# Research Papers — 2025-10 · bundle page 108 of 217

> Full-text markdown of papers #5351–5400 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/108
[← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/107.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-10/page/109.md)

### Unlocking Out-of-Distribution Generalization in Transformers via Recursive Latent Space Reasoning

Authors: Awni Altabaa, Siyu Chen, John Lafferty, Zhuoran Yang
Published: 2025-10-15
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.14095v1

Systematic, compositional generalization beyond the training distribution remains a core challenge in machine learning -- and a critical bottleneck for the emergent reasoning abilities of modern language models. This work investigates out-of-distribution (OOD) generalization in Transformer networks using a GSM8K-style modular arithmetic on computational graphs task as a testbed. We introduce and explore a set of four architectural mechanisms aimed at enhancing OOD generalization: (i) input-adaptive recurrence; (ii) algorithmic supervision; (iii) anchored latent representations via a discrete bottleneck; and (iv) an explicit error-correction mechanism. Collectively, these mechanisms yield an architectural approach for native and scalable latent space reasoning in Transformer networks with robust algorithmic generalization capabilities. We complement these empirical results with a detailed mechanistic interpretability analysis that reveals how these mechanisms give rise to robust OOD generalization abilities.

---

### Neural Network approximation power on homogeneous and heterogeneous reaction-diffusion equations

Authors: Haotian Feng
Published: 2025-10-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.14094v1

Reaction-diffusion systems represent one of the most fundamental formulations used to describe a wide range of physical, chemical, and biological processes. With the increasing adoption of neural networks, recent research has focused on solving differential equations using machine learning techniques. However, the theoretical foundation explaining why neural networks can effectively approximate such solutions remains insufficiently explored. This paper provides a theoretical analysis of the approximation power of neural networks for one- and two-dimensional reaction-diffusion equations in both homogeneous and heterogeneous media. Building upon the universal approximation theorem, we demonstrate that a two-layer neural network can approximate the one-dimensional reaction-diffusion equation, while a three-layer neural network can approximate its two-dimensional counterpart. The theoretical framework presented here can be further extended to elliptic and parabolic equations. Overall, this work highlights the expressive power of neural networks in approximating solutions to reaction-diffusion equations and related PDEs, providing a theoretical foundation for neural network-based differential equation solvers.

---

### deFOREST: Fusing Optical and Radar satellite data for Enhanced Sensing of Tree-loss

Authors: Julio Enrique Castrillon-Candas, Hanfeng Gu, Caleb Meredith, Yulin Li, Xiaojing Tang, Pontus Olofsson, Mark Kon
Published: 2025-10-15
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2510.14092v2

In this paper we develop a deforestation detection pipeline that incorporates optical and Synthetic Aperture Radar (SAR) data. A crucial component of the pipeline is the construction of anomaly maps of the optical data, which is done using the residual space of a discrete Karhunen-Loéve (KL) expansion. Anomalies are quantified using a concentration bound on the distribution of the residual components for the nominal state of the forest. This bound does not require prior knowledge on the distribution of the data. This is in contrast to statistical parametric methods that assume knowledge of the data distribution, an impractical assumption that is especially infeasible for high dimensional data such as ours. Once the optical anomaly maps are computed they are combined with SAR data, and the state of the forest is classified by using a Hidden Markov Model (HMM). We test our approach with Sentinel-1 (SAR) and Sentinel-2 (Optical) data on a $92\,km \times 92\,km$ region in the Amazon forest. The results show that both the hybrid optical-radar and optical only methods achieve high accuracy that is superior to the recent state-of-the-art hybrid method. Moreover, the hybrid method is significantly more robust in the case of sparse optical data that are common in highly cloudy regions.

---

### Every Language Model Has a Forgery-Resistant Signature

Authors: Matthew Finlayson, Xiang Ren, Swabha Swayamdipta
Published: 2025-10-15
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2510.14086v2

The ubiquity of closed-weight language models with public-facing APIs has generated interest in forensic methods, both for extracting hidden model details (e.g., parameters) and for identifying models by their outputs. One successful approach to these goals has been to exploit the geometric constraints imposed by the language model architecture and parameters. In this work, we show that a lesser-known geometric constraint -- namely, that language model outputs lie on the surface of a high-dimensional ellipse -- functions as a signature for the model and can be used to identify the source model of a given output. This ellipse signature has unique properties that distinguish it from existing model-output association methods like language model fingerprints. In particular, the signature is hard to forge: without direct access to model parameters, it is practically infeasible to produce log-probabilities (logprobs) on the ellipse using currently known methods. Secondly, the signature is naturally occurring, since all language models have these elliptical constraints. Thirdly, the signature is self-contained, in that it is detectable without access to the model inputs or the full weights. Finally, the signature is compact and redundant, as it is independently detectable in each logprob output from the model. We evaluate a novel technique for extracting the ellipse from small models and discuss the practical hurdles that make it infeasible for production-scale models. Finally, we use ellipse signatures to propose a protocol for language model output verification, analogous to cryptographic symmetric-key message authentication systems.

---

### Capture, Canonicalize, Splat: Zero-Shot 3D Gaussian Avatars from Unstructured Phone Images

Authors: Emanuel Garbin, Guy Adam, Oded Krams, Zohar Barzelay, Eran Guendelman, Michael Schwarz, Matteo Presutto, Moran Vatelmacher, Yigal Shenkman, Eli Peker, Itai Druker, Uri Patish et al.
Published: 2025-10-15
Categories: cs.CV, cs.GR
arXiv: https://arxiv.org/abs/2510.14081v3

We present a novel, zero-shot pipeline for creating hyperrealistic, identity-preserving 3D avatars from a few unstructured phone images. Existing methods face several challenges: single-view approaches suffer from geometric inconsistencies and hallucinations, degrading identity preservation, while models trained on synthetic data fail to capture high-frequency details like skin wrinkles and fine hair, limiting realism. Our method introduces two key contributions: (1) a generative canonicalization module that processes multiple unstructured views into a standardized, consistent representation, and (2) a transformer-based model trained on a new, large-scale dataset of high-fidelity Gaussian splatting avatars derived from dome captures of real people. This "Capture, Canonicalize, Splat" pipeline produces static quarter-body avatars with compelling realism and robust identity preservation from unstructured photos.

---

### ERGO: Entropy-guided Resetting for Generation Optimization in Multi-turn Language Models

Authors: Haziq Mohammad Khalid, Athikash Jeyaganthan, Timothy Do, Yicheng Fu, Sean O'Brien, Vasu Sharma, Kevin Zhu
Published: 2025-10-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.14077v1

Large Language Models (LLMs) suffer significant performance degradation in multi-turn conversations when information is presented incrementally. Given that multi-turn conversations characterize everyday interactions with LLMs, this degradation poses a severe challenge to real world usability. We hypothesize that abrupt increases in model uncertainty signal misalignment in multi-turn LLM interactions, and we exploit this insight to dynamically realign conversational context. We introduce ERGO (Entropy-guided Resetting for Generation Optimization), which continuously quantifies internal uncertainty via Shannon entropy over next token distributions and triggers adaptive prompt consolidation when a sharp spike in entropy is detected. By treating uncertainty as a first class signal rather than a nuisance to eliminate, ERGO embraces variability in language and modeling, representing and responding to uncertainty. In multi-turn tasks with incrementally revealed instructions, ERGO yields a 56.6% average performance gain over standard baselines, increases aptitude (peak performance capability) by 24.7%, and decreases unreliability (variability in performance) by 35.3%, demonstrating that uncertainty aware interventions can improve both accuracy and reliability in conversational AI.

---

### DiffOPF: Diffusion Solver for Optimal Power Flow

Authors: Milad Hoseinpour, Vladimir Dvorkin
Published: 2025-10-15
Categories: eess.SY, cs.AI, stat.CO, stat.ML
arXiv: https://arxiv.org/abs/2510.14075v2

The optimal power flow (OPF) is a multi-valued, non-convex mapping from loads to dispatch setpoints. The variability of system parameters (e.g., admittances, topology) further contributes to the multiplicity of dispatch setpoints for a given load. Existing deep learning OPF solvers are single-valued and thus fail to capture the variability of system parameters unless fully represented in the feature space, which is prohibitive. To solve this problem, we introduce a diffusion-based OPF solver, termed \textit{DiffOPF}, that treats OPF as a conditional sampling problem. The solver learns the joint distribution of loads and dispatch setpoints from operational history, and returns the marginal dispatch distributions conditioned on loads. Unlike single-valued solvers, DiffOPF enables sampling statistically credible warm starts with favorable cost and constraint satisfaction trade-offs. We explore the sample complexity of DiffOPF to ensure the OPF solution within a prescribed distance from the optimization-based solution, and verify this experimentally on power system benchmarks.

---

### Exact Dynamics of Multi-class Stochastic Gradient Descent

Authors: Elizabeth Collins-Woodfin, Inbar Seroussi
Published: 2025-10-15
Categories: stat.ML, cs.LG, math.OC, math.PR
arXiv: https://arxiv.org/abs/2510.14074v2

We develop a framework for analyzing the learning dynamics of high-dimensional problems trained using one-pass stochastic gradient descent (SGD) with data from multiple anisotropic classes. Our main theorem provides exact expressions for quantities of interest, including the risk and the overlap with the true signal, in terms of a deterministic system of ODEs, valid in the high-dimensional limit. The theorem holds for a broad class of optimization problems and extends to settings where the number of classes grows with dimension. To illustrate its utility, we investigate in detail the effect of the data's anisotropic structure on the problems of binary logistic regression and least-squares (LS) loss. We study the LS in a linear multiclass setup and derive a learning-rate threshold that depends on the average eigenvalue of the covariance matrices. In the binary logistic regression, we study three cases: isotropic covariances, data covariance matrices with a large fraction of zero eigenvalues (denoted as the zero-one model), and covariance matrices with power-law spectra. We show that a structural phase transition occurs. In particular, for the zero-one model and the power-law model with sufficiently large power, SGD aligns more closely with values of the class mean that are projected onto the ``clean directions'' (i.e., directions of smaller variance). This is supported by analytical studies and numerical simulations, which show the exact asymptotic behavior of the loss in the high-dimensional limit. The effects of data anisotropy that we demonstrate are likely to hold beyond these examples and illustrate one application of the broader theorem that we prove.

---

### Exploratory Causal Inference in SAEnce

Authors: Tommaso Mencattini, Riccardo Cadei, Francesco Locatello
Published: 2025-10-15
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.14073v2

Randomized Controlled Trials are one of the pillars of science; nevertheless, they rely on hand-crafted hypotheses and expensive analysis. Such constraints prevent causal effect estimation at scale, potentially anchoring on popular yet incomplete hypotheses. We propose to discover the unknown effects of a treatment directly from data. For this, we turn unstructured data from a trial into meaningful representations via pretrained foundation models and interpret them via a sparse autoencoder. However, discovering significant causal effects at the neural level is not trivial due to multiple-testing issues and effects entanglement. To address these challenges, we introduce Neural Effect Search, a novel recursive procedure solving both issues by progressive stratification. After assessing the robustness of our algorithm on semi-synthetic experiments, we showcase, in the context of experimental ecology, the first successful unsupervised causal effect identification on a real-world scientific trial.

---

### On the expressivity of sparse maxout networks

Authors: Moritz Grillo, Tobias Hofmann
Published: 2025-10-15
Categories: cs.LG, cs.AI, math.CO
arXiv: https://arxiv.org/abs/2510.14068v1

We study the expressivity of sparse maxout networks, where each neuron takes a fixed number of inputs from the previous layer and employs a, possibly multi-argument, maxout activation. This setting captures key characteristics of convolutional or graph neural networks. We establish a duality between functions computable by such networks and a class of virtual polytopes, linking their geometry to questions of network expressivity. In particular, we derive a tight bound on the dimension of the associated polytopes, which serves as the central tool for our analysis. Building on this, we construct a sequence of depth hierarchies. While sufficiently deep sparse maxout networks are universal, we prove that if the required depth is not reached, width alone cannot compensate for the sparsity of a fixed indegree constraint.

---

### Optical Computation-in-Communication enables low-latency, high-fidelity perception in telesurgery

Authors: Rui Yang, Jiaming Hu, Jian-Qing Zheng, Yue-Zhen Lu, Jian-Wei Cui, Qun Ren, Yi-Jie Yu, John Edward Wu, Zhao-Yu Wang, Xiao-Li Lin, Dandan Zhang, Mingchu Tang et al.
Published: 2025-10-15
Categories: physics.optics, cs.AI, eess.IV
arXiv: https://arxiv.org/abs/2510.14058v1

Artificial intelligence (AI) holds significant promise for enhancing intraoperative perception and decision-making in telesurgery, where physical separation impairs sensory feedback and control. Despite advances in medical AI and surgical robotics, conventional electronic AI architectures remain fundamentally constrained by the compounded latency from serial processing of inference and communication. This limitation is especially critical in latency-sensitive procedures such as endovascular interventions, where delays over 200 ms can compromise real-time AI reliability and patient safety. Here, we introduce an Optical Computation-in-Communication (OCiC) framework that reduces end-to-end latency significantly by performing AI inference concurrently with optical communication. OCiC integrates Optical Remote Computing Units (ORCUs) directly into the optical communication pathway, with each ORCU experimentally achieving up to 69 tera-operations per second per channel through spectrally efficient two-dimensional photonic convolution. The system maintains ultrahigh inference fidelity within 0.1% of CPU/GPU baselines on classification and coronary angiography segmentation, while intrinsically mitigating cumulative error propagation, a longstanding barrier to deep optical network scalability. We validated the robustness of OCiC through outdoor dark fibre deployments, confirming consistent and stable performance across varying environmental conditions. When scaled globally, OCiC transforms long-haul fibre infrastructure into a distributed photonic AI fabric with exascale potential, enabling reliable, low-latency telesurgery across distances up to 10,000 km and opening a new optical frontier for distributed medical intelligence.

---

### FedHFT: Efficient Federated Finetuning with Heterogeneous Edge Clients

Authors: Fatih Ilhan, Selim Furkan Tekin, Tiansheng Huang, Gaowen Liu, Ramana Kompella, Greg Eisenhauer, Yingyan Celine Lin, Calton Pu, Ling Liu
Published: 2025-10-15
Categories: cs.LG, cs.DC
arXiv: https://arxiv.org/abs/2510.14054v1

Fine-tuning pre-trained large language models (LLMs) has become a common practice for personalized natural language understanding (NLU) applications on downstream tasks and domain-specific datasets. However, there are two main challenges: (i) limited and/or heterogeneous data for fine-tuning due to proprietary data confidentiality or privacy requirements, and (ii) varying computation resources available across participating clients such as edge devices. This paper presents FedHFT - an efficient and personalized federated fine-tuning framework to address both challenges. First, we introduce a mixture of masked adapters to handle resource heterogeneity across participating clients, enabling high-performance collaborative fine-tuning of pre-trained language model(s) across multiple clients in a distributed setting, while keeping proprietary data local. Second, we introduce a bi-level optimization approach to handle non-iid data distribution based on masked personalization and client clustering. Extensive experiments demonstrate significant performance and efficiency improvements over various natural language understanding tasks under data and resource heterogeneity compared to representative heterogeneous federated learning methods.

---

### Position: Require Frontier AI Labs To Release Small "Analog" Models

Authors: Shriyash Upadhyay, Chaithanya Bandi, Narmeen Oozeer, Philip Quirke
Published: 2025-10-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.14053v1

Recent proposals for regulating frontier AI models have sparked concerns about the cost of safety regulation, and most such regulations have been shelved due to the safety-innovation tradeoff. This paper argues for an alternative regulatory approach that ensures AI safety while actively promoting innovation: mandating that large AI laboratories release small, openly accessible analog models (scaled-down versions) trained similarly to and distilled from their largest proprietary models. Analog models serve as public proxies, allowing broad participation in safety verification, interpretability research, and algorithmic transparency without forcing labs to disclose their full-scale models. Recent research demonstrates that safety and interpretability methods developed using these smaller models generalize effectively to frontier-scale systems. By enabling the wider research community to directly investigate and innovate upon accessible analogs, our policy substantially reduces the regulatory burden and accelerates safety advancements. This mandate promises minimal additional costs, leveraging reusable resources like data and infrastructure, while significantly contributing to the public good. Our hope is not only that this policy be adopted, but that it illustrates a broader principle supporting fundamental research in machine learning: deeper understanding of models relaxes the safety-innovation tradeoff and lets us have more of both.

---

### Synchronization of Multiple Videos

Authors: Avihai Naaman, Ron Shapira Weber, Oren Freifeld
Published: 2025-10-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.14051v1

Synchronizing videos captured simultaneously from multiple cameras in the same scene is often easy and typically requires only simple time shifts. However, synchronizing videos from different scenes or, more recently, generative AI videos, poses a far more complex challenge due to diverse subjects, backgrounds, and nonlinear temporal misalignment. We propose Temporal Prototype Learning (TPL), a prototype-based framework that constructs a shared, compact 1D representation from high-dimensional embeddings extracted by any of various pretrained models. TPL robustly aligns videos by learning a unified prototype sequence that anchors key action phases, thereby avoiding exhaustive pairwise matching. Our experiments show that TPL improves synchronization accuracy, efficiency, and robustness across diverse datasets, including fine-grained frame retrieval and phase classification tasks. Importantly, TPL is the first approach to mitigate synchronization issues in multiple generative AI videos depicting the same action. Our code and a new multiple video synchronization dataset are available at https://bgu-cs-vil.github.io/TPL/

---

### CausalVerse: Benchmarking Causal Representation Learning with Configurable High-Fidelity Simulations

Authors: Guangyi Chen, Yunlong Deng, Peiyuan Zhu, Yan Li, Yifan Shen, Zijian Li, Kun Zhang
Published: 2025-10-15
Categories: cs.LG, cs.MS
arXiv: https://arxiv.org/abs/2510.14049v2

Causal Representation Learning (CRL) aims to uncover the data-generating process and identify the underlying causal variables and relations, whose evaluation remains inherently challenging due to the requirement of known ground-truth causal variables and causal structure. Existing evaluations often rely on either simplistic synthetic datasets or downstream performance on real-world tasks, generally suffering a dilemma between realism and evaluative precision. In this paper, we introduce a new benchmark for CRL using high-fidelity simulated visual data that retains both realistic visual complexity and, more importantly, access to ground-truth causal generating processes. The dataset comprises around 200 thousand images and 3 million video frames across 24 sub-scenes in four domains: static image generation, dynamic physical simulations, robotic manipulations, and traffic situation analysis. These scenarios range from static to dynamic settings, simple to complex structures, and single to multi-agent interactions, offering a comprehensive testbed that hopefully bridges the gap between rigorous evaluation and real-world applicability. In addition, we provide flexible access to the underlying causal structures, allowing users to modify or configure them to align with the required assumptions in CRL, such as available domain labels, temporal dependencies, or intervention histories. Leveraging this benchmark, we evaluated representative CRL methods across diverse paradigms and offered empirical insights to assist practitioners and newcomers in choosing or extending appropriate CRL frameworks to properly address specific types of real problems that can benefit from the CRL perspective. Welcome to visit our: Project page:https://causal-verse.github.io/, Dataset:https://huggingface.co/CausalVerse.

---

### Cyber-Resilient System Identification for Power Grid through Bayesian Integration

Authors: Shimiao Li, Guannan Qu, Bryan Hooi, Vyas Sekar, Soummya Kar, Larry Pileggi
Published: 2025-10-15
Categories: eess.SY, cs.AI, cs.CR
arXiv: https://arxiv.org/abs/2510.14043v1

Power grids increasingly need real-time situational awareness under the ever-evolving cyberthreat landscape. Advances in snapshot-based system identification approaches have enabled accurately estimating states and topology from a snapshot of measurement data, under random bad data and topology errors. However, modern interactive, targeted false data can stay undetectable to these methods, and significantly compromise estimation accuracy. This work advances system identification that combines snapshot-based method with time-series model via Bayesian Integration, to advance cyber resiliency against both random and targeted false data. Using a distance-based time-series model, this work can leverage historical data of different distributions induced by changes in grid topology and other settings. The normal system behavior captured from historical data is integrated into system identification through a Bayesian treatment, to make solutions robust to targeted false data. We experiment on mixed random anomalies (bad data, topology error) and targeted false data injection attack (FDIA) to demonstrate our method's 1) cyber resilience: achieving over 70% reduction in estimation error under FDIA; 2) anomalous data identification: being able to alarm and locate anomalous data; 3) almost linear scalability: achieving comparable speed with the snapshot-based baseline, both taking <1min per time tick on the large 2,383-bus system using a laptop CPU.

---

### Quantifying Phonosemantic Iconicity Distributionally in 6 Languages

Authors: George Flint, Kaustubh Kislay
Published: 2025-10-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.14040v2

Language is, as commonly theorized, largely arbitrary. Yet, systematic relationships between phonetics and semantics have been observed in many specific cases. To what degree could those systematic relationships manifest themselves in large scale, quantitative investigations--both in previously identified and unidentified phenomena? This work undertakes a distributional approach to quantifying phonosemantic iconicity at scale across 6 diverse languages (English, Spanish, Hindi, Finnish, Turkish, and Tamil). In each language, we analyze the alignment of morphemes' phonetic and semantic similarity spaces with a suite of statistical measures, and discover an array of interpretable phonosemantic alignments not previously identified in the literature, along with crosslinguistic patterns. We also analyze 5 previously hypothesized phonosemantic alignments, finding support for some such alignments and mixed results for others.

---

### One Bug, Hundreds Behind: LLMs for Large-Scale Bug Discovery

Authors: Qiushi Wu, Yue Xiao, Dhilung Kirat, Kevin Eykholt, Jiyong Jang, Douglas Lee Schales
Published: 2025-10-15
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2510.14036v1

Fixing bugs in large programs is a challenging task that demands substantial time and effort. Once a bug is found, it is reported to the project maintainers, who work with the reporter to fix it and eventually close the issue. However, across the program, there are often similar code segments, which may also contain the bug, but were missed during discovery. Finding and fixing each recurring bug instance individually is labor intensive. Even more concerning, bug reports can inadvertently widen the attack surface as they provide attackers with an exploitable pattern that may be unresolved in other parts of the program. In this paper, we explore these Recurring Pattern Bugs (RPBs) that appear repeatedly across various code segments of a program or even in different programs, stemming from a same root cause, but are unresolved. Our investigation reveals that RPBs are widespread and can significantly compromise the security of software programs. This paper introduces BugStone, a program analysis system empowered by LLVM and a Large Language Model (LLM). The key observation is that many RPBs have one patched instance, which can be leveraged to identify a consistent error pattern, such as a specific API misuse. By examining the entire program for this pattern, it is possible to identify similar sections of code that may be vulnerable. Starting with 135 unique RPBs, BugStone identified more than 22K new potential issues in the Linux kernel. Manual analysis of 400 of these findings confirmed that 246 were valid. We also created a dataset from over 1.9K security bugs reported by 23 recent top-tier conference works. We manually annotate the dataset, identify 80 recurring patterns and 850 corresponding fixes. Even with a cost-efficient model choice, BugStone achieved 92.2% precision and 79.1% pairwise accuracy on the dataset.

---

### GammaZero: Learning To Guide POMDP Belief Space Search With Graph Representations

Authors: Rajesh Mangannavar, Prasad Tadepalli
Published: 2025-10-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.14035v2

We introduce an uncertainty-aware graph representation framework for learning to guide planning in Partially Observable Markov Decision Processes (POMDPs). Unlike existing approaches that require domain or problem size specific neural architectures, GammaZero leverages a unified graph-based belief representation that enables generalization across problem sizes within a domain. Our key insight is that belief states can be systematically transformed into uncertainty-aware graphs where structural patterns learned on small problems transfer to larger instances. We employ a graph neural network with a decoder architecture to learn value functions and policies from expert demonstrations on computationally tractable problems, then apply these learned heuristics to guide Monte Carlo tree search on larger problems. Experimental results on standard POMDP benchmarks demonstrate that GammaZero achieves comparable performance to BetaZero when trained and tested on the same-sized problems, while enabling zero-shot generalization to problems 2-6x larger than those seen during training.

---

### Vgent: Graph-based Retrieval-Reasoning-Augmented Generation For Long Video Understanding

Authors: Xiaoqian Shen, Wenxuan Zhang, Jun Chen, Mohamed Elhoseiny
Published: 2025-10-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.14032v1

Understanding and reasoning over long videos pose significant challenges for large video language models (LVLMs) due to the difficulty in processing intensive video tokens beyond context window and retaining long-term sequential information. Retrieval-Augmented Generation (RAG) has demonstrated effectiveness in processing long context for Large Language Models (LLMs); however, applying RAG to long video faces challenges such as disrupted temporal dependencies and inclusion of irrelevant information that can hinder accurate reasoning. To address these limitations, we propose Vgent, a novel graph-based retrieval-reasoning-augmented generation framework to enhance LVLMs for long video understanding. Our approach introduces two key innovations: (i) It represents videos by structured graphs with semantic relationships across video clips preserved to improve retrieval effectiveness. (ii) It introduces an intermediate reasoning step to mitigate the reasoning limitation of LVLMs, which leverages structured verification to reduce retrieval noise and facilitate the explicit aggregation of relevant information across clips, resulting in more accurate and context-aware responses. We comprehensively evaluate our framework with various open-source LVLMs on three long-video understanding benchmarks. Our approach yielded an overall performance improvement of $3.0\%\sim 5.4\%$ over base models on MLVU, and outperformed state-of-the-art video RAG methods by $8.6\%$. Our code is publicly available at https://xiaoqian-shen.github.io/Vgent.

---

### Think Globally, Group Locally: Evaluating LLMs Using Multi-Lingual Word Grouping Games

Authors: César Guerra-Solano, Zhuochun Li, Xiang Lorraine Li
Published: 2025-10-15
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.14030v1

Large language models (LLMs) can exhibit biases in reasoning capabilities due to linguistic modality, performing better on tasks in one language versus another, even with similar content. Most previous works evaluate this through reasoning tasks where reliance on strategies or knowledge can ensure success, such as in commonsense or math tasks. However, abstract reasoning is vital to reasoning for everyday life, where people apply "out-of-the-box thinking" to identify and use patterns for solutions, without a reliance on formulaic approaches. Comparatively, little work has evaluated linguistic biases in this task type. In this paper, we propose a task inspired by the New York Times Connections: GlobalGroup, that evaluates models in an abstract reasoning task across several languages. We constructed a game benchmark with five linguistic backgrounds -- English, Spanish, Chinese, Hindi, and Arabic -- in both the native language and an English translation for comparison. We also proposed game difficulty measurements to evaluate models on games with similar difficulty, enabling a more controlled comparison, which is particularly important in reasoning evaluations. Through experimentation, we find English modalities largely lead to better performance in this abstract reasoning task, and performance disparities between open- and closed-source models.

---

### Context-Selective State Space Models: Feedback is All You Need

Authors: Riccardo Zattra, Giacomo Baggio, Umberto Casti, Augusto Ferrante, Francesco Ticozzi
Published: 2025-10-15
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.14027v2

Transformers, powered by the attention mechanism, are the backbone of most foundation models, yet they suffer from quadratic complexity and difficulties in dealing with long-range dependencies in the input sequence. Recent work has shown that state space models (SSMs) provide a promising alternative. In this paper, we introduce the COFFEE (COntext From FEEdback) model, a novel time-varying SSM that incorporates state feedback to enable context-dependent selectivity, while still allowing for parallel implementation. This idea allows the model to regulate its dynamics based on the context described by the internal state, which embodies a compact representation of the input history. State feedback allows COFFEE to improve its ability to capture long-range dependencies: on the induction head task, it achieves near-perfect accuracy with two orders of magnitude fewer parameters and training sequences compared to S6 (the SSM of Mamba). On MNIST, COFFEE largely outperforms S6 within the same architecture, reaching 97% accuracy with only 3585 parameters. These results showcase the role of state feedback as a key mechanism for building scalable and efficient sequence models.

---

### NAPPure: Adversarial Purification for Robust Image Classification under Non-Additive Perturbations

Authors: Junjie Nan, Jianing Li, Wei Chen, Mingkun Zhang, Xueqi Cheng
Published: 2025-10-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.14025v2

Adversarial purification has achieved great success in combating adversarial image perturbations, which are usually assumed to be additive. However, non-additive adversarial perturbations such as blur, occlusion, and distortion are also common in the real world. Under such perturbations, existing adversarial purification methods are much less effective since they are designed to fit the additive nature. In this paper, we propose an extended adversarial purification framework named NAPPure, which can further handle non-additive perturbations. Specifically, we first establish the generation process of an adversarial image, and then disentangle the underlying clean image and perturbation parameters through likelihood maximization. Experiments on GTSRB and CIFAR-10 datasets show that NAPPure significantly boosts the robustness of image classification models against non-additive perturbations.

---

### CRaFT: An Explanation-Based Framework for Evaluating Cultural Reasoning in Multilingual Language Models

Authors: Shehenaz Hossain, Haithem Afli
Published: 2025-10-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.14014v1

Correct answers do not necessarily reflect cultural understanding. We introduce CRaFT, an explanation-based multilingual evaluation framework designed to assess how large language models (LLMs) reason across cultural contexts. Rather than scoring outputs solely based on accuracy, CRaFT evaluates model explanations using four interpretable metrics: Cultural Fluency, Deviation, Consistency, and Linguistic Adaptation. We apply the framework to 50 culturally grounded questions from the World Values Survey, translated into Arabic, Bengali, and Spanish, and evaluate three models (GPT, DeepSeek, and FANAR) across over 2,100 answer-explanation pairs. Results reveal significant cross-lingual variation in reasoning: Arabic reduces fluency, Bengali enhances it, and Spanish remains largely stable. While GPT adapts more effectively across languages, it exhibits lower consistency; FANAR shows stable but rigid reasoning. These findings suggest that cultural awareness in LLMs is not intrinsic but emerges through linguistic framing. CRaFT offers a new lens for evaluating cross-cultural reasoning in multilingual settings, providing actionable insights for building culturally adaptive language models.

---

### Noise-Adaptive Layerwise Learning Rates: Accelerating Geometry-Aware Optimization for Deep Neural Network Training

Authors: Jie Hao, Xiaochuan Gong, Jie Xu, Zhengdao Wang, Mingrui Liu
Published: 2025-10-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.14009v2

Geometry-aware optimization algorithms, such as Muon, have achieved remarkable success in training deep neural networks (DNNs). These methods leverage the underlying geometry of DNNs by selecting appropriate norms for different layers and updating parameters via norm-constrained linear minimization oracles (LMOs). However, even within a group of layers associated with the same norm, the local curvature can be heterogeneous across layers and vary dynamically over the course of training. For example, recent work shows that sharpness varies substantially across transformer layers and throughout training, yet standard geometry-aware optimizers impose fixed learning rates to layers within the same group, which may be inefficient for DNN training. In this paper, we introduce a noise-adaptive layerwise learning rate scheme on top of geometry-aware optimization algorithms and substantially accelerate DNN training compared to methods that use fixed learning rates within each group. Our method estimates gradient variance in the dual norm induced by the chosen LMO on the fly, and uses it to assign time-varying noise-adaptive layerwise learning rates within each group. We provide a theoretical analysis showing that our algorithm achieves a sharp convergence rate. Empirical results on transformer architectures such as LLaMA and GPT demonstrate that our approach achieves faster convergence than state-of-the-art optimizers.

---

### Conditional Clifford-Steerable CNNs for PDE Modeling

Authors: Bálint László Szarvas, Maksim Zhdanov
Published: 2025-10-15
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.14007v2

We introduce Conditional Clifford-Steerable CNNs (C-CSCNNs), a unified framework that incorporates equivariance to arbitrary pseudo-Euclidean groups and significantly improves the expressivity of standard CSCNNs. We show that the kernel basis of the standard formulation is incomplete, limiting model capacity. To address this, we augment the kernels with equivariant representations of the input feature field. We derive the equivariance constraint for these input-dependent kernels and show how it can be solved efficiently via implicit parameterization. We empirically validate on multiple PDE forecasting tasks, including fluid dynamics and relativistic electrodynamics, where our method consistently outperforms standard CSCNNs and performs on par with state-of-the-art baselines.

---

### PIShield: Detecting Prompt Injection Attacks via Intrinsic LLM Features

Authors: Wei Zou, Yupei Liu, Yanting Wang, Ying Chen, Neil Gong, Jinyuan Jia
Published: 2025-10-15
Categories: cs.CR, cs.LG
arXiv: https://arxiv.org/abs/2510.14005v3

LLM-integrated applications are vulnerable to prompt injection attacks, where an attacker contaminates the input to inject malicious instructions, causing the LLM to follow the attacker's intent instead of the original user's. Existing prompt injection detection methods often have sub-optimal performance and/or high computational overhead. In this work, we propose PIShield, an effective and efficient detection method based on the observation that instruction-tuned LLMs internally encode distinguishable signals for prompts containing injected instructions. PIShield leverages residual-stream representations and a simple linear classifier to detect prompt injection, without expensive model fine-tuning or response generation. We conduct extensive evaluations on a diverse set of short- and long-context benchmarks. The results show that PIShield consistently achieves low false positive and false negative rates, significantly outperforming existing baselines. These findings demonstrate that internal representations of instruction-tuned LLMs provide a powerful and practical foundation for prompt injection detection in real-world applications.

---

### REAP the Experts: Why Pruning Prevails for One-Shot MoE compression

Authors: Mike Lasby, Ivan Lazarevich, Nish Sinnadurai, Sean Lie, Yani Ioannou, Vithursan Thangarasa
Published: 2025-10-15
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.13999v3

Sparsely-activated Mixture-of-Experts (SMoE) models offer efficient pre-training and low latency but their large parameter counts create significant memory overhead, motivating research into expert compression. Contrary to recent findings favouring expert merging on discriminative benchmarks, we find that expert pruning is a superior strategy for generative tasks. We demonstrate that existing merging techniques introduce an irreducible error due to the loss of fine-grained routing control over experts. Leveraging this insight, we propose Router-weighted Expert Activation Pruning (REAP), a novel pruning criterion that considers both router gate-values and expert activation norms to minimize the reconstruction error bound. Across a diverse set of SMoE models ranging from 20B to 1T parameters, REAP consistently outperforms merging and other pruning methods on generative benchmarks, especially at 50% compression. Notably, our method achieves near-lossless compression on code generation tasks with Qwen3-Coder-480B and Kimi-K2, even after pruning 50% of experts.

---

### BitNet Distillation

Authors: Xun Wu, Shaohan Huang, Wenhui Wang, Ting Song, Li Dong, Yan Xia, Furu Wei
Published: 2025-10-15
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2510.13998v1

In this paper, we present BitNet Distillation (BitDistill), a lightweight pipeline that fine-tunes off-the-shelf full-precision LLMs (e.g., Qwen) into 1.58-bit precision (i.e., ternary weights {-1, 0, 1}) for specific downstream tasks, achieving strong task-specific performance with minimal computational cost. Specifically, BitDistill incorporates three key techniques: the SubLN module, as introduced in BitNet; multi-head attention distillation, based on MiniLM; and continual pre-training, which serves as a crucial warm-up step to mitigate the scalability issue of the performance gap between finetuned full-precision and 1.58-bit LLMs on specific tasks. Experimental results show that BitDistill achieves performance comparable to the full-precision counterpart models across model size, while enabling up to 10x memory savings and 2.65x faster inference on CPUs. Code is available at https://github.com/microsoft/BitNet.

---

### Dynamic SBI: Round-free Sequential Simulation-Based Inference with Adaptive Datasets

Authors: Huifang Lyu, James Alvey, Noemi Anau Montel, Mauro Pieroni, Christoph Weniger
Published: 2025-10-15
Categories: astro-ph.IM, astro-ph.CO, cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2510.13997v1

Simulation-based inference (SBI) is emerging as a new statistical paradigm for addressing complex scientific inference problems. By leveraging the representational power of deep neural networks, SBI can extract the most informative simulation features for the parameters of interest. Sequential SBI methods extend this approach by iteratively steering the simulation process towards the most relevant regions of parameter space. This is typically implemented through an algorithmic structure, in which simulation and network training alternate over multiple rounds. This strategy is particularly well suited for high-precision inference in high-dimensional settings, which are commonplace in physics applications with growing data volumes and increasing model fidelity. Here, we introduce dynamic SBI, which implements the core ideas of sequential methods in a round-free, asynchronous, and highly parallelisable manner. At its core is an adaptive dataset that is iteratively transformed during inference to resemble the target observation. Simulation and training proceed in parallel: trained networks are used both to filter out simulations incompatible with the data and to propose new, more promising ones. Compared to round-based sequential methods, this asynchronous structure can significantly reduce simulation costs and training overhead. We demonstrate that dynamic SBI achieves significant improvements in simulation and training efficiency while maintaining inference performance. We further validate our framework on two challenging astrophysical inference tasks: characterising the stochastic gravitational wave background and analysing strong gravitational lensing systems. Overall, this work presents a flexible and efficient new paradigm for sequential SBI.

---

### The German Commons - 154 Billion Tokens of Openly Licensed Text for German Language Models

Authors: Lukas Gienapp, Christopher Schröder, Stefan Schweter, Christopher Akiki, Ferdinand Schlatt, Arden Zimmermann, Phillipe Genêt, Martin Potthast
Published: 2025-10-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.13996v1

Large language model development relies on large-scale training corpora, yet most contain data of unclear licensing status, limiting the development of truly open models. This problem is exacerbated for non-English languages, where openly licensed text remains critically scarce. We introduce the German Commons, the largest collection of openly licensed German text to date. It compiles data from 41 sources across seven domains, encompassing legal, scientific, cultural, political, news, economic, and web text. Through systematic sourcing from established data providers with verifiable licensing, it yields 154.56 billion tokens of high-quality text for language model training. Our processing pipeline implements comprehensive quality filtering, deduplication, and text formatting fixes, ensuring consistent quality across heterogeneous text sources. All domain subsets feature licenses of at least CC-BY-SA 4.0 or equivalent, ensuring legal compliance for model training and redistribution. The German Commons therefore addresses the critical gap in openly licensed German pretraining data, and enables the development of truly open German language models. We also release code for corpus construction and data filtering tailored to German language text, rendering the German Commons fully reproducible and extensible.

---

### Finding Holes: Pathologist Level Performance Using AI for Cribriform Morphology Detection in Prostate Cancer

Authors: Kelvin Szolnoky, Anders Blilie, Nita Mulliqi, Toyonori Tsuzuki, Hemamali Samaratunga, Matteo Titus, Xiaoyi Ji, Sol Erika Boman, Einar Gudlaugsson, Svein Reidar Kjosavik, José Asenjo, Marcello Gambacorta et al.
Published: 2025-10-15
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.13995v1

Background: Cribriform morphology in prostate cancer is a histological feature that indicates poor prognosis and contraindicates active surveillance. However, it remains underreported and subject to significant interobserver variability amongst pathologists. We aimed to develop and validate an AI-based system to improve cribriform pattern detection. Methods: We created a deep learning model using an EfficientNetV2-S encoder with multiple instance learning for end-to-end whole-slide classification. The model was trained on 640 digitised prostate core needle biopsies from 430 patients, collected across three cohorts. It was validated internally (261 slides from 171 patients) and externally (266 slides, 104 patients from three independent cohorts). Internal validation cohorts included laboratories or scanners from the development set, while external cohorts used completely independent instruments and laboratories. Annotations were provided by three expert uropathologists with known high concordance. Additionally, we conducted an inter-rater analysis and compared the model's performance against nine expert uropathologists on 88 slides from the internal validation cohort. Results: The model showed strong internal validation performance (AUC: 0.97, 95% CI: 0.95-0.99; Cohen's kappa: 0.81, 95% CI: 0.72-0.89) and robust external validation (AUC: 0.90, 95% CI: 0.86-0.93; Cohen's kappa: 0.55, 95% CI: 0.45-0.64). In our inter-rater analysis, the model achieved the highest average agreement (Cohen's kappa: 0.66, 95% CI: 0.57-0.74), outperforming all nine pathologists whose Cohen's kappas ranged from 0.35 to 0.62. Conclusion: Our AI model demonstrates pathologist-level performance for cribriform morphology detection in prostate cancer. This approach could enhance diagnostic reliability, standardise reporting, and improve treatment decisions for prostate cancer patients.

---

### Efficient Few-Shot Learning in Remote Sensing: Fusing Vision and Vision-Language Models

Authors: Jia Yun Chua, Argyrios Zolotas, Miguel Arana-Catania
Published: 2025-10-15
Categories: cs.CV, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.13993v1

Remote sensing has become a vital tool across sectors such as urban planning, environmental monitoring, and disaster response. While the volume of data generated has increased significantly, traditional vision models are often constrained by the requirement for extensive domain-specific labelled data and their limited ability to understand the context within complex environments. Vision Language Models offer a complementary approach by integrating visual and textual data; however, their application to remote sensing remains underexplored, particularly given their generalist nature. This work investigates the combination of vision models and VLMs to enhance image analysis in remote sensing, with a focus on aircraft detection and scene understanding. The integration of YOLO with VLMs such as LLaVA, ChatGPT, and Gemini aims to achieve more accurate and contextually aware image interpretation. Performance is evaluated on both labelled and unlabelled remote sensing data, as well as degraded image scenarios which are crucial for remote sensing. The findings show an average MAE improvement of 48.46% across models in the accuracy of aircraft detection and counting, especially in challenging conditions, in both raw and degraded scenarios. A 6.17% improvement in CLIPScore for comprehensive understanding of remote sensing images is obtained. The proposed approach combining traditional vision models and VLMs paves the way for more advanced and efficient remote sensing image analysis, especially in few-shot learning scenarios.

---

### Signature in Code Backdoor Detection, how far are we?

Authors: Quoc Hung Le, Thanh Le-Cong, Bach Le, Bowen Xu
Published: 2025-10-15
Categories: cs.SE, cs.LG
arXiv: https://arxiv.org/abs/2510.13992v1

As Large Language Models (LLMs) become increasingly integrated into software development workflows, they also become prime targets for adversarial attacks. Among these, backdoor attacks are a significant threat, allowing attackers to manipulate model outputs through hidden triggers embedded in training data. Detecting such backdoors remains a challenge, and one promising approach is the use of Spectral Signature defense methods that identify poisoned data by analyzing feature representations through eigenvectors. While some prior works have explored Spectral Signatures for backdoor detection in neural networks, recent studies suggest that these methods may not be optimally effective for code models. In this paper, we revisit the applicability of Spectral Signature-based defenses in the context of backdoor attacks on code models. We systematically evaluate their effectiveness under various attack scenarios and defense configurations, analyzing their strengths and limitations. We found that the widely used setting of Spectral Signature in code backdoor detection is often suboptimal. Hence, we explored the impact of different settings of the key factors. We discovered a new proxy metric that can more accurately estimate the actual performance of Spectral Signature without model retraining after the defense.

---

### Do Large Language Models Show Biases in Causal Learning? Insights from Contingency Judgment

Authors: María Victoria Carro, Denise Alejandra Mester, Francisca Gauna Selasco, Giovanni Franco Gabriel Marraffini, Mario Alejandro Leiva, Gerardo I. Simari, María Vanina Martinez
Published: 2025-10-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.13985v1

Causal learning is the cognitive process of developing the capability of making causal inferences based on available information, often guided by normative principles. This process is prone to errors and biases, such as the illusion of causality, in which people perceive a causal relationship between two variables despite lacking supporting evidence. This cognitive bias has been proposed to underlie many societal problems, including social prejudice, stereotype formation, misinformation, and superstitious thinking. In this work, we examine whether large language models are prone to developing causal illusions when faced with a classic cognitive science paradigm: the contingency judgment task. To investigate this, we constructed a dataset of 1,000 null contingency scenarios (in which the available information is not sufficient to establish a causal relationship between variables) within medical contexts and prompted LLMs to evaluate the effectiveness of potential causes. Our findings show that all evaluated models systematically inferred unwarranted causal relationships, revealing a strong susceptibility to the illusion of causality. While there is ongoing debate about whether LLMs genuinely understand causality or merely reproduce causal language without true comprehension, our findings support the latter hypothesis and raise concerns about the use of language models in domains where accurate causal reasoning is essential for informed decision-making.

---

### Static Sandboxes Are Inadequate: Modeling Societal Complexity Requires Open-Ended Co-Evolution in LLM-Based Multi-Agent Simulations

Authors: Jinkun Chen, Sher Badshah, Xuemin Yu, Sijia Han
Published: 2025-10-15
Categories: cs.MA, cs.AI
arXiv: https://arxiv.org/abs/2510.13982v3

What if artificial agents could not just communicate, but also evolve, adapt, and reshape their worlds in ways we cannot fully predict? With llm now powering multi-agent systems and social simulations, we are witnessing new possibilities for modeling open-ended, ever-changing environments. Yet, most current simulations remain constrained within static sandboxes, characterized by predefined tasks, limited dynamics, and rigid evaluation criteria. These limitations prevent them from capturing the complexity of real-world societies. In this paper, we argue that static, task-specific benchmarks are fundamentally inadequate and must be rethought. We critically review emerging architectures that blend llm with multi-agent dynamics, highlight key hurdles such as balancing stability and diversity, evaluating unexpected behaviors, and scaling to greater complexity, and introduce a fresh taxonomy for this rapidly evolving field. Finally, we present a research roadmap centered on open-endedness, continuous co-evolution, and the development of resilient, socially aligned AI ecosystems. We call on the community to move beyond static paradigms and help shape the next generation of adaptive, socially-aware multi-agent simulations.

---

### Do Slides Help? Multi-modal Context for Automatic Transcription of Conference Talks

Authors: Supriti Sinhamahapatra, Jan Niehues
Published: 2025-10-15
Categories: cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2510.13979v1

State-of-the-art (SOTA) Automatic Speech Recognition (ASR) systems primarily rely on acoustic information while disregarding additional multi-modal context. However, visual information are essential in disambiguation and adaptation. While most work focus on speaker images to handle noise conditions, this work also focuses on integrating presentation slides for the use cases of scientific presentation. In a first step, we create a benchmark for multi-modal presentation including an automatic analysis of transcribing domain-specific terminology. Next, we explore methods for augmenting speech models with multi-modal information. We mitigate the lack of datasets with accompanying slides by a suitable approach of data augmentation. Finally, we train a model using the augmented dataset, resulting in a relative reduction in word error rate of approximately 34%, across all words and 35%, for domain-specific terms compared to the baseline model.

---

### Classifying and Addressing the Diversity of Errors in Retrieval-Augmented Generation Systems

Authors: Kin Kwan Leung, Mouloud Belbahri, Yi Sui, Alex Labach, Xueying Zhang, Stephen Anthony Rose, Jesse C. Cresswell
Published: 2025-10-15
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2510.13975v2

Retrieval-augmented generation (RAG) is a prevalent approach for building LLM-based question-answering systems that can take advantage of external knowledge databases. Due to the complexity of real-world RAG systems, there are many potential causes for erroneous outputs. Understanding the range of errors that can occur in practice is crucial for robust deployment. We present a new taxonomy of the error types that can occur in realistic RAG systems, examples of each, and practical advice for addressing them. Additionally, we curate a dataset of erroneous RAG responses annotated by error types. We then propose an auto-evaluation method aligned with our taxonomy that can be used in practice to track and address errors during development. Code and data are available at https://github.com/layer6ai-labs/rag-error-classification.

---

### Distributional Consistency Loss: Beyond Pointwise Data Terms in Inverse Problems

Authors: George Webber, Andrew J. Reader
Published: 2025-10-15
Categories: cs.LG, cs.CV, physics.med-ph
arXiv: https://arxiv.org/abs/2510.13972v2

Recovering true signals from noisy measurements is a central challenge in inverse problems spanning medical imaging, geophysics, and signal processing. Current methods balance prior signal priors (regularization) with agreement with noisy data (data-fidelity). Conventional data-fidelity loss functions, such as mean-squared error (MSE) or negative log-likelihood, seek pointwise agreement with noisy measurements, often leading to overfitting to noise. In this work, we instead evaluate data-fidelity collectively by testing whether the observed measurements are statistically consistent with the noise distributions implied by the current estimate. We introduce distributional consistency (DC) loss, a data-fidelity objective that replaces pointwise matching with distribution-level calibration. DC loss acts as a direct and practical plug-in replacement for standard data consistency terms: i) it is compatible with modern unsupervised regularizers that operate without paired measurement-ground-truth data, ii) it is optimized in the same way as traditional losses, and iii) it avoids overfitting to measurement noise without early stopping or priors. Its scope naturally fits many practical inverse problems where the measurement-noise distribution is known and where the measured dataset consists of many independent noisy values. We demonstrate efficacy in two key example application areas: i) in image denoising with deep image prior, using DC instead of MSE loss removes the need for early stopping and achieves higher PSNR; ii) in medical image reconstruction from Poisson-noisy data, DC loss reduces artifacts in highly-iterated reconstructions and enhances the efficacy of hand-crafted regularization. These results position DC loss as a statistically grounded, performance-enhancing alternative to conventional fidelity losses for an important class of unsupervised noise-dominated inverse problems.

---

### Less is More: Improving LLM Reasoning with Minimal Test-Time Intervention

Authors: Zhen Yang, Mingyang Zhang, Feng Chen, Ganggui Ding, Liang Hou, Xin Tao, Ying-Cong Chen
Published: 2025-10-15
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.13940v4

Recent progress in large language models (LLMs) has focused on test-time scaling to improve reasoning via increased inference computation, but often at the cost of efficiency. We revisit test-time behavior and uncover a simple yet underexplored phenomenon: reasoning uncertainty is highly localized-only a small subset of high-entropy tokens dominantly affects output correctness. Motivated by this, we propose Minimal Test-Time Intervention (MTI), a training-free framework that enhances reasoning accuracy and stability with minimal overhead. MTI includes: (i) Selective CFG intervention, applying classifier-free guidance only at uncertain positions; and (ii) Lightweight negative-prompt guidance, reusing the main model's KV cache to approximate unconditional decoding efficiently. MTI yields consistent gains across general, coding, and STEM tasks-e.g., +9.28% average improvement on six benchmarks for DeepSeek-R1-7B and +11.25% on AIME2024 using Ling-mini-2.0-while remaining highly efficient.

---

### Readers Prefer Outputs of AI Trained on Copyrighted Books over Expert Human Writers

Authors: Tuhin Chakrabarty, Jane C. Ginsburg, Paramveer Dhillon
Published: 2025-10-15
Categories: cs.CL, cs.AI, cs.CY
arXiv: https://arxiv.org/abs/2510.13939v4

The use of copyrighted books for training AI has sparked lawsuits from authors concerned about AI generating derivative content. Yet whether these models can produce high-quality literary text emulating authors' voices remains unclear. We conducted a preregistered study comparing MFA-trained writers with three frontier models (ChatGPT, Claude, Gemini) writing up to 450-word excerpts emulating 50 award-winning authors' styles. In blind pairwise evaluations by 28 MFA-trained readers and 516 college-educated general readers, AI text from in-context prompting was strongly disfavored by MFA readers for stylistic fidelity (OR=0.16) and quality (OR=0.13), while general readers showed no fidelity preference (OR=1.06) but favored AI for quality (OR=1.82). Fine-tuning ChatGPT on authors' complete works reversed these results: MFA readers favored AI for fidelity (OR=8.16) and quality (OR=1.87), with general readers showing even stronger preference (fidelity OR=16.65; quality OR=5.42). Both groups preferred fine-tuned AI, but the writer-type X reader-type interaction remained significant (p=0.021 for fidelity; p<10^-4 for quality), indicating general readers favored AI by a wider margin. Effects are robust under cluster-robust inference and generalize across authors in heterogeneity analyses. Fine-tuned outputs were rarely flagged as AI-generated (3% vs. 97% for prompting) by leading detectors. Mediation analysis shows fine-tuning eliminates detectable AI quirks that penalize in-context outputs, altering the nexus between detectability and preference. While not accounting for effort to transform AI output into publishable prose, the median fine-tuning cost of $81 per author represents a 99.7% reduction versus typical writer compensation. Author-specific fine-tuning enables non-verbatim AI writing preferred over expert human writing, providing evidence relevant to copyright's fourth fair-use factor.

---

### FinDeepResearch: Evaluating Deep Research Agents in Rigorous Financial Analysis

Authors: Fengbin Zhu, Xiang Yao Ng, Ziyang Liu, Chang Liu, Xianwei Zeng, Chao Wang, Tianhui Tan, Xuan Yao, Pengyang Shao, Min Xu, Zixuan Wang, Jing Wang et al.
Published: 2025-10-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.13936v2

Deep Research (DR) agents, powered by advanced Large Language Models (LLMs), have recently garnered increasing attention for their capability in conducting complex research tasks. However, existing literature lacks a rigorous and systematic evaluation of DR Agent's capabilities in critical research analysis. To address this gap, we first propose HisRubric, a novel evaluation framework with a hierarchical analytical structure and a fine-grained grading rubric for rigorously assessing DR agents' capabilities in corporate financial analysis. This framework mirrors the professional analyst's workflow, progressing from data recognition to metric calculation, and finally to strategic summarization and interpretation. Built on this framework, we construct a FinDeepResearch benchmark that comprises 64 listed companies from 8 financial markets across 4 languages, encompassing a total of 15,808 grading items. We further conduct extensive experiments on the FinDeepResearch using 16 representative methods, including 6 DR agents, 5 LLMs equipped with both deep reasoning and search capabilities, and 5 LLMs with deep reasoning capabilities only. The results reveal the strengths and limitations of these approaches across diverse capabilities, financial markets, and languages, offering valuable insights for future research and development. The benchmark and evaluation code is publicly available at https://OpenFinArena.com/.

---

### Instruction Retrieval at Inference Time for Small Language Models

Authors: Kenan Alkiek, David Jurgens, Vinod Vydiswaran
Published: 2025-10-15
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.13935v3

The facts a language model stores are tied to its parameter count, so small models that fit on edge devices fail on expert problems, which need specialized knowledge and follow multi-step procedures. Fine-tuning for a specific domain or task writes the knowledge into the parameters but must be repeated for every model and domain, and a retrieved passage leaves the model to find the relevant fact and apply it on its own. We introduce instruction retrieval, which distills a teacher model's expertise into a corpus of instructions tailored so that a small model can follow. For each cluster of a domain's problems, the teacher writes one instruction with the background knowledge the cluster depends on, a procedure for that kind of problem, and the common mistakes made on it. This reusable corpus needs only to be built once per domain and requires no run-time teacher access. At inference, a frozen small model retrieves the instructions nearest its question and follows them, with no fine-tuning. Across medicine, law, and mathematics benchmarks, we demonstrate the corpus improves over zero-shot on every task and over few-shot prompting, self-consistency, and other retrieved text on medicine and law. On MedQA the corpus raises mean accuracy by 10.6 points, where retrieved textbook passages raise it by 3.9 and few-shot examples from the same teacher lower it. An error analysis shows that only the background knowledge fixes the questions a small model always gets wrong, and that the procedure and common mistakes fix only the questions where it wavers between options. Our results show that automatically retrieved inference-time procedural guidance and domain knowledge can yield substantial gains for small models.

---

### Robust or Suggestible? Exploring Non-Clinical Induction in LLM Drug-Safety Decisions

Authors: Siying Liu, Shisheng Zhang, Indu Bala
Published: 2025-10-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.13931v1

Large language models (LLMs) are increasingly applied in biomedical domains, yet their reliability in drug-safety prediction remains underexplored. In this work, we investigate whether LLMs incorporate socio-demographic information into adverse event (AE) predictions, despite such attributes being clinically irrelevant. Using structured data from the United States Food and Drug Administration Adverse Event Reporting System (FAERS) and a persona-based evaluation framework, we assess two state-of-the-art models, ChatGPT-4o and Bio-Medical-Llama-3.8B, across diverse personas defined by education, marital status, employment, insurance, language, housing stability, and religion. We further evaluate performance across three user roles (general practitioner, specialist, patient) to reflect real-world deployment scenarios where commercial systems often differentiate access by user type. Our results reveal systematic disparities in AE prediction accuracy. Disadvantaged groups (e.g., low education, unstable housing) were frequently assigned higher predicted AE likelihoods than more privileged groups (e.g., postgraduate-educated, privately insured). Beyond outcome disparities, we identify two distinct modes of bias: explicit bias, where incorrect predictions directly reference persona attributes in reasoning traces, and implicit bias, where predictions are inconsistent, yet personas are not explicitly mentioned. These findings expose critical risks in applying LLMs to pharmacovigilance and highlight the urgent need for fairness-aware evaluation protocols and mitigation strategies before clinical deployment.

---

### LLMs Can Get "Brain Rot": A Pilot Study on Twitter/X

Authors: Shuo Xing, Junyuan Hong, Yifan Wang, Runjin Chen, Zhenyu Zhang, Ananth Grama, Zhengzhong Tu, Zhangyang Wang
Published: 2025-10-15
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.13928v3

We propose and test the LLM Brain Rot Hypothesis: continual exposure to junk web text induces lasting cognitive decline in large language models (LLMs). To unveil junk effects, we designed a novel controlled experiment on real Twitter/X corpora, by constructing junk and reverse-controlled datasets via two orthogonal operationalizations: M1 (engagement degree) and M2 (semantic quality), with matched token scale and training operations across conditions. Compared to the control group, continual pre-training of 4 LLMs on the junk dataset causes non-trivial declines (Hedges' g>0.3) on reasoning, long-context understanding, safety, and inflating "dark traits" (e.g., psychopathy, narcissism). The gradual mixtures of junk and control datasets also yield dose-response cognition decay: for example, under M1, ARC-Challenge with Chain-of-Thought drops 72.1 -> 57.2 and RULER-CWE 83.7 -> 52.3 as junk ratio rises from 0% to 100%. Error forensics reveal several key insights. First, we identify thought-skipping as the primary lesion in reasoning: models increasingly truncate or skip chains. Second, partial but incomplete healing is observed: scaling instruction tuning and clean continual pre-training improve the declined cognition, yet cannot restore baseline capability, suggesting persistent representational drift rather than format mismatch. Finally, we discover that the popularity, a non-semantic metric, of a tweet is a better indicator of the Brain Rot effect than the length in M1. Together, the results provide significant, multi-perspective evidence that social effects of data could be a causal driver of LLM capability decay in continual pre-training, thereby motivating routine "cognitive health checks" for deployed and evolving LLMs.

---

### Long-Term Spatio-Temporal Forecasting of Monthly Rainfall in West Bengal Using Ensemble Learning Approaches

Authors: Jishu Adhikary, Raju Maiti
Published: 2025-10-15
Categories: stat.AP, cs.LG
arXiv: https://arxiv.org/abs/2510.13927v2

Rainfall forecasting plays a critical role in climate adaptation, agriculture, and water resource management. This study develops long-term forecasts of monthly rainfall across 19 districts of West Bengal using a century-scale dataset spanning 1900-2019. Daily rainfall records are aggregated into monthly series, resulting in 120 years of observations for each district. The forecasting task involves predicting the next 108 months (9 years, 2011-2019) while accounting for temporal dependencies and spatial interactions among districts. To address the nonlinear and complex structure of rainfall dynamics, we propose a hierarchical modeling framework that combines regression-based forecasting of yearly features with multi-layer perceptrons (MLPs) for monthly prediction. Yearly features, such as annual totals, quarterly proportions, variability measures, skewness, and extremes, are first forecasted using regression models that incorporate both own lags and neighboring-district lags. These forecasts are then integrated as auxiliary inputs into an MLP model, which captures nonlinear temporal patterns and spatial dependencies in the monthly series. The results demonstrate that the hierarchical regression-MLP architecture provides robust long-term spatio-temporal forecasts, offering valuable insights for agriculture, irrigation planning, and water conservation strategies.

---

### BioMedSearch: A Multi-Source Biomedical Retrieval Framework Based on LLMs

Authors: Congying Liu, Xingyuan Wei, Peipei Liu, Yiqing Shen, Yanxu Mao, Tiehan Cui
Published: 2025-10-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.13926v1

Biomedical queries often rely on a deep understanding of specialized knowledge such as gene regulatory mechanisms and pathological processes of diseases. They require detailed analysis of complex physiological processes and effective integration of information from multiple data sources to support accurate retrieval and reasoning. Although large language models (LLMs) perform well in general reasoning tasks, their generated biomedical content often lacks scientific rigor due to the inability to access authoritative biomedical databases and frequently fabricates protein functions, interactions, and structural details that deviate from authentic information. Therefore, we present BioMedSearch, a multi-source biomedical information retrieval framework based on LLMs. The method integrates literature retrieval, protein database and web search access to support accurate and efficient handling of complex biomedical queries. Through sub-queries decomposition, keywords extraction, task graph construction, and multi-source information filtering, BioMedSearch generates high-quality question-answering results. To evaluate the accuracy of question answering, we constructed a multi-level dataset, BioMedMCQs, consisting of 3,000 questions. The dataset covers three levels of reasoning: mechanistic identification, non-adjacent semantic integration, and temporal causal reasoning, and is used to assess the performance of BioMedSearch and other methods on complex QA tasks. Experimental results demonstrate that BioMedSearch consistently improves accuracy over all baseline models across all levels. Specifically, at Level 1, the average accuracy increases from 59.1% to 91.9%; at Level 2, it rises from 47.0% to 81.0%; and at the most challenging Level 3, the average accuracy improves from 36.3% to 73.4%. The code and BioMedMCQs are available at: https://github.com/CyL-ucas/BioMed_Search

---

### An LLM-Powered AI Agent Framework for Holistic IoT Traffic Interpretation

Authors: Daniel Adu Worae, Spyridon Mastorakis
Published: 2025-10-15
Categories: cs.CL, cs.CR, cs.NI
arXiv: https://arxiv.org/abs/2510.13925v1

Internet of Things (IoT) networks generate diverse and high-volume traffic that reflects both normal activity and potential threats. Deriving meaningful insight from such telemetry requires cross-layer interpretation of behaviors, protocols, and context rather than isolated detection. This work presents an LLM-powered AI agent framework that converts raw packet captures into structured and semantically enriched representations for interactive analysis. The framework integrates feature extraction, transformer-based anomaly detection, packet and flow summarization, threat intelligence enrichment, and retrieval-augmented question answering. An AI agent guided by a large language model performs reasoning over the indexed traffic artifacts, assembling evidence to produce accurate and human-readable interpretations. Experimental evaluation on multiple IoT captures and six open models shows that hybrid retrieval, which combines lexical and semantic search with reranking, substantially improves BLEU, ROUGE, METEOR, and BERTScore results compared with dense-only retrieval. System profiling further indicates low CPU, GPU, and memory overhead, demonstrating that the framework achieves holistic and efficient interpretation of IoT network traffic.

---

### LTR-ICD: A Ranking-Aware Framework for Automatic ICD Coding

Authors: Mohammad Mansoori, Amira Soliman, Farzaneh Etminani
Published: 2025-10-15
Categories: cs.LG, cs.CL, cs.IR
arXiv: https://arxiv.org/abs/2510.13922v3

Clinical notes contain unstructured text provided by clinicians during patient encounters. These notes are usually accompanied by a sequence of diagnostic codes following the International Classification of Diseases (ICD). Correctly assigning and ordering ICD codes is essential for medical diagnosis and reimbursement. However, automating this task remains challenging. State-of-the-art methods treated this problem as a classification task, leading to ignoring the order of ICD codes that is essential for different purposes. In this work, as a first attempt, we approach this task from a retrieval system perspective to consider the order of codes, thus formulating this problem as a classification and ranking task. Our results and analysis show that the proposed framework has a superior ability to identify high-priority codes compared to other methods. For instance, our model's accuracy in correctly ranking primary diagnosis codes is 47%, compared to 20% for the state-of-the-art classifier. Additionally, in terms of classification metrics, the proposed model achieves a micro- and macro-F1 scores of 0.6065 and 0.2904, respectively, surpassing the previous best model with scores of 0.6035 and 0.2741.

---

### Weight Weaving: Parameter Pooling for Data-Free Model Merging

Authors: Levy Chaves, Eduardo Valle, Sandra Avila
Published: 2025-10-15
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2510.13921v1

Model merging provides a cost-effective and data-efficient combination of specialized deep neural networks through parameter integration. This technique leverages expert models across downstream tasks without requiring retraining. Most model merging approaches critically depend on scaling hyper-parameters $λ$, which weight each model's contribution globally or individually. Principled approaches for setting scaling factors without accessing any data (data-free) are scarce, often leading researchers to tune $λ$ using privileged data from the evaluation set, which is obviously unfeasible in practice. To address this limitation, we introduce Weight Weaving, a plug-and-play technique that pools model weights across $λ$ values search space using user-defined pooling functions, such as averaging, random selection, or even existing model merging methods. Our method demonstrates high modularity, imposing minimal constraints on the search space. It operates orthogonally to existing model merging methods and eliminates evaluation data requirements. We validate Weight Weaving across three ViT variants in three experimental setups: vision multi-task learning, vision continual learning, and domain generalization. Our method consistently improves the performance of several model merging methods, achieving average accuracy gains of up to 15.9 percentage points in a data-free setting.

---