D flow matching priors for joint modeling of multimodal, seamless equirectangular representations. Furthermore, we construct a large-scale synthetic panorama dataset comprising high-quality multimodal panoramas from diverse indoor and outdoor scenes. Extensive experiments demonstrate the effectiveness and generality of OmniX as a unified framework for panorama generation and perception across geometry, lighting, and semantics, enabling graphics-ready 3D scene generation and opening new possibilities for immersive and physically realistic virtual world creation.

---

### Masked Diffusion Captioning for Visual Feature Learning

Authors: Chao Feng, Zihao Wei, Andrew Owens
Published: 2025-10-30
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.26799v1

We learn visual features by captioning images with an image-conditioned masked diffusion language model, a formulation we call masked diffusion captioning (MDC). During training, text tokens in each image-caption pair are masked at a randomly chosen ratio, and a decoder conditioned on visual features is trained to reconstruct the original text. After training, the learned visual features can be applied to downstream vision tasks. Unlike autoregressive captioning, the strength of the visual learning signal in MDC does not depend on each token's position in the sequence, reducing the need for auxiliary objectives. Linear probing experiments across a variety of academic-scale models and datasets show that the learned visual features are competitive with those produced by autoregressive and contrastive approaches.

---

### See4D: Pose-Free 4D Generation via Auto-Regressive Video Inpainting

Authors: Dongyue Lu, Ao Liang, Tianxin Huang, Xiao Fu, Yuyang Zhao, Baorui Ma, Liang Pan, Wei Yin, Lingdong Kong, Wei Tsang Ooi, Ziwei Liu
Published: 2025-10-30
Categories: cs.CV, cs.GR
arXiv: https://arxiv.org/abs/2510.26796v2

Immersive applications call for synthesizing spatiotemporal 4D content from casual videos without costly 3D supervision. Existing video-to-4D methods typically rely on manually annotated camera poses, which are labor-intensive and brittle for in-the-wild footage. Recent warp-then-inpaint approaches mitigate the need for pose labels by warping input frames along a novel camera trajectory and using an inpainting model to fill missing regions, thereby depicting the 4D scene from diverse viewpoints. However, this trajectory-to-trajectory formulation often entangles camera motion with scene dynamics and complicates both modeling and inference. We introduce See4D, a pose-free, trajectory-to-camera framework that replaces explicit trajectory prediction with rendering to a bank of fixed virtual cameras, thereby separating camera control from scene modeling. A view-conditional video inpainting model is trained to learn a robust geometry prior by denoising realistically synthesized warped images and to inpaint occluded or missing regions across virtual viewpoints, eliminating the need for explicit 3D annotations. Building on this inpainting core, we design a spatiotemporal autoregressive inference pipeline that traverses virtual-camera splines and extends videos with overlapping windows, enabling coherent generation at bounded per-step complexity. We validate See4D on cross-view video generation and sparse reconstruction benchmarks. Across quantitative metrics and qualitative assessments, our method achieves superior generalization and improved performance relative to pose- or trajectory-conditioned baselines, advancing practical 4D world modeling from casual videos.

---

### Scaling Image Geo-Localization to Continent Level

Authors: Philipp Lindenberger, Paul-Edouard Sarlin, Jan Hosang, Matteo Balice, Marc Pollefeys, Simon Lynen, Eduard Trulls
Published: 2025-10-30
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2510.26795v1

Determining the precise geographic location of an image at a global scale remains an unsolved challenge. Standard image retrieval techniques are inefficient due to the sheer volume of images (>100M) and fail when coverage is insufficient. Scalable solutions, however, involve a trade-off: global classification typically yields coarse results (10+ kilometers), while cross-view retrieval between ground and aerial imagery suffers from a domain gap and has been primarily studied on smaller regions. This paper introduces a hybrid approach that achieves fine-grained geo-localization across a large geographic expanse the size of a continent. We leverage a proxy classification task during training to learn rich feature representations that implicitly encode precise location information. We combine these learned prototypes with embeddings of aerial imagery to increase robustness to the sparsity of ground-level data. This enables direct, fine-grained retrieval over areas spanning multiple countries. Our extensive evaluation demonstrates that our approach can localize within 200m more than 68\% of queries of a dataset covering a large part of Europe. The code is publicly available at https://scaling-geoloc.github.io.

---

### The Quest for Generalizable Motion Generation: Data, Model, and Evaluation

Authors: Jing Lin, Ruisi Wang, Junzhe Lu, Ziqi Huang, Guorui Song, Ailing Zeng, Xian Liu, Chen Wei, Wanqi Yin, Qingping Sun, Zhongang Cai, Lei Yang et al.
Published: 2025-10-30
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.26794v2

Despite recent advances in 3D human motion generation (MoGen) on standard benchmarks, existing text-to-motion models still face a fundamental bottleneck in their generalization capability. In contrast, adjacent generative fields, most notably video generation (ViGen), have demonstrated remarkable generalization in modeling human behaviors, highlighting transferable insights that MoGen can leverage. Motivated by this observation, we present a comprehensive framework that systematically transfers knowledge from ViGen to MoGen across three key pillars: data, modeling, and evaluation. First, we introduce ViMoGen-228K, a large-scale dataset comprising 228,000 high-quality motion samples that integrates high-fidelity optical MoCap data with semantically annotated motions from web videos and synthesized samples generated by state-of-the-art ViGen models. The dataset includes both text-motion pairs and text-video-motion triplets, substantially expanding semantic diversity. Second, we propose ViMoGen, a flow-matching-based diffusion transformer that unifies priors from MoCap data and ViGen models through gated multimodal conditioning. To enhance efficiency, we further develop ViMoGen-light, a distilled variant that eliminates video generation dependencies while preserving strong generalization. Finally, we present MBench, a hierarchical benchmark designed for fine-grained evaluation across motion quality, prompt fidelity, and generalization ability. Extensive experiments show that our framework significantly outperforms existing approaches in both automatic and human evaluations. The code, data, and benchmark will be made publicly available. Homepage: https://motrixlab.github.io/2026_iclr_vimogen.

---

### Learning Pseudorandom Numbers with Transformers: Permuted Congruential Generators, Curricula, and Interpretability

Authors: Tao Tao, Maissam Barkeshli
Published: 2025-10-30
Categories: cs.LG, cond-mat.dis-nn, cs.CR
arXiv: https://arxiv.org/abs/2510.26792v2

We study the ability of Transformer models to learn sequences generated by Permuted Congruential Generators (PCGs), a widely used family of pseudo-random number generators (PRNGs). PCGs introduce substantial additional difficulty over linear congruential generators (LCGs) by applying a series of bit-wise shifts, XORs, rotations and truncations to the hidden state. We show that Transformers can nevertheless successfully perform in-context prediction on unseen sequences from diverse PCG variants, in tasks that are beyond published classical attacks. In our experiments we scale moduli up to $2^{22}$ using up to $50$ million model parameters and datasets with up to $5$ billion tokens. Surprisingly, we find even when the output is truncated to a single bit, it can be reliably predicted by the model. When multiple distinct PRNGs are presented together during training, the model can jointly learn them, identifying structures from different permutations. We demonstrate a scaling law with modulus $m$: the number of in-context sequence elements required for near-perfect prediction grows as $\sqrt{m}$. For larger moduli, optimization enters extended stagnation phases; in our experiments, learning moduli $m \geq 2^{20}$ requires incorporating training data from smaller moduli, demonstrating a critical necessity for curriculum learning. Finally, we analyze embedding layers and uncover a novel clustering phenomenon: the top principal components spontaneously group the integer inputs into bitwise rotationally-invariant clusters, revealing how representations can transfer from smaller to larger moduli.

---

### Gistify! Codebase-Level Understanding via Runtime Execution

Authors: Hyunji Lee, Minseon Kim, Chinmay Singh, Matheus Pereira, Atharv Sonwane, Isadora White, Elias Stengel-Eskin, Mohit Bansal, Zhengyan Shi, Alessandro Sordoni, Marc-Alexandre Côté, Xingdi Yuan et al.
Published: 2025-10-30
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.26790v1

As coding agents are increasingly deployed in large codebases, the need to automatically design challenging, codebase-level evaluation is central. We propose Gistify, a task where a coding LLM must create a single, minimal, self-contained file that can reproduce a specific functionality of a codebase. The coding LLM is given full access to a codebase along with a specific entrypoint (e.g., a python command), and the generated file must replicate the output of the same command ran under the full codebase, while containing only the essential components necessary to execute the provided command. Success on Gistify requires both structural understanding of the codebase, accurate modeling of its execution flow as well as the ability to produce potentially large code patches. Our findings show that current state-of-the-art models struggle to reliably solve Gistify tasks, especially ones with long executions traces.

---

### Defeating the Training-Inference Mismatch via FP16

Authors: Penghui Qi, Zichen Liu, Xiangxin Zhou, Tianyu Pang, Chao Du, Wee Sun Lee, Min Lin
Published: 2025-10-30
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2510.26788v1

Reinforcement learning (RL) fine-tuning of large language models (LLMs) often suffers from instability due to the numerical mismatch between the training and inference policies. While prior work has attempted to mitigate this issue through algorithmic corrections or engineering alignments, we show that its root cause lies in the floating point precision itself. The widely adopted BF16, despite its large dynamic range, introduces large rounding errors that breaks the consistency between training and inference. In this work, we demonstrate that simply reverting to \textbf{FP16} effectively eliminates this mismatch. The change is simple, fully supported by modern frameworks with only a few lines of code change, and requires no modification to the model architecture or learning algorithm. Our results suggest that using FP16 uniformly yields more stable optimization, faster convergence, and stronger performance across diverse tasks, algorithms and frameworks. We hope these findings motivate a broader reconsideration of precision trade-offs in RL fine-tuning.

---

### Remote Labor Index: Measuring AI Automation of Remote Work

Authors: Mantas Mazeika, Alice Gatti, Cristina Menghini, Udari Madhushani Sehwag, Shivam Singhal, Yury Orlovskiy, Steven Basart, Manasi Sharma, Denis Peskoff, Elaine Lau, Jaehyuk Lim, Lachlan Carroll et al.
Published: 2025-10-30
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2510.26787v1

AIs have made rapid progress on research-oriented benchmarks of knowledge and reasoning, but it remains unclear how these gains translate into economic value and automation. To measure this, we introduce the Remote Labor Index (RLI), a broadly multi-sector benchmark comprising real-world, economically valuable projects designed to evaluate end-to-end agent performance in practical settings. AI agents perform near the floor on RLI, with the highest-performing agent achieving an automation rate of 2.5%. These results help ground discussions of AI automation in empirical evidence, setting a common basis for tracking AI impacts and enabling stakeholders to proactively navigate AI-driven labor automation.

---

### HEIR: Learning Graph-Based Motion Hierarchies

Authors: Cheng Zheng, William Koch, Baiang Li, Felix Heide
Published: 2025-10-30
Categories: cs.CV, cs.GR, cs.LG
arXiv: https://arxiv.org/abs/2510.26786v1

Hierarchical structures of motion exist across research fields, including computer vision, graphics, and robotics, where complex dynamics typically arise from coordinated interactions among simpler motion components. Existing methods to model such dynamics typically rely on manually-defined or heuristic hierarchies with fixed motion primitives, limiting their generalizability across different tasks. In this work, we propose a general hierarchical motion modeling method that learns structured, interpretable motion relationships directly from data. Our method represents observed motions using graph-based hierarchies, explicitly decomposing global absolute motions into parent-inherited patterns and local motion residuals. We formulate hierarchy inference as a differentiable graph learning problem, where vertices represent elemental motions and directed edges capture learned parent-child dependencies through graph neural networks. We evaluate our hierarchical reconstruction approach on three examples: 1D translational motion, 2D rotational motion, and dynamic 3D scene deformation via Gaussian splatting. Experimental results show that our method reconstructs the intrinsic motion hierarchy in 1D and 2D cases, and produces more realistic and interpretable deformations compared to the baseline on dynamic 3D Gaussian splatting scenes. By providing an adaptable, data-driven hierarchical modeling paradigm, our method offers a formulation applicable to a broad range of motion-centric tasks. Project Page: https://light.princeton.edu/HEIR/

---

### LLMs Process Lists With General Filter Heads

Authors: Arnab Sen Sharma, Giordano Rogers, Natalie Shapira, David Bau
Published: 2025-10-30
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.26784v3

We investigate the mechanisms underlying a range of list-processing tasks in LLMs, and we find that LLMs have learned to encode a compact, causal representation of a general filtering operation that mirrors the generic "filter" function of functional programming. Using causal mediation analysis on a diverse set of list-processing tasks, we find that a small number of attention heads, which we dub filter heads, encode a compact representation of the filtering predicate in their query states at certain tokens. We demonstrate that this predicate representation is general and portable: it can be extracted and reapplied to execute the same filtering operation on different collections, presented in different formats, languages, or even in tasks. However, we also identify situations where transformer LMs can exploit a different strategy for filtering: eagerly evaluating if an item satisfies the predicate and storing this intermediate result as a flag directly in the item representations. Our results reveal that transformer LMs can develop human-interpretable implementations of abstract computational operations that generalize in ways that are surprisingly similar to strategies used in traditional functional programming patterns.

---

### A Unified Theory for Causal Inference: Direct Debiased Machine Learning via Bregman-Riesz Regression

Authors: Masahiro Kato
Published: 2025-10-30
Categories: stat.ML, cs.LG, econ.EM, math.ST, stat.ME
arXiv: https://arxiv.org/abs/2510.26783v1

This note introduces a unified theory for causal inference that integrates Riesz regression, covariate balancing, density-ratio estimation (DRE), targeted maximum likelihood estimation (TMLE), and the matching estimator in average treatment effect (ATE) estimation. In ATE estimation, the balancing weights and the regression functions of the outcome play important roles, where the balancing weights are referred to as the Riesz representer, bias-correction term, and clever covariates, depending on the context. Riesz regression, covariate balancing, DRE, and the matching estimator are methods for estimating the balancing weights, where Riesz regression is essentially equivalent to DRE in the ATE context, the matching estimator is a special case of DRE, and DRE is in a dual relationship with covariate balancing. TMLE is a method for constructing regression function estimators such that the leading bias term becomes zero. Nearest Neighbor Matching is equivalent to Least Squares Density Ratio Estimation and Riesz Regression.

---

### Cloning Deterministic Worlds: The Critical Role of Latent Geometry in Long-Horizon World Models

Authors: Zaishuo Xia, Yukuan Lu, Xinyi Li, Yifan Xu, Yubei Chen
Published: 2025-10-30
Categories: cs.LG, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2510.26782v3

A world model is an internal model that simulates how the world evolves. Given past observations and actions, it predicts the future physical state of both the embodied agent and its environment. Accurate world models are essential for enabling agents to think, plan, and reason effectively in complex, dynamic settings. However, existing world models often focus on random generation of open worlds, but neglect the need for high-fidelity modeling of deterministic scenarios (such as fixed-map mazes and static space robot navigation). In this work, we take a step toward building a truly accurate world model by addressing a fundamental yet open problem: constructing a model that can fully clone a deterministic 3D world. 1) Through diagnostic experiment, we quantitatively demonstrate that high-fidelity cloning is feasible and the primary bottleneck for long-horizon fidelity is the geometric structure of the latent representation, not the dynamics model itself. 2) Building on this insight, we show that applying temporal contrastive learning principle as a geometric regularization can effectively curate a latent space that better reflects the underlying physical state manifold, demonstrating that contrastive constraints can serve as a powerful inductive bias for stable world modeling; we call this approach Geometrically-Regularized World Models (GRWM). At its core is a lightweight geometric regularization module that can be seamlessly integrated into standard autoencoders, reshaping their latent space to provide a stable foundation for effective dynamics modeling. By focusing on representation quality, GRWM offers a simple yet powerful pipeline for improving world model fidelity.

---

### ChartAB: A Benchmark for Chart Grounding & Dense Alignment

Authors: Aniruddh Bansal, Davit Soselia, Dang Nguyen, Tianyi Zhou
Published: 2025-10-30
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.26781v2

Charts play an important role in visualization, reasoning, data analysis, and the exchange of ideas among humans. However, existing vision-language models (VLMs) still lack accurate perception of details and struggle to extract fine-grained structures from charts. Such limitations in chart grounding also hinder their ability to compare multiple charts and reason over them. In this paper, we introduce a novel "ChartAlign Benchmark (ChartAB)" to provide a comprehensive evaluation of VLMs in chart grounding tasks, i.e., extracting tabular data, localizing visualization elements, and recognizing various attributes from charts of diverse types and complexities. We design a JSON template to facilitate the calculation of evaluation metrics specifically tailored for each grounding task. By incorporating a novel two-stage inference workflow, the benchmark can further evaluate VLMs capability to align and compare elements/attributes across two charts. Our analysis of evaluations on several recent VLMs reveals new insights into their perception biases, weaknesses, robustness, and hallucinations in chart understanding. These findings highlight the fine-grained discrepancies among VLMs in chart understanding tasks and point to specific skills that need to be strengthened in current models.

---

### Surpassing state of the art on AMD area estimation from RGB fundus images through careful selection of U-Net architectures and loss functions for class imbalance

Authors: Valentyna Starodub, Mantas Lukoševičius
Published: 2025-10-30
Categories: cs.CV, cs.LG, eess.IV
arXiv: https://arxiv.org/abs/2510.26778v1

Age-related macular degeneration (AMD) is one of the leading causes of irreversible vision impairment in people over the age of 60. This research focuses on semantic segmentation for AMD lesion detection in RGB fundus images, a non-invasive and cost-effective imaging technique. The results of the ADAM challenge - the most comprehensive AMD detection from RGB fundus images research competition and open dataset to date - serve as a benchmark for our evaluation. Taking the U-Net connectivity as a base of our framework, we evaluate and compare several approaches to improve the segmentation model's architecture and training pipeline, including pre-processing techniques, encoder (backbone) deep network types of varying complexity, and specialized loss functions to mitigate class imbalances on image and pixel levels. The main outcome of this research is the final configuration of the AMD detection framework, which outperforms all the prior ADAM challenge submissions on the multi-class segmentation of different AMD lesion types in non-invasive RGB fundus images. The source code used to conduct the experiments presented in this paper is made freely available.

---

### Pre-trained Forecasting Models: Strong Zero-Shot Feature Extractors for Time Series Classification

Authors: Andreas Auer, Daniel Klotz, Sebastinan Böck, Sepp Hochreiter
Published: 2025-10-30
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.26777v1

Recent research on time series foundation models has primarily focused on forecasting, leaving it unclear how generalizable their learned representations are. In this study, we examine whether frozen pre-trained forecasting models can provide effective representations for classification. To this end, we compare different representation extraction strategies and introduce two model-agnostic embedding augmentations. Our experiments show that the best forecasting models achieve classification accuracy that matches or even surpasses that of state-of-the-art models pre-trained specifically for classification. Moreover, we observe a positive correlation between forecasting and classification performance. These findings challenge the assumption that task-specific pre-training is necessary, and suggest that learning to forecast may provide a powerful route toward constructing general-purpose time series foundation models.

---

### Faithful and Fast Influence Function via Advanced Sampling

Authors: Jungyeon Koh, Hyeonsu Lyu, Jonggyu Jang, Hyun Jong Yang
Published: 2025-10-30
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.26776v2

How can we explain the influence of training data on black-box models? Influence functions (IFs) offer a post-hoc solution by utilizing gradients and Hessians. However, computing the Hessian for an entire dataset is resource-intensive, necessitating a feasible alternative. A common approach involves randomly sampling a small subset of the training data, but this method often results in highly inconsistent IF estimates due to the high variance in sample configurations. To address this, we propose two advanced sampling techniques based on features and logits. These samplers select a small yet representative subset of the entire dataset by considering the stochastic distribution of features or logits, thereby enhancing the accuracy of IF estimations. We validate our approach through class removal experiments, a typical application of IFs, using the F1-score to measure how effectively the model forgets the removed class while maintaining inference consistency on the remaining classes. Our method reduces computation time by 30.1% and memory usage by 42.2%, or improves the F1-score by 2.5% compared to the baseline.

---

### STaMP: Sequence Transformation and Mixed Precision for Low-Precision Activation Quantization

Authors: Marco Federici, Riccardo Del Chiaro, Boris van Breugel, Paul Whatmough, Markus Nagel
Published: 2025-10-30
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.26771v1

Quantization is the key method for reducing inference latency, power and memory footprint of generative AI models. However, accuracy often degrades sharply when activations are quantized below eight bits. Recent work suggests that invertible linear transformations (e.g. rotations) can aid quantization, by reparameterizing feature channels and weights. In this paper, we propose \textit{Sequence Transformation and Mixed Precision} (STaMP) quantization, a novel strategy that applies linear transformations along the \textit{sequence} dimension to exploit the strong local correlation in language and visual data. By keeping a small number of tokens in each intermediate activation at higher precision, we can maintain model accuracy at lower (average) activations bit-widths. We evaluate STaMP on recent LVM and LLM architectures, demonstrating that it significantly improves low bit width activation quantization and complements established activation and weight quantization methods including recent feature transformations.

---

### SteerVLM: Robust Model Control through Lightweight Activation Steering for Vision Language Models

Authors: Anushka Sivakumar, Andrew Zhang, Zaber Hakim, Chris Thomas
Published: 2025-10-30
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2510.26769v1

This work introduces SteerVLM, a lightweight steering module designed to guide Vision-Language Models (VLMs) towards outputs that better adhere to desired instructions. Our approach learns from the latent embeddings of paired prompts encoding target and converse behaviors to dynamically adjust activations connecting the language modality with image context. This allows for fine-grained, inference-time control over complex output semantics without modifying model weights while preserving performance on off-target tasks. Our steering module requires learning parameters equal to 0.14% of the original VLM's size. Our steering module gains model control through dimension-wise activation modulation and adaptive steering across layers without requiring pre-extracted static vectors or manual tuning of intervention points. Furthermore, we introduce VNIA (Visual Narrative Intent Alignment), a multimodal dataset specifically created to facilitate the development and evaluation of VLM steering techniques. Our method outperforms existing intervention techniques on steering and hallucination mitigation benchmarks for VLMs and proposes a robust solution for multimodal model control through activation engineering.

---

### AMO-Bench: Large Language Models Still Struggle in High School Math Competitions

Authors: Shengnan An, Xunliang Cai, Xuezhi Cao, Xiaoyu Li, Yehao Lin, Junlin Liu, Xinxuan Lv, Dan Ma, Xuanlin Wang, Ziwen Wang, Shuang Zhou
Published: 2025-10-30
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.26768v1

We present AMO-Bench, an Advanced Mathematical reasoning benchmark with Olympiad level or even higher difficulty, comprising 50 human-crafted problems. Existing benchmarks have widely leveraged high school math competitions for evaluating mathematical reasoning capabilities of large language models (LLMs). However, many existing math competitions are becoming less effective for assessing top-tier LLMs due to performance saturation (e.g., AIME24/25). To address this, AMO-Bench introduces more rigorous challenges by ensuring all 50 problems are (1) cross-validated by experts to meet at least the International Mathematical Olympiad (IMO) difficulty standards, and (2) entirely original problems to prevent potential performance leakages from data memorization. Moreover, each problem in AMO-Bench requires only a final answer rather than a proof, enabling automatic and robust grading for evaluation. Experimental results across 26 LLMs on AMO-Bench show that even the best-performing model achieves only 52.4% accuracy on AMO-Bench, with most LLMs scoring below 40%. Beyond these poor performances, our further analysis reveals a promising scaling trend with increasing test-time compute on AMO-Bench. These results highlight the significant room for improving the mathematical reasoning in current LLMs. We release AMO-Bench to facilitate further research into advancing the reasoning abilities of language models. https://amo-bench.github.io/

---

### MORE: Multi-Organ Medical Image REconstruction Dataset

Authors: Shaokai Wu, Yapan Guo, Yanbiao Ji, Jing Tong, Yuxiang Lu, Mei Li, Suizhi Huang, Yue Ding, Hongtao Lu
Published: 2025-10-30
Categories: eess.IV, cs.CV, cs.MM
arXiv: https://arxiv.org/abs/2510.26759v1

CT reconstruction provides radiologists with images for diagnosis and treatment, yet current deep learning methods are typically limited to specific anatomies and datasets, hindering generalization ability to unseen anatomies and lesions. To address this, we introduce the Multi-Organ medical image REconstruction (MORE) dataset, comprising CT scans across 9 diverse anatomies with 15 lesion types. This dataset serves two key purposes: (1) enabling robust training of deep learning models on extensive, heterogeneous data, and (2) facilitating rigorous evaluation of model generalization for CT reconstruction. We further establish a strong baseline solution that outperforms prior approaches under these challenging conditions. Our results demonstrate that: (1) a comprehensive dataset helps improve the generalization capability of models, and (2) optimization-based methods offer enhanced robustness for unseen anatomies. The MORE dataset is freely accessible under CC-BY-NC 4.0 at our project page https://more-med.github.io/

---

### The Oversight Game: Learning to Cooperatively Balance an AI Agent's Safety and Autonomy

Authors: William Overman, Mohsen Bayati
Published: 2025-10-30
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.26752v2

As increasingly capable agents are deployed, a central safety challenge is how to retain meaningful human control without modifying the underlying system. We study a minimal control interface in which an agent chooses whether to act autonomously (play) or defer (ask), while a human simultaneously chooses whether to be permissive (trust) or engage in oversight (oversee), and model this interaction as a two-player Markov game. When this game forms a Markov Potential Game, we prove an alignment guarantee: any increase in the agent's utility from acting more autonomously cannot decrease the human's value. This establishes a form of intrinsic alignment where the agent's incentive to seek autonomy is structurally coupled to the human's welfare. Practically, the framework induces a transparent control layer that encourages the agent to defer when risky and act when safe. While we use gridworld simulations to illustrate the emergence of this collaboration, our primary validation involves an agentic tool-use task in which two 30B parameter language models are fine-tuned via independent policy gradient. We demonstrate that even as the agents learn to coordinate on the fly, this framework effectively reduces safety violations in realistic, open-ended environments.

---

### Deep sequence models tend to memorize geometrically; it is unclear why

Authors: Shahriar Noroozizadeh, Vaishnavh Nagarajan, Elan Rosenfeld, Sanjiv Kumar
Published: 2025-10-30
Categories: cs.LG, cs.AI, cs.CL, stat.ML
arXiv: https://arxiv.org/abs/2510.26745v3

Deep sequence models are said to store atomic facts predominantly in the form of associative memory: a brute-force lookup of co-occurring entities. We identify a dramatically different form of storage of atomic facts that we term as geometric memory. Here, the model has synthesized embeddings encoding novel global relationships between all entities, including ones that do not co-occur in training. Such storage is powerful: for instance, we show how it transforms a hard reasoning task involving an $\ell$-fold composition into an easy-to-learn $1$-step navigation task. From this phenomenon, we extract fundamental aspects of neural embedding geometries that are hard to explain. We argue that the rise of such a geometry, as against a lookup of local associations, cannot be straightforwardly attributed to typical supervisory, architectural, or optimizational pressures. Counterintuitively, a geometry is learned even when it is more complex than the brute-force lookup. Then, by analyzing a connection to Node2Vec, we demonstrate how the geometry stems from a spectral bias that -- in contrast to prevailing theories -- indeed arises naturally despite the lack of various pressures. This analysis also points out to practitioners a visible headroom to make Transformer memory more strongly geometric. We hope the geometric view of parametric memory encourages revisiting the default intuitions that guide researchers in areas like knowledge acquisition, capacity, discovery, and unlearning.

---

### A General Incentives-Based Framework for Fairness in Multi-agent Resource Allocation

Authors: Ashwin Kumar, William Yeoh
Published: 2025-10-30
Categories: cs.MA, cs.AI
arXiv: https://arxiv.org/abs/2510.26740v1

We introduce the General Incentives-based Framework for Fairness (GIFF), a novel approach for fair multi-agent resource allocation that infers fair decision-making from standard value functions. In resource-constrained settings, agents optimizing for efficiency often create inequitable outcomes. Our approach leverages the action-value (Q-)function to balance efficiency and fairness without requiring additional training. Specifically, our method computes a local fairness gain for each action and introduces a counterfactual advantage correction term to discourage over-allocation to already well-off agents. This approach is formalized within a centralized control setting, where an arbitrator uses the GIFF-modified Q-values to solve an allocation problem. Empirical evaluations across diverse domains, including dynamic ridesharing, homelessness prevention, and a complex job allocation task-demonstrate that our framework consistently outperforms strong baselines and can discover far-sighted, equitable policies. The framework's effectiveness is supported by a theoretical foundation; we prove its fairness surrogate is a principled lower bound on the true fairness improvement and that its trade-off parameter offers monotonic tuning. Our findings establish GIFF as a robust and principled framework for leveraging standard reinforcement learning components to achieve more equitable outcomes in complex multi-agent systems.

---

### Cross-Platform Evaluation of Reasoning Capabilities in Foundation Models

Authors: J. de Curtò, I. de Zarzà, Pablo García, Jordi Cabot
Published: 2025-10-30
Categories: cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2510.26732v1

This paper presents a comprehensive cross-platform evaluation of reasoning capabilities in contemporary foundation models, establishing an infrastructure-agnostic benchmark across three computational paradigms: HPC supercomputing (MareNostrum 5), cloud platforms (Nebius AI Studio), and university clusters (a node with eight H200 GPUs). We evaluate 15 foundation models across 79 problems spanning eight academic domains (Physics, Mathematics, Chemistry, Economics, Biology, Statistics, Calculus, and Optimization) through three experimental phases: (1) Baseline establishment: Six models (Mixtral-8x7B, Phi-3, LLaMA 3.1-8B, Gemma-2-9b, Mistral-7B, OLMo-7B) evaluated on 19 problems using MareNostrum 5, establishing methodology and reference performance; (2) Infrastructure validation: The 19-problem benchmark repeated on university cluster (seven models including Falcon-Mamba state-space architecture) and Nebius AI Studio (nine state-of-the-art models: Hermes-4 70B/405B, LLaMA 3.1-405B/3.3-70B, Qwen3 30B/235B, DeepSeek-R1, GPT-OSS 20B/120B) to confirm infrastructure-agnostic reproducibility; (3) Extended evaluation: Full 79-problem assessment on both university cluster and Nebius platforms, probing generalization at scale across architectural diversity. The findings challenge conventional scaling assumptions, establish training data quality as more critical than model size, and provide actionable guidelines for model selection across educational, production, and research contexts. The tri-infrastructure methodology and 79-problem benchmark enable longitudinal tracking of reasoning capabilities as foundation models evolve.

---

### ExpertFlow: Adaptive Expert Scheduling and Memory Coordination for Efficient MoE Inference

Authors: Zixu Shen, Kexin Chu, Yifan Zhang, Dawei Xiang, Runxin Wu, Wei Zhang
Published: 2025-10-30
Categories: cs.DC, cs.AI, cs.PF
arXiv: https://arxiv.org/abs/2510.26730v1

The expansion of large language models is increasingly limited by the constrained memory capacity of modern GPUs. To mitigate this, Mixture-of-Experts (MoE) architectures activate only a small portion of parameters during inference, significantly lowering both memory demand and computational overhead. However, conventional MoE inference approaches, which select active experts independently at each layer, often introduce considerable latency because of frequent parameter transfers between host and GPU memory. In addition, current cross-layer prediction strategies, which are typically based on fixed steps, lack adaptability across different hardware platforms and workloads, thereby reducing their robustness and effectiveness. To address these challenges, we present ExpertFlow, a runtime system for MoE inference that combines adaptive expert prefetching and cache-aware routing. ExpertFlow continuously adjusts its prediction horizon for expert activation by leveraging runtime statistics such as transfer bandwidth, parameter dimensionality, and model feedback signals. Furthermore, it incorporates a hybrid cross-layer prediction scheme that fuses pregating information with intermediate computational states to anticipate future expert needs. By adaptively refining prefetching decisions and aligning them with actual usage behavior, ExpertFlow effectively decreases cache misses and removes latency caused by expert swap-ins. Our evaluation demonstrates that ExpertFlow reduces model stall time to less than 0.1% of the baseline, highlighting its capability to optimize MoE inference under stringent memory constraints.

---

### Bridging the Gap between Empirical Welfare Maximization and Conditional Average Treatment Effect Estimation in Policy Learning

Authors: Masahiro Kato
Published: 2025-10-30
Categories: stat.ML, cs.LG, econ.EM, math.ST, stat.ME
arXiv: https://arxiv.org/abs/2510.26723v2

The goal of policy learning is to train a policy function that recommends a treatment given covariates to maximize population welfare. There are two major approaches in policy learning: the empirical welfare maximization (EWM) approach and the plug-in approach. The EWM approach is analogous to a classification problem, where one first builds an estimator of the population welfare, which is a functional of policy functions, and then trains a policy by maximizing the estimated welfare. In contrast, the plug-in approach is based on regression, where one first estimates the conditional average treatment effect (CATE) and then recommends the treatment with the highest estimated outcome. This study bridges the gap between the two approaches by showing that both are based on essentially the same optimization problem. In particular, we prove an exact equivalence between EWM and least squares over a reparameterization of the policy class. As a consequence, the two approaches are interchangeable in several respects and share the same theoretical guarantees under common conditions. Leveraging this equivalence, we propose a regularization method for policy learning. The reduction to least squares yields a smooth surrogate that is typically easier to optimize in practice. At the same time, for many natural policy classes the inherent combinatorial hardness of exact EWM generally remains, so the reduction should be viewed as an optimization aid rather than a universal bypass of NP-hardness.

---

### Non-Convex Over-the-Air Heterogeneous Federated Learning: A Bias-Variance Trade-off

Authors: Muhammad Faraz Ul Abrar, Nicolò Michelusi
Published: 2025-10-30
Categories: cs.LG, cs.AI, cs.DC, eess.SP, eess.SY
arXiv: https://arxiv.org/abs/2510.26722v4

Over-the-air (OTA) federated learning (FL) has been well recognized as a scalable paradigm that exploits the waveform superposition of the wireless multiple-access channel to aggregate model updates in a single use. Existing OTA-FL designs largely enforce zero-bias model updates by either assuming \emph{homogeneous} wireless conditions (equal path loss across devices) or forcing zero-bias updates to guarantee convergence. Under \emph{heterogeneous} wireless scenarios, however, such designs are constrained by the weakest device and inflate the update variance. Moreover, prior analyses of biased OTA-FL largely address convex objectives, while most modern AI models are highly non-convex. Motivated by these gaps, we study OTA-FL with stochastic gradient descent (SGD) for general smooth non-convex objectives under wireless heterogeneity. We develop novel OTA-FL SGD updates that allow a structured, time-invariant model bias while facilitating reduced variance updates. We derive a finite-time stationarity bound (expected time average squared gradient norm) that explicitly reveals a bias-variance trade-off. To optimize this trade-off, we pose a non-convex joint OTA power-control design and develop an efficient successive convex approximation (SCA) algorithm that requires only statistical CSI at the base station. Experiments on a non-convex image classification task validate the approach: the SCA-based design accelerates convergence via an optimized bias and improves generalization over prior OTA-FL baselines.

---

### MaLoRA: Gated Modality LoRA for Key-Space Alignment in Multimodal LLM Fine-Tuning

Authors: Xinhan Zheng, Huyu Wu, Xueting Wang, Duo Su, Haiyun Jiang
Published: 2025-10-30
Categories: cs.AI, cs.MM
arXiv: https://arxiv.org/abs/2510.26721v2

Multimodal large language models (MLLMs) exhibit a pronounced preference for textual inputs when processing vision-language data, limiting their ability to reason effectively from visual evidence. Unlike prior studies that attribute this text bias to external factors such as data imbalance or instruction tuning, we propose that the bias originates from the model's internal architecture. Specifically, we hypothesize that visual key vectors (Visual Keys) are out-of-distribution (OOD) relative to the text key space learned during language-only pretraining. Consequently, these visual keys receive systematically lower similarity scores during attention computation, leading to their under-utilization in the context representation. To validate this hypothesis, we extract key vectors from LLaVA and Qwen2.5-VL and analyze their distributional structures using qualitative (t-SNE) and quantitative (Jensen-Shannon divergence) methods. The results provide direct evidence that visual and textual keys occupy markedly distinct subspaces within the attention space. The inter-modal divergence is statistically significant, exceeding intra-modal variation by several orders of magnitude. These findings reveal that text bias arises from an intrinsic misalignment within the attention key space rather than solely from external data factors.

---

### Reducing base drag on road vehicles using pulsed jets optimized by hybrid genetic algorithms

Authors: Isaac Robledo, Juan Alfaro, Víctor Duro, Alberto Solera-Rico, Rodrigo Castellanos, Carlos Sanmiguel Vila
Published: 2025-10-30
Categories: physics.flu-dyn, cs.NE, math.OC
arXiv: https://arxiv.org/abs/2510.26718v2

Aerodynamic drag on flat-backed vehicles like vans and trucks is dominated by a low-pressure wake, whose control is critical for reducing fuel consumption. This paper presents an experimental study at $Re_W\approx 78,300$ on active flow control using four pulsed jets at the rear edges of a bluff body model. A hybrid genetic algorithm, combining a global search with a local gradient-based optimizer, was used to determine the best-performing jet actuation parameters in an experiment-in-the-loop setup. The cost function was designed to achieve a net energy saving by simultaneously minimizing aerodynamic drag and penalizing the actuation's energy consumption. The optimization campaign successfully identified a control strategy that yields a drag reduction of approximately 8.8%. The best-performing control law features a strong, low-frequency actuation from the bottom jet, which targets the main vortex shedding, while the top and lateral jets address higher-frequency, less energetic phenomena. Particle Image Velocimetry analysis reveals a significant upward shift and stabilization of the wake, leading to substantial pressure recovery on the model's lower base. Ultimately, this work demonstrates that a model-free optimization approach can successfully identify non-intuitive, multi-faceted actuation strategies that yield significant and energetically efficient drag reduction.

---