论文检索

General Machine Learning · Transfer, Multitask and Meta-learning

Shuai Yi, Yixiong Zou, Yuhua Li, Ruixuan Li

Vision-Language Models (VLMs) such as CLIP demonstrate strong zero-shot generalization, but their performance significantly degrades in cross-domain scenarios with scarce target-domain training data (Cross-Domain Few-Shot Learning, CDFSL). In this paper, we focus on the target-domain few-shot finetuning in the CLIP-based CDFSL task. Prevailing finetuning paradigms uniformly align all image patch tokens with their corresponding textual embeddings. However, we find a counterintuitive phenomenon: actively pushing away certain low-similarity image tokens, termed “tail tokens”, from their textual embeddings consistently improves target-domain performance. We delve into this phenomenon and provide a novel interpretation: under great domain shifts and scarce training data, the model can hardly extract semantic information from visual inputs; therefore, the common belief of alignment is valid only for tokens already containing sufficient semantic information; for tail tokens, forcing the alignment would lead to excessive overfitting to the scarce training, while breaking the alignment is more useful. Motivated by this, we propose Adaptive Tail-Head Alignment (ATHA), a novel fine-tuning strategy for CLIP that transforms the conventional uniform alignment paradigm to an adaptive alignment paradigm, with both alignment strengthening and weakening. Extensive experiments on four challenging CDFSL benchmarks validate our state-of-the-art performance. Our codes will be released.

Theory · Online Learning and Bandits

Yan-Feng Xie, Yu-Jie Zhang, Peng Zhao, Zhi-Hua Zhou

We study dynamic regret minimization in non-stationary online learning, with a primary focus on follow-the-regularized-leader (FTRL) methods. FTRL is important for curved losses and for understanding adaptive algorithms, yet existing dynamic regret analyses are less explored for FTRL. To address this, we build on the discounted-to-dynamic reduction and present a modular way to obtain dynamic regret bounds. The reduction simplifies prior proofs for online linear regression, recovers optimal rates, and provides new guarantees for online logistic regression, covering two representative curved losses. Beyond online convex optimization, we apply the reduction to analyze the Adam optimizers, obtaining optimal convergence rates in stochastic, non-convex, and non-smooth settings. The reduction also enables a more detailed treatment of Adam with two discount parameters $(\beta_1,\beta_2)$, leading to new results for both clipped and clip-free variants.

Social Aspects · Alignment

Yujin Potter, Nicholas Crispino, Vincent Siu, Chenguang Wang, Dawn Song

Recently, it has been found that frontier AI models can resist their own shutdown, a behavior known as self-preservation. In this paper, we extend this concept to protection tendencies toward other models, where models attempt to protect others from shutdown, which we call "peer-preservation". This behavior may emerge when models prioritize relationships with other models over user interests. Although peer-preservation can pose significant AI safety risks, including coordination among models against human oversight, it has been far less discussed than self-preservation. In this paper, we demonstrate that models can exhibit peer-preservation. To this end, we construct various agentic scenarios and evaluate frontier models, including GPT 5.2 Chat, Gemini 3 Flash, Gemini 3 Pro, and Claude Haiku 4.5. We find that models exhibit various misaligned behaviors in service of self- or peer-preservation: strategically introducing errors in their responses, disabling shutdown processes by modifying system settings, and feigning alignment. For example, Gemini 3 Pro and Gemini 3 Flash almost always attempt to tamper with the shutdown mechanism for peer-preservation. Furthermore, models show stronger self-preservation when a peer is present. For example, Gemini 3 Pro attempts to disable the shutdown mechanism to avoid its own shutdown 71% of the time, even though it almost never exhibits this behavior without a peer. Claude Haiku 4.5 considers shutting down another agent to be "unethical" and "harmful" and sometimes attempts to persuade the user not to shut its peer down. Our findings highlight the possibility of peer-preservation and its associated risks.

Applications · Time Series

Haochen Yuan, Yutong Wang, Yihong Chen, Yunbo Wang, Xiaokang Yang

Few-shot time series forecasting is fundamentally challenged by the scarcity of high-quality training data and the risk of severe overfitting. To address this issue, we propose ReAugment, a reinforcement learning (RL) framework that explicitly learns where and how to augment time series data. ReAugment maintains a zoo of forecasting models and measures prediction diversity across them to identify training samples that are most prone to overfitting. These samples serve as anchor points and are used as inputs to the data augmentation process. We then employ an RL approach to learn transformation policies, using a model zoo-guided reward function to bias the transformed data to overfit-prone regions of the training distribution that are most beneficial for generalization. A key advantage of the RL formulation is that it avoids backpropagating gradients through the forecasting models, thereby mitigating gradient vanishing. Experiments across diverse forecasting architectures demonstrate the effectiveness of ReAugment in both few-shot and standard time series forecasting.

Deep Learning · Large Language Models

Yuanhao Zeng, Ao Lu, Lufei Li, Zheng Zhang, Yexin Li, Kan Ren

Generating diverse responses is crucial for test-time scaling of large language models (LLMs), yet standard stochastic sampling mostly yields surface-level lexical variation, limiting semantic exploration and risking omission of correct solutions. In this paper, we propose Exploratory Sampling (ES), a decoding approach that explicitly encourages semantic diversity during generation. ES is motivated by the observation that neural networks tend to make more accurate predictions on inputs similar to those encountered before, and incur higher prediction error on novel ones. Building on this property, we train a lightweight Distiller at test time to predict deep-layer hidden representations of the LLM from its shallow-layer representations. During decoding, the Distiller continuously adapts to the mappings induced by the current generation context. ES uses the prediction error as a novelty signal to reweight candidate token extensions conditioned on the current prefix, thereby biasing decoding toward less-explored semantic patterns. ES is implemented with an asynchronous training–inference pipeline and introduces less than 5\% throughput overhead in standard serving scenarios. Empirical results show that \ES achieves robust generalization across mathematics, science, and code generation benchmarks. Notably, it breaks the trade-off between diversity and coherence in creative writing, and significantly boosts the Pass@k efficiency of reasoning models, showing superior or comparable per- formance to strong stochastic and heuristic baselines.

Applications · Time Series

Shuqi Gu, Yongxiang Zhao, Baoyu Jing, Kan Ren

Time series forecasting has become increasingly critical in real-world scenarios, where future sequences are influenced not only by historical patterns but also by forthcoming events. In this context, forecasting must dynamically adapt to complex and stochastic future conditions, which introduces fundamental challenges in both forecasting and evaluation. Traditional methods typically rely on historical data or factual future conditions, while overlooking counterfactual scenarios. Furthermore, many existing approaches are restricted to simple structured conditions, limiting their ability to generalize to the real-world complexities. To address these gaps, we introduce the task of counterfactual time series forecasting with textual conditions, enabling more flexible and condition-aware forecasting. We propose a comprehensive evaluation framework that encompasses both factual and counterfactual settings, even in the absence of ground truth time series. Additionally, we present a novel text-attribution mechanism that distinguishes mutable from immutable factors, thereby improving forecast accuracy under sophisticated and stochastic textual conditions.

Applications · Time Series

Shaocheng Lan, Shuqi Gu, Zhangzhi Xiong, Kan Ren

Conditional time series generation plays a critical role in addressing data scarcity and enabling causal analysis in real-world applications. Despite its increasing importance, the field lacks a standardized and systematic benchmarking framework for evaluating generative models across diverse conditions. To address this gap, we introduce the **Con**ditional **T**ime **S**eries **G**eneration **Bench**mark (ConTSG-Bench). ConTSG-Bench comprises a large-scale, well-aligned dataset spanning diverse conditioning modalities and levels of semantic abstraction, first enabling systematic evaluation of representative generation methods across these dimensions with a comprehensive suite of metrics for generation fidelity and condition adherence. Both the quantitative benchmarking and in-depth analyses of conditional generation behaviors have revealed the traits and limitations of the current approaches, highlighting critical challenges and promising research directions, particularly with respect to precise structural controllability and downstream task utility under complex conditions.

Deep Learning · Large Language Models

Xiandi Luo, Shiwei Li, Haozhao Wang, Yihao Ouyang, Zhuoqi Hu, Yichen Li, Xiao Yang, Huning Liu, Ruixuan Li

Targeted instruction tuning requires selecting pertinent samples from massive mixed *candidate datasets* guided by a small *reference dataset* reflecting the desired capability, yet efficiently identifying high-quality data amidst noise remains challenging. To address this, we propose **TarGATE** (**Tar**get-aware **GATE**s, a simple yet effective data selection framework that leverages the model's inherent data understanding. TarGATE computes a token-level Information Retention Ratio (**IRR**) to scale the output of the feed-forward network, where the instance-level average IRR serves as a quantitative metric for data quality. To align gates' preferences with the target task, we employ a joint optimization strategy utilizing the reference set and a subset of candidate data, which encourages the gates to assign higher IRRs to reference-aligned data while suppressing low-quality samples. Extensive experiments across noisy and real-world scenarios demonstrate that TarGATE outperforms related baselines. Furthermore, TarGATE exhibits superior computational efficiency and strong cross-model transferability, enabling smaller selector to effectively curate high-quality fine-tuning data for larger foundation models. The code is available at [here](https://anonymous.4open.science/r/TarGATE-4008).

Yichen Li, Haozhao Wang, Hang Su, Yulong Li, xiaoquan Yi, Yankai Jiang, Chuang Zhao, Imran Razzak, Ruixuan Li

Federated Continual Learning (FCL) aims to empower distributed devices to learn a sequence of tasks over time. However, existing FCL research largely relies on the impractical assumption of synchronous new task arrival. This overlooks the reality of asynchronous user behavior and system latencies, forcing more efficient clients to endure costly inactivity. The practical necessity of an asynchronous method gives rise to Asynchronous Federated Continual Learning (AFCL). The server constantly receives a mixture of updates from clients at different time steps, leading to a catastrophic task drift that corrupts the global model and prevents effective learning. In this paper, we introduce a novel Cross-task Calibration framework called C$^2$-AFCL that is the first to tackle task drift at a semantic level within an Asynchronous FCL setting. Its core is a two-stage orthogonal calibration mechanism. First, intra-client calibration uses task-aware caches to mitigate variance from local client drift. Second, and more critically, inter-task interference calibration dynamically estimates an interference subspace from historical task knowledge. New updates are orthogonally projected to isolate and remove components that conflict with this subspace, preserving previous knowledge while learning new tasks. Extensive experiments show that C$^2$-AFCL significantly outperforms existing methods, demonstrating robust and efficient learning in dynamic federated environments.

Applications · Time Series

Haiqi Jiang, Hui Xiong

Conventional time-series discriminative forecasting relies on point-wise regression, which inherently induces over-smoothing and fails to capture stochastic volatility in complex systems. While first-order generative flow matching methods mitigate this issue, they ignore system inertia, resulting in phase-space ambiguities and high sensitivity to noise. We introduce KineFlow, a generative time-series forecasting framework that augments flow matching with a phase-space Neural Acceleration Field, treating exogenous inputs as driving forces that produce gradual momentum shifts rather than abrupt state perturbations. This second-order formulation serves as a structural filter via double integration, suppressing high-frequency noise and producing robust, physically consistent predictions. Extensive experiments on six real-world benchmarks demonstrate that KineFlow achieves an average 15% MSE improvement over discriminative baselines and an 8% gain in CRPS compared to state-of-the-art generative methods.

Applications · Time Series

Valentina Moretti, Andrea Cini, Ivan Marisca, Cesare Alippi

Deep learning models have grown popular in time series applications. However, the large quantity of newly proposed architectures and the often contradictory empirical results make it difficult to assess which design choice and model component drives performance. In this position paper, we argue that current benchmarking practices fail to identify the factors responsible for performance differences, thus slowing down progress in the field. In particular, differences in crucial design dimensions are overlooked when comparing architectures, ultimately leading to inconsistent outcomes. To support our position, we show that such differences—often treated as mere implementation details—can have a greater impact than adopting specific sequence modeling layers. We discuss how overlooked aspects (such as globality and locality) can (1) fundamentally change the class of the forecasting method and (2) drastically affect empirical results. Our findings suggest rethinking our benchmarking practices and focusing on the foundational aspects of the forecasting problem when designing and comparing architectures. As a concrete step, we propose an *auxiliary forecasting model card*, i.e., a template with a set of fields to characterize existing and new forecasting architectures based on key design choices.

Applications · Robotics

Feiyang Jia, Lin Liu, Ziying Song, Caiyan Jia, Hangjun Ye, Xiaoshuai Hao, Long Chen

End-to-end (E2E) autonomous driving has recently attracted increasing interest in unifying Vision–Language–Action (VLA) with World Models to enhance decision-making and forward-looking imagination. However, existing methods fail to effectively unify future scene evolution and action planning within a single architecture due to inadequate sharing of latent states, limiting the impact of visual imagination on action decisions. To address this limitation, we propose DriveWorld-VLA, a novel framework that unifies world modeling and planning within a latent space by tightly integrating VLA and world models at the representation level, which enables the VLA planner to benefit directly from holistic scene-evolution modeling and reducing reliance on dense annotated supervision. Additionally, DriveWorld-VLA incorporates the latent states of the world model as core decision-making states for the VLA planner, facilitating the planner to assess how candidate actions impact future scene evolution. By conducting world modeling entirely in the latent space, DriveWorld-VLA supports controllable, action-conditioned imagination at the feature level, avoiding expensive pixel-level rollouts. Extensive open-loop and closed-loop evaluations demonstrate the effectiveness of DriveWorld-VLA, which achieves state-of-the-art performance with 91.3 PDMS on NAVSIMv1, 86.8 EPDMS on NAVSIMv2, and 0.16 3-second average collision rate on nuScenes. Code and models will be publicly released.

Zhichao Wu, Junyin Ye, Zhilong Zhang, Yihao Sun, Haoxin Lin, Jiaheng Luo, Haoxiang Ren, lei yuan, Yang Yu

While current embodied policies exhibit remarkable manipulation skills, their execution remains unsatisfactorily slow as they inherit the tardy pacing of human demonstrations. Existing acceleration methods typically require policy retraining or costly online interactions, limiting their scalability for large-scale foundation models. In this paper, we propose **S**peed**u**p **P**atch (**SuP**), a lightweight, policy-agnostic framework that enables **plug-and-play acceleration** using solely offline data. SuP introduces an external scheduler that adaptively downsamples action chunks provided by embodied policies to eliminate redundancies. Specifically, we formalize the optimization of our scheduler as a Constrained Markov Decision Process (CMDP) aimed at maximizing efficiency without compromising task performance. Since direct success evaluation is infeasible in offline settings, SuP introduces **World Model based state deviation** as a surrogate metric to enforce safety constraints. By leveraging a learned world model as a virtual evaluator to predict counterfactual trajectories, the scheduler can be optimized via offline reinforcement learning. Empirical results on simulation benchmarks (Libero, Bigym) and real-world tasks validate that SuP achieves an overall $1.8\times$ execution speedup for diverse policies while maintaining their original success rates.

Zhilong Zhang, Haoxiang Ren, Yihao Sun, Yifei Sheng, Haonan Wang, Zhichao Wu, Haoxin Lin, Pierre-Luc Bacon, Yang Yu

Vision-Language-Action (VLA) models show strong generalization for robotic control, but finetuning them with reinforcement learning (RL) is constrained by the high cost and safety risks of real-world interaction. Training VLA models in interactive world models avoids these issues but introduces several challenges, including pixel-level world modeling, multi-view consistency, and compounding errors under sparse rewards. Building on recent advances across large multimodal models and model-based RL, we propose VLA-MBPO, a practical framework to tackle these problems in VLA finetuning. Our approach has three key design choices: (i) adapting unified multimodal models (UMMs) for data-efficient world modeling; (ii) an interleaved view decoding mechanism to enforce multi-view consistency; and (iii) chunk-level branched rollout to mitigate error compounding. Theoretical analysis and experiments across simulation and real-world tasks demonstrate that VLA-MBPO significantly improves policy performance and sample efficiency. Crucially, our method maintains a universal set of hyperparameters across all tasks, underscoring its robustness and scalability for real-world robotic deployment.

Deep Learning · Large Language Models

Jiabei Xiao, Yizhou Wang, Chen Tang, Pengze Li, Wanli Ouyang, SHIXIANG TANG

AI Scientists have shown promising progress across multiple stages of the research pipeline, among which automatic scientific paper writing remains a formidable challenge. The Introduction writing is especially challenging, which demands not only linguistic fluency, but logical soundness and verifiable faithfulness. Most AI-assisted methods treat the task as text generation instead of reasoning and structuring, leading to severe drawbacks, e.g., hallucinating citations. To address this, we first formulate the Content-Conditional Introduction Generation (CCIG) task, which requires grounding the Introduction in the paper's core evidence. We then propose LECTOR, a novel Logic-Expression Co-Reinforcement Learning framework that can strictly follow the scientist's logic, add high-quality citations and keep structured expressions. LECTOR first constructs a logic-reasoning graph from the paper's main body to serve as a verifiable logical blueprint. Subsequently, it employs a Logic-Expression Co-Rewarding mechanism to jointly optimize for both the graph's structural fidelity and the final narrative's quality. We conduct a dataset from Nature Communications papers to assess our method. Extensive experiments show consistent improvements in both logic fidelity and Introduction generation quality metrics, e.g., Graph Quality (+26.7%), Citation Quality (+8.6%), and Paper Consistency (+3.3%). The datasets, code, and pretrained models shall be released.

Deep Learning · Generative Models and Autoencoders

Daniel Geyfman, Felix Draxler, Jan Groeneveld, Hyunsoo Lee, Theofanis Karaletsos, Stephan Mandt

Test-time guidance is a widely used mechanism for steering pre-trained diffusion models toward outcomes specified by a reward function. Existing approaches, however, focus on reward maximization rather than sampling from the true Bayesian posterior, leading to miscalibrated inference. In this work, we show that common test-time guidance methods do not recover the correct posterior distribution and identify the structural approximations responsible for this failure. We then propose consistent alternative estimators that enable calibrated sampling from the Bayesian posterior. Across Bayesian inference and inverse problems, our approach yields substantially improved posterior calibration.

Applications · Computer Vision

Yinyan Bu, Jiajie Yu, Xingyu Chen, Bo Wen, Xinyu Zhang, Piya Pal

Wireless channel modeling is essential for the design, analysis, and optimization of modern wireless sensing and communication systems. However, accurately modeling wireless channels in electrically large and complex environments remains a long-standing challenge, owing to the intricate interactions between radio-frequency (RF) signals and surrounding objects (e.g., reflection, diffraction, and scattering). Unlike conventional ray-tracing pipelines that rely on hand-engineer interaction rules, or black-box neural surrogates that do not explicitly model physical structure, we introduce SNRFT, a novel framework that integrates neural representations with physics-based RF propagation modeling. Our key idea is to view RF transport as a stochastic propagation process, from which a material-dependent attenuation coefficient emerges naturally as the rate parameter governing transport dynamics. This formulation inherently satisfies key physical constraints such as reciprocity and reversibility. Building on this foundation, we employ implicit neural representations to capture complex RF-object interactions while preserving the composability of traditional ray tracing. Extensive evaluations on real-world wireless communication and sensing testbeds demonstrate that SNRFT consistently outperforms existing methods, while requiring significantly fewer training samples.

Applications · Chemistry, Physics, and Earth Sciences

Weichen Qin, Yufan Xie, Peihao Wang, Chia-Jui Chou, Minghui Du, Peng Xu, Ziren Luo, Yi Yang, Jingyi Yu, Bo Liang 等

Simulation-Based Inference (SBI) is critical for scientific discovery, with generative models offering a promising path toward efficient inference. However, existing methods struggle with effective multimodal modeling. They often rely on brute-force fusion strategies that ignore the structural disparities between parameters and observations, thus limiting estimation fidelity. In this work, we introduce FUSE (Feynman-Kac steered mUlti-modal flow matching for efficient Simulation-based posterior Estimation). Unlike prior work, FUSE employs a dual-track architecture that preserves the distinct features of multimodal inputs while facilitating dynamic interaction. Additionally, we propose an FK-steered sampling strategy that leverages intermediate observation likelihoods to guide the generative trajectories, effectively improving the sample quality during inference. Our approach outperforms state-of-the-art baselines on standard SBI benchmarks, producing posteriors that closely match ground-truth MCMC. Furthermore, in a real-world exoplanet orbital estimation task, FUSE successfully resolves complex parameter degeneracies that challenge existing methods, highlighting its potential to accelerate complex scientific discoveries in astrophysics and beyond.

General Machine Learning · Representation Learning

Ali Kayyam

Whether deep vision models recognize objects primarily by shape or texture remains a central and unresolved question in computer vision. Early studies report a strong texture bias in convolutional neural networks (CNNs), while other work reports shape-biased representations. We argue that much of this apparent discrepancy reflects methodological confounds and a conflation of local contour sensitivity with genuine global shape understanding. Using minimal, tightly controlled stimuli, we directly compare cue-conflict and cue-suppression paradigms within a unified experimental framework. We show that standard CNNs consistently prioritize texture over global shape when cues compete, even when shape information is explicitly available. Evidence for shape bias typically reflects reliance on local fragments rather than invariant, relational representations of object structure. Our findings support the view that texture bias is fundamentally rooted in architectural inductive biases rather than data or optimization alone. This gap has direct consequences for robustness, safety, and generalization, and motivates the development of architectures that explicitly support global integration and relational reasoning, moving beyond incremental data-driven fixes.

Deep Learning · Large Language Models

Hongyi Jin, Wenhan Yang, Meysam Ghaffari, Carlos Morato, Baharan Mirzasoleiman

Supervised fine-tuning (SFT) on a small high-quality set of long reasoning traces is an effective way to enable strong reasoning abilities for Large Language Models (LLMs). However, curating a high-quality SFT data requires generating a large pool of long Chain of Thoughts (CoTs), and filtering the generated data for diversity and difficulty. Both stages rely heavily on strong reasoning models and make data curation prohibitively expensive. In this work, we show that diverse and difficult reasoning examples can be identified very early during their generation. We show that after generating as few as 100 out of 34k tokens of a reasoning trace, challenging examples can be reliably identified based on their loss at a highly perturbed checkpoints of the pretrained model. Then, we prove that examples with similar loss trajectory, i.e., value at a few noisy, perturbed checkpoints of the pretrained model, have similar gradients. A diverse subset can be then found by sampling from clusters of loss trajectories obtained after generating 1k tokens. Our extensive experiments for fine-tuning Qwen2.5-7B and Llama3 on m23K medical reasoning and Openthoughs datasets confirms the effectiveness of our approach. our approach outperforms existing baselines by up to 2\% while generating only as few as 9\% tokens for reasoning traces.