论文检索

Social Aspects · Alignment

Jing Yao, Xiaoyuan Yi, Jindong Wang, Zhicheng Dou, Xing Xie

As Large Language Models (LLMs) more deeply integrate into human life across various regions, aligning them with pluralistic cultures is crucial for improving user engagement and mitigating cultural conflicts. For this purpose, recently, different culture-specific corpora have been carefully curated, either synthesized or manually annotated. Nevertheless, inspired by culture theories, we identify two key challenges faced by these datasets: (1) Representativeness: These corpora fail to fully capture the target culture's core characteristics, causing insufficient cultural coverage with redundancy; (2) Distinctiveness: They struggle to distinguish the unique nuances of a given culture from shared patterns across other relevant ones, hindering precise cultural modelling. To handle these challenges, we introduce CAReDiO, a novel data optimization framework, which alternatively refines culture-sensitive questions and responses according to information-theoretic objectives in an in-context optimization manner, enhancing the cultural informativeness and distinguishability of constructed data. Extensive experiments on 15 distinct cultures demonstrate that CAReDiO can create high-quality data with richer cultural information and enable efficient alignment of small open-source or large proprietary LLMs with as few as 200 training samples, consistently outperforming previous datasets in both multi-choice and open-ended cultural benchmarks.

Applications · Computer Vision

Dong Liang, yuhao liu, Jinyuan Jia, Youjun Zhao, Rynson Lau

Being able to edit panoramic images is crucial for creating realistic 360° visual experiences. However, existing perspective-based image editing methods fail to model the spatial structure of panoramas. Conventional cube-map decompositions attempt to overcome this problem but inevitably break global consistency due to their mismatch with spherical geometry. Motivated by this insight, we reformulate panoramic editing directly in the equirectangular projection (ERP) domain and present World-Shaper, a unified geometry-aware framework that bridges panoramic generation and editing within a single editing-centric design. To overcome the scarcity of paired data, we adopt a generate-then-edit paradigm, where controllable panoramic generation serves as an auxiliary stage to synthesize diverse paired examples for supervised editing learning. To address geometric distortion, we introduce a geometry-aware learning strategy that explicitly enforces position-aware shape supervision and implicitly internalizes panoramic priors through progressive training. Extensive experiments on our new benchmark, PEBench, demonstrate that our method achieves superior geometric consistency, editing fidelity, and text controllability compared to SOTA methods, enabling coherent and flexible 360° visual world creation with unified editing control. Code, model, and data will be released.

Applications · Computer Vision

Shitian Zhao, Shaoheng Lin, Ming Li, Haoquan Zhang, Wenshuo Peng, Kaipeng Zhang, Chen Wei

Reinforcement learning for agentic multimodal models often suffers from interaction collapse, where models learn to reduce tool usage and multi-turn reasoning, limiting the benefits of agentic behavior. We introduce PyVision-RL, a reinforcement learning framework for open-weight multimodal models that stabilizes training and sustains interaction. Our approach combines an oversampling–filtering–ranking rollout strategy with an accumulative tool reward to prevent collapse and encourage multi-turn tool use. Using a unified training pipeline, we develop PyVision-Image and PyVision-Video for image and video understanding. For video reasoning, PyVision-Video employs on-demand context construction, selectively sampling task-relevant frames during reasoning to significantly reduce visual token usage. Experiments show strong performance and improved efficiency, demonstrating that sustained interaction and on-demand visual processing are critical for scalable multimodal agents.

Applications · Computer Vision

Ioana Simion, Mohammadreza Salehi, Shashanka Venkataramanan, Cees Snoek, Yuki Asano

Visual foundation models have achieved remarkable progress in scale and versatility, yet understanding the 3D world remains a fundamental challenge. While 2D images contain cues about 3D structure that humans readily interpret, deep models often fail to exploit them, underperforming on tasks such as multiview semantic consistency--crucial for applications including robotics and autonomous driving. We propose a self-supervised approach to enhance the 3D understanding of vision foundation models by (i) introducing a temporal nearest-neighbor consistency loss that finds corresponding points across video frames and enforces consistency between their nearest neighbors, (ii) incorporating reference-guided ordering that requires patch-level features to be not only expressive but also consistently aligned, and (iii) constructing a mixture of video datasets tailored to these objectives, thereby leveraging rich 3D information. Our method, 3DPoV, achieves state-of-the-art performance in keypoint matching under viewpoint variation, as well as in depth and surface normal estimation, and consistently improves a diverse set of backbones, including DINOv3.

Social Aspects · Accountability, Transparency, and Interpretability

Vikram Natarajan, Devina Jain, Shivam Arora, Satvik Golechha, Joseph Bloom

Linear probes are a promising approach for monitoring AI systems for deceptive behaviour. Previous work has shown that a linear classifier trained on a contrastive instruction pair and a simple dataset can achieve good performance. However, these probes exhibit notable failures even in straightforward scenarios, including spurious correlations and false positives on non-deceptive responses. In this paper, we identify the importance of the instruction pair used during training. Furthermore, we show that targeting specific deceptive behaviors through a human-interpretable taxonomy of deception leads to improved results on evaluation datasets. Our findings reveal that instruction pairs capture deceptive \emph{intent} rather than content-specific patterns, explaining why prompt choice dominates probe performance (70.6\% of variance). Given the heterogeneity of deception types across datasets, we conclude that organizations should design specialized probes targeting their specific threat models rather than seeking a universal deception detector.

Deep Learning · Large Language Models

Tessa Han, Sebastian Bordt, Hanlin Zhang, Sham Kakade

The prevailing paradigm in large language model (LLM) development is to pretrain a base model, then perform further training to improve performance and model behavior. However, hyperparameter optimization and scaling laws have been studied primarily from the perspective of the base model's validation loss, ignoring downstream adaptability. In this work, we study pretraining from the perspective of model plasticity, that is, the ability of the base model to successfully adapt to downstream tasks through fine-tuning. We focus on the role of weight decay, a key regularization parameter during pretraining. Through systematic experiments, we show that models trained with larger weight decay values are more plastic, meaning they show larger performance gains when fine-tuned on downstream tasks. This phenomenon can lead to counterintuitive trade-offs where base models that perform worse after pretraining can perform better after fine-tuning. Further investigation of weight decay's mechanistic effects on model behavior reveals that it encourages linearly separable representations, regularizes attention matrices, and reduces overfitting on the training data. In conclusion, this work casts light on the multifaceted role that a single optimization hyperparameter can play in shaping model behavior and demonstrates the importance of using evaluation metrics beyond the cross-entropy loss for hyperparameter optimization.

Deep Learning · Foundation Models

Wonbong Jang, Shikun Liu, Soubhik Sanyal, Juan Perez, Kam Woh Ng, Sanskar Agrawal, Juan-Manuel Perez-Rua, Yiannis Douratsos, Tao Xiang

Can we bridge the gap between perceiving camera trajectories and rendering novel views within a single generative framework? Recovering camera parameters from images and rendering scenes from novel viewpoints are considered the forward and inverse problems in the field of computer vision and graphics. Previous approaches treat these problems in isolation, often failing when image coverage is sparse or camera poses are ambiguous. In this work, we propose Rays as Pixels, a specialized Video Diffusion Model (VDM) that learns a joint distribution of videos and camera trajectories. We represent cameras as dense ray pixels (raxels) and simultaneously denoise them alongside video frames using a novel Decoupled Self-Cross Attention. This joint formulation enables us to: i) generate a video from multiple input images following a defined camera trajectory, ii) perform novel view synthesis from sparse views (without necessarily requiring camera poses), and iii) predict the camera trajectory from a raw video. We evaluate our model on pose estimation, camera-controlled video generation and validate its self-consistency. Please reference supplementary material for more qualitative results.

Optimization · Discrete and Combinatorial Optimization

Achref Jaziri, Thibaut Cuvelier, Bruno De Backer

Large-scale Set Cover Problems (SCP) with millions of variables and complex cost structures require high-quality solutions within seconds, yet remain beyond the reach of exact solvers and pose severe generalization challenges for neural methods. Such problems necessitate decomposition into bounded subproblems; however, when the induced subproblem topology differs from that observed during training, existing neural approaches often fail to transfer reliably. We introduce Generative Large Neighborhood Search (GLNS), which reframes neighborhood selection as generation using a discrete diffusion model. Our key insight is that the diffusion denoising trajectory exposes variables exhibiting high prediction instability across timesteps and identifies regions where local repair yields downstream improvement. GLNS exploits this trajectory-level signal to construct high-impact neighborhoods via a localized, bounded-complexity generative sampling procedure, enabling robust neighborhood selection without retraining. As a result, GLNS transfers effectively across cost regimes and instance scales within SCP. Under tight and equal wall-clock budgets, GLNS consistently outperforms established neural baselines and achieves competitive performance with state-of-the-art MIP solvers. These results demonstrate trajectory-guided generation as a scalable framework for large-scale SCP and suggest potential relevance to other constrained optimization settings.

Applications · Chemistry, Physics, and Earth Sciences

Zhenyuan Zhao, Yu Xing, Tianyang Xue, Lingxin Cao, Xin Yan, Lin Lu

Designing microstructures that satisfy coupled cross-physics objectives is a fundamental challenge in material science. This inverse design problem involves a vast, discontinuous search space where traditional topology optimization is computationally prohibitive, and deep generative models often suffer from "physical hallucinations," lacking the capability to ensure rigorous validity. To address this limitation, we introduce AutoMS, a multi-agent neuro-symbolic framework that reformulates inverse design as an LLM-driven evolutionary search. Unlike methods that treat LLMs merely as interfaces, AutoMS integrates them as "semantic navigators" to initialize search spaces and break local optima, while our novel Simulation-Aware Evolutionary Search (SAES) addresses the "blindness" of traditional evolutionary strategies. Specifically, SAES utilizes simulation feedback to perform local gradient approximation and directed parameter updates, effectively guiding the search toward physically valid Pareto frontiers. Orchestrating specialized agents (Manager, Parser, Generator, and Simulator), AutoMS achieves a state-of-the-art 83.8\% success rate on 17 diverse cross-physics tasks, nearly doubling the performance of traditional NSGA-II (43.7\%) and significantly outperforming ReAct-based LLM baselines (53.3\%). Furthermore, our hierarchical architecture reduces total execution time by 23.3\%. AutoMS demonstrates that autonomous agent systems can effectively navigate complex physical landscapes, bridging the gap between semantic design intent and rigorous physical validity.

Deep Learning · Large Language Models

Zichao Li, Jie Lou, Fangchen Dong, Zhiyuan Fan, Mengjie Ren, Hongyu Lin, Xianpei Han, Debing Zhang, Le Sun, Yaojie Lu 等

Reinforcement learning significantly enhances LLM capabilities but suffers from a critical issue: length inflation, where models adopt verbosity or inefficient reasoning to maximize rewards. Prior approaches struggle to address this challenge in a general and lossless manner, primarily because additive penalties introduce a compensatory effect that creates optimization shortcuts, while heuristic gating strategies lack generality beyond binary feedback. To bridge this gap, we present Group Relative Reward Rescaling (GR$^3$), which reframes length control as a multiplicative rescaling paradigm, effectively establishing a generalized, continuous, and reward-dependent gating mechanism. To further ensure lossless optimization, we incorporate group-relative regularization and advantage-aware calibration, which dynamically adapt length budgets to instance difficulty and preserve the advantage signal of high-quality trajectories. Empirically, across both RLHF and RLVR settings, GR$^3$ maintains training dynamics and downstream performance comparable to standard GRPO while significantly mitigating length inflation, outperforming state-of-the-art length-regularized baselines.

Deep Learning · Large Language Models

Yewei Liu, Xiyuan Wang, Yansheng Mao, Yoav Gelberg, Haggai Maron, Muhan Zhang

We propose SHINE (Scalable Hyper In-context NEtwork), a scalable hypernetwork that can map diverse meaningful contexts into high-quality LoRA adapters for large language models (LLM). By reusing the frozen LLM's own parameters in an in-context hypernetwork design and introducing architectural innovations, SHINE overcomes key limitations of prior hypernetworks and achieves strong expressive power with a relatively small number of parameters. We introduce a pretraining and instruction fine-tuning pipeline, and train our hypernetwork to generate high quality LoRA adapters from diverse meaningful contexts in a single forward pass. It updates LLM parameters without any fine-tuning, and immediately enables complex question answering tasks related to the context without directly accessing the context, effectively transforming in-context knowledge to in-parameter knowledge in one pass. Our work achieves outstanding results on various tasks, greatly saves time, computation and memory costs compared to SFT-based LLM adaptation, and shows great potential for scaling. Our code is available at https://anonymous.4open.science/r/metalora-734E

Optimization · Zero-order and Black-box Optimization

Wei LIN, Yining Jiang, Qingyu Song, Qiao Xiang, Hong Xu

Zeroth-Order (ZO) optimization has emerged as a promising solution for fine-tuning LLMs under strict memory constraints, as it avoids the prohibitive memory cost of storing activations for backpropagation. However, existing ZO methods typically employ isotropic perturbations, neglecting the rich structural information available during the forward pass. In this paper, we identify a crucial link between gradient formation and activation structure: the gradient of a linear layer is confined to the subspace spanned by its input activations. Leveraging this insight, we propose Activation-Guided Zeroth-Order optimization (AGZO). Unlike prior methods, AGZO extracts a compact, activation-informed subspace on the fly during the forward pass and restricts perturbations to this low-rank subspace. We provide a theoretical framework showing that AGZO optimizes a subspace-smoothed objective and provably yields update directions with higher cosine similarity to the true gradient than isotropic baselines. Empirically, we evaluate AGZO on Qwen3 and Pangu models across various benchmarks. AGZO consistently outperforms state-of-the-art ZO baselines and significantly narrows the performance gap with first-order fine-tuning, while maintaining almost the same peak memory footprint as other ZO methods.

Deep Learning · Large Language Models

Zhuoqun Li, Boxi Cao, Guiping Jiang, Fangrui Lv, Ruotong Pan, Jianan Wang, Xiangyu Wu, Hongyu Lin, Yaojie Lu, Yong Du 等

Metaphorical videos are prevalent across various real-world scenarios to convey complex ideas, and understanding them typically requires high-order cognitive capabilities. The lack of systematic studies on metaphorical video understanding not only constrains the real-world applicability of MLLMs but also impedes the thorough assessment of their high-order cognitive capabilities. To bridge this gap, we propose MetaphorVU-Bench, the first systematic and comprehensive benchmark dedicated to metaphorical video understanding. Through experiments, we find current MLLMs struggle with accurate metaphorical video understanding, lagging far behind human level, primarily due to defective cross-domain mapping. Motivated by this finding, we construct a metaphor knowledge graph as mapping augmentation and propose MetaphorBoost, an inference-time enhancement framework achieving consistent performance improvement. Our benchmark, analysis, and method provide useful insights and a foundation for future research on advancing MLLMs.

Danning Zhang, Yijing Lin, Shuhan Zhuang, Mengqi Huang, Shaojin Wu, Shancheng Fang, Zhendong Mao

Multi-modal image generation, particularly subject-driven customization, has garnered growing attention in recent years. Despite the rapid advancement of generative models, their evaluation remains largely lagging. Existing methods, whether embedding-based or Multi-modal Large Language Model (MLLM)-based, evaluate alignment with each modal condition in isolation, which contradicts the simultaneous condition alignment objective of multi-modal image generation, leading to poor consistency with human judgments. To address this challenge, we propose \textbf{UFO}, the first \textbf{U}ni\textbf{F}ied framework for \textbf{O}mini-condition alignment simultaneous evaluation. Specifically, UFO introduces a novel Atomized Chain-of-Evaluation paradigm, \emph{i.e.}, it first decomposes omni-condition alignment into a sequential chain of fine-grained, disentangled Atomic Evaluation Units (AEUs), categorizes them into distinct modality-relevance classes, and then employs general or dedicated functional calls for accurate verification of different AEU types. Experimental results demonstrate that UFO achieves the highest correlation with human evaluation preferences, delivering an average improvement of 15.25\%. Furthermore, we present UFO-Bench, a dedicated benchmark designed to holistically evaluate the performance of existing customization models under the diverse mutual interactions of textual and visual conditions.

General Machine Learning · Kernel methods

Seok-Jin Kim, Kaizheng Wang

We consider the problem of estimating the effect function of a continuous treatment, which maps each treatment value to its average effect over the population. A central challenge in this setting is confounding: treatment assignment often depends on covariates, creating selection bias that makes direct regression of the response on the treatment unreliable. To address this issue, we propose a two-stage kernel ridge regression method. In the first stage, we learn a model for the response as a function of both treatment and covariates; in the second stage, we use this model to construct pseudo-outcomes that correct for the distribution shift and then fit a second model to estimate the treatment effect. Although the response varies with both treatment and covariates, the induced effect function obtained by averaging over covariates is typically much simpler, and our estimator adapts to this structure. Furthermore, we introduce a fully data-driven model selection procedure that achieves provable adaptivity to both the unknown degree of overlap and the regularity (eigenvalue decay) of the underlying kernel.

Deep Learning · Large Language Models

Xueru Wen, Jie Lou, Yanjiang Liu, Hongyu Lin, Ben He, Xianpei Han, Le Sun, Yaojie Lu, Debing Zhang

While reinforcement learning has achieved impressive progress in language model reasoning, it is constrained by the requirement for verifiable rewards. Recent verifier-free RL methods address this limitation by utilizing the probabilities that LLMs generate reference answers as reward signals. However, these approaches typically sample reasoning traces conditioned only on the question. This design decouples reasoning-trace sampling from answer information, leading to inefficient exploration and incoherence between traces and final answers. In this paper, we propose \textit{\b{Co}upled \b{V}ariational \b{R}einforcement \b{L}earning} (CoVRL), which bridges variational inference and reinforcement learning by coupling prior and posterior distributions through a hybrid sampling strategy. By constructing and optimizing a composite distribution that integrates these two distributions, CoVRL enables efficient exploration while preserving strong thought-answer coherence. Extensive experiments on mathematical and general reasoning benchmarks show that CoVRL improves performance by 12.4\% over the base model and achieves an additional 2.3\% improvement over state-of-the-art verifier-free RL baselines, providing a principled framework for enhancing the general reasoning capabilities of language models.

Deep Learning · Large Language Models

Yansheng Mao, Yufei Xu, Jiaqi Li, Fanxu Meng, Haotong Yang, Zilong Zheng, Xiyuan Wang, Muhan Zhang

Long context understanding remains challenging for large language models due to their limited context windows. This paper introduces Long Input Fine-Tuning (LIFT), a novel framework for long-context modeling that can enhance the long-context performance of arbitrary short-context LLMs by dynamically adapting their parameters to the given long input. Importantly, rather than endlessly extending the context window size to accommodate increasingly longer inputs in context, LIFT stores and absorbs the long input in parameters. By fine-tuning the long input into model parameters, LIFT allows short-context LLMs to answer questions even when the required information is not provided in the context during inference, avoiding the quadratic complexity w.r.t. input length of a normal long context model. Furthermore, LIFT does not simply perform continued pretraining on new, long contexts, but leverages carefully designed LLM-generated synthetic tasks to enhance the comprehension of long contexts, moving beyond mere memorization. To accommodate the additional cost of fine-tuning, we design a highly optimized pipeline that reduces the Time to First Token (TTFT) to less than 10 seconds for 8k context. We further provide a comprehensive analysis of LIFT's strengths and limitations in long-context understanding, discuss its feasibility for large-scale real-world deployment, and highlight valuable directions for future research.

Applications · Computer Vision

Yuchen Xian, Yunqiu Xu, Yang He, Yi Yang

Multimodal image fusion (MMIF) aims to integrate complementary information from different modalities into a single fused image that preserves *fine local details* while maintaining *globally consistent appearance*. Most existing approaches build shared representations on 2D feature grids, which excel at modeling local structures but offer limited leverage over image-level appearance factors. To better optimize two objectives jointly, we redesign the shared representation by mapping inputs into a compact sequence of discrete *1D image tokens*, and instantiate this design with TiTok as a lightweight tokenizer, decoupling the shared representation from fixed pixel locations and concentrating image-level attributes into a small set of global tokens. We propose *Selective Token Editing* (STE): we sparsely update/replace only a small set of critical shared tokens, providing a lightweight token-level mechanism to steer global appearance coherence while keeping the fusion backbone unchanged and avoiding complex loss designs. Experiments on multiple benchmarks show that our method delivers *consistent, multi-metric improvements*—enhancing global coherence and local fidelity simultaneously—and achieves the *best overall performance* under comprehensive evaluation.

Deep Learning · Large Language Models

Zhengzhao Ma, Xueru Wen, Boxi Cao, Yaojie Lu, Hongyu Lin, Jinglin Yang, Min He, Xianpei Han, Le Sun

Reinforcement Learning from Verifiable Rewards (RLVR) significantly enhances large language models (LLMs) reasoning but severely suffers from calibration degeneration, where models become excessively over-confident in incorrect answers. Previous studies devote to directly incorporating calibration objective into existing optimization target. However, our theoretical analysis demonstrates that there exists a fundamental gradient conflict between the optimization for maximizing policy accuracy and minimizing calibration error. Building on this insight, we propose DCPO, a simple yet effective framework that systematically decouples reasoning and calibration objectives. Extensive experiments demonstrate that our DCPO not only preserves accuracy on par with GRPO but also achieves the best calibration performance and substantially mitigates the over-confidence issue. Our study provides valuable insights and practical solution for more reliable LLM deployment.

Deep Learning · Generative Models and Autoencoders

Marta Gentiloni Silveri, Giovanni Conforti, Alain Oliviero Durmus

Diffusion Flow Matching (DFM) has recently emerged as a versatile framework for generative modeling, yet its theoretical convergence properties remain only partially understood. In this work, we provide refined and novel convergence guarantees for Brownian motion based DFMs, focusing on the discretization error. Our analysis is conducted under the Kullback–Leibler (KL) divergence and the 2-Wasserstein distance. Under finite-moment and mild integrability assumptions, we derive KL convergence bounds with improved dimensional dependence compared to prior work, achieving, up to our knowledge, state-of-the-art scaling under minimal conditions. We further extend the analysis to the 2-Wasserstein distance: assuming weak log-concavity and one-sided Lipschitz continuity, we obtain convergence guarantees with dimensional dependence consistent with the KL case.