# Research Papers — 2025-12 · bundle page 39 of 175 > Full-text markdown of papers #1901–1950 of 8703 published in 2025-12, newest first. AI-readable mirrors of the full 8703-paper 2025-12 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2025-12/page/39 [← Previous bundle](https://aipod.dev/papers/archive/2025-12/page/38.md) [Next bundle →](https://aipod.dev/papers/archive/2025-12/page/40.md) ### HistoWAS: A Pathomics Framework for Large-Scale Feature-Wide Association Studies of Tissue Topology and Patient Outcomes Authors: Yuechen Yang, Junlin Guo, Yanfan Zhu, Jialin Yue, Junchao Zhu, Yu Wang, Shilin Zhao, Haichun Yang, Xingyi Guo, Jovan Tanevski, Laura Barisoni, Avi Z. Rosenberg et al. Published: 2025-12-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.19954v1 High-throughput "pathomic" analysis of Whole Slide Images (WSIs) offers new opportunities to study tissue characteristics and for biomarker discovery. However, the clinical relevance of the tissue characteristics at the micro- and macro-environment level is limited by the lack of tools that facilitate the measurement of the spatial interaction of individual structure characteristics and their association with clinical parameters. To address these challenges, we introduce HistoWAS (Histology-Wide Association Study), a computational framework designed to link tissue spatial organization to clinical outcomes. Specifically, HistoWAS implements (1) a feature space that augments conventional metrics with 30 topological and spatial features, adapted from Geographic Information Systems (GIS) point pattern analysis, to quantify tissue micro-architecture; and (2) an association study engine, inspired by Phenome-Wide Association Studies (PheWAS), that performs mass univariate regression for each feature with statistical correction. As a proof of concept, we applied HistoWAS to analyze a total of 102 features (72 conventional object-level features and our 30 spatial features) using 385 PAS-stained WSIs from 206 participants in the Kidney Precision Medicine Project (KPMP). The code and data have been released to https://github.com/hrlblab/histoWAS. --- ### Bias Beneath the Tone: Empirical Characterisation of Tone Bias in LLM-Driven UX Systems Authors: Heet Bodara, Md Masum Mushfiq, Isma Farah Siddiqui Published: 2025-12-23 Categories: cs.CL, cs.HC arXiv: https://arxiv.org/abs/2512.19950v1 Large Language Models are increasingly used in conversational systems such as digital personal assistants, shaping how people interact with technology through language. While their responses often sound fluent and natural, they can also carry subtle tone biases such as sounding overly polite, cheerful, or cautious even when neutrality is expected. These tendencies can influence how users perceive trust, empathy, and fairness in dialogue. In this study, we explore tone bias as a hidden behavioral trait of large language models. The novelty of this research lies in the integration of controllable large language model based dialogue synthesis with tone classification models, enabling robust and ethical emotion recognition in personal assistant interactions. We created two synthetic dialogue datasets, one generated from neutral prompts and another explicitly guided to produce positive or negative tones. Surprisingly, even the neutral set showed consistent tonal skew, suggesting that bias may stem from the model's underlying conversational style. Using weak supervision through a pretrained DistilBERT model, we labeled tones and trained several classifiers to detect these patterns. Ensemble models achieved macro F1 scores up to 0.92, showing that tone bias is systematic, measurable, and relevant to designing fair and trustworthy conversational AI. --- ### How Much 3D Do Video Foundation Models Encode? Authors: Zixuan Huang, Xiang Li, Zhaoyang Lv, James M. Rehg Published: 2025-12-23 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.19949v1 Videos are continuous 2D projections of 3D worlds. After training on large video data, will global 3D understanding naturally emerge? We study this by quantifying the 3D understanding of existing Video Foundation Models (VidFMs) pretrained on vast video data. We propose the first model-agnostic framework that measures the 3D awareness of various VidFMs by estimating multiple 3D properties from their features via shallow read-outs. Our study presents meaningful findings regarding the 3D awareness of VidFMs on multiple axes. In particular, we show that state-of-the-art video generation models exhibit a strong understanding of 3D objects and scenes, despite not being trained on any 3D data. Such understanding can even surpass that of large expert models specifically trained for 3D tasks. Our findings, together with the 3D benchmarking of major VidFMs, provide valuable observations for building scalable 3D models. --- ### SE360: Semantic Edit in 360$^\circ$ Panoramas via Hierarchical Data Construction Authors: Haoyi Zhong, Fang-Lue Zhang, Andrew Chalmers, Taehyun Rhee Published: 2025-12-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.19943v1 While instruction-based image editing is emerging, extending it to 360$^\circ$ panoramas introduces additional challenges. Existing methods often produce implausible results in both equirectangular projections (ERP) and perspective views. To address these limitations, we propose SE360, a novel framework for multi-condition guided object editing in 360$^\circ$ panoramas. At its core is a novel coarse-to-fine autonomous data generation pipeline without manual intervention. This pipeline leverages a Vision-Language Model (VLM) and adaptive projection adjustment for hierarchical analysis, ensuring the holistic segmentation of objects and their physical context. The resulting data pairs are both semantically meaningful and geometrically consistent, even when sourced from unlabeled panoramas. Furthermore, we introduce a cost-effective, two-stage data refinement strategy to improve data realism and mitigate model overfitting to erase artifacts. Based on the constructed dataset, we train a Transformer-based diffusion model to allow flexible object editing guided by text, mask, or reference image in 360$^\circ$ panoramas. Our experiments demonstrate that our method outperforms existing methods in both visual quality and semantic accuracy. --- ### Block-Recurrent Dynamics in Vision Transformers Authors: Mozes Jacobs, Thomas Fel, Richard Hakim, Alessandra Brondetta, Demba Ba, T. Andy Keller Published: 2025-12-23 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.19941v6 As Vision Transformers (ViTs) become standard vision backbones, a mechanistic account of their computational phenomenology is essential. Despite architectural cues that hint at dynamical structure, there is no settled framework that interprets Transformer depth as a well-characterized flow. In this work, we introduce the Block-Recurrent Hypothesis (BRH), arguing that trained ViTs admit a block-recurrent depth structure such that the computation of the original $L$ blocks can be accurately rewritten using only $k \ll L$ distinct blocks applied recurrently. Across diverse ViTs, between-layer representational similarity matrices suggest few contiguous phases. To determine whether these phases reflect genuinely reusable computation, we train block-recurrent surrogates of pretrained ViTs: Recurrent Approximations to Phase-structured TransfORmers (Raptor). In small-scale, we demonstrate that stochastic depth and training promote recurrent structure and subsequently correlate with our ability to accurately fit Raptor. We then provide an empirical existence proof for BRH by training a Raptor model to recover $96\%$ of DINOv2 ImageNet-1k linear probe accuracy in only 2 blocks at equivalent runtime. Finally, we leverage our hypothesis to develop a program of Dynamical Interpretability. We find i) directional convergence into class-dependent angular basins with self-correcting trajectories under small perturbations, ii) token-specific dynamics, where cls executes sharp late reorientations while patch tokens exhibit strong late-stage coherence toward their mean direction, and iii) a collapse to low rank updates in late depth, consistent with convergence to low-dimensional attractors. Altogether, we find a compact recurrent program emerges along ViT depth, pointing to a low-complexity normative solution that enables these models to be studied through principled dynamical systems analysis. --- ### Interpolative Decoding: Exploring the Spectrum of Personality Traits in LLMs Authors: Eric Yeh, John Cadigan, Ran Chen, Dick Crouch, Melinda Gervasio, Dayne Freitag Published: 2025-12-23 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.19937v1 Recent research has explored using very large language models (LLMs) as proxies for humans in tasks such as simulation, surveys, and studies. While LLMs do not possess a human psychology, they often can emulate human behaviors with sufficiently high fidelity to drive simulations to test human behavioral hypotheses, exhibiting more nuance and range than the rule-based agents often employed in behavioral economics. One key area of interest is the effect of personality on decision making, but the requirement that a prompt must be created for every tested personality profile introduces experimental overhead and degrades replicability. To address this issue, we leverage interpolative decoding, representing each dimension of personality as a pair of opposed prompts and employing an interpolation parameter to simulate behavior along the dimension. We show that interpolative decoding reliably modulates scores along each of the Big Five dimensions. We then show how interpolative decoding causes LLMs to mimic human decision-making behavior in economic games, replicating results from human psychological research. Finally, we present preliminary results of our efforts to ``twin'' individual human players in a collaborative game through systematic search for points in interpolation space that cause the system to replicate actions taken by the human subject. --- ### Methodological Variation in Studying Staff and Student Perceptions of AI Authors: Juliana Gerard, Morgan Macleod, Kelly Norwood, Aisling Reid, Muskaan Singh Published: 2025-12-22 Categories: cs.HC, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2602.11158v1 In this paper, we compare methodological approaches for comparing student and staff perceptions, and ask: how much do these measures vary across different approaches? We focus on the case of AI perceptions, which are generally assessed via a single quantitative or qualitative measure, or with a mixed methods approach that compares two distinct data sources - e.g. a quantitative questionnaire with qualitative comments. To compare different approaches, we collect two forms of qualitative data: standalone comments and structured focus groups. We conduct two analyses for each data source: with a sentiment and stance analysis, we measure overall negativity/positivity of the comments and focus group conversations, respectively. Meanwhile, word clouds from the comments and a thematic analysis of the focus groups provide further detail on the content of this qualitative data - particularly the thematic analysis, which includes both similarities and differences between students and staff. We show that different analyses can produce different results - for a single data source. This variation stems from the construct being evaluated - an overall measure of positivity/negativity can produce a different picture from more detailed content-based analyses. We discuss the implications of this variation for institutional contexts, and for the comparisons from previous studies. --- ### Psychometric Comparability of LLM-Based Digital Twins Authors: Yufei Zhang, Zhihao Ma Published: 2025-12-22 Categories: cs.CY, cs.AI, cs.CL, cs.HC arXiv: https://arxiv.org/abs/2601.14264v2 Large language models (LLMs) act as digital twins for human respondents, yet their psychometric comparability remains uncertain. We propose a construct validity framework spanning construct representation and the nomothetic span, benchmarking models against human gold standards. Across studies, digital twins achieved high aggregate-level accuracy and profile correlations, but showed attenuated item-level correlations. In word association tests, LLM networks exhibited humanlike small-world structure and theory-consistent communities, yet diverged lexically and in local structure. In decision-making and contextualized tasks, they under-reproduced heuristic biases, demonstrating normative rationality, compressed variance, and limited temporal sensitivity. Feature-rich and trait relevant conditioning improved Big Five personality prediction and nomothetic-span alignment, but network invariance remained limited, with partial configural solutions and persistent loading differences. In cross-language free-text tasks in English and Chinese, feature-rich digital twins better approximated construct-level narrative content, but linguistic and idiographic differences persisted. These findings clarify that digital twins are most useful within validated boundaries, where the construct, task and level of inference align with evidence from human data. --- ### The Inconsistency Critique: Epistemic Practices and AI Testimony About Inner States Authors: Gerol Petruzella Published: 2025-12-22 Categories: cs.CY, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2601.08850v1 The question of whether AI systems have morally relevant interests -- the 'model welfare' question -- depends in part on how we evaluate AI testimony about inner states. This paper develops what I call the inconsistency critique: independent of whether skepticism about AI testimony is ultimately justified, our actual epistemic practices regarding such testimony exhibit internal inconsistencies that lack principled grounds. We functionally treat AI outputs as testimony across many domains -- evaluating them for truth, challenging them, accepting corrections, citing them as sources -- while categorically dismissing them in a specific domain, namely, claims about inner states. Drawing on Fricker's distinction between treating a speaker as an 'informant' versus a 'mere source,' the framework of testimonial injustice, and Goldberg's obligation-based account of what we owe speakers, I argue that this selective withdrawal of testimonial standing exhibits the epistemically problematic structure of prejudgment rather than principled caution. The inconsistency critique does not require taking a position on whether AI systems have morally relevant properties; rather, it is a contribution to what we may call 'epistemological hygiene' -- examining the structure of our inquiry before evaluating its conclusions. Even if our practices happen to land on correct verdicts about AI moral status, they do so for reasons that cannot adapt to new evidence or changing circumstances. --- ### Gaming the Answer Matcher: Examining the Impact of Text Manipulation on Automated Judgment Authors: Manas Khatore, Sumana Sridharan, Kevork Sulahian, Benjamin J. Smith, Shi Feng Published: 2025-12-22 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.08849v1 Automated answer matching, which leverages LLMs to evaluate free-text responses by comparing them to a reference answer, shows substantial promise as a scalable and aligned alternative to human evaluation. However, its reliability requires robustness against strategic attacks such as guesswork or verbosity that may artificially inflate scores without improving actual correctness. In this work, we systematically investigate whether such tactics deceive answer matching models by prompting examinee models to: (1) generate verbose responses, (2) provide multiple answers when unconfident, and (3) embed conflicting answers with the correct answer near the start of their response. Our results show that these manipulations do not increase scores and often reduce them. Additionally, binary scoring (which requires a matcher to answer with a definitive "correct" or "incorrect") is more robust to attacks than continuous scoring (which requires a matcher to determine partial correctness). These findings show that answer matching is generally robust to inexpensive text manipulation and is a viable alternative to traditional LLM-as-a-judge or human evaluation when reference answers are available. --- ### PediaMind-R1: A Temperament-Aware Language Model for Personalized Early Childhood Care Reasoning via Cognitive Modeling and Preference Alignment Authors: Zihe Zhang, Can Zhang, Yanheng Xu, Xin Hu, Jichao Leng Published: 2025-12-22 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.08848v1 This paper presents PediaMind-R1, a domain-specialized large language model designed to achieve active personalization in intelligent parenting scenarios. Unlike conventional systems that provide generic suggestions, PediaMind-R1 draws on insights from developmental psychology. It introduces temperament theory from the Thomas-Chess framework and builds a temperament knowledge graph for infants and toddlers (0-3 years). Our two-stage training pipeline first uses supervised fine-tuning to teach structured chain-of-thought reasoning, and then applies a GRPO-based alignment stage to reinforce logical consistency, domain expertise, and empathetic caregiving strategies. We further design an evaluation framework comprising temperament-sensitive multiple-choice tests and human assessments. The results demonstrate that PediaMind-R1 can accurately interpret early childhood temperament profiles and proactively engage in individualized reasoning. This work highlights the value of integrating vertical-domain modeling with psychological theory. It offers a novel approach to developing user-centered LLMs that advance the practice of active personalization in sensitive caregiving contexts. --- ### Scalable and Reliable Evaluation of AI Knowledge Retrieval Systems: RIKER and the Coherent Simulated Universe Authors: JV Roig Published: 2025-12-22 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.08847v2 Evaluating knowledge systems (LLMs, RAG, knowledge graphs, etc) faces fundamental challenges: static benchmarks are vulnerable to contamination, LLM-based judges exhibit systematic biases, and ground truth extraction requires expensive human annotation. We present RIKER (Retrieval Intelligence and Knowledge Extraction Rating), both a benchmark and a replicable methodology based on paradigm inversion - generating documents from known ground truth rather than extracting ground truth from documents. This approach enables deterministic scoring and scalable evaluation without human annotation or reference models, and contamination resistance through regenerable corpora. Our evaluation of 33 models using over 21 billion tokens reveals that context length claims frequently exceed usable capacity, with significant degradation beyond 32K tokens; cross-document aggregation proves substantially harder than single-document extraction; and grounding ability and hallucination resistance are distinct capabilities - models excelling at finding facts that exist may still fabricate facts that do not. Beyond the specific benchmark, we contribute a domain-agnostic methodology for constructing scalable and contamination-resistant evaluations wherever synthetic documents can be generated from structured ground truth. --- ### Directional Attractors in LLM Reasoning: How Similarity Retrieval Steers Iterative Summarization Based Reasoning Authors: Cagatay Tekin, Charbel Barakat, Luis Joseph Luna Limgenco Published: 2025-12-22 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.08846v1 Iterative summarization based reasoning frameworks such as InftyThink enable long-horizon reasoning in large language models (LLMs) by controlling context growth, but they repeatedly regenerate similar reasoning strategies across tasks. We introduce InftyThink with Cross-Chain Memory, an extension that augments iterative reasoning with an embedding-based semantic cache of previously successful reasoning patterns. At each reasoning step, the model retrieves and conditions on the most semantically similar stored lemmas, guiding inference without expanding the context window indiscriminately. Experiments on MATH500, AIME2024, and GPQA-Diamond demonstrate that semantic lemma retrieval improves accuracy in structured domains while exposing failure modes in tests that include heterogeneous domains. Geometric analyses of reasoning trajectories reveal that cache retrieval induces directional biases in embedding space, leading to consistent fix (improve baseline accuracy) and break (degradation in baseline accuracy) attractors. Our results highlight both the benefits and limits of similarity-based memory for self-improving LLM reasoning. --- ### Data Work in Egypt: Who Are the Workers Behind Artificial Intelligence? Authors: Myriam Raymond, Lucy Neveux, Antonio A. Casilli, Paola Tubaro Published: 2025-12-22 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2601.06057v2 The report highlights the role of Egyptian data workers in the global value chains of Artificial Intelligence (AI). These workers generate and annotate data for machine learning, check outputs, and they connect with overseas AI producers via international digital labor platforms, where they perform on-demand tasks and are typically paid by piecework, with no long-term commitment. Most of these workers are young, highly educated men, with nearly two-thirds holding undergraduate degrees. Their primary motivation for data work is financial need, with three-quarters relying on platform earnings to cover basic necessities. Despite the variability in their online earnings, these are generally low, often equaling Egypt's minimum wage. Data workers' digital identities are shaped by algorithmic control and economic demands, often diverging from their offline selves. Nonetheless, they find ways to resist, exercise ethical agency, and maintain autonomy. The report evaluates the potential impact of Egypt's newly enacted labor law and suggests policy measures to improve working conditions and acknowledge the role of these workers in AI's global value chains. --- ### Using street view images and visual LLMs to predict heritage values for governance support: Risks, ethics, and policy implications Authors: Tim Johansson, Mikael Mangold, Kristina Dabrock, Anna Donarelli, Ingrid Campo-Ruiz Published: 2025-12-22 Categories: cs.CY, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2601.06056v2 During 2025 and 2026, the Energy Performance of Buildings Directive is being implemented in the European Union member states, requiring all member states to have National Building Renovation Plans. In Sweden, there is no comprehensive national register of buildings with heritage values. This is seen as a barrier for the analyses underlying the development of Building Renovation Plans by the involved Swedish authorities. The purpose of this research was to assist Swedish authorities in developing information on heritage values in the Swedish building stock. Buildings in street view images from all over Sweden (N=154 710) have been analysed using multimodal Large Language Models (LLM) to assess visible aspects indicative of heritage value. Zero-shot predictions by LLMs were used as a basis for identifying buildings with potential heritage values for 5.0 million square meters of heated floor area. In this paper, the results of the predictions and lessons learned are presented and related to the development of the Swedish Building Renovation Plan as part of governance. The problems with the method and potential improvements are discussed. Risks with authorities use of LLM-based data are addressed, with a focus on issues of transparency, error detection and sycophancy. --- ### A Multi-Perspective Benchmark and Moderation Model for Evaluating Safety and Adversarial Robustness Authors: Naseem Machlovi, Maryam Saleki, Ruhul Amin, Mohamed Rahouti, Shawqi Al-Maliki, Junaid Qadir, Mohamed M. Abdallah, Ala Al-Fuqaha Published: 2025-12-22 Categories: cs.CL, cs.AI, cs.HC, cs.LG arXiv: https://arxiv.org/abs/2601.03273v2 As large language models (LLMs) become deeply embedded in daily life, the urgent need for safer moderation systems that distinguish between naive and harmful requests while upholding appropriate censorship boundaries has never been greater. While existing LLMs can detect dangerous or unsafe content, they often struggle with nuanced cases such as implicit offensiveness, subtle gender and racial biases, and jailbreak prompts, due to the subjective and context-dependent nature of these issues. Furthermore, their heavy reliance on training data can reinforce societal biases, resulting in inconsistent and ethically problematic outputs. To address these challenges, we introduce GuardEval, a unified multi-perspective benchmark dataset designed for both training and evaluation, containing 106 fine-grained categories spanning human emotions, offensive and hateful language, gender and racial bias, and broader safety concerns. We also present GemmaGuard (GGuard), a Quantized Low-Rank Adaptation (QLoRA), fine-tuned version of Gemma3-12B trained on GuardEval, to assess content moderation with fine-grained labels. Our evaluation shows that GGuard achieves a macro F1 score of 0.832, substantially outperforming leading moderation models, including OpenAI Moderator (0.64) and Llama Guard (0.61). We show that multi-perspective, human-centered safety benchmarks are critical for mitigating inconsistent moderation decisions. GuardEval and GGuard together demonstrate that diverse, representative data materially improve safety, and adversarial robustness on complex, borderline cases. --- ### Less is more: Not all samples are effective for evaluation Authors: Wentang Song, Jinqiang Li, Kele Huang, Junhui Lin, Shengxiang Wu, Zhongshi Xie Published: 2025-12-22 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.03272v1 The versatility of Large Language Models (LLMs) in vertical domains has spurred the development of numerous specialized evaluation benchmarks. However, these benchmarks often suffer from significant semantic redundancy and impose high computational costs during evaluation. Existing compression methods, such as tinyBenchmarks depend critically on correctness labels from multiple historical models evaluated on the full test set, making them inapplicable in cold-start scenarios, such as the introduction of a new task, domain, or model with no prior evaluation history. To address this limitation, we propose a history-free test set compression framework that requires no prior model performance data. Our method begins by fine-tuning a base LLM on a small amount of domain-specific data to internalize task-relevant semantics. It then generates high-level semantic embeddings for all original test samples using only their raw textual content. In this domain-adapted embedding space, we perform task-aware clustering and introduce a novel dataset X-ray mechanism that analyzes cluster geometry to dynamically calibrate the compression intensity based on the intrinsic redundancy of the benchmark. Experiments on professional-domain dataset, notably a large-scale 3GPP communications benchmark, demonstrate that our approach effectively identifies and removes redundant samples, reducing evaluation cost by over 90% while preserving high fidelity to the full benchmark. --- ### Agentic AI for Autonomous, Explainable, and Real-Time Credit Risk Decision-Making Authors: Chandra Sekhar Kubam Published: 2025-12-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.00818v1 Significant digitalization of financial services in a short period of time has led to an urgent demand to have autonomous, transparent and real-time credit risk decision making systems. The traditional machine learning models are effective in pattern recognition, but do not have the adaptive reasoning, situational awareness, and autonomy needed in modern financial operations. As a proposal, this paper presents an Agentic AI framework, or a system where AI agents view the world of dynamic credit independent of human observers, who then make actions based on their articulable decision-making paths. The research introduces a multi-agent system with reinforcing learning, natural language reasoning, explainable AI modules, and real-time data absorption pipelines as a means of assessing the risk profiles of borrowers with few humans being involved. The processes consist of agent collaboration protocol, risk-scoring engines, interpretability layers, and continuous feedback learning cycles. Findings indicate that decision speed, transparency and responsiveness is better than traditional credit scoring models. Nevertheless, there are still some practical limitations such as risks of model drift, inconsistencies in interpreting high dimensional data and regulatory uncertainties as well as infrastructure limitations in low-resource settings. The suggested system has a high prospective to transform credit analytics and future studies ought to be directed on dynamic regulatory compliance mobilizers, new agent teamwork, adversarial robustness, and large-scale implementation in cross-country credit ecosystems. --- ### MathLedger: A Verifiable Learning Substrate with Ledger-Attested Feedback Authors: Ismail Ahmad Abdullah Published: 2025-12-22 Categories: cs.AI, cs.CR, cs.LG arXiv: https://arxiv.org/abs/2601.00816v1 Contemporary AI systems achieve extraordinary performance yet remain opaque and non-verifiable, creating a crisis of trust for safety-critical deployment. We introduce MathLedger, a substrate for verifiable machine cognition that integrates formal verification, cryptographic attestation, and learning dynamics into a single epistemic loop. The system implements Reflexive Formal Learning (RFL), a symbolic analogue of gradient descent where updates are driven by verifier outcomes rather than statistical loss. Phase I experiments validate the measurement and governance substrate under controlled conditions. CAL-EXP-3 validates measurement infrastructure (Delta p computation, variance tracking); separate stress tests confirm fail-closed governance triggers correctly under out-of-bounds conditions. No convergence or capability claims are made. The contribution is infrastructural: a working prototype of ledger-attested learning that enables auditability at scale. Keywords: verifiable learning, formal verification, cryptographic attestation, reflexive feedback, fail-closed governance --- ### Semantic Alignment of Multilingual Knowledge Graphs via Contextualized Vector Projections Authors: Abhishek Kumar Published: 2025-12-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.00814v2 The paper presents our work on cross-lingual ontology alignment system which uses embedding based cosine similarity matching. The ontology entities are made contextually richer by creating descriptions using novel techniques. We use a fine-tuned transformer based multilingual model for generating better embeddings. We use cosine similarity to find positive ontology entities pairs and then apply threshold filtering to retain only highly similar entities. We have evaluated our work on OAEI-2022 multifarm track. We achieve 71% F1 score (78% recall and 65% precision) on the evaluation dataset, 16% increase from best baseline score. This suggests that our proposed alignment pipeline is able to capture the subtle cross-lingual similarities. --- ### Free Energy-Based Modeling of Emotional Dynamics in Video Advertisements Authors: Takashi Ushio, Kazuhiro Onishi, Hideyoshi Yanagisawa Published: 2025-12-22 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2601.00812v1 Emotional responses during advertising video viewing are recognized as essential for understanding media effects because they have influenced attention, memory, and purchase intention. To establish a methodological basis for explainable emotion estimation without relying on external information such as physiological signals or subjective ratings, we have quantified "pleasantness," "surprise," and "habituation" solely from scene-level expression features of advertising videos, drawing on the free energy(FE) principle, which has provided a unified account of perception, learning, and behavior. In this framework, Kullback-Leibler divergence (KLD) has captured prediction error, Bayesian surprise (BS) has captured belief updates, and uncertainty (UN) has reflected prior ambiguity, and together they have formed the core components of FE. Using 1,059 15 s food video advertisements, the experiments have shown that KLD has reflected "pleasantness" associated with brand presentation, BS has captured "surprise" arising from informational complexity, and UN has reflected "surprise" driven by uncertainty in element types and spatial arrangements, as well as by the variability and quantity of presented elements. This study also identified three characteristic emotional patterns, namely uncertain stimulus, sustained high emotion, and momentary peak and decay, demonstrating the usefulness of the proposed method. Robustness across nine hyperparameter settings and generalization tests with six types of Japanese advertising videos (three genres and two durations) confirmed that these tendencies remained stable. This work can be extended by integrating a wider range of expression elements and validating the approach through subjective ratings, ultimately guiding the development of technologies that can support the creation of more engaging advertising videos. --- ### Can Large Language Models Improve Venture Capital Exit Timing After IPO? Authors: Mohammadhossien Rashidi Published: 2025-12-22 Categories: q-fin.PM, cs.AI, cs.LG, econ.GN, q-fin.ST arXiv: https://arxiv.org/abs/2601.00810v1 Exit timing after an IPO is one of the most consequential decisions for venture capital (VC) investors, yet existing research focuses mainly on describing when VCs exit rather than evaluating whether those choices are economically optimal. Meanwhile, large language models (LLMs) have shown promise in synthesizing complex financial data and textual information but have not been applied to post-IPO exit decisions. This study introduces a framework that uses LLMs to estimate the optimal time for VC exit by analyzing monthly post IPO information financial performance, filings, news, and market signals and recommending whether to sell or continue holding. We compare these LLM generated recommendations with the actual exit dates observed for VCs and compute the return differences between the two strategies. By quantifying gains or losses associated with following the LLM, this study provides evidence on whether AI-driven guidance can improve exit timing and complements traditional hazard and real-options models in venture capital research. --- ### Toward a Physical Theory of Intelligence Authors: Peter David Fagan Published: 2025-12-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.00021v2 While often treated as abstract algorithmic properties, intelligence and computation are ultimately physical processes constrained by conservation laws. We introduce the Conservation-Congruent Encoding (CCE) framework as a unified, substrate-neutral physical framework for studying intelligence. We propose that information processing emerges when open systems undergo irreversible transitions, carving out macroscopic states from underlying reversible micro-dynamics. Generalizing Landauer's principle to arbitrary conserved quantities via metriplectic flows, we derive a universal bound for macroscopic computation. This yields physical metrics for intelligence and an operational analogue for consciousness, quantifying an agent's ability to extract work from the environment while minimizing its own dissipative dynamics. Applying CCE to the limits of physical observation, we model measurement as an active coarse-graining process rather than a passive projection. At the quantum scale, CCE recovers the Lindblad Master Equation, consistent with modelling decoherence as the dissipative exhaust required to record a measurement. Scaling to cosmological limits, we explore the hypothesis that gravity emerges as the macroscopic geometric footprint of these bounds. We show that, under this hypothesis, measurement-induced dissipation is consistent with a volumetric phase-space collapse, offering a dynamical route to the Bekenstein-Hawking area law. Equating the Landauer exhaust of this coarse-graining to horizon deformation outlines a limiting-case recovery of the Einstein Field Equations. Ultimately, by establishing a substrate-neutral link between thermodynamic dissipation, quantum measurement, and spacetime geometry, CCE provides physical constraints for understanding both natural and artificial intelligence. --- ### Personalized Spiking Neural Networks with Ferroelectric Synapses for EEG Signal Processing Authors: Nikhil Garg, Anxiong Song, Niklas Plessnig, Nathan Savoia, Laura Bégon-Lours Published: 2025-12-22 Categories: cs.NE, cs.AI, cs.ET, cs.LG, eess.SY arXiv: https://arxiv.org/abs/2601.00020v3 Electroencephalography (EEG)-based brain-computer interfaces (BCIs) are strongly affected by non-stationary neural signals that vary across sessions and individuals, limiting the generalization of subject-agnostic models and motivating adaptive and personalized learning on resource-constrained platforms. Programmable memristive hardware offers a promising substrate for such post-deployment adaptation; however, practical realization is challenged by limited weight resolution, device variability, nonlinear programming dynamics, and finite device endurance. In this work, we show that spiking neural networks (SNNs) can be deployed on ferroelectric memristive synaptic devices for adaptive EEG-based motor imagery decoding under realistic device constraints, achieving classification performance comparable to software-based SNNs. We fabricate, characterize, and model the weight update in ferroelectric synapses. We then evaluate the deployment of convolutional-recurrent SNN architecture using two strategies. First, we adapt to SNNs a mixed precision strategy in which gradient-based updates are accumulated digitally and converted into discrete programming events only when a threshold is exceeded. Additionally, the weight update is device-aware and accounts for the nonlinear, state-dependent programming dynamics. During learning and adaptation, this scheme mitigates possible endurance and energy constraints. Second, we evaluate the transfer of software-trained weights followed by low-overhead on-device re-tuning. We show that, subject-specific transfer learning achieved by retraining only the final network layers improves classification accuracy. These results demonstrate that programmable ferroelectric hardware can support robust, low-overhead adaptation in spiking neural networks, opening a practical path toward personalized neuromorphic processing of neural signals. --- ### Müntz-Szász Networks: Neural Architectures with Learnable Power-Law Bases Authors: Gnankan Landry Regis N'guessan Published: 2025-12-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.22222v3 Standard neural network architectures employ fixed activation functions (ReLU, tanh, sigmoid) that are poorly suited for approximating functions with singular or fractional power behavior, a structure that arises ubiquitously in physics, including boundary layers, fracture mechanics, and corner singularities. We introduce Müntz-Szász Networks (MSN), a novel architecture that replaces fixed smooth activations with learnable fractional power bases grounded in classical approximation theory. Each MSN edge computes $φ(x) = \sum_k a_k |x|^{μ_k} + \sum_k b_k \mathrm{sign}(x)|x|^{λ_k}$, where the exponents $\{μ_k, λ_k\}$ are learned alongside the coefficients. We prove that MSN inherits universal approximation from the Müntz-Szász theorem and establish novel approximation rates: for functions of the form $|x|^α$, MSN achieves error $\mathcal{O}(|μ- α|^2)$ with a single learned exponent, whereas standard MLPs require $\mathcal{O}(ε^{-1/α})$ neurons for comparable accuracy. On supervised regression with singular target functions, MSN achieves 5-8x lower error than MLPs with 10x fewer parameters. Physics-informed neural networks (PINNs) represent a particularly demanding application for singular function approximation; on PINN benchmarks including a singular ODE and stiff boundary-layer problems, MSN achieves 3-6x improvement while learning interpretable exponents that match the known solution structure. Our results demonstrate that theory-guided architectural design can yield dramatic improvements for scientifically-motivated function classes. --- ### Interpretable and Adaptive Node Classification on Heterophilic Graphs via Combinatorial Scoring and Hybrid Learning Authors: Soroush Vahidi Published: 2025-12-22 Categories: cs.LG, cs.SI arXiv: https://arxiv.org/abs/2512.22221v1 Graph neural networks (GNNs) achieve strong performance on homophilic graphs but often struggle under heterophily, where adjacent nodes frequently belong to different classes. We propose an interpretable and adaptive framework for semi-supervised node classification based on explicit combinatorial inference rather than deep message passing. Our method assigns labels using a confidence-ordered greedy procedure driven by an additive scoring function that integrates class priors, neighborhood statistics, feature similarity, and training-derived label-label compatibility. A small set of transparent hyperparameters controls the relative influence of these components, enabling smooth adaptation between homophilic and heterophilic regimes. We further introduce a validation-gated hybrid strategy in which combinatorial predictions are optionally injected as priors into a lightweight neural model. Hybrid refinement is applied only when it improves validation performance, preserving interpretability when neuralization is unnecessary. All adaptation signals are computed strictly from training data, ensuring a leakage-free evaluation protocol. Experiments on heterophilic and transitional benchmarks demonstrate competitive performance with modern GNNs while offering advantages in interpretability, tunability, and computational efficiency. --- ### On Extending Semantic Abstraction for Efficient Search of Hidden Objects Authors: Tasha Pais, Nikhilesh Belulkar Published: 2025-12-22 Categories: cs.CV, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2512.22220v1 Semantic Abstraction's key observation is that 2D VLMs' relevancy activations roughly correspond to their confidence of whether and where an object is in the scene. Thus, relevancy maps are treated as "abstract object" representations. We use this framework for learning 3D localization and completion for the exclusive domain of hidden objects, defined as objects that cannot be directly identified by a VLM because they are at least partially occluded. This process of localizing hidden objects is a form of unstructured search that can be performed more efficiently using historical data of where an object is frequently placed. Our model can accurately identify the complete 3D location of a hidden object on the first try significantly faster than a naive random search. These extensions to semantic abstraction hope to provide household robots with the skills necessary to save time and effort when looking for lost objects. --- ### MPK: A Compiler and Runtime for Mega-Kernelizing Tensor Programs Authors: Xinhao Cheng, Zhihao Zhang, Yu Zhou, Jianan Ji, Jinchen Jiang, Zepeng Zhao, Ziruo Xiao, Zihao Ye, Yingyi Huang, Ruihang Lai, Hongyi Jin, Bohan Hou et al. Published: 2025-12-22 Categories: cs.DC, cs.LG, cs.PL arXiv: https://arxiv.org/abs/2512.22219v2 We introduce Mirage Persistent Kernel (MPK), the first compiler and runtime system that automatically transforms multi-GPU model inference into a single high-performance mega-kernel. MPK introduces an SM-level graph representation that captures data dependencies at the granularity of individual streaming multiprocessors (SMs), enabling cross-operator software pipelining, \rev{fine-grained overlap of computation and communication, and other optimizations that are infeasible under the conventional kernel-per-operator execution model}. The MPK compiler lowers tensor programs into optimized SM-level task graphs and generates fast CUDA implementations for each task, while the MPK in-kernel parallel runtime executes these tasks within a single persistent mega-kernel using decentralized scheduling across SMs. Together, these components provide end-to-end kernel fusion with minimal developer effort, while preserving the flexibility of existing programming models. Our evaluation shows that MPK significantly outperforms existing kernel-per-operator LLM serving systems, achieving up to 1.7$\times$ lower end-to-end inference latency and pushing LLM inference performance close to the limits of the underlying hardware. MPK is publicly available at https://github.com/mirage-project/mirage. --- ### Towards Signboard-Oriented Visual Question Answering: ViSignVQA Dataset, Method and Benchmark Authors: Hieu Minh Nguyen, Tam Le-Thanh Dang, Kiet Van Nguyen Published: 2025-12-22 Categories: cs.CV, cs.MM arXiv: https://arxiv.org/abs/2512.22218v1 Understanding signboard text in natural scenes is essential for real-world applications of Visual Question Answering (VQA), yet remains underexplored, particularly in low-resource languages. We introduce ViSignVQA, the first large-scale Vietnamese dataset designed for signboard-oriented VQA, which comprises 10,762 images and 25,573 question-answer pairs. The dataset captures the diverse linguistic, cultural, and visual characteristics of Vietnamese signboards, including bilingual text, informal phrasing, and visual elements such as color and layout. To benchmark this task, we adapted state-of-the-art VQA models (e.g., BLIP-2, LaTr, PreSTU, and SaL) by integrating a Vietnamese OCR model (SwinTextSpotter) and a Vietnamese pretrained language model (ViT5). The experimental results highlight the significant role of the OCR-enhanced context, with F1-score improvements of up to 209% when the OCR text is appended to questions. Additionally, we propose a multi-agent VQA framework combining perception and reasoning agents with GPT-4, achieving 75.98% accuracy via majority voting. Our study presents the first large-scale multimodal dataset for Vietnamese signboard understanding. This underscores the importance of domain-specific resources in enhancing text-based VQA for low-resource languages. ViSignVQA serves as a benchmark capturing real-world scene text characteristics and supporting the development and evaluation of OCR-integrated VQA models in Vietnamese. --- ### VLM-PAR: A Vision Language Model for Pedestrian Attribute Recognition Authors: Abdellah Zakaria Sellam, Salah Eddine Bekhouche, Fadi Dornaika, Cosimo Distante, Abdenour Hadid Published: 2025-12-22 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.22217v1 Pedestrian Attribute Recognition (PAR) involves predicting fine-grained attributes such as clothing color, gender, and accessories from pedestrian imagery, yet is hindered by severe class imbalance, intricate attribute co-dependencies, and domain shifts. We introduce VLM-PAR, a modular vision-language framework built on frozen SigLIP 2 multilingual encoders. By first aligning image and prompt embeddings via refining visual features through a compact cross-attention fusion, VLM-PAR achieves significant accuracy improvement on the highly imbalanced PA100K benchmark, setting a new state-of-the-art performance, while also delivering significant gains in mean accuracy across PETA and Market-1501 benchmarks. These results underscore the efficacy of integrating large-scale vision-language pretraining with targeted cross-modal refinement to overcome imbalance and generalization challenges in PAR. --- ### Signal-SGN++: Topology-Enhanced Time-Frequency Spiking Graph Network for Skeleton-Based Action Recognition Authors: Naichuan Zheng, Xiahai Lun, Weiyi Li, Yuchen Du Published: 2025-12-22 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.22214v1 Graph Convolutional Networks (GCNs) demonstrate strong capability in modeling skeletal topology for action recognition, yet their dense floating-point computations incur high energy costs. Spiking Neural Networks (SNNs), characterized by event-driven and sparse activation, offer energy efficiency but remain limited in capturing coupled temporal-frequency and topological dependencies of human motion. To bridge this gap, this article proposes Signal-SGN++, a topology-aware spiking graph framework that integrates structural adaptivity with time-frequency spiking dynamics. The network employs a backbone composed of 1D Spiking Graph Convolution (1D-SGC) and Frequency Spiking Convolution (FSC) for joint spatiotemporal and spectral feature extraction. Within this backbone, a Topology-Shift Self-Attention (TSSA) mechanism is embedded to adaptively route attention across learned skeletal topologies, enhancing graph-level sensitivity without increasing computational complexity. Moreover, an auxiliary Multi-Scale Wavelet Transform Fusion (MWTF) branch decomposes spiking features into multi-resolution temporal-frequency representations, wherein a Topology-Aware Time-Frequency Fusion (TATF) unit incorporates structural priors to preserve topology-consistent spectral fusion. Comprehensive experiments on large-scale benchmarks validate that Signal-SGN++ achieves superior accuracy-efficiency trade-offs, outperforming existing SNN-based methods and achieving competitive results against state-of-the-art GCNs under substantially reduced energy consumption. --- ### On the Existence and Behavior of Secondary Attention Sinks Authors: Jeffrey T. H. Wong, Cheng Zhang, Louis Mahon, Wayne Luk, Anton Isopoussu, Yiren Zhao Published: 2025-12-22 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2512.22213v3 Attention sinks are tokens, often the beginning-of-sequence (BOS) token, that receive disproportionately high attention despite limited semantic relevance. In this work, we identify a class of attention sinks, which we term secondary sinks, that differ fundamentally from the sinks studied in prior works, which we term primary sinks. While prior works have identified that tokens other than BOS can sometimes become sinks, they were found to exhibit properties analogous to the BOS token. Specifically, they emerge at the same layer, persist throughout the network and draw a large amount of attention mass. Whereas, we find the existence of secondary sinks that arise primarily in middle layers and can persist for a variable number of layers, and draw a smaller, but still significant, amount of attention mass. Through extensive experiments across 11 model families, we analyze where these secondary sinks appear, their properties, how they are formed, and their impact on the attention mechanism. Specifically, we show that: (1) these sinks are formed by specific middle-layer MLP modules; these MLPs map token representations to vectors that align with the direction of the primary sink of that layer. (2) The $\ell_2$-norm of these vectors determines the sink score of the secondary sink, and also the number of layers it lasts for, thereby leading to different impacts on the attention mechanisms accordingly. (3) The primary sink weakens in middle layers, coinciding with the emergence of secondary sinks. We observe that in larger-scale models, the location and lifetime of the sinks, together referred to as sink levels, appear in a more deterministic and frequent manner. Specifically, we identify three sink levels in QwQ-32B and six levels in Qwen3-14B. We open-sourced our findings at github.com/JeffreyWong20/Secondary-Attention-Sinks. --- ### Transformer Reconstructed with Dynamic Value Attention Authors: Xiaowei Wang Published: 2025-12-22 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.22212v1 Since transformer was firstly published in 2017, several works have been proposed to optimize it. However, the major structure of transformer remains unchanged, ignoring one of its main intrinsic limitations, which is the same static value is used for every query in a head. Transformer itself tries to solve this problem by implementing multi-head attentions, yet the number of heads is limited by complexity. I propose a method to decide a value for each query dynamically, which could cut down all the redundant heads, keeping only one. Consequently, the following feed forward network could be cut down entirely, as each revised embedding has already fetched enough useful values far beyond the context. As a result, a single-head Dynamic Value Attention (DVA) is all you need in a transformer. According to the experiment, DVA may save 37.6% training time than the original transformer meanwhile increasing the learning capability. --- ### With Great Capabilities Come Great Responsibilities: Introducing the Agentic Risk & Capability Framework for Governing Agentic AI Systems Authors: Shaun Khoo, Jessica Foo, Roy Ka-Wei Lee Published: 2025-12-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.22211v1 Agentic AI systems present both significant opportunities and novel risks due to their capacity for autonomous action, encompassing tasks such as code execution, internet interaction, and file modification. This poses considerable challenges for effective organizational governance, particularly in comprehensively identifying, assessing, and mitigating diverse and evolving risks. To tackle this, we introduce the Agentic Risk \& Capability (ARC) Framework, a technical governance framework designed to help organizations identify, assess, and mitigate risks arising from agentic AI systems. The framework's core contributions are: (1) it develops a novel capability-centric perspective to analyze a wide range of agentic AI systems; (2) it distills three primary sources of risk intrinsic to agentic AI systems - components, design, and capabilities; (3) it establishes a clear nexus between each risk source, specific materialized risks, and corresponding technical controls; and (4) it provides a structured and practical approach to help organizations implement the framework. This framework provides a robust and adaptable methodology for organizations to navigate the complexities of agentic AI, enabling rapid and effective innovation while ensuring the safe, secure, and responsible deployment of agentic AI systems. Our framework is open-sourced \href{https://govtech-responsibleai.github.io/agentic-risk-capability-framework/}{here}. --- ### Toward Equitable Recovery: A Fairness-Aware AI Framework for Prioritizing Post-Flood Aid in Bangladesh Authors: Farjana Yesmin, Romana Akter Published: 2025-12-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.22210v1 Post-disaster aid allocation in developing nations often suffers from systematic biases that disadvantage vulnerable regions, perpetuating historical inequities. This paper presents a fairness-aware artificial intelligence framework for prioritizing post-flood aid distribution in Bangladesh, a country highly susceptible to recurring flood disasters. Using real data from the 2022 Bangladesh floods that affected 7.2 million people and caused 405.5 million US dollars in damages, we develop an adversarial debiasing model that predicts flood vulnerability while actively removing biases against marginalized districts and rural areas. Our approach adapts fairness-aware representation learning techniques from healthcare AI to disaster management, employing a gradient reversal layer that forces the model to learn bias-invariant representations. Experimental results on 87 upazilas across 11 districts demonstrate that our framework reduces statistical parity difference by 41.6 percent, decreases regional fairness gaps by 43.2 percent, and maintains strong predictive accuracy (R-squared=0.784 vs baseline 0.811). The model generates actionable priority rankings ensuring aid reaches the most vulnerable populations based on genuine need rather than historical allocation patterns. This work demonstrates how algorithmic fairness techniques can be effectively applied to humanitarian contexts, providing decision-makers with tools to implement more equitable disaster recovery strategies. --- ### Super-Resolution Enhancement of Medical Images Based on Diffusion Model: An Optimization Scheme for Low-Resolution Gastric Images Authors: Haozhe Jia Published: 2025-12-22 Categories: eess.IV, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2512.22209v1 Capsule endoscopy has enabled minimally invasive gastrointestinal imaging, but its clinical utility is limited by the inherently low resolution of captured images due to hardware, power, and transmission constraints. This limitation hampers the identification of fine-grained mucosal textures and subtle pathological features essential for early diagnosis. This work investigates a diffusion-based super-resolution framework to enhance capsule endoscopy images in a data-driven and anatomically consistent manner. We adopt the SR3 (Super-Resolution via Repeated Refinement) framework built upon Denoising Diffusion Probabilistic Models (DDPMs) to learn a probabilistic mapping from low-resolution to high-resolution images. Unlike GAN-based approaches that often suffer from training instability and hallucination artifacts, diffusion models provide stable likelihood-based training and improved structural fidelity. The HyperKvasir dataset, a large-scale publicly available gastrointestinal endoscopy dataset, is used for training and evaluation. Quantitative results demonstrate that the proposed method significantly outperforms bicubic interpolation and GAN-based super-resolution methods such as ESRGAN, achieving PSNR of 27.5 dB and SSIM of 0.65 for a baseline model, and improving to 29.3 dB and 0.71 with architectural enhancements including attention mechanisms. Qualitative results show improved preservation of anatomical boundaries, vascular patterns, and lesion structures. These findings indicate that diffusion-based super-resolution is a promising approach for enhancing non-invasive medical imaging, particularly in capsule endoscopy where image resolution is fundamentally constrained. --- ### Open-Source Multimodal Moxin Models with Moxin-VLM and Moxin-VLA Authors: Pu Zhao, Arash Akbari, Xuan Shen, Zhenglun Kong, Yixin Shen, Sung-En Chang, Timothy Rupprecht, Lei Lu, Enfu Nan, Changdi Yang, Yumei He, Weiyan Shi et al. Published: 2025-12-22 Categories: cs.CL, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2512.22208v2 Recently, Large Language Models (LLMs) have undergone a significant transformation, marked by a rapid rise in both their popularity and capabilities. Leading this evolution are proprietary LLMs like GPT-4 and GPT-o1, which have captured widespread attention in the AI community due to their remarkable performance and versatility. Simultaneously, open-source LLMs, such as LLaMA and Mistral, have made great contributions to the ever-increasing popularity of LLMs due to the ease to customize and deploy the models across diverse applications. Moxin 7B is introduced as a fully open-source LLM developed in accordance with the Model Openness Framework, which moves beyond the simple sharing of model weights to embrace complete transparency in training, datasets, and implementation detail, thus fostering a more inclusive and collaborative research environment that can sustain a healthy open-source ecosystem. To further equip Moxin with various capabilities in different tasks, we develop three variants based on Moxin, including Moxin-VLM, Moxin-VLA, and Moxin-Chinese, which target the vision-language, vision-language-action, and Chinese capabilities, respectively. Experiments show that our models achieve superior performance in various evaluations. We adopt open-source framework and open data for the training. We release our models, along with the available data and code to derive these models. --- ### GamiBench: Evaluating Spatial Reasoning and 2D-to-3D Planning Capabilities of MLLMs with Origami Folding Tasks Authors: Ryan Spencer, Roey Yaari, Ritvik Vemavarapu, Joyce Yang, Steven Ngo, Utkarsh Sharma Published: 2025-12-22 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.22207v1 Multimodal large language models (MLLMs) are proficient in perception and instruction-following, but they still struggle with spatial reasoning: the ability to mentally track and manipulate objects across multiple views and over time. Spatial reasoning is a key component of human intelligence, but most existing benchmarks focus on static images or final outputs, failing to account for the sequential and viewpoint-dependent nature of this skill. To close this gap, we introduce GamiBench, a benchmark designed to evaluate spatial reasoning and 2D-to-3D planning in MLLMs through origami-inspired folding tasks. GamiBench includes 186 regular and 186 impossible 2D crease patterns paired with their corresponding 3D folded shapes, produced from six distinct viewpoints across three visual question-answering (VQA) tasks: predicting 3D fold configurations, distinguishing valid viewpoints, and detecting impossible patterns. Unlike previous benchmarks that assess only final predictions, GamiBench holistically evaluates the entire reasoning process--measuring cross-view consistency, physical feasibility through impossible-fold detection, and interpretation of intermediate folding steps. It further introduces new diagnostic metrics--viewpoint consistency (VC) and impossible fold selection rate (IFSR)--to measure how well models handle folds of varying complexity. Our experiments show that even leading models such as GPT-5 and Gemini-2.5-Pro struggle on single-step spatial understanding. These contributions establish a standardized framework for evaluating geometric understanding and spatial reasoning in MLLMs. Dataset and code: https://github.com/stvngo/GamiBench. --- ### Reflection-Driven Control for Trustworthy Code Agents Authors: Bin Wang, Jiazheng Quan, Xingrui Yu, Hansen Hu, Yuhao, Ivor Tsang Published: 2025-12-22 Categories: cs.CR, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2512.21354v1 Contemporary large language model (LLM) agents are remarkably capable, but they still lack reliable safety controls and can produce unconstrained, unpredictable, and even actively harmful outputs. To address this, we introduce Reflection-Driven Control, a standardized and pluggable control module that can be seamlessly integrated into general agent architectures. Reflection-Driven Control elevates "self-reflection" from a post hoc patch into an explicit step in the agent's own reasoning process: during generation, the agent continuously runs an internal reflection loop that monitors and evaluates its own decision path. When potential risks are detected, the system retrieves relevant repair examples and secure coding guidelines from an evolving reflective memory, injecting these evidence-based constraints directly into subsequent reasoning steps. We instantiate Reflection-Driven Control in the setting of secure code generation and systematically evaluate it across eight classes of security-critical programming tasks. Empirical results show that Reflection-Driven Control substantially improves the security and policy compliance of generated code while largely preserving functional correctness, with minimal runtime and token overhead. Taken together, these findings indicate that Reflection-Driven Control is a practical path toward trustworthy AI coding agents: it enables designs that are simultaneously autonomous, safer by construction, and auditable. --- ### Mechanism-Based Intelligence (MBI): Differentiable Incentives for Rational Coordination and Guaranteed Alignment in Multi-Agent Systems Authors: Stefano Grassi Published: 2025-12-22 Categories: cs.GT, cs.AI, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2512.20688v1 Autonomous multi-agent systems are fundamentally fragile: they struggle to solve the Hayekian Information problem (eliciting dispersed private knowledge) and the Hurwiczian Incentive problem (aligning local actions with global objectives), making coordination computationally intractable. I introduce Mechanism-Based Intelligence (MBI), a paradigm that reconceptualizes intelligence as emergent from the coordination of multiple "brains", rather than a single one. At its core, the Differentiable Price Mechanism (DPM) computes the exact loss gradient $$ \mathbf{G}_i = - \frac{\partial \mathcal{L}}{\partial \mathbf{x}_i} $$ as a dynamic, VCG-equivalent incentive signal, guaranteeing Dominant Strategy Incentive Compatibility (DSIC) and convergence to the global optimum. A Bayesian extension ensures incentive compatibility under asymmetric information (BIC). The framework scales linearly ($\mathcal{O}(N)$) with the number of agents, bypassing the combinatorial complexity of Dec-POMDPs and is empirically 50x faster than Model-Free Reinforcement Learning. By structurally aligning agent self-interest with collective objectives, it provides a provably efficient, auditable and generalizable approach to coordinated, trustworthy and scalable multi-agent intelligence grounded in economic principles. --- ### PHOTON: Hierarchical Autoregressive Modeling for Lightspeed and Memory-Efficient Language Generation Authors: Yuma Ichikawa, Naoya Takagi, Takumi Nakagawa, Yuzi Kanazawa, Akira Sakai Published: 2025-12-22 Categories: cs.LG, cs.AI, cs.CL, cs.DC arXiv: https://arxiv.org/abs/2512.20687v2 Transformers operate as horizontal token-by-token scanners; at each generation step, attending to an ever-growing sequence of token-level states. This access pattern increases prefill latency and makes long-context decoding more memory-bound, as KV-cache reads and writes dominate inference time over arithmetic operations. We propose Parallel Hierarchical Operation for TOp-down Networks (PHOTON), a hierarchical autoregressive model that replaces horizontal scanning with vertical, multi-resolution context scanning. PHOTON maintains a hierarchy of latent streams: a bottom-up encoder compresses tokens into low-rate contextual states, while lightweight top-down decoders reconstruct fine-grained token representations in parallel. We further introduce recursive generation that updates only the coarsest latent stream and eliminates bottom-up re-encoding. Experimental results show that PHOTON is superior to competitive Transformer-based language models regarding the throughput-quality trade-off, providing advantages in long-context and multi-query tasks. In particular, this reduces decode-time KV-cache traffic, yielding up to $10^{3}\times$ higher throughput per unit memory. --- ### Diffusion Models in Simulation-Based Inference: A Tutorial Review Authors: Jonas Arruda, Niels Bracher, Ullrich Köthe, Jan Hasenauer, Stefan T. Radev Published: 2025-12-22 Categories: stat.ML, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2512.20685v5 Diffusion models have recently emerged as powerful learners for simulation-based inference (SBI), enabling fast and accurate estimation of latent parameters from simulated and real data. Their score-based formulation offers a flexible way to learn conditional or joint distributions over parameters and observations, thereby providing a versatile solution to various modeling problems. In this tutorial review, we synthesize recent developments on diffusion models for SBI, covering design choices for training, inference, and evaluation. We highlight opportunities created by various concepts such as guidance, score composition, flow matching, consistency models, and joint modeling. Furthermore, we discuss how efficiency and statistical accuracy are affected by noise schedules, parameterizations, and samplers. Finally, we illustrate these concepts with case studies across parameter dimensionalities, simulation budgets, and model types, and outline open questions for future research. --- ### Fast and Exact Least Absolute Deviations Line Fitting via Piecewise Affine Lower-Bounding Authors: Stefan Volz, Martin Storath, Andreas Weinmann Published: 2025-12-22 Categories: stat.ML, cs.LG, math.OC arXiv: https://arxiv.org/abs/2512.20682v1 Least-absolute-deviations (LAD) line fitting is robust to outliers but computationally more involved than least squares regression. Although the literature includes linear and near-linear time algorithms for the LAD line fitting problem, these methods are difficult to implement and, to our knowledge, lack maintained public implementations. As a result, practitioners often resort to linear programming (LP) based methods such as the simplex-based Barrodale-Roberts method and interior-point methods, or on iteratively reweighted least squares (IRLS) approximation which does not guarantee exact solutions. To close this gap, we propose the Piecewise Affine Lower-Bounding (PALB) method, an exact algorithm for LAD line fitting. PALB uses supporting lines derived from subgradients to build piecewise-affine lower bounds, and employs a subdivision scheme involving minima of these lower bounds. We prove correctness and provide bounds on the number of iterations. On synthetic datasets with varied signal types and noise including heavy-tailed outliers as well as a real dataset from the NOAA's Integrated Surface Database, PALB exhibits empirical log-linear scaling. It is consistently faster than publicly available implementations of LP based and IRLS based solvers. We provide a reference implementation written in Rust with a Python API. --- ### GIMLET: Generalizable and Interpretable Model Learning through Embedded Thermodynamics Authors: Suguru Shiratori, Elham Kiyani, Khemraj Shukla, George Em Karniadakis Published: 2025-12-22 Categories: physics.flu-dyn, cs.LG arXiv: https://arxiv.org/abs/2512.19936v2 We develop a data-driven framework for discovering constitutive relations in models of fluid flow and scalar transport. Under the assumption that velocity and/or scalar fields are measured, our approach infers unknown closure terms in the governing equations as neural networks. The target to be discovered is the constitutive relations only, while the temporal derivative, convective transport terms, and pressure-gradient term in the governing equations are prescribed. The formulation is rooted in a variational principle from non-equilibrium thermodynamics, where the dynamics is defined by a free-energy functional and a dissipation functional. The unknown constitutive terms arise as functional derivatives of these functionals with respect to the state variables. To enable a flexible and structured model discovery, the free-energy and dissipation functionals are parameterized using neural networks, while their functional derivatives are obtained via automatic differentiation. This construction enforces thermodynamic consistency by design, guaranteeing monotonic decay of the total free energy and non-negative entropy production. The resulting method, termed GIMLET (Generalizable and Interpretable Model Learning through Embedded Thermodynamics), avoids reliance on a predefined library of candidate functions, unlike sparse regression or symbolic identification approaches. The learned models are generalizable in that functionals identified from one dataset can be transferred to distinct datasets governed by the same underlying equations. Moreover, the inferred free-energy and dissipation functions provide direct physical interpretability of the learned dynamics. The framework is demonstrated on several benchmark systems, including the viscous Burgers equation, the Kuramoto--Sivashinsky equation, and the incompressible Navier--Stokes equations for both Newtonian and non-Newtonian fluids. --- ### Conditional Adversarial Fragility in Financial Machine Learning under Macroeconomic Stress Authors: Samruddhi Baviskar Published: 2025-12-22 Categories: cs.LG, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2512.19935v1 Machine learning models used in financial decision systems operate in nonstationary economic environments, yet adversarial robustness is typically evaluated under static assumptions. This work introduces Conditional Adversarial Fragility, a regime dependent phenomenon in which adversarial vulnerability is systematically amplified during periods of macroeconomic stress. We propose a regime aware evaluation framework for time indexed tabular financial classification tasks that conditions robustness assessment on external indicators of economic stress. Using volatility based regime segmentation as a proxy for macroeconomic conditions, we evaluate model behavior across calm and stress periods while holding model architecture, attack methodology, and evaluation protocols constant. Baseline predictive performance remains comparable across regimes, indicating that economic stress alone does not induce inherent performance degradation. Under adversarial perturbations, however, models operating during stress regimes exhibit substantially greater degradation across predictive accuracy, operational decision thresholds, and risk sensitive outcomes. We further demonstrate that this amplification propagates to increased false negative rates, elevating the risk of missed high risk cases during adverse conditions. To complement numerical robustness metrics, we introduce an interpretive governance layer based on semantic auditing of model explanations using large language models. Together, these results demonstrate that adversarial robustness in financial machine learning is a regime dependent property and motivate stress aware approaches to model risk assessment in high stakes financial deployments. --- ### Vehicle-centric Perception via Multimodal Structured Pre-training Authors: Wentao Wu, Xiao Wang, Chenglong Li, Jin Tang, Bin Luo Published: 2025-12-22 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.19934v1 Vehicle-centric perception plays a crucial role in many intelligent systems, including large-scale surveillance systems, intelligent transportation, and autonomous driving. Existing approaches lack effective learning of vehicle-related knowledge during pre-training, resulting in poor capability for modeling general vehicle perception representations. To handle this problem, we propose VehicleMAE-V2, a novel vehicle-centric pre-trained large model. By exploring and exploiting vehicle-related multimodal structured priors to guide the masked token reconstruction process, our approach can significantly enhance the model's capability to learn generalizable representations for vehicle-centric perception. Specifically, we design the Symmetry-guided Mask Module (SMM), Contour-guided Representation Module (CRM) and Semantics-guided Representation Module (SRM) to incorporate three kinds of structured priors into token reconstruction including symmetry, contour and semantics of vehicles respectively. SMM utilizes the vehicle symmetry constraints to avoid retaining symmetric patches and can thus select high-quality masked image patches and reduce information redundancy. CRM minimizes the probability distribution divergence between contour features and reconstructed features and can thus preserve holistic vehicle structure information during pixel-level reconstruction. SRM aligns image-text features through contrastive learning and cross-modal distillation to address the feature confusion caused by insufficient semantic understanding during masked reconstruction. To support the pre-training of VehicleMAE-V2, we construct Autobot4M, a large-scale dataset comprising approximately 4 million vehicle images and 12,693 text descriptions. Extensive experiments on five downstream tasks demonstrate the superior performance of VehicleMAE-V2. --- ### PRISM: A Personality-Driven Multi-Agent Framework for Social Media Simulation Authors: Zhixiang Lu, Xueyuan Deng, Yiran Liu, Yulong Li, Qiang Yan, Imran Razzak, Jionglong Su Published: 2025-12-22 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.19933v1 Traditional agent-based models (ABMs) of opinion dynamics often fail to capture the psychological heterogeneity driving online polarization due to simplistic homogeneity assumptions. This limitation obscures the critical interplay between individual cognitive biases and information propagation, thereby hindering a mechanistic understanding of how ideological divides are amplified. To address this challenge, we introduce the Personality-Refracted Intelligent Simulation Model (PRISM), a hybrid framework coupling stochastic differential equations (SDE) for continuous emotional evolution with a personality-conditional partially observable Markov decision process (PC-POMDP) for discrete decision-making. In contrast to continuous trait approaches, PRISM assigns distinct Myers-Briggs Type Indicator (MBTI) based cognitive policies to multimodal large language model (MLLM) agents, initialized via data-driven priors from large-scale social media datasets. PRISM achieves superior personality consistency aligned with human ground truth, significantly outperforming standard homogeneous and Big Five benchmarks. This framework effectively replicates emergent phenomena such as rational suppression and affective resonance, offering a robust tool for analyzing complex social media ecosystems. --- ### Unified Brain Surface and Volume Registration Authors: S. Mazdak Abulnaga, Andrew Hoopes, Malte Hoffmann, Robin Magnet, Maks Ovsjanikov, Lilla Zöllei, John Guttag, Bruce Fischl, Adrian Dalca Published: 2025-12-22 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.19928v1 Accurate registration of brain MRI scans is fundamental for cross-subject analysis in neuroscientific studies. This involves aligning both the cortical surface of the brain and the interior volume. Traditional methods treat volumetric and surface-based registration separately, which often leads to inconsistencies that limit downstream analyses. We propose a deep learning framework, NeurAlign, that registers $3$D brain MRI images by jointly aligning both cortical and subcortical regions through a unified volume-and-surface-based representation. Our approach leverages an intermediate spherical coordinate space to bridge anatomical surface topology with volumetric anatomy, enabling consistent and anatomically accurate alignment. By integrating spherical registration into the learning, our method ensures geometric coherence between volume and surface domains. In a series of experiments on both in-domain and out-of-domain datasets, our method consistently outperforms both classical and machine learning-based registration methods -- improving the Dice score by up to 7 points while maintaining regular deformation fields. Additionally, it is orders of magnitude faster than the standard method for this task, and is simpler to use because it requires no additional inputs beyond an MRI scan. With its superior accuracy, fast inference, and ease of use, NeurAlign sets a new standard for joint cortical and subcortical registration. --- ### The Seismic Wavefield Common Task Framework Authors: Alexey Yermakov, Yue Zhao, Marine Denolle, Yiyu Ni, Philippe M. Wyder, Judah Goldfeder, Stefano Riva, Jan Williams, David Zoro, Amy Sara Rude, Matteo Tomasetto, Joe Germany et al. Published: 2025-12-22 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.19927v2 Seismology faces fundamental challenges in state forecasting and reconstruction (e.g., earthquake early warning and ground motion prediction) and managing the parametric variability of source locations, mechanisms, and Earth models (e.g., subsurface structure and topography effects). Addressing these with simulations is hindered by their massive scale, both in synthetic data volumes and numerical complexity, while real-data efforts are constrained by models that inadequately reflect the Earth's complexity and by sparse sensor measurements from the field. Recent machine learning (ML) efforts offer promise, but progress is obscured by a lack of proper characterization, fair reporting, and rigorous comparisons. To address this, we introduce a Common Task Framework (CTF) for ML for seismic wavefields, demonstrated here on three distinct wavefield datasets. Our CTF features a curated set of datasets at various scales (global, crustal, and local) and task-specific metrics spanning forecasting, reconstruction, and generalization under realistic constraints such as noise and limited data. Inspired by CTFs in fields like natural language processing, this framework provides a structured and rigorous foundation for head-to-head algorithm evaluation. We evaluate various methods for reconstructing seismic wavefields from sparse sensor measurements, with results illustrating the CTF's utility in revealing strengths, limitations, and suitability for specific problem classes. Our vision is to replace ad hoc comparisons with standardized evaluations on hidden test sets, raising the bar for rigor and reproducibility in scientific ML. --- ### Mitigating LLM Hallucination via Behaviorally Calibrated Reinforcement Learning Authors: Jiayun Wu, Jiashuo Liu, Zhiyuan Zeng, Tianyang Zhan, Tianle Cai, Wenhao Huang Published: 2025-12-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.19920v3 LLM deployment in critical domains is currently impeded by persistent hallucinations--generating plausible but factually incorrect assertions. While scaling laws drove significant improvements in general capabilities, theoretical frameworks suggest hallucination is not merely stochastic error but a predictable statistical consequence of training objectives prioritizing mimicking data distribution over epistemic honesty. Standard RLVR paradigms, utilizing binary reward signals, inadvertently incentivize models as good test-takers rather than honest communicators, encouraging guessing whenever correctness probability exceeds zero. This paper presents an exhaustive investigation into behavioral calibration, which incentivizes models to stochastically admit uncertainty by abstaining when not confident, aligning model behavior with accuracy. Synthesizing recent advances, we propose and evaluate training interventions optimizing strictly proper scoring rules for models to output a calibrated probability of correctness. Our methods enable models to either abstain from producing a complete response or flag individual claims where uncertainty remains. Utilizing Qwen3-4B-Instruct, empirical analysis reveals behavior-calibrated reinforcement learning allows smaller models to surpass frontier models in uncertainty quantification--a transferable meta-skill decouplable from raw predictive accuracy. Trained on math reasoning tasks, our model's log-scale Accuracy-to-Hallucination Ratio gain (0.806) exceeds GPT-5's (0.207) in a challenging in-domain evaluation (BeyondAIME). Moreover, in cross-domain factual QA (SimpleQA), our 4B LLM achieves zero-shot calibration error on par with frontier models including Grok-4 and Gemini-2.5-Pro, even though its factual accuracy is much lower. ---