# Research Papers — 2025-12 · bundle page 106 of 175 > Full-text markdown of papers #5251–5300 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/106 [← Previous bundle](https://aipod.dev/papers/archive/2025-12/page/105.md) [Next bundle →](https://aipod.dev/papers/archive/2025-12/page/107.md) ### Topology-Guided Quantum GANs for Constrained Graph Generation Authors: Tobias Rohe, Markus Baumann, Michael Poppel, Gerhard Stenzel, Maximilian Zorn, Claudia Linnhoff-Popien Published: 2025-12-11 Categories: quant-ph, cs.LG arXiv: https://arxiv.org/abs/2512.10582v1 Quantum computing (QC) promises theoretical advantages, benefiting computational problems that would not be efficiently classically simulatable. However, much of this theoretical speedup depends on the quantum circuit design solving the problem. We argue that QC literature has yet to explore more domain specific ansatz-topologies, instead of relying on generic, one-size-fits-all architectures. In this work, we show that incorporating task-specific inductive biases -- specifically geometric priors -- into quantum circuit design can enhance the performance of hybrid Quantum Generative Adversarial Networks (QuGANs) on the task of generating geometrically constrained K4 graphs. We evaluate a portfolio of entanglement topologies and loss-function designs to assess their impact on both statistical fidelity and compliance with geometric constraints, including the Triangle and Ptolemaic inequalities. Our results show that aligning circuit topology with the underlying problem structure yields substantial benefits: the Triangle-topology QuGAN achieves the highest geometric validity among quantum models and matches the performance of classical Generative Adversarial Networks (GAN). Additionally, we showcase how specific architectural choices, such as entangling gate types, variance regularization and output-scaling govern the trade-off between geometric consistency and distributional accuracy, thus emphasizing the value of structured, task-aware quantum ansatz-topologies. --- ### Simpler and Stronger Baselines for All-in-One Image Restoration Authors: Hongyu Du, Yuqi Wang, Wenlong Jiao, Wenwen Li, DongWei Ren Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10581v2 Current methods for all-in-one image restoration increasingly rely on complex mechanisms, such as Mixture-of-Experts or prompt learning, to handle diverse degradations.In this work, we challenge this trend by showing that structural symmetry provides a strong and effective inductive bias, which is often sufficient to support high-quality restoration in many practical settings. We propose SymUNet, a minimalist baseline that enforces strict encoder-decoder alignment via simple additive fusion, which outperforms complex state-of-the-art models on standard benchmarks while significantly reducing computational costs. Furthermore, to address the negative transfer observed in highly diverse degradation scenarios (e.g., 5-task settings), we introduce SE-SymUNet, which leverages frozen semantic priors (CLIP) to disentangle conflicting restoration tasks. Extensive experiments demonstrate that our approach establishes a simpler, stronger, and more efficient foundation for multi-degradation learning. --- ### Unleashing Degradation-Carrying Features in Symmetric U-Net: Simpler and Stronger Baselines for All-in-One Image Restoration Authors: Wenlong Jiao, Heyang Lee, Ping Wang, Pengfei Zhu, Qinghua Hu, Dongwei Ren Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10581v1 All-in-one image restoration aims to handle diverse degradations (e.g., noise, blur, adverse weather) within a unified framework, yet existing methods increasingly rely on complex architectures (e.g., Mixture-of-Experts, diffusion models) and elaborate degradation prompt strategies. In this work, we reveal a critical insight: well-crafted feature extraction inherently encodes degradation-carrying information, and a symmetric U-Net architecture is sufficient to unleash these cues effectively. By aligning feature scales across encoder-decoder and enabling streamlined cross-scale propagation, our symmetric design preserves intrinsic degradation signals robustly, rendering simple additive fusion in skip connections sufficient for state-of-the-art performance. Our primary baseline, SymUNet, is built on this symmetric U-Net and achieves better results across benchmark datasets than existing approaches while reducing computational cost. We further propose a semantic enhanced variant, SE-SymUNet, which integrates direct semantic injection from frozen CLIP features via simple cross-attention to explicitly amplify degradation priors. Extensive experiments on several benchmarks validate the superiority of our methods. Both baselines SymUNet and SE-SymUNet establish simpler and stronger foundations for future advancements in all-in-one image restoration. The source code is available at https://github.com/WenlongJiao/SymUNet. --- ### RoleRMBench & RoleRM: Towards Reward Modeling for Profile-Based Role Play in Dialogue Systems Authors: Hang Ding, Qiming Feng, Dongqi Liu, Qi Zhao, Tao Yao, Shuo Wang, Dongsheng Chen, Jian Li, Zhenye Gan, Jiangning Zhang, Chengjie Wang, Yabiao Wang Published: 2025-12-11 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.10575v1 Reward modeling has become a cornerstone of aligning large language models (LLMs) with human preferences. Yet, when extended to subjective and open-ended domains such as role play, existing reward models exhibit severe degradation, struggling to capture nuanced and persona-grounded human judgments. To address this gap, we introduce RoleRMBench, the first systematic benchmark for reward modeling in role-playing dialogue, covering seven fine-grained capabilities from narrative management to role consistency and engagement. Evaluation on RoleRMBench reveals large and consistent gaps between general-purpose reward models and human judgment, particularly in narrative and stylistic dimensions. We further propose RoleRM, a reward model trained with Continuous Implicit Preferences (CIP), which reformulates subjective evaluation as continuous consistent pairwise supervision under multiple structuring strategies. Comprehensive experiments show that RoleRM surpasses strong open- and closed-source reward models by over 24% on average, demonstrating substantial gains in narrative coherence and stylistic fidelity. Our findings highlight the importance of continuous preference representation and annotation consistency, establishing a foundation for subjective alignment in human-centered dialogue systems. --- ### Is the Information Bottleneck Robust Enough? Towards Label-Noise Resistant Information Bottleneck Learning Authors: Yi Huang, Qingyun Sun, Yisen Gao, Haonan Yuan, Xingcheng Fu, Jianxin Li Published: 2025-12-11 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.10573v1 The Information Bottleneck (IB) principle facilitates effective representation learning by preserving label-relevant information while compressing irrelevant information. However, its strong reliance on accurate labels makes it inherently vulnerable to label noise, prevalent in real-world scenarios, resulting in significant performance degradation and overfitting. To address this issue, we propose LaT-IB, a novel Label-Noise ResistanT Information Bottleneck method which introduces a "Minimal-Sufficient-Clean" (MSC) criterion. Instantiated as a mutual information regularizer to retain task-relevant information while discarding noise, MSC addresses standard IB's vulnerability to noisy label supervision. To achieve this, LaT-IB employs a noise-aware latent disentanglement that decomposes the latent representation into components aligned with to the clean label space and the noise space. Theoretically, we first derive mutual information bounds for each component of our objective including prediction, compression, and disentanglement, and moreover prove that optimizing it encourages representations invariant to input noise and separates clean and noisy label information. Furthermore, we design a three-phase training framework: Warmup, Knowledge Injection and Robust Training, to progressively guide the model toward noise-resistant representations. Extensive experiments demonstrate that LaT-IB achieves superior robustness and efficiency under label noise, significantly enhancing robustness and applicability in real-world scenarios with label noise. --- ### AVI-Edit: Audio-sync Video Instance Editing with Granularity-Aware Mask Refiner Authors: Haojie Zheng, Shuchen Weng, Jingqi Liu, Siqi Yang, Boxin Shi, Xinlong Wang Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10571v4 Recent advancements in video generation highlight that realistic audio-visual synchronization is crucial for engaging content creation. However, existing video editing methods largely overlook audio-visual synchronization and lack the fine-grained spatial and temporal controllability required for precise instance-level edits. In this paper, we propose AVI-Edit, a framework for audio-sync video instance editing. We propose a granularity-aware mask refiner that iteratively refines coarse user-provided masks into precise instance-level regions. We further design a self-feedback audio agent to curate high-quality audio guidance, providing fine-grained temporal control. To facilitate this task, we additionally construct a large-scale dataset with instance-centric correspondence and comprehensive annotations. Extensive experiments demonstrate that AVI-Edit outperforms state-of-the-art methods in visual quality, condition following, and audio-visual synchronization. Project page: https://hjzheng.net/projects/AVI-Edit/. --- ### Audio-sync Video Instance Editing with Granularity-Aware Mask Refiner Authors: Haojie Zheng, Shuchen Weng, Jingqi Liu, Siqi Yang, Boxin Shi, Xinlong Wang Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10571v1 Recent advancements in video generation highlight that realistic audio-visual synchronization is crucial for engaging content creation. However, existing video editing methods largely overlook audio-visual synchronization and lack the fine-grained spatial and temporal controllability required for precise instance-level edits. In this paper, we propose AVI-Edit, a framework for audio-sync video instance editing. We propose a granularity-aware mask refiner that iteratively refines coarse user-provided masks into precise instance-level regions. We further design a self-feedback audio agent to curate high-quality audio guidance, providing fine-grained temporal control. To facilitate this task, we additionally construct a large-scale dataset with instance-centric correspondence and comprehensive annotations. Extensive experiments demonstrate that AVI-Edit outperforms state-of-the-art methods in visual quality, condition following, and audio-visual synchronization. Project page: https://hjzheng.net/projects/AVI-Edit/. --- ### Flexible Deep Neural Networks for Partially Linear Survival Data: Estimation and Survival Inference Authors: Asaf Ben Arie, Malka Gorfine Published: 2025-12-11 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2512.10570v2 We propose a flexible deep neural network (DNN) framework for modeling survival data within a partially linear regression structure. The approach preserves interpretability through a parametric linear component for covariates of primary interest, while a nonparametric DNN component captures complex time-covariate interactions among nuisance variables. We refer to the method as FLEXI-Haz, a FLEXIble Hazard model with a partially linear structure. In contrast to existing DNN approaches for partially linear Cox models, FLEXI-Haz does not rely on the proportional hazards assumption. We establish theoretical guarantees: the neural network component attains minimax-optimal convergence rates over composite Hölder classes, the linear estimator is sqrt-n-consistent, asymptotically normal, and semiparametrically efficient, and we develop a cross-fitted one-step estimator of the cumulative hazard and survival function for a new subject, together with pointwise asymptotic confidence intervals. To the best of our knowledge, this is the first frequentist asymptotic pointwise inference result for a survival function in a DNN survival model, with or without a linear component. Simulations and real-data analyses demonstrate the utility of FLEXI-Haz as a principled and interpretable alternative to methods based on proportional hazards. --- ### Flexible Deep Neural Networks for Partially Linear Survival Data Authors: Asaf Ben Arie, Malka Gorfine Published: 2025-12-11 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2512.10570v1 We propose a flexible deep neural network (DNN) framework for modeling survival data within a partially linear regression structure. The approach preserves interpretability through a parametric linear component for covariates of primary interest, while a nonparametric DNN component captures complex time-covariate interactions among nuisance variables. We refer to the method as FLEXI-Haz, a flexible hazard model with a partially linear structure. In contrast to existing DNN approaches for partially linear Cox models, FLEXI-Haz does not rely on the proportional hazards assumption. We establish theoretical guarantees: the neural network component attains minimax-optimal convergence rates based on composite Holder classes, and the linear estimator is root-n consistent, asymptotically normal, and semiparametrically efficient. Extensive simulations and real-data analyses demonstrate that FLEXI-Haz provides accurate estimation of the linear effect, offering a principled and interpretable alternative to modern methods based on proportional hazards. Code for implementing FLEXI-Haz, as well as scripts for reproducing data analyses and simulations, is available at: https://github.com/AsafBanana/FLEXI-Haz --- ### NormCode: A Semi-Formal Language for Auditable AI Planning Authors: Xin Guan, Yunshan Li, Zekun Wu, Ruibo Zhang Published: 2025-12-11 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.10563v3 As AI systems move into high stakes domains such as legal reasoning, medical diagnosis, and financial decision making, regulators and practitioners increasingly demand auditability. Auditability means the ability to trace exactly what each step in a multi step workflow saw and did. Current large language model based workflows are fundamentally opaque. Context pollution, defined as the accumulation of information across reasoning steps, causes models to hallucinate and lose track of constraints. At the same time, implicit data flow makes it impossible to reconstruct what any given step actually received as input. We present NormCode, a semi formal language that makes AI workflows auditable by construction. Each inference step operates in enforced data isolation and can access only explicitly passed inputs. This eliminates cross step contamination and ensures that every intermediate state can be inspected. A strict separation between semantic operations, meaning probabilistic language model reasoning, and syntactic operations, meaning deterministic data flow, allows auditors to clearly distinguish inference from mechanical restructuring. The multi format ecosystem, consisting of NCDS, NCD, NCN, and NCDN files, allows developers, domain experts, and auditors to inspect the same plan in formats suited to their individual needs. A four phase compilation pipeline transforms natural language intent into executable JSON repositories. A visual Canvas application provides real time graph visualization and breakpoint debugging. We validate the approach by achieving full accuracy on base X addition and by self hosted execution of the NormCode compiler itself. These results demonstrate that structured intermediate representations can bridge human intuition and machine rigor while maintaining full transparency. --- ### NormCode: A Semi-Formal Language for Context-Isolated AI Planning Authors: Xin Guan Published: 2025-12-11 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.10563v1 Multistep workflows that chain large language model (LLM) calls suffer from context pollution: as information accumulates across steps, models hallucinate, confuse intermediate outputs, and lose track of task constraints. We present NormCode, a semiformal language for constructing plans of inferences, structured decompositions where each step operates in data isolation and receives only explicitly passed inputs, which eliminates crossstep contamination by design. NormCode enforces a strict separation between semantic operations (LLMdriven reasoning, nondeterministic) and syntactic operations (deterministic data restructuring), enabling precise cost and reliability tracing. The language exists in three isomorphic formats: .ncds for human authoring, .ncd for machine execution, and .ncn for human verification, supporting progressive formalization from sketch to production. We validate NormCode through two demonstrations: (1) a base X addition algorithm achieving 100 percent accuracy on arbitrary length inputs, and (2) self hosted execution of NormCode's own five phase compiler pipeline. The working orchestrator provides dependency driven scheduling, SQLite backed checkpointing, and loop management, making AI workflows auditable by design and addressing a critical need for transparency in high stakes domains such as legal reasoning, medical decision making, and financial analysis. --- ### Data-Efficient American Sign Language Recognition via Few-Shot Prototypical Networks Authors: Meher Md Saad Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10562v1 Isolated Sign Language Recognition (ISLR) is critical for bridging the communication gap between the Deaf and Hard-of-Hearing (DHH) community and the hearing world. However, robust ISLR is fundamentally constrained by data scarcity and the long-tail distribution of sign vocabulary, where gathering sufficient examples for thousands of unique signs is prohibitively expensive. Standard classification approaches struggle under these conditions, often overfitting to frequent classes while failing to generalize to rare ones. To address this bottleneck, we propose a Few-Shot Prototypical Network framework adapted for a skeleton based encoder. Unlike traditional classifiers that learn fixed decision boundaries, our approach utilizes episodic training to learn a semantic metric space where signs are classified based on their proximity to dynamic class prototypes. We integrate a Spatiotemporal Graph Convolutional Network (ST-GCN) with a novel Multi-Scale Temporal Aggregation (MSTA) module to capture both rapid and fluid motion dynamics. Experimental results on the WLASL dataset demonstrate the superiority of this metric learning paradigm: our model achieves 43.75% Top-1 and 77.10% Top-5 accuracy on the test set. Crucially, this outperforms a standard classification baseline sharing the identical backbone architecture by over 13%, proving that the prototypical training strategy effectively outperforms in a data scarce situation where standard classification fails. Furthermore, the model exhibits strong zero-shot generalization, achieving nearly 30% accuracy on the unseen SignASL dataset without fine-tuning, offering a scalable pathway for recognizing extensive sign vocabularies with limited data. --- ### Causal Reasoning Favors Encoders: On The Limits of Decoder-Only Models Authors: Amartya Roy, Elamparithy M, Kripabandhu Ghosh, Ponnurangam Kumaraguru, Adrian de Wynter Published: 2025-12-11 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2512.10561v1 In context learning (ICL) underpins recent advances in large language models (LLMs), although its role and performance in causal reasoning remains unclear. Causal reasoning demands multihop composition and strict conjunctive control, and reliance on spurious lexical relations of the input could provide misleading results. We hypothesize that, due to their ability to project the input into a latent space, encoder and encoder decoder architectures are better suited for said multihop conjunctive reasoning versus decoder only models. To do this, we compare fine-tuned versions of all the aforementioned architectures with zero and few shot ICL in both natural language and non natural language scenarios. We find that ICL alone is insufficient for reliable causal reasoning, often overfocusing on irrelevant input features. In particular, decoder only models are noticeably brittle to distributional shifts, while finetuned encoder and encoder decoder models can generalize more robustly across our tests, including the non natural language split. Both architectures are only matched or surpassed by decoder only architectures at large scales. We conclude by noting that for cost effective, short horizon robust causal reasoning, encoder or encoder decoder architectures with targeted finetuning are preferable. --- ### Grounding Everything in Tokens for Multimodal Large Language Models Authors: Xiangxuan Ren, Zhongdao Wang, Liping Hou, Pin Tang, Guoqing Wang, Chao Ma Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10554v2 Multimodal large language models (MLLMs) have made significant advancements in vision understanding and reasoning. However, the autoregressive Transformer architecture used by MLLMs requries tokenization on input images, which limits their ability to accurately ground objects within the 2D image space. This raises an important question: how can sequential language tokens be improved to better ground objects in 2D spatial space for MLLMs? To address this, we present a spatial representation method for grounding objects, namely GETok, that integrates a specialized vocabulary of learnable tokens into MLLMs. GETok first uses grid tokens to partition the image plane into structured spatial anchors, and then exploits offset tokens to enable precise and iterative refinement of localization predictions. By embedding spatial relationships directly into tokens, GETok significantly advances MLLMs in native 2D space reasoning without modifying the autoregressive architecture. Extensive experiments demonstrate that GETok achieves superior performance over the state-of-the-art methods across various referring tasks in both supervised fine-tuning and reinforcement learning settings. --- ### Grounding Everything in Tokens for Multimodal Large Language Models Authors: Xiangxuan Ren, Zhongdao Wang, Liping Hou, Pin Tang, Guoqing Wang, Chao Ma Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10554v1 Multimodal large language models (MLLMs) have made significant advancements in vision understanding and reasoning. However, the autoregressive Transformer architecture used by MLLMs requries tokenization on input images, which limits their ability to accurately ground objects within the 2D image space. This raises an important question: how can sequential language tokens be improved to better ground objects in 2D spatial space for MLLMs? To address this, we present a spatial representation method for grounding objects, namely GETok, that integrates a specialized vocabulary of learnable tokens into MLLMs. GETok first uses grid tokens to partition the image plane into structured spatial anchors, and then exploits offset tokens to enable precise and iterative refinement of localization predictions. By embedding spatial relationships directly into tokens, GETok significantly advances MLLMs in native 2D space reasoning without modifying the autoregressive architecture. Extensive experiments demonstrate that GETok achieves superior performance over the state-of-the-art methods across various referring tasks in both supervised fine-tuning and reinforcement learning settings. --- ### LLM-Auction: Generative Auction towards LLM-Native Advertising Authors: Chujie Zhao, Qun Hu, Shiping Song, Dagui Chen, Han Zhu, Jian Xu, Bo Zheng Published: 2025-12-11 Categories: cs.GT, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.10551v2 The commercialization of LLM applications is the next frontier in online advertising, with LLM-native advertising emerging as a promising paradigm by integrating ads into LLM-generated content. However, classic mechanisms are no longer applicable in this setting where the auction object is shifted from discrete ad slots to distributions over LLM outputs, and existing methods are impractical in industrial scenarios due to ignored externalities or high inference costs. To address these issues, we propose LLM-Auction, the first learning-based generative auction mechanism that integrates auction and generation. By formulating the allocation as preference alignment between LLM outputs and a mechanism objective that balances advertisers' value and user experience, we optimize the LLMs to inherently model allocation externalities without extra inference cost. Theoretically, we identify the allocation monotonicity and continuity of LLM-Auction, and prove that a simple first-price payment rule exhibits favorable incentive properties. Furthermore, we build an LLM-as-a-judge simulation environment for quantitative evaluation, and experiments demonstrate that LLM-Auction achieves the state-of-the-art allocation efficiency while satisfying key mechanism properties. --- ### LLM-Auction: Generative Auction towards LLM-Native Advertising Authors: Chujie Zhao, Qun Hu, Shiping Song, Dagui Chen, Han Zhu, Jian Xu, Bo Zheng Published: 2025-12-11 Categories: cs.GT, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.10551v1 The rapid advancement of large language models (LLMs) necessitates novel monetization strategies, among which LLM-native advertising has emerged as a promising paradigm by naturally integrating advertisement within LLM-generated responses. However, this paradigm fundamentally shifts the auction object from discrete ad slots to the distribution over LLM outputs, posing new challenges for designing auction mechanisms. Existing mechanisms for LLM-native advertising adopt frameworks that decouple auction and generation, which either ignore externalities or require multiple LLM inferences for ad allocation, rendering them impractical for industrial scenarios. To address these challenges, we propose LLM-Auction, which to the best of our knowledge is the first learning-based generative auction mechanism that integrates auction and LLM generation for LLM-native advertising. By formulating the allocation optimization as a preference alignment problem between LLM outputs and the mechanism's objective which reflects both advertisers' expected value and user experience, we introduce Iterative Reward-Preference Optimization (IRPO) algorithm that alternately optimizes the reward model and the LLM. This approach enables the LLM to inherently model allocation externalities without any extra inference cost. We further identify the allocation monotonicity and continuity of LLM-Auction, which allows us to prove that a simple first-price payment rule exhibits favorable incentive properties. Additionally, we design an LLM-as-a-judge simulation environment to facilitate large-scale data construction and enable comprehensive quantitative evaluation of the mechanism's performance. Extensive quantitative and qualitative experiments demonstrate that LLM-Auction significantly outperforms existing baselines in allocation efficiency, while achieving the desired mechanism properties. --- ### Blink: Dynamic Visual Token Resolution for Enhanced Multimodal Understanding Authors: Yuchen Feng, Zhenyu Zhang, Naibin Gu, Yilong Chen, Peng Fu, Zheng Lin, Shuohuan Wang, Yu Sun, Hua Wu, Weiping Wang, Haifeng Wang Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10548v3 Multimodal large language models (MLLMs) have achieved remarkable progress on various vision-language tasks, yet their visual perception remains limited. Humans, in comparison, perceive complex scenes efficiently by dynamically scanning and focusing on salient regions in a sequential "blink-like" process. Motivated by this strategy, we first investigate whether MLLMs exhibit similar behavior. Our pilot analysis reveals that MLLMs naturally attend to different visual regions across layers and that selectively allocating more computation to salient tokens can enhance visual perception. Building on this insight, we propose Blink, a dynamic visual token resolution framework that emulates the human-inspired process within a single forward pass. Specifically, Blink includes two modules: saliency-guided scanning and dynamic token resolution. It first estimates the saliency of visual tokens in each layer based on the attention map, and extends important tokens through a plug-and-play token super-resolution (TokenSR) module. In the next layer, it drops the extended tokens when they lose focus. This dynamic mechanism balances broad exploration and fine-grained focus, thereby enhancing visual perception adaptively and efficiently. Extensive experiments validate Blink, demonstrating its effectiveness in enhancing visual perception and multimodal understanding. --- ### Blink: Dynamic Visual Token Resolution for Enhanced Multimodal Understanding Authors: Yuchen Feng, Zhenyu Zhang, Naibin Gu, Yilong Chen, Peng Fu, Zheng Lin, Shuohuan Wang, Yu Sun, Hua Wu, Weiping Wang, Haifeng Wang Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10548v1 Multimodal large language models (MLLMs) have achieved remarkable progress on various vision-language tasks, yet their visual perception remains limited. Humans, in comparison, perceive complex scenes efficiently by dynamically scanning and focusing on salient regions in a sequential "blink-like" process. Motivated by this strategy, we first investigate whether MLLMs exhibit similar behavior. Our pilot analysis reveals that MLLMs naturally attend to different visual regions across layers and that selectively allocating more computation to salient tokens can enhance visual perception. Building on this insight, we propose Blink, a dynamic visual token resolution framework that emulates the human-inspired process within a single forward pass. Specifically, Blink includes two modules: saliency-guided scanning and dynamic token resolution. It first estimates the saliency of visual tokens in each layer based on the attention map, and extends important tokens through a plug-and-play token super-resolution (TokenSR) module. In the next layer, it drops the extended tokens when they lose focus. This dynamic mechanism balances broad exploration and fine-grained focus, thereby enhancing visual perception adaptively and efficiently. Extensive experiments validate Blink, demonstrating its effectiveness in enhancing visual perception and multimodal understanding. --- ### Unlocking the Address Book: Dissecting the Sparse Semantic Structure of LLM Key-Value Caches via Sparse Autoencoders Authors: Qingsen Ma, Dianyun Wang, Jiaming Lyu, Yaoye Wang, Lechen Ning, Sujie Zhu, Zhenbo Xu, Liuyu Xiang, Huining Li, Huijia Wu, Zhaofeng He Published: 2025-12-11 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.10547v1 The Key-Value (KV) cache is the primary memory bottleneck in long-context Large Language Models, yet it is typically treated as an opaque numerical tensor. In this work, we propose \textbf{STA-Attention}, a framework that utilizes Top-K Sparse Autoencoders (SAEs) to decompose the KV cache into interpretable ``semantic atoms.'' Unlike standard $L_1$-regularized SAEs, our Top-K approach eliminates shrinkage bias, preserving the precise dot-product geometry required for attention. Our analysis uncovers a fundamental \textbf{Key-Value Asymmetry}: while Key vectors serve as highly sparse routers dominated by a ``Semantic Elbow,'' deep Value vectors carry dense content payloads requiring a larger budget. Based on this structure, we introduce a Dual-Budget Strategy that selectively preserves the most informative semantic components while filtering representational noise. Experiments on Yi-6B, Mistral-7B, Qwen2.5-32B, and others show that our semantic reconstructions maintain perplexity and zero-shot performance comparable to the original models, effectively bridging the gap between mechanistic interpretability and faithful attention modeling. --- ### XDoGE: Multilingual Data Reweighting to Enhance Language Inclusivity in LLMs Authors: Iñaki Lacunza, José Javier Saiz, Alexander Shvets, Aitor Gonzalez-Agirre, Marta Villegas Published: 2025-12-11 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.10545v1 Current large language models (LLMs) are trained on massive amounts of text data, primarily from a few dominant languages. Studies suggest that this over-reliance on high-resource languages, such as English, hampers LLM performance in mid- and low-resource languages. To mitigate this problem, we propose to (i) optimize the language distribution by training a small proxy model within a domain-reweighing DoGE algorithm that we extend to XDoGE for a multilingual setup, and (ii) rescale the data and train a full-size model with the established language weights either from scratch or within a continual pre-training phase (CPT). We target six languages possessing a variety of geographic and intra- and inter-language-family relations, namely, English and Spanish (high-resource), Portuguese and Catalan (mid-resource), Galician and Basque (low-resource). We experiment with Salamandra-2b, which is a promising model for these languages. We investigate the effects of substantial data repetition on minor languages and under-sampling on dominant languages using the IberoBench framework for quantitative evaluation. Finally, we release a new promising IberianLLM-7B-Instruct model centering on Iberian languages and English that we pretrained from scratch and further improved using CPT with the XDoGE weights. --- ### Achieving Olympiad-Level Geometry Large Language Model Agent via Complexity Boosting Reinforcement Learning Authors: Haiteng Zhao, Junhao Shen, Yiming Zhang, Songyang Gao, Kuikun Liu, Tianyou Ma, Fan Zheng, Dahua Lin, Wenwei Zhang, Kai Chen Published: 2025-12-11 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.10534v5 Large language model (LLM) agents exhibit strong mathematical problem-solving abilities and can even solve International Mathematical Olympiad (IMO) level problems with the assistance of formal proof systems. However, due to weak heuristics for auxiliary constructions, AI for geometry problem solving remains dominated by expert models such as AlphaGeometry 2, which rely heavily on large-scale data synthesis and search for both training and evaluation. In this work, we make the first attempt to build a medalist-level LLM agent for geometry and present InternGeometry. InternGeometry overcomes the heuristic limitations in geometry by iteratively proposing propositions and auxiliary constructions, verifying them with a symbolic engine, and reflecting on the engine's feedback to guide subsequent proposals. A dynamic memory mechanism enables InternGeometry to conduct more than two hundred interactions with the symbolic engine per problem. To further accelerate learning, we introduce Complexity-Boosting Reinforcement Learning (CBRL), which gradually increases the complexity of synthesized problems across training stages. Built on InternThinker-32B, InternGeometry solves 44 of 50 IMO geometry problems (2000-2024), exceeding the average gold medalist score (40.9), using only 13K training examples, just 0.004% of the data used by AlphaGeometry 2, demonstrating the potential of LLM agents on expert-level geometry tasks. InternGeometry can also propose novel auxiliary constructions for IMO problems that do not appear in human solutions. --- ### Achieving Olympia-Level Geometry Large Language Model Agent via Complexity Boosting Reinforcement Learning Authors: Haiteng Zhao, Junhao Shen, Yiming Zhang, Songyang Gao, Kuikun Liu, Tianyou Ma, Fan Zheng, Dahua Lin, Wenwei Zhang, Kai Chen Published: 2025-12-11 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.10534v1 Large language model (LLM) agents exhibit strong mathematical problem-solving abilities and can even solve International Mathematical Olympiad (IMO) level problems with the assistance of formal proof systems. However, due to weak heuristics for auxiliary constructions, AI for geometry problem solving remains dominated by expert models such as AlphaGeometry 2, which rely heavily on large-scale data synthesis and search for both training and evaluation. In this work, we make the first attempt to build a medalist-level LLM agent for geometry and present InternGeometry. InternGeometry overcomes the heuristic limitations in geometry by iteratively proposing propositions and auxiliary constructions, verifying them with a symbolic engine, and reflecting on the engine's feedback to guide subsequent proposals. A dynamic memory mechanism enables InternGeometry to conduct more than two hundred interactions with the symbolic engine per problem. To further accelerate learning, we introduce Complexity-Boosting Reinforcement Learning (CBRL), which gradually increases the complexity of synthesized problems across training stages. Built on InternThinker-32B, InternGeometry solves 44 of 50 IMO geometry problems (2000-2024), exceeding the average gold medalist score (40.9), using only 13K training examples, just 0.004% of the data used by AlphaGeometry 2, demonstrating the potential of LLM agents on expert-level geometry tasks. InternGeometry can also propose novel auxiliary constructions for IMO problems that do not appear in human solutions. We will release the model, data, and symbolic engine to support future research. --- ### Inverse problems with diffusion models: MAP estimation via mode-seeking loss Authors: Sai Bharath Chandra Gutha, Ricardo Vinuesa, Hossein Azizpour Published: 2025-12-11 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2512.10524v2 A pre-trained unconditional diffusion model, combined with posterior sampling or maximum a posteriori (MAP) estimation techniques, can solve arbitrary inverse problems without task-specific training or fine-tuning. However, existing posterior sampling and MAP estimation methods often rely on modeling approximations and can also be computationally demanding. In this work, we propose a new MAP estimation strategy for solving inverse problems with a pre-trained unconditional diffusion model. Specifically, we introduce the variational mode-seeking loss (VML) and show that its minimization at each reverse diffusion step guides the generated sample towards the MAP estimate (modes in practice). VML arises from a novel perspective of minimizing the Kullback-Leibler (KL) divergence between the diffusion posterior $p(\mathbf{x}_0|\mathbf{x}_t)$ and the measurement posterior $p(\mathbf{x}_0|\mathbf{y})$, where $\mathbf{y}$ denotes the measurement. Importantly, for linear inverse problems, VML can be analytically derived without any modeling approximations. Based on further theoretical insights, we propose VML-MAP, an empirically effective algorithm for solving inverse problems via VML minimization, and validate its efficacy in both performance and computational time through extensive experiments on diverse image-restoration tasks across multiple datasets. --- ### Mode-Seeking for Inverse Problems with Diffusion Models Authors: Sai Bharath Chandra Gutha, Ricardo Vinuesa, Hossein Azizpour Published: 2025-12-11 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2512.10524v1 A pre-trained unconditional diffusion model, combined with posterior sampling or maximum a posteriori (MAP) estimation techniques, can solve arbitrary inverse problems without task-specific training or fine-tuning. However, existing posterior sampling and MAP estimation methods often rely on modeling approximations and can be computationally demanding. In this work, we propose the variational mode-seeking loss (VML), which, when minimized during each reverse diffusion step, guides the generated sample towards the MAP estimate. VML arises from a novel perspective of minimizing the Kullback-Leibler (KL) divergence between the diffusion posterior $p(\mathbf{x}_0|\mathbf{x}_t)$ and the measurement posterior $p(\mathbf{x}_0|\mathbf{y})$, where $\mathbf{y}$ denotes the measurement. Importantly, for linear inverse problems, VML can be analytically derived and need not be approximated. Based on further theoretical insights, we propose VML-MAP, an empirically effective algorithm for solving inverse problems, and validate its efficacy over existing methods in both performance and computational time, through extensive experiments on diverse image-restoration tasks across multiple datasets. --- ### Disentangled and Distilled Encoder for Out-of-Distribution Reasoning with Rademacher Guarantees Authors: Zahra Rahiminasab, Michael Yuhas, Arvind Easwaran Published: 2025-12-11 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.10522v1 Recently, the disentangled latent space of a variational autoencoder (VAE) has been used to reason about multi-label out-of-distribution (OOD) test samples that are derived from different distributions than training samples. Disentangled latent space means having one-to-many maps between latent dimensions and generative factors or important characteristics of an image. This paper proposes a disentangled distilled encoder (DDE) framework to decrease the OOD reasoner size for deployment on resource-constrained devices while preserving disentanglement. DDE formalizes student-teacher distillation for model compression as a constrained optimization problem while preserving disentanglement with disentanglement constraints. Theoretical guarantees for disentanglement during distillation based on Rademacher complexity are established. The approach is evaluated empirically by deploying the compressed model on an NVIDIA --- ### Take a Peek: Efficient Encoder Adaptation for Few-Shot Semantic Segmentation via LoRA Authors: Pasquale De Marinis, Gennaro Vessio, Giovanna Castellano Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10521v3 Few-shot semantic segmentation (FSS) aims to segment novel classes in query images using only a small annotated support set. While prior research has mainly focused on improving decoders, the encoder's limited ability to extract meaningful features for unseen classes remains a key bottleneck. In this work, we introduce \textit{Take a Peek} (TaP), a simple yet effective method that enhances encoder adaptability for both FSS and cross-domain FSS \rev{by inducing a lightweight \textit{feature-space shift} conditioned on the support set}. TaP leverages Low-Rank Adaptation to fine-tune the encoder on the support set with minimal computational overhead, enabling fast adaptation to novel classes while mitigating catastrophic forgetting. Our method is model-agnostic and can be seamlessly integrated into existing FSS pipelines. Extensive experiments across multiple benchmarks--including COCO $20^i$, Pascal $5^i$, and cross-domain datasets such as DeepGlobe, ISIC, and Chest X-ray--demonstrate that TaP consistently improves segmentation performance across diverse models and shot settings. Notably, TaP delivers significant gains in complex multi-class scenarios, highlighting its practical effectiveness in realistic settings. A rank sensitivity analysis also shows that strong performance can be achieved even with low-rank adaptations, thereby ensuring computational efficiency. By addressing a critical limitation in FSS--the encoder's generalization to novel classes--TaP paves the way toward more robust, efficient, and generalizable segmentation systems. The code is available at https://github.com/pasqualedem/TakeAPeek. --- ### Take a Peek: Efficient Encoder Adaptation for Few-Shot Semantic Segmentation via LoRA Authors: Pasquale De Marinis, Gennaro Vessio, Giovanna Castellano Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10521v1 Few-shot semantic segmentation (FSS) aims to segment novel classes in query images using only a small annotated support set. While prior research has mainly focused on improving decoders, the encoder's limited ability to extract meaningful features for unseen classes remains a key bottleneck. In this work, we introduce \textit{Take a Peek} (TaP), a simple yet effective method that enhances encoder adaptability for both FSS and cross-domain FSS (CD-FSS). TaP leverages Low-Rank Adaptation (LoRA) to fine-tune the encoder on the support set with minimal computational overhead, enabling fast adaptation to novel classes while mitigating catastrophic forgetting. Our method is model-agnostic and can be seamlessly integrated into existing FSS pipelines. Extensive experiments across multiple benchmarks--including COCO $20^i$, Pascal $5^i$, and cross-domain datasets such as DeepGlobe, ISIC, and Chest X-ray--demonstrate that TaP consistently improves segmentation performance across diverse models and shot settings. Notably, TaP delivers significant gains in complex multi-class scenarios, highlighting its practical effectiveness in realistic settings. A rank sensitivity analysis also shows that strong performance can be achieved even with low-rank adaptations, ensuring computational efficiency. By addressing a critical limitation in FSS--the encoder's generalization to novel classes--TaP paves the way toward more robust, efficient, and generalizable segmentation systems. The code is available at https://github.com/pasqualedem/TakeAPeek. --- ### 3D Blood Pulsation Maps Authors: Maurice Rohr, Tobias Reinhardt, Tizian Dege, Justus Thies, Christoph Hoog Antink Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10517v1 We present Pulse3DFace, the first dataset of its kind for estimating 3D blood pulsation maps. These maps can be used to develop models of dynamic facial blood pulsation, enabling the creation of synthetic video data to improve and validate remote pulse estimation methods via photoplethysmography imaging. Additionally, the dataset facilitates research into novel multi-view-based approaches for mitigating illumination effects in blood pulsation analysis. Pulse3DFace consists of raw videos from 15 subjects recorded at 30 Hz with an RGB camera from 23 viewpoints, blood pulse reference measurements, and facial 3D scans generated using monocular structure-from-motion techniques. It also includes processed 3D pulsation maps compatible with the texture space of the 3D head model FLAME. These maps provide signal-to-noise ratio, local pulse amplitude, phase information, and supplementary data. We offer a comprehensive evaluation of the dataset's illumination conditions, map consistency, and its ability to capture physiologically meaningful features in the facial and neck skin regions. --- ### Adaptive Replay Buffer for Offline-to-Online Reinforcement Learning Authors: Chihyeon Song, Jaewoo Lee, Jinkyoo Park Published: 2025-12-11 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.10510v2 Offline-to-Online Reinforcement Learning (O2O RL) faces a critical dilemma in balancing the use of a fixed offline dataset with newly collected online experiences. Standard methods, often relying on a fixed data-mixing ratio, struggle to manage the trade-off between early learning stability and asymptotic performance. To overcome this, we introduce the Adaptive Replay Buffer (ARB), a novel approach that dynamically prioritizes data sampling based on a lightweight metric we call 'on-policyness'. Unlike prior methods that rely on complex learning procedures or fixed ratios, ARB is designed to be learning-free and simple to implement, seamlessly integrating into existing O2O RL algorithms. It assesses how closely collected trajectories align with the current policy's behavior and assigns a proportional sampling weight to each transition within that trajectory. This strategy effectively leverages offline data for initial stability while progressively focusing learning on the most relevant, high-rewarding online experiences. Our extensive experiments on D4RL benchmarks demonstrate that ARB consistently mitigates early performance degradation and significantly improves the final performance of various O2O RL algorithms, highlighting the importance of an adaptive, behavior-aware replay buffer design. Our code is publicly available at https://github.com/song970407/ARB. --- ### Adaptive Replay Buffer for Offline-to-Online Reinforcement Learning Authors: Chihyeon Song, Jaewoo Lee, Jinkyoo Park Published: 2025-12-11 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.10510v1 Offline-to-Online Reinforcement Learning (O2O RL) faces a critical dilemma in balancing the use of a fixed offline dataset with newly collected online experiences. Standard methods, often relying on a fixed data-mixing ratio, struggle to manage the trade-off between early learning stability and asymptotic performance. To overcome this, we introduce the Adaptive Replay Buffer (ARB), a novel approach that dynamically prioritizes data sampling based on a lightweight metric we call 'on-policyness'. Unlike prior methods that rely on complex learning procedures or fixed ratios, ARB is designed to be learning-free and simple to implement, seamlessly integrating into existing O2O RL algorithms. It assesses how closely collected trajectories align with the current policy's behavior and assigns a proportional sampling weight to each transition within that trajectory. This strategy effectively leverages offline data for initial stability while progressively focusing learning on the most relevant, high-rewarding online experiences. Our extensive experiments on D4RL benchmarks demonstrate that ARB consistently mitigates early performance degradation and significantly improves the final performance of various O2O RL algorithms, highlighting the importance of an adaptive, behavior-aware replay buffer design. --- ### Hyperspectral Image Data Reduction for Endmember Extraction Authors: Tomohiko Mizutani Published: 2025-12-11 Categories: eess.IV, cs.LG, eess.SP arXiv: https://arxiv.org/abs/2512.10506v3 Endmember extraction from hyperspectral images aims to identify the spectral signatures of materials present in a scene. Recent studies have shown that self-dictionary methods can achieve high extraction accuracy; however, their high computational cost limits their applicability to large-scale hyperspectral images. Although several approaches have been proposed to mitigate this issue, it remains a major challenge. Motivated by this situation, this paper pursues a data reduction approach. Assuming that a hyperspectral image follows the linear mixing model with the pure-pixel assumption, we develop a data reduction technique to remove pixels corresponding to mixtures of multiple endmember signatures. We analyze the theoretical properties of this reduction step and show that it preserves pixels that lie close to the endmembers. Building on this result, we propose a data-reduced self-dictionary method that integrates the data reduction with a self-dictionary method based on a linear programming formulation. Numerical experiments demonstrate that the proposed method can substantially reduce the computational time of the original self-dictionary method without sacrificing endmember extraction accuracy. --- ### Zero-shot 3D Map Generation with LLM Agents: A Dual-Agent Architecture for Procedural Content Generation Authors: Lim Chien Her, Ming Yan, Yunshu Bai, Ruihao Li, Hao Zhang Published: 2025-12-11 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.10501v2 Procedural Content Generation (PCG) offers scalable methods for algorithmically creating complex, customizable worlds. However, controlling these pipelines requires the precise configuration of opaque technical parameters. We propose a training-free architecture that utilizes LLM agents for zero-shot PCG parameter configuration. While Large Language Models (LLMs) promise a natural language interface for PCG tools, off-the-shelf models often fail to bridge the semantic gap between abstract user instructions and strict parameter specifications. Our system pairs an Actor agent with a Critic agent, enabling an iterative workflow where the system autonomously reasons over tool parameters and refines configurations to progressively align with human design preferences. We validate this approach on the generation of various 3D maps, establishing a new benchmark for instruction-following in PCG. Experiments demonstrate that our approach outperforms single-agent baselines, producing diverse and structurally valid environments from natural language descriptions. These results demonstrate that off-the-shelf LLMs can be effectively repurposed as generalized agents for arbitrary PCG tools. By shifting the burden from model training to architectural reasoning, our method offers a scalable framework for mastering complex software without task-specific fine-tuning. --- ### Robust Shape from Focus via Multiscale Directional Dilated Laplacian and Recurrent Network Authors: Khurram Ashfaq, Muhammad Tariq Mahmood Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10498v1 Shape-from-Focus (SFF) is a passive depth estimation technique that infers scene depth by analyzing focus variations in a focal stack. Most recent deep learning-based SFF methods typically operate in two stages: first, they extract focus volumes (a per pixel representation of focus likelihood across the focal stack) using heavy feature encoders; then, they estimate depth via a simple one-step aggregation technique that often introduces artifacts and amplifies noise in the depth map. To address these issues, we propose a hybrid framework. Our method computes multi-scale focus volumes traditionally using handcrafted Directional Dilated Laplacian (DDL) kernels, which capture long-range and directional focus variations to form robust focus volumes. These focus volumes are then fed into a lightweight, multi-scale GRU-based depth extraction module that iteratively refines an initial depth estimate at a lower resolution for computational efficiency. Finally, a learned convex upsampling module within our recurrent network reconstructs high-resolution depth maps while preserving fine scene details and sharp boundaries. Extensive experiments on both synthetic and real-world datasets demonstrate that our approach outperforms state-of-the-art deep learning and traditional methods, achieving superior accuracy and generalization across diverse focal conditions. --- ### UACER: An Uncertainty-Adaptive Critic Ensemble Framework for Robust Adversarial Reinforcement Learning Authors: Jiaxi Wu, Tiantian Zhang, Yuxing Wang, Yongzhe Chang, Xueqian Wang Published: 2025-12-11 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2512.10492v2 Robust adversarial reinforcement learning has emerged as an effective paradigm for training agents to handle uncertain disturbance in real environments, with critical applications in sequential decision-making domains such as autonomous driving and robotic control. Within this paradigm, agent training is typically formulated as a zero-sum Markov game between a protagonist and an adversary to enhance policy robustness. However, the trainable nature of the adversary inevitably induces non-stationarity in the learning dynamics, leading to exacerbated training instability and convergence difficulties, particularly in high-dimensional complex environments. In this paper, we propose a novel approach, Uncertainty-Adaptive Critic Ensemble for robust adversarial Reinforcement learning (UACER), which consists of two components: 1) Diversified critic ensemble: A diverse set of K critic networks is employed in parallel to stabilize Q-value estimation in robust adversarial reinforcement learning, reducing variance and enhancing robustness compared to conventional single-critic designs. 2) Time-varying Decay Uncertainty (TDU) mechanism: Moving beyond simple linear combinations, we propose a variance-derived Q-value aggregation strategy that explicitly incorporates epistemic uncertainty to adaptively regulate the exploration-exploitation trade-off while stabilizing the training process. Comprehensive experiments across several challenging MuJoCo control problems validate the superior effectiveness of UACER, outperforming state-of-the-art methods in terms of overall performance, stability, and efficiency. --- ### From Lab to Reality: A Practical Evaluation of Deep Learning Models and LLMs for Vulnerability Detection Authors: Chaomeng Lu, Bert Lagaisse Published: 2025-12-11 Categories: cs.CR, cs.LG, cs.SE arXiv: https://arxiv.org/abs/2512.10485v3 Vulnerability detection methods based on deep learning (DL) have shown strong performance on benchmark datasets, yet their real-world effectiveness remains underexplored. Recent work suggests that graph neural network-based and transformer-based models, including large language models (LLMs), yield promising results when evaluated on curated benchmark datasets. These datasets are typically characterized by similar data distributions and may contain synthetic samples, heuristic labels, or labeling noise. In this study, we systematically evaluate four representative DL models---Devign, ReVeal, LineVul, and VulBERTa---across four representative datasets: Juliet, Devign, BigVul, and ICVul. Each model is trained independently on each dataset, and the graph-based and CodeBERT representations adopted by these models are analyzed using t-SNE and centroid distance to examine vulnerability-related patterns. To assess realistic applicability, we further evaluate trained ReVeal and LineVul models, along with four open-weight LLMs, on VentiVul, our newly constructed temporally separated out-of-distribution (OOD) dataset comprising 200 recent vulnerabilities from Linux and Chromium. Our experiments reveal that current representation methods struggle to distinguish vulnerable from non-vulnerable code and that trained models generalize poorly across datasets with differing distributions and characteristics. When evaluated on VentiVul, performance drops sharply, with most models failing to detect vulnerabilities reliably or distinguish vulnerable functions from their patched counterparts. These results expose a persistent gap between academic benchmarks and real-world deployment, emphasizing the value of our deployment-oriented evaluation framework and the need for more robust code representations, higher-quality datasets, and evaluation methods that account for vulnerability-fixing changes. --- ### Symphony: A Heuristic Normalized Calibrated Advantage Actor and Critic Algorithm in application for Humanoid Robots Authors: Timur Ishuov, Michele Folgheraiter, Madi Nurmanov, Goncalo Gordo, Richárd Farkas, József Dombi Published: 2025-12-11 Categories: cs.RO, cs.NE arXiv: https://arxiv.org/abs/2512.10477v8 In our work we implicitly suggest that it is a misconception to think that humans learn fast. The learning process takes time. Babies start learning to move in the restricted fluid environment of the womb. Children are often limited by underdeveloped body. Even adults are not allowed to participate in complex competitions right away. However, with robots, when learning from scratch, we often don't have the privilege of waiting for tens of millions of steps. "Swaddling" regularization is responsible for restraining an agent in rapid but unstable development penalizing action strength in a specific way not affecting actions directly. The Symphony, Transitional-policy Deterministic Actor and Critic algorithm, is a concise combination of different ideas for possibility of training humanoid robots from scratch with Sample Efficiency, Sample Proximity and Safety of Actions in mind. It is well known that continuous increase in Gaussian noise without appropriate smoothing is harmful for motors and gearboxes. Compared to Stochastic algorithms, we set limited parametric noise and promote a reduced strength of actions, safely increasing entropy, since the actions are submerged in weaker noise. When actions require more extreme values, actions rise above the weak noise. Training becomes empirically much safer for both the environment around and the robot's mechanisms. We use Fading Replay Buffer: using a fixed formula containing the hyperbolic tangent, we adjust the batch sampling probability: the memory contains a recent memory and a long-term memory trail. Fading Replay Buffer allows us to use Temporal Advantage when we improve the current Critic Network prediction compared to the exponential moving average. Temporal Advantage allows us to update the Actor and Critic in one pass, as well as combine the Actor and Critic in one Object and implement their Losses in one line. --- ### T-SKM-Net: Trainable Neural Network Framework for Linear Constraint Satisfaction via Sampling Kaczmarz-Motzkin Method Authors: Haoyu Zhu, Yao Zhang, Jiashen Ren, Qingchun Hou Published: 2025-12-11 Categories: cs.LG, cs.AI, math.OC arXiv: https://arxiv.org/abs/2512.10461v1 Neural network constraint satisfaction is crucial for safety-critical applications such as power system optimization, robotic path planning, and autonomous driving. However, existing constraint satisfaction methods face efficiency-applicability trade-offs, with hard constraint methods suffering from either high computational complexity or restrictive assumptions on constraint structures. The Sampling Kaczmarz-Motzkin (SKM) method is a randomized iterative algorithm for solving large-scale linear inequality systems with favorable convergence properties, but its argmax operations introduce non-differentiability, posing challenges for neural network applications. This work proposes the Trainable Sampling Kaczmarz-Motzkin Network (T-SKM-Net) framework and, for the first time, systematically integrates SKM-type methods into neural network constraint satisfaction. The framework transforms mixed constraint problems into pure inequality problems through null space transformation, employs SKM for iterative solving, and maps solutions back to the original constraint space, efficiently handling both equality and inequality constraints. We provide theoretical proof of post-processing effectiveness in expectation and end-to-end trainability guarantees based on unbiased gradient estimators, demonstrating that despite non-differentiable operations, the framework supports standard backpropagation. On the DCOPF case118 benchmark, our method achieves 4.27ms/item GPU serial forward inference with 0.0025% max optimality gap with post-processing mode and 5.25ms/item with 0.0008% max optimality gap with joint training mode, delivering over 25$\times$ speedup compared to the pandapower solver while maintaining zero constraint violations under given tolerance. --- ### Hybrid Physics-ML Model for Forward Osmosis Flux with Complete Uncertainty Quantification Authors: Shiv Ratn, Shivang Rampriyan, Bahni Ray Published: 2025-12-11 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2512.10457v1 Forward Osmosis (FO) is a promising low-energy membrane separation technology, but challenges in accurately modelling its water flux (Jw) persist due to complex internal mass transfer phenomena. Traditional mechanistic models struggle with empirical parameter variability, while purely data-driven models lack physical consistency and rigorous uncertainty quantification (UQ). This study introduces a novel Robust Hybrid Physics-ML framework employing Gaussian Process Regression (GPR) for highly accurate, uncertainty-aware Jw prediction. The core innovation lies in training the GPR on the residual error between the detailed, non-linear FO physical model prediction (Jw_physical) and the experimental water flux (Jw_actual). Crucially, we implement a full UQ methodology by decomposing the total predictive variance (sigma2_total) into model uncertainty (epistemic, from GPR's posterior variance) and input uncertainty (aleatoric, analytically propagated via the Delta method for multi-variate correlated inputs). Leveraging the inherent strength of GPR in low-data regimes, the model, trained on a meagre 120 data points, achieved a state-of-the-art Mean Absolute Percentage Error (MAPE) of 0.26% and an R2 of 0.999 on the independent test data, validating a truly robust and reliable surrogate model for advanced FO process optimization and digital twin development. --- ### Grammaticality Judgments in Humans and Language Models: Revisiting Generative Grammar with LLMs Authors: Lars G. B. Johnsen Published: 2025-12-11 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.10453v1 What counts as evidence for syntactic structure? In traditional generative grammar, systematic contrasts in grammaticality such as subject-auxiliary inversion and the licensing of parasitic gaps are taken as evidence for an internal, hierarchical grammar. In this paper, we test whether large language models (LLMs), trained only on surface forms, reproduce these contrasts in ways that imply an underlying structural representation. We focus on two classic constructions: subject-auxiliary inversion (testing recognition of the subject boundary) and parasitic gap licensing (testing abstract dependency structure). We evaluate models including GPT-4 and LLaMA-3 using prompts eliciting acceptability ratings. Results show that LLMs reliably distinguish between grammatical and ungrammatical variants in both constructions, and as such support that they are sensitive to structure and not just linear order. Structural generalizations, distinct from cognitive knowledge, emerge from predictive training on surface forms, suggesting functional sensitivity to syntax without explicit encoding. --- ### Metacognitive Sensitivity for Test-Time Dynamic Model Selection Authors: Le Tuan Minh Trinh, Le Minh Vu Pham, Thi Minh Anh Pham, An Duc Nguyen Published: 2025-12-11 Categories: cs.LG arXiv: https://arxiv.org/abs/2512.10451v1 A key aspect of human cognition is metacognition - the ability to assess one's own knowledge and judgment reliability. While deep learning models can express confidence in their predictions, they often suffer from poor calibration, a cognitive bias where expressed confidence does not reflect true competence. Do models truly know what they know? Drawing from human cognitive science, we propose a new framework for evaluating and leveraging AI metacognition. We introduce meta-d', a psychologically-grounded measure of metacognitive sensitivity, to characterise how reliably a model's confidence predicts its own accuracy. We then use this dynamic sensitivity score as context for a bandit-based arbiter that performs test-time model selection, learning which of several expert models to trust for a given task. Our experiments across multiple datasets and deep learning model combinations (including CNNs and VLMs) demonstrate that this metacognitive approach improves joint-inference accuracy over constituent models. This work provides a novel behavioural account of AI models, recasting ensemble selection as a problem of evaluating both short-term signals (confidence prediction scores) and medium-term traits (metacognitive sensitivity). --- ### Error-Propagation-Free Learned Video Compression With Dual-Domain Progressive Temporal Alignment Authors: Han Li, Shaohui Li, Wenrui Dai, Chenglin Li, Xinlong Pan, Haipeng Wang, Junni Zou, Hongkai Xiong Published: 2025-12-11 Categories: cs.CV arXiv: https://arxiv.org/abs/2512.10450v1 Existing frameworks for learned video compression suffer from a dilemma between inaccurate temporal alignment and error propagation for motion estimation and compensation (ME/MC). The separate-transform framework employs distinct transforms for intra-frame and inter-frame compression to yield impressive rate-distortion (R-D) performance but causes evident error propagation, while the unified-transform framework eliminates error propagation via shared transforms but is inferior in ME/MC in shared latent domains. To address this limitation, in this paper, we propose a novel unifiedtransform framework with dual-domain progressive temporal alignment and quality-conditioned mixture-of-expert (QCMoE) to enable quality-consistent and error-propagation-free streaming for learned video compression. Specifically, we propose dualdomain progressive temporal alignment for ME/MC that leverages coarse pixel-domain alignment and refined latent-domain alignment to significantly enhance temporal context modeling in a coarse-to-fine fashion. The coarse pixel-domain alignment efficiently handles simple motion patterns with optical flow estimated from a single reference frame, while the refined latent-domain alignment develops a Flow-Guided Deformable Transformer (FGDT) over latents from multiple reference frames to achieve long-term motion refinement (LTMR) for complex motion patterns. Furthermore, we design a QCMoE module for continuous bit-rate adaptation that dynamically assigns different experts to adjust quantization steps per pixel based on target quality and content rather than relies on a single quantization step. QCMoE allows continuous and consistent rate control with appealing R-D performance. Experimental results show that the proposed method achieves competitive R-D performance compared with the state-of-the-arts, while successfully eliminating error propagation. --- ### When Reject Turns into Accept: Quantifying the Vulnerability of LLM-Based Scientific Reviewers to Indirect Prompt Injection Authors: Devanshu Sahoo, Manish Prasad, Vasudev Majhi, Jahnvi Singh, Vinay Chamola, Yash Sinha, Murari Mandal, Dhruv Kumar Published: 2025-12-11 Categories: cs.AI, cs.CL, cs.CR arXiv: https://arxiv.org/abs/2512.10449v3 Driven by surging submission volumes, scientific peer review has catalyzed two parallel trends: individual over-reliance on LLMs and institutional AI-powered assessment systems. This study investigates the robustness of "LLM-as-a-Judge" systems to adversarial PDF manipulation via invisible text injections and layout aware encoding attacks. We specifically target the distinct incentive of flipping "Reject" decisions to "Accept," a vulnerability that fundamentally compromises scientific integrity. To measure this, we introduce the Weighted Adversarial Vulnerability Score (WAVS), a novel metric that quantifies susceptibility by weighting score inflation against the severity of decision shifts relative to ground truth. We adapt 15 domain-specific attack strategies, ranging from semantic persuasion to cognitive obfuscation, and evaluate them across 13 diverse language models (including GPT-5 and DeepSeek) using a curated dataset of 200 official and real-world accepted and rejected submissions (e.g., ICLR OpenReview). Our results demonstrate that obfuscation techniques like "Maximum Mark Magyk" and "Symbolic Masking & Context Redirection" successfully manipulate scores, achieving decision flip rates of up to 86.26% in open-source models, while exposing distinct "reasoning traps" in proprietary systems. We release our complete dataset and injection framework to facilitate further research on the topic (https://anonymous.4open.sciencer/llm-jailbreak-FC9E/). --- ### Maximum Risk Minimization with Random Forests Authors: Francesco Freni, Anya Fries, Linus Kühne, Markus Reichstein, Jonas Peters Published: 2025-12-11 Categories: stat.ML, cs.AI, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2512.10445v2 We consider a regression setting where observations are collected in different environments modeled by different data distributions. The field of out-of-distribution (OOD) generalization aims to design methods that generalize better to test environments whose distributions differ from those observed during training. One line of such works has proposed to minimize the maximum risk across environments, a principle that we refer to as MaxRM (Maximum Risk Minimization). In this work, we introduce variants of random forests based on the principle of MaxRM. We provide computationally efficient algorithms and prove statistical consistency for our primary method. Our proposed method can be used with each of the following three risks: the mean squared error, the negative reward, and the regret (which quantifies the excess risk relative to the best predictor). For MaxRM with regret as the risk, we prove a novel out-of-sample guarantee over unseen test distributions. Finally, we evaluate the proposed methods on both simulated and real-world data. --- ### Clustered Federated Learning with Hierarchical Knowledge Distillation Authors: Sabtain Ahmad, Meerzhan Kanatbekova, Ivona Brandic, Atakan Aral Published: 2025-12-11 Categories: cs.DC, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2512.10443v1 Clustered Federated Learning (CFL) has emerged as a powerful approach for addressing data heterogeneity and ensuring privacy in large distributed IoT environments. By clustering clients and training cluster-specific models, CFL enables personalized models tailored to groups of heterogeneous clients. However, conventional CFL approaches suffer from fragmented learning for training independent global models for each cluster and fail to take advantage of collective cluster insights. This paper advocates a shift to hierarchical CFL, allowing bi-level aggregation to train cluster-specific models at the edge and a unified global model at the cloud. This shift improves training efficiency yet might introduce communication challenges. To this end, we propose CFLHKD, a novel personalization scheme for integrating hierarchical cluster knowledge into CFL. Built upon multi-teacher knowledge distillation, CFLHKD enables inter-cluster knowledge sharing while preserving cluster-specific personalization. CFLHKD adopts a bi-level aggregation to bridge the gap between local and global learning. Extensive evaluations of standard benchmark datasets demonstrate that CFLHKD outperforms representative baselines in cluster-specific and global model accuracy and achieves a performance improvement of 3.32-7.57\%. --- ### Decoding Student Minds: Leveraging Conversational Agents for Psychological and Learning Analysis Authors: Nour El Houda Ben Chaabene, Hamza Hammami, Laid Kahloul Published: 2025-12-11 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.10441v2 This paper presents a psychologically-aware conversational agent designed to enhance both learning performance and emotional well-being in educational settings. The system combines Large Language Models (LLMs), a knowledge graph-enhanced BERT (KG-BERT), and a bidirectional Long Short-Term Memory (LSTM) network with attention to classify students' cognitive and affective states in real time. Unlike prior chatbots limited to either tutoring or affective support, our approach leverages multimodal data-including textual semantics, prosodic speech features, and temporal behavioral trends-to infer engagement, stress, and conceptual understanding. A pilot study with 45 university students demonstrated improved motivation, reduced stress, and moderate academic gains compared to unimodal baselines. We explicitly discuss the exploratory nature of this small-sample pilot, report effect sizes and inter-rater reliability alongside significance tests, and provide an ablation analysis isolating the contribution of the knowledge-graph component and of each modality. These results underline the promise-while acknowledging the current limits in scale and modality balance-of integrating semantic reasoning, multimodal fusion, and temporal modeling to support adaptive, student-centered educational interventions. --- ### Enhancing Next-Generation Language Models with Knowledge Graphs: Extending Claude, Mistral IA, and GPT-4 via KG-BERT Authors: Nour El Houda Ben Chaabene, Hamza Hammami Published: 2025-12-11 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.10440v1 Large language models (LLMs) like Claude, Mistral IA, and GPT-4 excel in NLP but lack structured knowledge, leading to factual inconsistencies. We address this by integrating Knowledge Graphs (KGs) via KG-BERT to enhance grounding and reasoning. Experiments show significant gains in knowledge-intensive tasks such as question answering and entity linking. This approach improves factual reliability and enables more context-aware next-generation LLMs. --- ### An M-Health Algorithmic Approach to Identify and Assess Physiotherapy Exercises in Real Time Authors: Stylianos Kandylakis, Christos Orfanopoulos, Georgios Siolas, Panayiotis Tsanakas Published: 2025-12-11 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2512.10437v1 This work presents an efficient algorithmic framework for real-time identification, classification, and evaluation of human physiotherapy exercises using mobile devices. The proposed method interprets a kinetic movement as a sequence of static poses, which are estimated from camera input using a pose-estimation neural network. Extracted body keypoints are transformed into trigonometric angle-based features and classified with lightweight supervised models to generate frame-level pose predictions and accuracy scores. To recognize full exercise movements and detect deviations from prescribed patterns, we employ a dynamic-programming scheme based on a modified Levenshtein distance algorithm, enabling robust sequence matching and localization of inaccuracies. The system operates entirely on the client side, ensuring scalability and real-time performance. Experimental evaluation demonstrates the effectiveness of the methodology and highlights its applicability to remote physiotherapy supervision and m-health applications. --- ### Semantic Reconstruction of Adversarial Plagiarism: A Context-Aware Framework for Detecting and Restoring "Tortured Phrases" in Scientific Literature Authors: Agniva Maiti, Prajwal Panth, Suresh Chandra Satapathy Published: 2025-12-11 Categories: cs.CL arXiv: https://arxiv.org/abs/2512.10435v1 The integrity and reliability of scientific literature is facing a serious threat by adversarial text generation techniques, specifically from the use of automated paraphrasing tools to mask plagiarism. These tools generate "tortured phrases", statistically improbable synonyms (e.g. "counterfeit consciousness" for "artificial intelligence"), that preserve the local grammar while obscuring the original source. Most existing detection methods depend heavily on static blocklists or general-domain language models, which suffer from high false-negative rates for novel obfuscations and cannot determine the source of the plagiarized content. In this paper, we propose Semantic Reconstruction of Adversarial Plagiarism (SRAP), a framework designed not only to detect these anomalies but to mathematically recover the original terminology. We use a two-stage architecture: (1) statistical anomaly detection with a domain-specific masked language model (SciBERT) using token-level pseudo-perplexity, and (2) source-based semantic reconstruction using dense vector retrieval (FAISS) and sentence-level alignment (SBERT). Experiments on a parallel corpus of adversarial scientific text show that while zero-shot baselines fail completely (0.00 percent restoration accuracy), our retrieval-augmented approach achieves 23.67 percent restoration accuracy, significantly outperforming baseline methods. We also show that static decision boundaries are necessary for robust detection in jargon-heavy scientific text, since dynamic thresholding fails under high variance. SRAP enables forensic analysis by linking obfuscated expressions back to their most probable source documents. --- ### Targeted Data Protection for Diffusion Model by Matching Training Trajectory Authors: Hojun Lee, Mijin Koo, Yeji Song, Nojun Kwak Published: 2025-12-11 Categories: cs.AI arXiv: https://arxiv.org/abs/2512.10433v1 Recent advancements in diffusion models have made fine-tuning text-to-image models for personalization increasingly accessible, but have also raised significant concerns regarding unauthorized data usage and privacy infringement. Current protection methods are limited to passively degrading image quality, failing to achieve stable control. While Targeted Data Protection (TDP) offers a promising paradigm for active redirection toward user-specified target concepts, existing TDP attempts suffer from poor controllability due to snapshot-matching approaches that fail to account for complete learning dynamics. We introduce TAFAP (Trajectory Alignment via Fine-tuning with Adversarial Perturbations), the first method to successfully achieve effective TDP by controlling the entire training trajectory. Unlike snapshot-based methods whose protective influence is easily diluted as training progresses, TAFAP employs trajectory-matching inspired by dataset distillation to enforce persistent, verifiable transformations throughout fine-tuning. We validate our method through extensive experiments, demonstrating the first successful targeted transformation in diffusion models with simultaneous control over both identity and visual patterns. TAFAP significantly outperforms existing TDP attempts, achieving robust redirection toward target concepts while maintaining high image quality. This work enables verifiable safeguards and provides a new framework for controlling and tracing alterations in diffusion model outputs. ---