# Research Papers — 2025-11 · bundle page 44 of 197

> Full-text markdown of papers #2151–2200 of 9830 published in 2025-11, newest first. AI-readable mirrors of the full 9830-paper 2025-11 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2025-11/page/44
[← Previous bundle](https://aipod.dev/papers/archive/2025-11/page/43.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-11/page/45.md)

### Connecting the Dots: Training-Free Visual Grounding via Agentic Reasoning

Authors: Liqin Luo, Guangyao Chen, Xiawu Zheng, Yongxing Dai, Yixiong Zou, Yonghong Tian
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19516v2

Visual grounding, the task of linking textual queries to specific regions within images, plays a pivotal role in vision-language integration. Existing methods typically rely on extensive task-specific annotations and fine-tuning, limiting their ability to generalize effectively to novel or out-of-distribution scenarios. To address these limitations, we introduce GroundingAgent, a novel agentic visual grounding framework that operates without any task-specific fine-tuning. GroundingAgent employs a structured, iterative reasoning mechanism that integrates pretrained open-vocabulary object detectors, multimodal large language models (MLLMs), and large language models (LLMs) to progressively refine candidate regions through joint semantic and spatial analyses. Remarkably, GroundingAgent achieves an average zero-shot grounding accuracy of 65.1 % on widely-used benchmarks (RefCOCO, RefCOCO+, RefCOCOg), entirely without fine-tuning. Furthermore, by substituting MLLM-generated captions with the original query texts, the accuracy at the selection stage alone reaches approximately 90 %, closely matching supervised performance and underscoring the critical role of LLM reasoning capabilities. GroundingAgent also offers strong interpretability, transparently illustrating each reasoning step and providing clear insights into its decision-making process.

---

### Fewer Tokens, Greater Scaling: Self-Adaptive Visual Bases for Efficient and Expansive Representation Learning

Authors: Shawn Young, Xingyu Zeng, Lijian Xu
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19515v1

This paper investigates the fundamental relationship between model capacity and the minimal number of visual tokens required to preserve image semantics. Inspired by the Minimum Description Length principle, we reinterpret image tokens as vectors in a visual semantic space and define the intrinsic semantic complexity of an image as the smallest set of basis vectors needed to span this space. Building on this perspective, we propose Orthogonal Filtering, a lightweight module that adaptively clusters redundant tokens into a compact set of orthogonal bases. Through extensive experiments across a range of ViT models, we reveal a consistent token, model scaling law: larger models require significantly fewer tokens to span visual semantic space. Besides, we also contribute a visual long-context dataset.

---

### Row-Stochastic Matrices Can Provably Outperform Doubly Stochastic Matrices in Decentralized Learning

Authors: Bing Liu, Boao Kong, Limin Lu, Kun Yuan, Chengcheng Zhao
Published: 2025-11-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.19513v4

Decentralized learning often involves a weighted global loss with heterogeneous node weights $λ$. We revisit two natural strategies for incorporating these weights: (i) embedding them into the local losses to retain a uniform weight (and thus a doubly stochastic matrix), and (ii) keeping the original losses while employing a $λ$-induced row-stochastic matrix. Although prior work shows that both strategies target the same $λ$-weighted global loss, it remains unclear whether the Euclidean-space guarantees are tight and what fundamentally differentiates their behaviors. To clarify this, we develop a weighted Hilbert-space framework $L^2(λ;\mathbb{R}^d)$ and obtain convergence rates that are strictly tighter than those from standard Euclidean analysis. In this geometry, the row-stochastic matrix becomes \emph{self-adjoint} whereas the doubly stochastic one does not, creating additional \emph{penalty terms} that amplify consensus error, thereby slowing convergence. Consequently, the difference in convergence arises not only from spectral gaps but also from these penalty terms. We then derive sufficient conditions under which the row-stochastic design converges faster even with a smaller spectral gap. Finally, by using a Rayleigh-quotient and Loewner-order eigenvalue comparison, we further obtain topology conditions that guarantee this advantage and yield practical topology-design guidelines.

---

### Single Image to High-Quality 3D Object via Latent Features

Authors: Huanning Dong, Yinuo Huang, Fan Li, Ping Kuang
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19512v1

3D assets are essential in the digital age. While automatic 3D generation, such as image-to-3d, has made significant strides in recent years, it often struggles to achieve fast, detailed, and high-fidelity generation simultaneously. In this work, we introduce LatentDreamer, a novel framework for generating 3D objects from single images. The key to our approach is a pre-trained variational autoencoder that maps 3D geometries to latent features, which greatly reducing the difficulty of 3D generation. Starting from latent features, the pipeline of LatentDreamer generates coarse geometries, refined geometries, and realistic textures sequentially. The 3D objects generated by LatentDreamer exhibit high fidelity to the input images, and the entire generation process can be completed within a short time (typically in 70 seconds). Extensive experiments show that with only a small amount of training, LatentDreamer demonstrates competitive performance compared to contemporary approachs.

---

### The Determinant Ratio Matrix Approach to Solving 3D Matching and 2D Orthographic Projection Alignment Tasks

Authors: Andrew J. Hanson, Sonya M. Hanson
Published: 2025-11-24
Categories: cs.CV, eess.IV
arXiv: https://arxiv.org/abs/2511.19511v1

Pose estimation is a general problem in computer vision with wide applications. The relative orientation of a 3D reference object can be determined from a 3D rotated version of that object, or from a projection of the rotated object to a 2D planar image. This projection can be a perspective projection (the PnP problem) or an orthographic projection (the OnP problem). We restrict our attention here to the OnP problem and the full 3D pose estimation task (the EnP problem). Here we solve the least squares systems for both the error-free EnP and OnP problems in terms of the determinant ratio matrix (DRaM) approach. The noisy-data case can be addressed with a straightforward rotation correction scheme. While the SVD and optimal quaternion eigensystem methods solve the noisy EnP 3D-3D alignment exactly, the noisy 3D-2D orthographic (OnP) task has no known comparable closed form, and can be solved by DRaM-class methods. We note that while previous similar work has been presented in the literature exploiting both the QR decomposition and the Moore-Penrose pseudoinverse transformations, here we place these methods in a larger context that has not previously been fully recognized in the absence of the corresponding DRaM solution. We term this class of solutions as the DRaM family, and conduct comparisons of the behavior of the families of solutions for the EnP and OnP rotation estimation problems. Overall, this work presents both a new solution to the 3D and 2D orthographic pose estimation problems and provides valuable insight into these classes of problems. With hindsight, we are able to show that our DRaM solutions to the exact EnP and OnP problems possess derivations that could have been discovered in the time of Gauss, and in fact generalize to all analogous N-dimensional Euclidean pose estimation problems.

---

### TouchFormer: A Robust Transformer-based Framework for Multimodal Material Perception

Authors: Kailin Lyu, Long Xiao, Jianing Zeng, Junhao Dong, Xuexin Liu, Zhuojun Zou, Haoyue Yang, Lin Shu, Jie Hao
Published: 2025-11-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.19509v1

Traditional vision-based material perception methods often experience substantial performance degradation under visually impaired conditions, thereby motivating the shift toward non-visual multimodal material perception. Despite this, existing approaches frequently perform naive fusion of multimodal inputs, overlooking key challenges such as modality-specific noise, missing modalities common in real-world scenarios, and the dynamically varying importance of each modality depending on the task. These limitations lead to suboptimal performance across several benchmark tasks. In this paper, we propose a robust multimodal fusion framework, TouchFormer. Specifically, we employ a Modality-Adaptive Gating (MAG) mechanism and intra- and inter-modality attention mechanisms to adaptively integrate cross-modal features, enhancing model robustness. Additionally, we introduce a Cross-Instance Embedding Regularization(CER) strategy, which significantly improves classification accuracy in fine-grained subcategory material recognition tasks. Experimental results demonstrate that, compared to existing non-visual methods, the proposed TouchFormer framework achieves classification accuracy improvements of 2.48% and 6.83% on SSMC and USMC tasks, respectively. Furthermore, real-world robotic experiments validate TouchFormer's effectiveness in enabling robots to better perceive and interpret their environment, paving the way for its deployment in safety-critical applications such as emergency response and industrial automation. The code and datasets will be open-source, and the videos are available in the supplementary materials.

---

### LumiTex: Towards High-Fidelity PBR Texture Generation with Illumination Context

Authors: Jingzhi Bao, Hongze Chen, Lingting Zhu, Chenyu Liu, Runze Zhang, Keyang Luo, Zeyu Hu, Weikai Chen, Yingda Yin, Xin Wang, Zehong Lin, Jun Zhang et al.
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19437v1

Physically-based rendering (PBR) provides a principled standard for realistic material-lighting interactions in computer graphics. Despite recent advances in generating PBR textures, existing methods fail to address two fundamental challenges: 1) materials decomposition from image prompts under limited illumination cues, and 2) seamless and view-consistent texture completion. To this end, we propose LumiTex, an end-to-end framework that comprises three key components: (1) a multi-branch generation scheme that disentangles albedo and metallic-roughness under shared illumination priors for robust material understanding, (2) a lighting-aware material attention mechanism that injects illumination context into the decoding process for physically grounded generation of albedo, metallic, and roughness maps, and (3) a geometry-guided inpainting module based on a large view synthesis model that enriches texture coverage and ensures seamless, view-consistent UV completion. Extensive experiments demonstrate that LumiTex achieves state-of-the-art performance in texture quality, surpassing both existing open-source and commercial methods.

---

### VDC-Agent: When Video Detailed Captioners Evolve Themselves via Agentic Self-Reflection

Authors: Qiang Wang, Xinyuan Gao, Yuhang He, Jizhou Han, Jiangyang Li, SongLin Dong, Zhiheng Ma, Yihong Gong
Published: 2025-11-24
Categories: cs.CV, cs.AI, cs.LG, cs.MM
arXiv: https://arxiv.org/abs/2511.19436v2

Existing Video Detailed Captioning (VDC) methods predominantly rely on costly human annotations or distillation from powerful proprietary models, creating a dependency on external supervision. In this paper, we propose VDC-Agent, an autonomous self-evolving framework that empowers a single Multimodal Large Language Model (MLLM) to generate and refine high-quality captions through principle-guided self-reflection. To overcome the inference latency inherent in iterative refinement, we further propose to internalize this reflective capability into the model. Specifically, we construct VDC-Agent-19K, a preference dataset derived from the agent's self-scored trajectories, and introduce a Curriculum Direct Preference Optimization (DPO) strategy. This strategy leverages the quality gap between generated candidates to progressively align the model from easy to hard samples. Extensive experiments demonstrate that VDC-Agent achieves state-of-the-art performance on VDC and DREAM-1K benchmarks, generating captions with superior detail and faithfulness. Crucially, our internalization strategy retains the inference efficiency of the base model while significantly enhancing its generalization capabilities, as validated by both quantitative metrics and human evaluation.

---

### Are Image-to-Video Models Good Zero-Shot Image Editors?

Authors: Zechuan Zhang, Zhenyuan Chen, Zongxin Yang, Yi Yang
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19435v2

Large-scale video diffusion models show strong world simulation and temporal reasoning abilities, but their use as zero-shot image editors remains underexplored. We introduce IF-Edit, a tuning-free framework that repurposes pretrained image-to-video diffusion models for instruction-driven image editing. IF-Edit addresses three key challenges: prompt misalignment, redundant temporal latents, and blurry late-stage frames. It includes (1) a chain-of-thought prompt enhancement module that transforms static editing instructions into temporally grounded reasoning prompts; (2) a temporal latent dropout strategy that compresses frame latents after the expert-switch point, accelerating denoising while preserving semantic and temporal coherence; and (3) a self-consistent post-refinement step that sharpens late-stage frames using a short still-video trajectory. Experiments on four public benchmarks, covering non-rigid editing, physical and temporal reasoning, and general instruction edits, show that IF-Edit performs strongly on reasoning-centric tasks while remaining competitive on general-purpose edits. Our study provides a systematic view of video diffusion models as image editors and highlights a simple recipe for unified video-image generative reasoning.

---

### Breaking the Likelihood-Quality Trade-off in Diffusion Models by Merging Pretrained Experts

Authors: Yasin Esfandiari, Stefan Bauer, Sebastian U. Stich, Andrea Dittadi
Published: 2025-11-24
Categories: cs.CV, cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2511.19434v1

Diffusion models for image generation often exhibit a trade-off between perceptual sample quality and data likelihood: training objectives emphasizing high-noise denoising steps yield realistic images but poor likelihoods, whereas likelihood-oriented training overweights low-noise steps and harms visual fidelity. We introduce a simple plug-and-play sampling method that combines two pretrained diffusion experts by switching between them along the denoising trajectory. Specifically, we apply an image-quality expert at high noise levels to shape global structure, then switch to a likelihood expert at low noise levels to refine pixel statistics. The approach requires no retraining or fine-tuning -- only the choice of an intermediate switching step. On CIFAR-10 and ImageNet32, the merged model consistently matches or outperforms its base components, improving or preserving both likelihood and sample quality relative to each expert alone. These results demonstrate that expert switching across noise levels is an effective way to break the likelihood-quality trade-off in image diffusion models.

---

### Mixture of Horizons in Action Chunking

Authors: Dong Jing, Gang Wang, Jiaqi Liu, Weiliang Tang, Zelong Sun, Yunchao Yao, Zhenyu Wei, Yunhui Liu, Zhiwu Lu, Mingyu Ding
Published: 2025-11-24
Categories: cs.RO, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2511.19433v2

Vision-language-action (VLA) models have shown remarkable capabilities in robotic manipulation, but their performance is sensitive to the $\textbf{action chunk length}$ used during training, termed $\textbf{horizon}$. Our empirical study reveals an inherent trade-off: longer horizons provide stronger global foresight but degrade fine-grained accuracy, while shorter ones sharpen local control yet struggle on long-term tasks, implying fixed choice of single horizons being suboptimal. To mitigate the trade-off, we propose a $\textbf{mixture of horizons (MoH)}$ strategy. MoH rearranges the action chunk into several segments with different horizons, processes them in parallel with a shared action transformer, and fuses outputs with a light linear gate. It has three appealing benefits. 1) MoH exploits long-term foresight and short-term precision jointly within a single model, improving both performance and generalizability to complex tasks. 2) MoH is plug-and-play for full-attention action modules with minimal training or inference overhead. 3) MoH enables dynamic inference with adaptive horizons, which selects stable actions through cross-horizon consensus, achieving 2.5$\times$ higher throughput than baselines while preserving superior performance. Extensive experiments over flow-based policies $π_0$, $π_{0.5}$, and one-step regression policy $π_{\text{reg}}$ demonstrate that MoH yields consistent and significant gains on both simulations and real-world tasks. Notably, under mixed-task setting, $π_{0.5}$ with MoH reaches a new state-of-the-art with 99$\%$ average success rate on LIBERO after only $30k$ training iterations. Project page: https://timsty1.github.io/moh/

---

### Cloud4D: Estimating Cloud Properties at a High Spatial and Temporal Resolution

Authors: Jacob Lin, Edward Gryspeerdt, Ronald Clark
Published: 2025-11-24
Categories: cs.CV, physics.ao-ph
arXiv: https://arxiv.org/abs/2511.19431v2

There has been great progress in improving numerical weather prediction and climate models using machine learning. However, most global models act at a kilometer-scale, making it challenging to model individual clouds and factors such as extreme precipitation, wind gusts, turbulence, and surface irradiance. Therefore, there is a need to move towards higher-resolution models, which in turn require high-resolution real-world observations that current instruments struggle to obtain. We present Cloud4D, the first learning-based framework that reconstructs a physically consistent, four-dimensional cloud state using only synchronized ground-based cameras. Leveraging a homography-guided 2D-to-3D transformer, Cloud4D infers the full 3D distribution of liquid water content at 25 m spatial and 5 s temporal resolution. By tracking the 3D liquid water content retrievals over time, Cloud4D additionally estimates horizontal wind vectors. Across a two-month deployment comprising six skyward cameras, our system delivers an order-of-magnitude improvement in space-time resolution relative to state-of-the-art satellite measurements, while retaining single-digit relative error ($<10\%$) against collocated radar measurements. Code and data are available on our project page https://cloud4d.jacob-lin.com/.

---

### Cook and Clean Together: Teaching Embodied Agents for Parallel Task Execution

Authors: Dingkang Liang, Cheng Zhang, Xiaopeng Xu, Jianzhong Ju, Zhenbo Luo, Xiang Bai
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19430v1

Task scheduling is critical for embodied AI, enabling agents to follow natural language instructions and execute actions efficiently in 3D physical worlds. However, existing datasets often simplify task planning by ignoring operations research (OR) knowledge and 3D spatial grounding. In this work, we propose Operations Research knowledge-based 3D Grounded Task Scheduling (ORS3D), a new task that requires the synergy of language understanding, 3D grounding, and efficiency optimization. Unlike prior settings, ORS3D demands that agents minimize total completion time by leveraging parallelizable subtasks, e.g., cleaning the sink while the microwave operates. To facilitate research on ORS3D, we construct ORS3D-60K, a large-scale dataset comprising 60K composite tasks across 4K real-world scenes. Furthermore, we propose GRANT, an embodied multi-modal large language model equipped with a simple yet effective scheduling token mechanism to generate efficient task schedules and grounded actions. Extensive experiments on ORS3D-60K validate the effectiveness of GRANT across language understanding, 3D grounding, and scheduling efficiency. The code is available at https://github.com/H-EmbodVis/GRANT

---

### Flow Map Distillation Without Data

Authors: Shangyuan Tong, Nanye Ma, Saining Xie, Tommi Jaakkola
Published: 2025-11-24
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2511.19428v1

State-of-the-art flow models achieve remarkable quality but require slow, iterative sampling. To accelerate this, flow maps can be distilled from pre-trained teachers, a procedure that conventionally requires sampling from an external dataset. We argue that this data-dependency introduces a fundamental risk of Teacher-Data Mismatch, as a static dataset may provide an incomplete or even misaligned representation of the teacher's full generative capabilities. This leads us to question whether this reliance on data is truly necessary for successful flow map distillation. In this work, we explore a data-free alternative that samples only from the prior distribution, a distribution the teacher is guaranteed to follow by construction, thereby circumventing the mismatch risk entirely. To demonstrate the practical viability of this philosophy, we introduce a principled framework that learns to predict the teacher's sampling path while actively correcting for its own compounding errors to ensure high fidelity. Our approach surpasses all data-based counterparts and establishes a new state-of-the-art by a significant margin. Specifically, distilling from SiT-XL/2+REPA, our method reaches an impressive FID of 1.45 on ImageNet 256x256, and 1.49 on ImageNet 512x512, both with only 1 sampling step. We hope our work establishes a more robust paradigm for accelerating generative models and motivates the broader adoption of flow map distillation without data.

---

### Prompt Less, Smile More: MTP with Semantic Engineering in Lieu of Prompt Engineering

Authors: Jayanaka L. Dantanarayana, Savini Kashmira, Thakee Nathees, Zichen Zhang, Krisztian Flautner, Lingjia Tang, Jason Mars
Published: 2025-11-24
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2511.19427v1

AI-Integrated programming is emerging as a foundational paradigm for building intelligent systems with large language models (LLMs). Recent approaches such as Meaning Typed Programming (MTP) automate prompt generation by leveraging the semantics already present in code. However, many real-world applications depend on contextual cues, developer intent, and domain-specific reasoning that extend beyond what static code semantics alone can express. To address this limitation, we introduce Semantic Engineering, a lightweight method for enriching program semantics so that LLM-based systems can more accurately reflect developer intent without requiring full manual prompt design. We present Semantic Context Annotations (SemTexts), a language-level mechanism that allows developers to embed natural-language context directly into program constructs. Integrated into the Jac programming language, Semantic Engineering extends MTP to incorporate these enriched semantics during prompt generation. We further introduce a benchmark suite designed to reflect realistic AI-Integrated application scenarios. Our evaluation shows that Semantic Engineering substantially improves prompt fidelity, achieving performance comparable to Prompt Engineering while requiring significantly less developer effort.

---

### Ref-SAM3D: Bridging SAM3D with Text for Reference 3D Reconstruction

Authors: Yun Zhou, Yaoting Wang, Guangquan Jie, Jinyu Liu, Henghui Ding
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19426v1

SAM3D has garnered widespread attention for its strong 3D object reconstruction capabilities. However, a key limitation remains: SAM3D cannot reconstruct specific objects referred to by textual descriptions, a capability that is essential for practical applications such as 3D editing, game development, and virtual environments. To address this gap, we introduce Ref-SAM3D, a simple yet effective extension to SAM3D that incorporates textual descriptions as a high-level prior, enabling text-guided 3D reconstruction from a single RGB image. Through extensive qualitative experiments, we show that Ref-SAM3D, guided only by natural language and a single 2D view, delivers competitive and high-fidelity zero-shot reconstruction performance. Our results demonstrate that Ref-SAM3D effectively bridges the gap between 2D visual cues and 3D geometric understanding, offering a more flexible and accessible paradigm for reference-guided 3D reconstruction. Code is available at: https://github.com/FudanCVL/Ref-SAM3D.

---

### SAM3-Adapter: Efficient Adaptation of Segment Anything 3 for Camouflage Object Segmentation, Shadow Detection, and Medical Image Segmentation

Authors: Tianrun Chen, Runlong Cao, Xinda Yu, Lanyun Zhu, Chaotao Ding, Deyi Ji, Cheng Chen, Qi Zhu, Chunyan Xu, Papa Mao, Ying Zang
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19425v1

The rapid rise of large-scale foundation models has reshaped the landscape of image segmentation, with models such as Segment Anything achieving unprecedented versatility across diverse vision tasks. However, previous generations-including SAM and its successor-still struggle with fine-grained, low-level segmentation challenges such as camouflaged object detection, medical image segmentation, cell image segmentation, and shadow detection. To address these limitations, we originally proposed SAM-Adapter in 2023, demonstrating substantial gains on these difficult scenarios. With the emergence of Segment Anything 3 (SAM3)-a more efficient and higher-performing evolution with a redesigned architecture and improved training pipeline-we revisit these long-standing challenges. In this work, we present SAM3-Adapter, the first adapter framework tailored for SAM3 that unlocks its full segmentation capability. SAM3-Adapter not only reduces computational overhead but also consistently surpasses both SAM and SAM2-based solutions, establishing new state-of-the-art results across multiple downstream tasks, including medical imaging, camouflaged (concealed) object segmentation, and shadow detection. Built upon the modular and composable design philosophy of the original SAM-Adapter, SAM3-Adapter provides stronger generalizability, richer task adaptability, and significantly improved segmentation precision. Extensive experiments confirm that integrating SAM3 with our adapter yields superior accuracy, robustness, and efficiency compared to all prior SAM-based adaptations. We hope SAM3-Adapter can serve as a foundation for future research and practical segmentation applications. Code, pre-trained models, and data processing pipelines are available.

---

### Beyond Protein Language Models: An Agentic LLM Framework for Mechanistic Enzyme Design

Authors: Bruno Jacob, Khushbu Agarwal, Marcel Baer, Peter Rice, Simone Raugei
Published: 2025-11-24
Categories: q-bio.QM, cs.AI
arXiv: https://arxiv.org/abs/2511.19423v1

We present Genie-CAT, a tool-augmented large-language-model (LLM) system designed to accelerate scientific hypothesis generation in protein design. Using metalloproteins (e.g., ferredoxins) as a case study, Genie-CAT integrates four capabilities -- literature-grounded reasoning through retrieval-augmented generation (RAG), structural parsing of Protein Data Bank files, electrostatic potential calculations, and machine-learning prediction of redox properties -- into a unified agentic workflow. By coupling natural-language reasoning with data-driven and physics-based computation, the system generates mechanistically interpretable, testable hypotheses linking sequence, structure, and function. In proof-of-concept demonstrations, Genie-CAT autonomously identifies residue-level modifications near (Fe--S) clusters that affect redox tuning, reproducing expert-derived hypotheses in a fraction of the time. The framework highlights how AI agents combining language models with domain-specific tools can bridge symbolic reasoning and numerical simulation, transforming LLMs from conversational assistants into partners for computational discovery.

---

### SLMFix: Leveraging Small Language Models for Domain Specific Language Error Fixing with Reinforcement Learning

Authors: David Jiahao Fu, Aryan Gupta, Aaron Councilman, Yu-Xiong Wang, Vikram Adve
Published: 2025-11-24
Categories: cs.SE, cs.AI, cs.PL
arXiv: https://arxiv.org/abs/2511.19422v2

Large language models (LLMs) have shown impressive capabilities in code generation across many programming languages but even state-of-the-art LLMs generate programs that contain syntactic errors and fail to complete the given tasks, especially for low-resource programming languages (LRPLs). In addition, the high cost of training makes finetuning LLMs unaffordable for those with constrained computational resources, further weakening the effectiveness of LLMs for code generation. In this work, we propose SLMFix, a novel code generation pipeline that leverages a small language model (SLM) finetuned using reinforcement learning (RL) techniques to fix syntactic errors in LLM-generated programs for domain-specific languages (DSLs) based on interpreter feedback. Our experimental results demonstrate the effectiveness and generalizability of our approach across multiple DSLs, improving the validator pass rates by 40% on LRPLs and eliminating more than 50% of syntactic errors for high-resource DSLs. Notably, SLMFix brings substantial performance improvement to the base model and outperforms supervised finetuning approach even for 7B models on LRPLs including Ansible and Lean, showing the potential of our approach in improving the quality of LLM-generated programs.

---

### Chain-of-Visual-Thought: Teaching VLMs to See and Think Better with Continuous Visual Tokens

Authors: Yiming Qin, Bomin Wei, Jiaxin Ge, Konstantinos Kallidromitis, Stephanie Fu, Trevor Darrell, XuDong Wang
Published: 2025-11-24
Categories: cs.CV, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2511.19418v3

Vision-Language Models (VLMs) excel at reasoning in linguistic space but struggle with perceptual understanding that requires dense visual perception, e.g., spatial reasoning and geometric awareness. This limitation stems from the fact that current VLMs have limited mechanisms to capture dense visual information across spatial dimensions. We introduce Chain-of-Visual-Thought (COVT), a framework that enables VLMs to reason not only in words but also through continuous visual tokens-compact latent representations that encode rich perceptual cues. Within a small budget of roughly 20 tokens, COVT distills knowledge from lightweight vision experts, capturing complementary properties such as 2D appearance, 3D geometry, spatial layout, and edge structure. During training, the VLM with COVT autoregressively predicts these visual tokens to reconstruct dense supervision signals (e.g., depth, segmentation, edges, and DINO features). At inference, the model reasons directly in the continuous visual token space, preserving efficiency while optionally decoding dense predictions for interpretability. Evaluated across more than ten diverse perception benchmarks, including CV-Bench, MMVP, RealWorldQA, MMStar, WorldMedQA, and HRBench, integrating COVT into strong VLMs such as Qwen2.5-VL and LLaVA consistently improves performance by 3% to 16% and demonstrates that compact continuous visual thinking enables more precise, grounded, and interpretable multimodal intelligence.

---

### Be My Eyes: Extending Large Language Models to New Modalities Through Multi-Agent Collaboration

Authors: James Y. Huang, Sheng Zhang, Qianchu Liu, Guanghui Qin, Tinghui Zhu, Tristan Naumann, Muhao Chen, Hoifung Poon
Published: 2025-11-24
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2511.19417v1

Large Language Models (LLMs) have demonstrated remarkable capabilities in challenging, knowledge-intensive reasoning tasks. However, extending LLMs to perceive and reason over a new modality (e.g., vision), often requires costly development of large-scale vision language models (VLMs) with LLMs as backbones. Smaller VLMs are more efficient and adaptable but often lack the broad knowledge and reasoning capabilities of frontier LLMs. In this work, we propose BeMyEyes, a modular, multi-agent framework for extending LLMs to multimodal reasoning by orchestrating collaboration between efficient, adaptable VLMs as perceivers and powerful LLMs as reasoners through conversations. We then introduce a data synthesis and supervised fine-tuning pipeline to train the perceiver agent to effectively collaborate with the reasoner agent. By combining the complementary strengths of perception and reasoning agents, BeMyEyes avoids the need for training large-scale multimodal models, preserves the generalization and reasoning capabilities of LLMs, and allows flexible extension to new domains and modalities. Experiments show that our framework unlocks the multimodal reasoning capabilities for LLMs, enabling a lightweight and fully open-source solution, i.e. equipping text-only DeepSeek-R1 with Qwen2.5-VL-7B perceiver, to outperform large-scale proprietary VLMs such as GPT-4o on a wide range of knowledge-intensive multimodal tasks. These results demonstrate the effectiveness, modularity, and scalability of our multi-agent approach for building future multimodal reasoning systems.

---

### UniGame: Turning a Unified Multimodal Model Into Its Own Adversary

Authors: Zhaolong Su, Wang Lu, Hao Chen, Sharon Li, Jindong Wang
Published: 2025-11-24
Categories: cs.LG, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2511.19413v3

Unified Multimodal Models (UMMs) have shown impressive performance in both understanding and generation with a single architecture. However, UMMs still exhibit a fundamental inconsistency: understanding favors compact embeddings, whereas generation favors reconstruction-rich representations. This structural trade-off produces misaligned decision boundaries, degraded cross-modal coherence, and heightened vulnerability under distributional and adversarial shifts. In this paper, we present UniGame, a self-adversarial post-training framework that directly targets the inconsistencies. By applying a lightweight perturber at the shared token interface, UniGame enables the generation branch to actively seek and challenge fragile understanding, turning the model itself into its own adversary. Experiments demonstrate that UniGame significantly improves the consistency (+4.6%). Moreover, it also achieves substantial improvements in understanding (+3.6%), generation (+0.02)on GenEval, out-of-distribution and adversarial robustness (+4.8% and +6.2% on NaturalBench and AdVQA). The framework is architecture-agnostic, introduces less than 1% additional parameters, and is complementary to existing post-training methods. These results position adversarial self-play as a general and effective principle for enhancing the coherence, stability, and unified competence of future multimodal foundation models. The official code is available at: https://github.com/AIFrontierLab/TorchUMM

---

### Learning Robust Social Strategies with Large Language Models

Authors: Dereck Piche, Mohammed Muqeeth, Milad Aghajohari, Juan Duque, Michael Noukhovitch, Aaron Courville
Published: 2025-11-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.19405v2

As agentic AI becomes more widespread, agents with distinct and possibly conflicting goals will interact in complex ways. These multi-agent interactions pose a fundamental challenge, particularly in social dilemmas, where agents' individual incentives can undermine collective welfare. While reinforcement learning (RL) has been effective for aligning large language models (LLMs) in the single-agent regime, prior small-network results suggest that standard RL in multi-agent settings often converges to defecting, self-interested policies. We show the same effect in LLMs: despite cooperative priors, RL-trained LLM agents develop opportunistic behavior that can exploit even advanced closed-source models. To address this tendency of RL to converge to poor equilibria, we adapt a recent opponent-learning awareness algorithm, Advantage Alignment, to fine-tune LLMs toward multi-agent cooperation and non-exploitability. We then introduce a group-relative baseline that simplifies advantage computation in iterated games, enabling multi-agent training at LLM scale. We also contribute a novel social dilemma environment, Trust-and-Split, which requires natural language communication to achieve high collective welfare. Across a wide range of social dilemmas, policies learned with Advantage Alignment achieve higher collective payoffs while remaining robust against exploitation by greedy agents. We release all of our code to support future work on multi-agent RL training for LLMs.

---

### Nonparametric Instrumental Variable Regression with Observed Covariates

Authors: Zikai Shen, Zonghao Chen, Dimitri Meunier, Ingo Steinwart, Arthur Gretton, Zhu Li
Published: 2025-11-24
Categories: stat.ML, cs.LG, math.ST
arXiv: https://arxiv.org/abs/2511.19404v1

We study the problem of nonparametric instrumental variable regression with observed covariates, which we refer to as NPIV-O. Compared with standard nonparametric instrumental variable regression (NPIV), the additional observed covariates facilitate causal identification and enables heterogeneous causal effect estimation. However, the presence of observed covariates introduces two challenges for its theoretical analysis. First, it induces a partial identity structure, which renders previous NPIV analyses - based on measures of ill-posedness, stability conditions, or link conditions - inapplicable. Second, it imposes anisotropic smoothness on the structural function. To address the first challenge, we introduce a novel Fourier measure of partial smoothing; for the second challenge, we extend the existing kernel 2SLS instrumental variable algorithm with observed covariates, termed KIV-O, to incorporate Gaussian kernel lengthscales adaptive to the anisotropic smoothness. We prove upper $L^2$-learning rates for KIV-O and the first $L^2$-minimax lower learning rates for NPIV-O. Both rates interpolate between known optimal rates of NPIV and nonparametric regression (NPR). Interestingly, we identify a gap between our upper and lower bounds, which arises from the choice of kernel lengthscales tuned to minimize a projected risk. Our theoretical analysis also applies to proximal causal inference, an emerging framework for causal effect estimation that shares the same conditional moment restriction as NPIV-O.

---

### In-Video Instructions: Visual Signals as Generative Control

Authors: Gongfan Fang, Xinyin Ma, Xinchao Wang
Published: 2025-11-24
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2511.19401v1

Large-scale video generative models have recently demonstrated strong visual capabilities, enabling the prediction of future frames that adhere to the logical and physical cues in the current observation. In this work, we investigate whether such capabilities can be harnessed for controllable image-to-video generation by interpreting visual signals embedded within the frames as instructions, a paradigm we term In-Video Instruction. In contrast to prompt-based control, which provides textual descriptions that are inherently global and coarse, In-Video Instruction encodes user guidance directly into the visual domain through elements such as overlaid text, arrows, or trajectories. This enables explicit, spatial-aware, and unambiguous correspondences between visual subjects and their intended actions by assigning distinct instructions to different objects. Extensive experiments on three state-of-the-art generators, including Veo 3.1, Kling 2.5, and Wan 2.2, show that video models can reliably interpret and execute such visually embedded instructions, particularly in complex multi-object scenarios.

---

### DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research

Authors: Rulin Shao, Akari Asai, Shannon Zejiang Shen, Hamish Ivison, Varsha Kishore, Jingming Zhuo, Xinran Zhao, Molly Park, Samuel G. Finlayson, David Sontag, Tyler Murray, Sewon Min et al.
Published: 2025-11-24
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2511.19399v3

Deep research agents perform multi-step research to produce long-form, well-attributed answers. However, most open deep research agents are trained on easily verifiable short-form QA tasks via reinforcement learning with verifiable rewards, which does not extend to realistic long-form tasks. We address this with Reinforcement Learning with Evolving Rubrics (RLER), where rubrics are constructed and maintained to co-evolve with the policy model during training. This allows the rubrics to incorporate newly explored information from search and contrasting model responses, enabling better fact checking and more discriminative on-policy feedback. Using RLER, we develop Deep Research Tulu (DR Tulu-8B), the first fully open model that is directly trained for open-ended, long-form deep research. Across four long-form deep research benchmarks in science, healthcare, and general domains, DR Tulu substantially outperforms existing open deep research agents (by 15.6% over Tongyi DR on average) and matches or exceeds proprietary deep research agents (by 0.7% over OpenAI DR on average), while being significantly smaller and cheaper per query (1000x cheaper than OpenAI DR per query).

---

### PTF Testing Lower Bounds for Non-Gaussian Component Analysis

Authors: Ilias Diakonikolas, Daniel M. Kane, Sihan Liu, Thanasis Pittas
Published: 2025-11-24
Categories: cs.DS, cs.IT, cs.LG, math.ST, stat.ML
arXiv: https://arxiv.org/abs/2511.19398v1

This work studies information-computation gaps for statistical problems. A common approach for providing evidence of such gaps is to show sample complexity lower bounds (that are stronger than the information-theoretic optimum) against natural models of computation. A popular such model in the literature is the family of low-degree polynomial tests. While these tests are defined in such a way that make them easy to analyze, the class of algorithms that they rule out is somewhat restricted. An important goal in this context has been to obtain lower bounds against the stronger and more natural class of low-degree Polynomial Threshold Function (PTF) tests, i.e., any test that can be expressed as comparing some low-degree polynomial of the data to a threshold. Proving lower bounds against PTF tests has turned out to be challenging. Indeed, we are not aware of any non-trivial PTF testing lower bounds in the literature. In this paper, we establish the first non-trivial PTF testing lower bounds for a range of statistical tasks. Specifically, we prove a near-optimal PTF testing lower bound for Non-Gaussian Component Analysis (NGCA). Our NGCA lower bound implies similar lower bounds for a number of other statistical problems. Our proof leverages a connection to recent work on pseudorandom generators for PTFs and recent techniques developed in that context. At the technical level, we develop several tools of independent interest, including novel structural results for analyzing the behavior of low-degree polynomials restricted to random directions.

---

### Real-Time Object Tracking with On-Device Deep Learning for Adaptive Beamforming in Dynamic Acoustic Environments

Authors: Jorge Ortigoso-Narro, Jose A. Belloch, Adrian Amor-Martin, Sandra Roger, Maximo Cobos
Published: 2025-11-24
Categories: cs.SD, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2511.19396v1

Advances in object tracking and acoustic beamforming are driving new capabilities in surveillance, human-computer interaction, and robotics. This work presents an embedded system that integrates deep learning-based tracking with beamforming to achieve precise sound source localization and directional audio capture in dynamic environments. The approach combines single-camera depth estimation and stereo vision to enable accurate 3D localization of moving objects. A planar concentric circular microphone array constructed with MEMS microphones provides a compact, energy-efficient platform supporting 2D beam steering across azimuth and elevation. Real-time tracking outputs continuously adapt the array's focus, synchronizing the acoustic response with the target's position. By uniting learned spatial awareness with dynamic steering, the system maintains robust performance in the presence of multiple or moving sources. Experimental evaluation demonstrates significant gains in signal-to-interference ratio, making the design well-suited for teleconferencing, smart home devices, and assistive technologies.

---

### BackSplit: The Importance of Sub-dividing the Background in Biomedical Lesion Segmentation

Authors: Rachit Saluja, Asli Cihangir, Ruining Deng, Johannes C. Paetzold, Fengbei Liu, Mert R. Sabuncu
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19394v2

Segmenting small lesions in medical images remains notoriously difficult. Most prior work tackles this challenge by either designing better architectures, loss functions, or data augmentation schemes; and collecting more labeled data. We take a different view, arguing that part of the problem lies in how the background is modeled. Common lesion segmentation collapses all non-lesion pixels into a single "background" class, ignoring the rich anatomical context in which lesions appear. In reality, the background is highly heterogeneous-composed of tissues, organs, and other structures that can now be labeled manually or inferred automatically using existing segmentation models. In this paper, we argue that training with fine-grained labels that sub-divide the background class, which we call BackSplit, is a simple yet powerful paradigm that can offer a significant performance boost without increasing inference costs. From an information theoretic standpoint, we prove that BackSplit increases the expected Fisher Information relative to conventional binary training, leading to tighter asymptotic bounds and more stable optimization. With extensive experiments across multiple datasets and architectures, we empirically show that BackSplit consistently boosts small-lesion segmentation performance, even when auxiliary labels are generated automatically using pretrained segmentation models. Additionally, we demonstrate that auxiliary labels derived from interactive segmentation frameworks exhibit the same beneficial effect, demonstrating its robustness, simplicity, and broad applicability.

---

### Predicting partially observable dynamical systems via diffusion models with a multiscale inference scheme

Authors: Rudy Morel, Francesco Pio Ramunno, Jeff Shen, Alberto Bietti, Kyunghyun Cho, Miles Cranmer, Siavash Golkar, Olexandr Gugnin, Geraud Krawezik, Tanya Marwah, Michael McCabe, Lucas Meyer et al.
Published: 2025-11-24
Categories: cs.LG, astro-ph.SR, cs.AI, stat.ML
arXiv: https://arxiv.org/abs/2511.19390v1

Conditional diffusion models provide a natural framework for probabilistic prediction of dynamical systems and have been successfully applied to fluid dynamics and weather prediction. However, in many settings, the available information at a given time represents only a small fraction of what is needed to predict future states, either due to measurement uncertainty or because only a small fraction of the state can be observed. This is true for example in solar physics, where we can observe the Sun's surface and atmosphere, but its evolution is driven by internal processes for which we lack direct measurements. In this paper, we tackle the probabilistic prediction of partially observable, long-memory dynamical systems, with applications to solar dynamics and the evolution of active regions. We show that standard inference schemes, such as autoregressive rollouts, fail to capture long-range dependencies in the data, largely because they do not integrate past information effectively. To overcome this, we propose a multiscale inference scheme for diffusion models, tailored to physical processes. Our method generates trajectories that are temporally fine-grained near the present and coarser as we move farther away, which enables capturing long-range temporal dependencies without increasing computational cost. When integrated into a diffusion model, we show that our inference scheme significantly reduces the bias of the predicted distributions and improves rollout stability.

---

### UISearch: Graph-Based Embeddings for Multimodal Enterprise UI Screenshots Retrieval

Authors: Maroun Ayli, Youssef Bakouny, Tushar Sharma, Nader Jalloul, Hani Seifeddine, Rima Kilany
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19380v1

Enterprise software companies maintain thousands of user interface screens across products and versions, creating critical challenges for design consistency, pattern discovery, and compliance check. Existing approaches rely on visual similarity or text semantics, lacking explicit modeling of structural properties fundamental to user interface (UI) composition. We present a novel graph-based representation that converts UI screenshots into attributed graphs encoding hierarchical relationships and spatial arrangements, potentially generalizable to document layouts, architectural diagrams, and other structured visual domains. A contrastive graph autoencoder learns embeddings preserving multi-level similarity across visual, structural, and semantic properties. The comprehensive analysis demonstrates that our structural embeddings achieve better discriminative power than state-of-the-art Vision Encoders, representing a fundamental advance in the expressiveness of the UI representation. We implement this representation in UISearch, a multi-modal search framework that combines structural embeddings with semantic search through a composable query language. On 20,396 financial software UIs, UISearch achieves 0.92 Top-5 accuracy with 47.5ms median latency (P95: 124ms), scaling to 20,000+ screens. The hybrid indexing architecture enables complex queries and supports fine-grained UI distinction impossible with vision-only approaches.

---

### Efficiency vs. Fidelity: A Comparative Analysis of Diffusion Probabilistic Models and Flow Matching on Low-Resource Hardware

Authors: Srishti Gupta, Yashasvee Taiwade
Published: 2025-11-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.19379v1

Denoising Diffusion Probabilistic Models (DDPMs) have established a new state-of-the-art in generative image synthesis, yet their deployment is hindered by significant computational overhead during inference, often requiring up to 1,000 iterative steps. This study presents a rigorous comparative analysis of DDPMs against the emerging Flow Matching (Rectified Flow) paradigm, specifically isolating their geometric and efficiency properties on low-resource hardware. By implementing both frameworks on a shared Time-Conditioned U-Net backbone using the MNIST dataset, we demonstrate that Flow Matching significantly outperforms Diffusion in efficiency. Our geometric analysis reveals that Flow Matching learns a highly rectified transport path (Curvature $\mathcal{C} \approx 1.02$), which is near-optimal, whereas Diffusion trajectories remain stochastic and tortuous ($\mathcal{C} \approx 3.45$). Furthermore, we establish an ``efficiency frontier'' at $N=10$ function evaluations, where Flow Matching retains high fidelity while Diffusion collapses. Finally, we show via numerical sensitivity analysis that the learned vector field is sufficiently linear to render high-order ODE solvers (Runge-Kutta 4) unnecessary, validating the use of lightweight Euler solvers for edge deployment. \textbf{This work concludes that Flow Matching is the superior algorithmic choice for real-time, resource-constrained generative tasks.}

---

### LLM-Driven Stationarity-Aware Expert Demonstrations for Multi-Agent Reinforcement Learning in Mobile Systems

Authors: Tianyang Duan, Zongyuan Zhang, Zheng Lin, Songxiao Guo, Xiuxian Guan, Guangyu Wu, Zihan Fang, Haotian Meng, Xia Du, Ji-Zhe Zhou, Heming Cui, Jun Luo et al.
Published: 2025-11-24
Categories: cs.LG, cs.NI
arXiv: https://arxiv.org/abs/2511.19368v1

Multi-agent reinforcement learning (MARL) has been increasingly adopted in many real-world applications. While MARL enables decentralized deployment on resource-constrained edge devices, it suffers from severe non-stationarity due to the synchronous updates of agent policies. This non stationarity results in unstable training and poor policy con vergence, especially as the number of agents increases. In this paper, we propose RELED, a scalable MARL framework that integrates large language model (LLM)-driven expert demonstrations with autonomous agent exploration. RELED incorporates a Stationarity-Aware Expert Demonstration module, which leverages theoretical non-stationarity bounds to enhance the quality of LLM-generated expert trajectories, thus providing high reward and training-stable samples for each agent. Moreover, a Hybrid Expert-Agent Policy Optimization module adaptively balances each agent's learning from both expert-generated and agent-generated trajectories, accelerating policy convergence and improving generalization. Extensive experiments with real city networks based on OpenStreetMap demonstrate that RELED achieves superior performance compared to state-of-the-art MARL methods.

---

### AnatomicalNets: A Multi-Structure Segmentation and Contour-Based Distance Estimation Pipeline for Clinically Grounded Lung Cancer T-Staging

Authors: Saniah Kayenat Chowdhury, Rusab Sarmun, Muhammad E. H. Chowdhury, Sohaib Bassam Zoghoul, Israa Al-Hashimi, Adam Mushtak, Amith Khandakar
Published: 2025-11-24
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2511.19367v2

Accurate tumor staging in lung cancer is crucial for prognosis and treatment planning and is governed by explicit anatomical criteria under fixed guidelines. However, most existing deep learning approaches treat this spatially structured clinical decision as an uninterpretable image classification problem. Tumor stage depends on predetermined quantitative criteria, including the tumor's dimensions and its proximity to adjacent anatomical structures, and small variations can alter the staging outcome. To address this gap, we propose AnatomicalNets, a medically grounded, multi-stage pipeline that reformulates tumor staging as a measurement and rule-based inference problem rather than a learned mapping. We employ three dedicated encoder-decoder networks to precisely segment the lung parenchyma, tumor, and mediastinum. The diaphragm boundary is estimated via a lung-contour heuristic, while the tumor's largest dimension and its proximity to adjacent structures are computed through a contour-based distance estimation method. These features are passed through a deterministic decision module following the international association for the study of lung cancer guidelines. Evaluated on the Lung-PET-CT-Dx dataset, AnatomicalNets achieves an overall classification accuracy of 91.36%. We report the per-stage F1-scores of 0.93 (T1), 0.89 (T2), 0.96 (T3), and 0.90 (T4), a critical evaluation aspect often omitted in prior literature. We highlight that the representational bottleneck in prior work lies in feature design rather than classifier capacity. This work establishes a transparent and reliable staging paradigm that bridges the gap between deep learning performance and clinical interpretability.

---

### DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation

Authors: Zehong Ma, Longhui Wei, Shuai Wang, Shiliang Zhang, Qi Tian
Published: 2025-11-24
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2511.19365v2

Pixel diffusion aims to generate images directly in pixel space in an end-to-end fashion. This approach avoids the limitations of VAE in the two-stage latent diffusion, offering higher model capacity. Existing pixel diffusion models suffer from slow training and inference, as they usually model both high-frequency signals and low-frequency semantics within a single diffusion transformer (DiT). To pursue a more efficient pixel diffusion paradigm, we propose the frequency-DeCoupled pixel diffusion framework. With the intuition to decouple the generation of high and low frequency components, we leverage a lightweight pixel decoder to generate high-frequency details conditioned on semantic guidance from the DiT. This thus frees the DiT to specialize in modeling low-frequency semantics. In addition, we introduce a frequency-aware flow-matching loss that emphasizes visually salient frequencies while suppressing insignificant ones. Extensive experiments show that DeCo achieves superior performance among pixel diffusion models, attaining FID of 1.62 (256x256) and 2.22 (512x512) on ImageNet, closing the gap with latent diffusion methods. Furthermore, our pretrained text-to-image model achieves a leading overall score of 0.86 on GenEval in system-level comparison. Codes are publicly available at https://github.com/Zehong-Ma/DeCo.

---

### Neural surrogates for designing gravitational wave detectors

Authors: Carlos Ruiz-Gonzalez, Sören Arlt, Sebastian Lehner, Arturs Berzins, Yehonathan Drori, Rana X Adhikari, Johannes Brandstetter, Mario Krenn
Published: 2025-11-24
Categories: cs.LG, astro-ph.IM, gr-qc, quant-ph
arXiv: https://arxiv.org/abs/2511.19364v1

Physics simulators are essential in science and engineering, enabling the analysis, control, and design of complex systems. In experimental sciences, they are increasingly used to automate experimental design, often via combinatorial search and optimization. However, as the setups grow more complex, the computational cost of traditional, CPU-based simulators becomes a major limitation. Here, we show how neural surrogate models can significantly reduce reliance on such slow simulators while preserving accuracy. Taking the design of interferometric gravitational wave detectors as a representative example, we train a neural network to surrogate the gravitational wave physics simulator Finesse, which was developed by the LIGO community. Despite that small changes in physical parameters can change the output by orders of magnitudes, the model rapidly predicts the quality and feasibility of candidate designs, allowing an efficient exploration of large design spaces. Our algorithm loops between training the surrogate, inverse designing new experiments, and verifying their properties with the slow simulator for further training. Assisted by auto-differentiation and GPU parallelism, our method proposes high-quality experiments much faster than direct optimization. Solutions that our algorithm finds within hours outperform designs that take five days for the optimizer to reach. Though shown in the context of gravitational wave detectors, our framework is broadly applicable to other domains where simulator bottlenecks hinder optimization and discovery.

---

### Enhancing Conformal Prediction via Class Similarity

Authors: Ariel Fargion, Lahav Dabah, Tom Tirer
Published: 2025-11-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.19359v2

Conformal Prediction (CP) has emerged as a powerful statistical framework for high-stakes classification applications. Instead of predicting a single class, CP generates a prediction set, guaranteed to include the true label with a pre-specified probability. The performance of different CP methods is typically assessed by their average prediction set size. In setups where the classes can be partitioned into semantic groups, e.g., diseases that require similar treatment, users can benefit from prediction sets that are not only small on average, but also contain a small number of semantically different groups. This paper begins by addressing this problem and ultimately offers a widely applicable tool for boosting any CP method on any dataset. First, given a class partition, we propose augmenting the CP score function with a term that penalizes predictions with out-of-group errors. We theoretically analyze this strategy and prove its advantages for group-related metrics. Surprisingly, we show mathematically that, for common class partitions, it can also reduce the average set size of any CP score function. Our analysis reveals the class-similarity factors behind this improvement and motivates a variant that can further reduce prediction set size by leveraging the model's embeddings, without requiring any human semantic partition. Finally, we present an extensive empirical study, encompassing prominent CP methods, multiple models, and several datasets, which demonstrates that our class-similarity-based approach consistently enhances CP methods.

---

### Rethinking Reward Signals in Video GRPO: When Scores Become Targets

Authors: Rui Li, Yuanzhi Liang, Ziqi Ni, Haibin Huang, Chi Zhang, Xuelong Li
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19356v4

Group Relative Policy Optimization (GRPO) enables stable and preference-oriented updates via group-wise comparisons for post-training video generation. However, GRPO directly optimizes reward-induced advantages. Under sustained optimization, the reward score can lose fidelity as a proxy for true video quality, consistent with the phenomenon described by Goodhart's Law. This leads to two recurring issues: (i) shortcut-driven optimization under composite objectives and (ii) reward saturation within prompt groups. To address these issues, we introduce TaRoS, a Target-Robust Reward Signaling framework for Video generation GRPO. TaRoS leverages component level performance assessment together with intra-group sparsity to organize multi-aspect rewards towards optimization objectives. In addition, it adaptively downweights components that exhibit saturation, thereby preserving effective optimization directions and mitigating redundancy. This maintains meaningful optimization directions and preserves within-group ranking separation, thereby preventing reward hacking and leading to more reliable policy updates. Extensive experiments show consistent improvements in visual fidelity, motion coherence, and text-video alignment over strong baselines.

---

### Leveraging LLMs for reward function design in reinforcement learning control tasks

Authors: Franklin Cardenoso, Wouter Caarls
Published: 2025-11-24
Categories: cs.LG, cs.AI, cs.RO
arXiv: https://arxiv.org/abs/2511.19355v1

The challenge of designing effective reward functions in reinforcement learning (RL) represents a significant bottleneck, often requiring extensive human expertise and being time-consuming. Previous work and recent advancements in large language models (LLMs) have demonstrated their potential for automating the generation of reward functions. However, existing methodologies often require preliminary evaluation metrics, human-engineered feedback for the refinement process, or the use of environmental source code as context. To address these limitations, this paper introduces LEARN-Opt (LLM-based Evaluator and Analyzer for Reward functioN Optimization). This LLM-based, fully autonomous, and model-agnostic framework eliminates the need for preliminary metrics and environmental source code as context to generate, execute, and evaluate reward function candidates from textual descriptions of systems and task objectives. LEARN-Opt's main contribution lies in its ability to autonomously derive performance metrics directly from the system description and the task objective, enabling unsupervised evaluation and selection of reward functions. Our experiments indicate that LEARN-Opt achieves performance comparable to or better to that of state-of-the-art methods, such as EUREKA, while requiring less prior knowledge. We find that automated reward design is a high-variance problem, where the average-case candidate fails, requiring a multi-run approach to find the best candidates. Finally, we show that LEARN-Opt can unlock the potential of low-cost LLMs to find high-performing candidates that are comparable to, or even better than, those of larger models. This demonstrated performance affirms its potential to generate high-quality reward functions without requiring any preliminary human-defined metrics, thereby reducing engineering overhead and enhancing generalizability.

---

### CellFMCount: A Fluorescence Microscopy Dataset, Benchmark, and Methods for Cell Counting

Authors: Abdurahman Ali Mohammed, Catherine Fonder, Ying Wei, Wallapak Tavanapong, Donald S Sakaguchi, Qi Li, Surya K. Mallapragada
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19351v1

Accurate cell counting is essential in various biomedical research and clinical applications, including cancer diagnosis, stem cell research, and immunology. Manual counting is labor-intensive and error-prone, motivating automation through deep learning techniques. However, training reliable deep learning models requires large amounts of high-quality annotated data, which is difficult and time-consuming to produce manually. Consequently, existing cell-counting datasets are often limited, frequently containing fewer than $500$ images. In this work, we introduce a large-scale annotated dataset comprising $3{,}023$ images from immunocytochemistry experiments related to cellular differentiation, containing over $430{,}000$ manually annotated cell locations. The dataset presents significant challenges: high cell density, overlapping and morphologically diverse cells, a long-tailed distribution of cell count per image, and variation in staining protocols. We benchmark three categories of existing methods: regression-based, crowd-counting, and cell-counting techniques on a test set with cell counts ranging from $10$ to $2{,}126$ cells per image. We also evaluate how the Segment Anything Model (SAM) can be adapted for microscopy cell counting using only dot-annotated datasets. As a case study, we implement a density-map-based adaptation of SAM (SAM-Counter) and report a mean absolute error (MAE) of $22.12$, which outperforms existing approaches (second-best MAE of $27.46$). Our results underscore the value of the dataset and the benchmarking framework for driving progress in automated cell counting and provide a robust foundation for future research and development.

---

### Scalable Parameter-Light Spectral Method for Clustering Short Text Embeddings with a Cohesion-Based Evaluation Metric

Authors: Nikita Neveditsin, Pawan Lingras, Vijay Mago
Published: 2025-11-24
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2511.19350v2

Clustering short text embeddings is a foundational task in natural language processing, yet remains challenging due to the need to specify the number of clusters in advance. We introduce a scalable spectral method that estimates the number of clusters directly from the structure of the Laplacian eigenspectrum, constructed using cosine similarities and guided by an adaptive sampling strategy. This sampling approach enables our estimator to efficiently scale to large datasets without sacrificing reliability. To support intrinsic evaluation of cluster quality without ground-truth labels, we propose the Cohesion Ratio, a simple and interpretable evaluation metric that quantifies how much intra-cluster similarity exceeds the global similarity background. It has an information-theoretic motivation inspired by mutual information, and in our experiments it correlates closely with extrinsic measures such as normalized mutual information and homogeneity. Extensive experiments on six short-text datasets and four modern embedding models show that standard algorithms like K-Means and HAC, when guided by our estimator, significantly outperform popular parameter-light methods such as HDBSCAN, OPTICS, and Leiden. These results demonstrate the practical value of our spectral estimator and Cohesion Ratio for unsupervised organization and evaluation of short text data. Implementation of our estimator of k and Cohesion Ratio, along with code for reproducing the experiments, is available at https://anonymous.4open.science/r/towards_clustering-0C2E.

---

### Artificial Intelligence Driven Workflow for Accelerating Design of Novel Photosensitizers

Authors: Hongyi Wang, Xiuli Zheng, Weimin Liu, Zitian Tang, Sheng Gong
Published: 2025-11-24
Categories: cond-mat.mtrl-sci, cs.LG, physics.chem-ph
arXiv: https://arxiv.org/abs/2511.19347v1

The discovery of high-performance photosensitizers has long been hindered by the time-consuming and resource-intensive nature of traditional trial-and-error approaches. Here, we present \textbf{A}I-\textbf{A}ccelerated \textbf{P}hoto\textbf{S}ensitizer \textbf{I}nnovation (AAPSI), a closed-loop workflow that integrates expert knowledge, scaffold-based molecule generation, and Bayesian optimization to accelerate the design of novel photosensitizers. The scaffold-driven generation in AAPSI ensures structural novelty and synthetic feasibility, while the iterative AI-experiment loop accelerates the discovery of novel photosensitizers. AAPSI leverages a curated database of 102,534 photosensitizer-solvent pairs and generate 6,148 synthetically accessible candidates. These candidates are screened via graph transformers trained to predict singlet oxygen quantum yield ($φ_Δ$) and absorption maxima ($λ_{max}$), following experimental validation. This work generates several novel candidates for photodynamic therapy (PDT), among which the hypocrellin-based candidate HB4Ph exhibits exceptional performance at the Pareto frontier of high quantum yield of singlet oxygen and long absorption maxima among current photosensitizers ($φ_Δ$=0.85, $λ_{max}$=650nm).

---

### Annotation-Free Class-Incremental Learning

Authors: Hari Chandana Kuchibhotla, K S Ananth, Vineeth N Balasubramanian
Published: 2025-11-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.19344v1

Despite significant progress in continual learning ranging from architectural novelty to clever strategies for mitigating catastrophic forgetting most existing methods rest on a strong but unrealistic assumption the availability of labeled data throughout the learning process. In real-world scenarios, however, data often arrives sequentially and without annotations, rendering conventional approaches impractical. In this work, we revisit the fundamental assumptions of continual learning and ask: Can current systems adapt when labels are absent and tasks emerge incrementally over time? To this end, we introduce Annotation-Free Class-Incremental Learning (AFCIL), a more realistic and challenging paradigm where unlabeled data arrives continuously, and the learner must incrementally acquire new classes without any supervision. To enable effective learning under AFCIL, we propose CrossWorld CL, a Cross Domain World Guided Continual Learning framework that incorporates external world knowledge as a stable auxiliary source. The method retrieves semantically related ImageNet classes for each downstream category, maps downstream and ImageNet features through a cross domain alignment strategy and finally introduce a novel replay strategy. This design lets the model uncover semantic structure without annotations while keeping earlier knowledge intact. Across four datasets, CrossWorld-CL surpasses CLIP baselines and existing continual and unlabeled learning methods, underscoring the benefit of world knowledge for annotation free continual learning.

---

### Syn-GRPO: Self-Evolving Data Synthesis for MLLM Perception Reasoning

Authors: Qihan Huang, Haofei Zhang, Rong Wei, Yi Wang, Rui Tang, Mingli Song, Jie Song
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19343v1

RL (reinforcement learning) methods (e.g., GRPO) for MLLM (Multimodal LLM) perception ability has attracted wide research interest owing to its remarkable generalization ability. Nevertheless, existing reinforcement learning methods still face the problem of low data quality, where data samples cannot elicit diverse responses from MLLMs, thus restricting the exploration scope for MLLM reinforcement learning. Some methods attempt to mitigate this problem by imposing constraints on entropy, but none address it at its root. Therefore, to tackle this problem, this work proposes Syn-GRPO (Synthesis-GRPO), which employs an online data generator to synthesize high-quality training data with diverse responses in GRPO training. Specifically, Syn-GRPO consists of two components: (1) data server; (2) GRPO workflow. The data server synthesizes new samples from existing ones using an image generation model, featuring a decoupled and asynchronous scheme to achieve high generation efficiency. The GRPO workflow provides the data server with the new image descriptions, and it leverages a diversity reward to supervise the MLLM to predict image descriptions for synthesizing samples with diverse responses. Experiment results across three visual perception tasks demonstrate that Syn-GRPO improves the data quality by a large margin, achieving significant superior performance to existing MLLM perception methods, and Syn-GRPO presents promising potential for scaling long-term self-evolving RL. Our code is available at https://github.com/hqhQAQ/Syn-GRPO.

---

### Explicit Tonal Tension Conditioning via Dual-Level Beam Search for Symbolic Music Generation

Authors: Maral Ebrahimzadeh, Gilberto Bernardes, Sebastian Stober
Published: 2025-11-24
Categories: cs.SD, cs.AI
arXiv: https://arxiv.org/abs/2511.19342v1

State-of-the-art symbolic music generation models have recently achieved remarkable output quality, yet explicit control over compositional features, such as tonal tension, remains challenging. We propose a novel approach that integrates a computational tonal tension model, based on tonal interval vector analysis, into a Transformer framework. Our method employs a two-level beam search strategy during inference. At the token level, generated candidates are re-ranked using model probability and diversity metrics to maintain overall quality. At the bar level, a tension-based re-ranking is applied to ensure that the generated music aligns with a desired tension curve. Objective evaluations indicate that our approach effectively modulates tonal tension, and subjective listening tests confirm that the system produces outputs that align with the target tension. These results demonstrate that explicit tension conditioning through a dual-level beam search provides a powerful and intuitive tool to guide AI-generated music. Furthermore, our experiments demonstrate that our method can generate multiple distinct musical interpretations under the same tension condition.

---

### POUR: A Provably Optimal Method for Unlearning Representations via Neural Collapse

Authors: Anjie Le, Can Peng, Yuyuan Liu, J. Alison Noble
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19339v2

In computer vision, machine unlearning aims to remove the influence of specific visual concepts or training images without retraining from scratch. Studies show that existing approaches often modify the classifier while leaving internal representations intact, resulting in incomplete forgetting. In this work, we extend the notion of unlearning to the representation level, deriving a three-term interplay between forgetting efficacy, retention fidelity, and class separation. Building on Neural Collapse theory, we show that the orthogonal projection of a simplex Equiangular Tight Frame (ETF) remains an ETF in a lower dimensional space, yielding a provably optimal forgetting operator. We further introduce the Representation Unlearning Score (RUS) to quantify representation-level forgetting and retention fidelity. Building on this, we introduce POUR (Provably Optimal Unlearning of Representations), a geometric projection method with closed-form (POUR-P) and a feature-level unlearning variant under a distillation scheme (POUR-D). Experiments on CIFAR-10/100 and PathMNIST demonstrate that POUR achieves effective unlearning while preserving retained knowledge, outperforming state-of-the-art unlearning methods on both classification-level and representation-level metrics.

---

### High-throughput validation of phase formability and simulation accuracy of Cantor alloys

Authors: Changjun Cheng, Daniel Persaud, Kangming Li, Michael J. Moorehead, Natalie Page, Christian Lavoie, Beatriz Diaz Moreno, Adrien Couet, Samuel E Lofland, Jason Hattrick-Simpers
Published: 2025-11-24
Categories: cond-mat.mtrl-sci, cs.LG
arXiv: https://arxiv.org/abs/2511.19335v1

High-throughput methods enable accelerated discovery of novel materials in complex systems such as high-entropy alloys, which exhibit intricate phase stability across vast compositional spaces. Computational approaches, including Density Functional Theory (DFT) and calculation of phase diagrams (CALPHAD), facilitate screening of phase formability as a function of composition and temperature. However, the integration of computational predictions with experimental validation remains challenging in high-throughput studies. In this work, we introduce a quantitative confidence metric to assess the agreement between predictions and experimental observations, providing a quantitative measure of the confidence of machine learning models trained on either DFT or CALPHAD input in accounting for experimental evidence. The experimental dataset was generated via high-throughput in-situ synchrotron X-ray diffraction on compositionally varied FeNiMnCr alloy libraries, heated from room temperature to ~1000 °C. Agreement between the observed and predicted phases was evaluated using either temperature-independent phase classification or a model that incorporates a temperature-dependent probability of phase formation. This integrated approach demonstrates where strong overall agreement between computation and experiment exists, while also identifying key discrepancies, particularly in FCC/BCC predictions at Mn-rich regions to inform future model refinement.

---

### Learning to Reason: Training LLMs with GPT-OSS or DeepSeek R1 Reasoning Traces

Authors: Shaltiel Shmidman, Asher Fredman, Oleg Sudakov, Meriem Bendris
Published: 2025-11-24
Categories: cs.CL
arXiv: https://arxiv.org/abs/2511.19333v1

Test-time scaling, which leverages additional computation during inference to improve model accuracy, has enabled a new class of Large Language Models (LLMs) that are able to reason through complex problems by understanding the goal, turning this goal into a plan, working through intermediate steps, and checking their own work before answering . Frontier large language models with reasoning capabilities, such as DeepSeek-R1 and OpenAI's gpt-oss, follow the same procedure when solving complex problems by generating intermediate reasoning traces before giving the final answer. Today, these models are being increasingly used to generate reasoning traces that serve as high-quality supervised data for post-training of small and medium-sized language models to teach reasoning capabilities without requiring expensive human curation. In this work, we compare the performance of medium-sized LLMs on Math problems after post-training on two kinds of reasoning traces. We compare the impact of reasoning traces generated by DeepSeek-R1 and gpt-oss LLMs in terms of accuracy and inference efficiency.

---

### Targeted Manipulation: Slope-Based Attacks on Financial Time-Series Data

Authors: Dominik Luszczynski
Published: 2025-11-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.19330v1

A common method of attacking deep learning models is through adversarial attacks, which occur when an attacker specifically modifies the input of a model to produce an incorrect result. Adversarial attacks have been deeply investigated in the image domain; however, there is less research in the time-series domain and very little for forecasting financial data. To address these concerns, this study aims to build upon previous research on adversarial attacks for time-series data by introducing two new slope-based methods aimed to alter the trends of the predicted stock forecast generated by an N-HiTS model. Compared to the normal N-HiTS predictions, the two new slope-based methods, the General Slope Attack and Least-Squares Slope Attack, can manipulate N-HiTS predictions by doubling the slope. These new slope attacks can bypass standard security mechanisms, such as a discriminator that filters real and perturbed inputs, reducing a 4-layered CNN's specificity to 28% and accuracy to 57%. Furthermore, the slope based methods were incorporated into a GAN architecture as a means of generating realistic synthetic data, while simultaneously fooling the model. Finally, this paper also proposes a sample malware designed to inject an adversarial attack in the model inference library, proving that ML-security research should not only focus on making the model safe, but also securing the entire pipeline.

---

### Understanding the Staged Dynamics of Transformers in Learning Latent Structure

Authors: Rohan Saha, Farzane Aminmansour, Alona Fyshe
Published: 2025-11-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.19328v3

Language modeling has shown us that transformers can discover latent structure from context, but the dynamics of how they acquire different components of that structure remain poorly understood, leading to assertions that models just remix training data. In this work, we use the Alchemy benchmark in a controlled setting (Wang et al.,2021) to investigate latent structure learning. We train a small decoder-only transformer on three task variants: 1) inferring missing transitions from partial contextual information, 2) composing simple rules to solve multi-transition sequences, and 3) decomposing complex multi-step examples to infer intermediate transitions. By factorizing each task into interpretable components, we show that the model learns the different latent structure components in discrete stages. We also observe an asymmetry: the model composes fundamental transitions robustly, but struggles to decompose complex examples to discover the atomic transitions. Finally, using causal interventions, we identify layer-specific plasticity windows during which freezing substantially delays or prevents stage completion. These findings provide insight into how a transformer model acquires latent structure, offering a detailed view of how capabilities evolve during training.

---