论文检索

Deep Learning · Everything Else

Seonghwan Park, Jueun Mun, Donghyun Oh, Namhoon Lee

Concept bottleneck models (CBMs) ensure interpretability by decomposing predictions into human interpretable concepts. Yet the annotations used for training CBMs that enable this transparency are often noisy, and the impact of such corruption is not well understood. In this study, we present the first systematic study of noise in CBMs and show that even moderate corruption simultaneously impairs prediction performance, interpretability, and the intervention effectiveness. Our analysis identifies a susceptible subset of concepts whose accuracy declines far more than the average gap between noisy and clean supervision and whose corruption accounts for most performance loss. To mitigate this vulnerability we propose a two-stage framework. During training, sharpness-aware minimization stabilizes the learning of noise-sensitive concepts. During inference, where clean labels are unavailable, we rank concepts by predictive entropy and correct only the most uncertain ones, using uncertainty as a proxy for susceptibility. Theoretical analysis and extensive ablations elucidate why sharpness-aware training confers robustness and why uncertainty reliably identifies susceptible concepts, providing a principled basis that preserves both interpretability and resilience in the presence of noise.

Reinforcement Learning · Batch Offline

Matthew T Jackson, Uljad Berdica, Jarek Liesen, Shimon Whiteson, Jakob Foerster

Progress in offline reinforcement learning (RL) has been impeded by ambiguous problem definitions and entangled algorithmic designs, resulting in inconsistent implementations, insufficient ablations, and unfair evaluations. Although offline RL explicitly avoids environment interaction, prior methods frequently employ extensive, undocumented online evaluation for hyperparameter tuning, complicating method comparisons. Moreover, existing reference implementations differ significantly in boilerplate code, obscuring their core algorithmic contributions. We address these challenges by first introducing a rigorous taxonomy and a transparent evaluation protocol that explicitly quantifies online tuning budgets. To resolve opaque algorithmic design, we provide clean, minimalistic, single-file implementations of various model-free and model-based offline RL methods, significantly enhancing clarity and achieving substantial speed-ups. Leveraging these streamlined implementations, we propose Unifloral, a unified algorithm that encapsulates diverse prior approaches and enables development within a single, comprehensive hyperparameter space. Using Unifloral with our rigorous evaluation protocol, we develop two novel algorithms - TD3-AWR (model-free) and MoBRAC (model-based) - which substantially outperform established baselines. Our implementation is publicly available at https://github.com/EmptyJackson/unifloral.

Reinforcement Learning · Batch Offline

Da Wang, Yi Ma, Ting Guo, Hongyao Tang, Wei Wei, Jiye Liang

Offline reinforcement learning (RL) aims to learn optimal policies from static datasets while enhancing generalization to out-of-distribution (OOD) data. To mitigate overfitting to suboptimal behaviors in offline datasets, existing methods often relax constraints on policy and data or extract informative patterns through data-driven techniques. However, there has been limited exploration into structurally guiding the optimization process toward flatter regions of the solution space that offer better generalization. Motivated by this observation, we present \textit{FANS}, a generalization-oriented structured network framework that promotes flatter and robust policy learning by guiding the optimization trajectory through modular architectural design. FANS comprises four key components: (1) Residual Blocks, which facilitate compact and expressive representations; (2) Gaussian Activation, which promotes smoother gradients; (3) Layer Normalization, which mitigates overfitting; and (4) Ensemble Modeling, which reduces estimation variance. By integrating FANS into a standard actor-critic framework, we highlight that this remarkably simple architecture achieves superior performance across various tasks compared to many existing advanced methods. Moreover, we validate the effectiveness of FANS in mitigating overestimation and promoting generalization, demonstrating the promising potential of architectural design in advancing offline RL.

Wenhao Li, Yuxin Zhang, Gen Luo, Haiyuan Wan, Ziyang Gong, Fei Chao, Rongrong Ji

Reducing the key-value (KV) cache burden in Large Language Models (LLMs) significantly accelerates inference. Dynamically selecting critical KV caches during decoding helps maintain performance. Existing methods use random linear hashing to identify important tokens, but this approach is inefficient due to the orthogonal distribution of queries and keys within two narrow cones in LLMs. We introduce Spotlight Attention, a novel method that employs non-linear hashing functions to optimize the embedding distribution of queries and keys, enhancing coding efficiency and robustness. We also developed a lightweight, stable training framework using a Bradley-Terry ranking-based loss, enabling optimization of the non-linear hashing module on GPUs with 16GB memory in 8 hours. Experimental results show that Spotlight Attention drastically improves retrieval precision while shortening the length of the hash code at least 5$\times$ compared to traditional linear hashing. Finally, we exploit the computational advantages of bitwise operations by implementing specialized CUDA kernels, achieving hashing retrieval for 512K tokens in under 100$\mu$s on a single A100 GPU, with end-to-end throughput up to 3$\times$ higher than vanilla decoding.

Deep Learning · Generative Models and Autoencoders

Zander W. Blasingame, Chen Liu

Training-free guided generation is a widely used and powerful technique that allows the end user to exert further control over the generative process of flow/diffusion models. Generally speaking, two families of techniques have emerged for solving this problem for *gradient-based guidance*: namely, *posterior guidance* (*i.e.*, guidance via projecting the current sample to the target distribution via the target prediction model) and *end-to-end guidance* (*i.e.*, guidance by performing backpropagation throughout the entire ODE solve). In this work, we show that these two seemingly separate families can actually be *unified* by looking at posterior guidance as a *greedy strategy* of *end-to-end guidance*. We explore the theoretical connections between these two families and provide an in-depth theoretical of these two techniques relative to the *continuous ideal gradients*. Motivated by this analysis we then show a method for *interpolating* between these two families enabling a trade-off between compute and accuracy of the guidance gradients. We then validate this work on several inverse image problems and property-guided molecular generation.

Theory · Reinforcement Learning and Planning

James Chapman, Kedar Karhadkar, Guido Montufar

Deep reinforcement learning (DRL) has achieved remarkable success across multiple domains, including competitive games, natural language processing, and robotics. Despite these advancements, policies trained via DRL often struggle to generalize to evaluation environments with different parameters. This challenge is typically addressed by training with multiple contexts and/or by leveraging additional structure in the problem. However, obtaining sufficient training data across diverse contexts can be impractical in real-world applications. In this work, we consider contextual Markov decision processes (CMDPs) with transition and reward functions that exhibit regularity in context parameters. We introduce the context-enhanced Bellman equation (CEBE) to improve generalization when training on a single context. We prove both analytically and empirically that the CEBE yields a first-order approximation to the Q function trained across multiple contexts. We then derive context sample enhancement (CSE) as an efficient data augmentation method for approximating the CEBE in deterministic control environments. We numerically validate the performance of CSE in simulation environments, showcasing its potential to improve generalization in DRL.

Optimization

Weiming Liu, Xinting Liao, Jun Dan, Fan Wang, Hua Yu, Junhao Dong, Shunjie Dong, Lianyong Qi, Yew Soon Ong

Semi-Unbalanced Optimal Transport (SemiUOT) shows great promise in matching two probability measures by relaxing one of the marginal constraints. Previous solvers often incorporate an entropy regularization term, which can result in inaccurate matching solutions. To address this issue, we focus on determining the marginal probability distribution of SemiUOT with KL divergence using the proposed Equivalent Transformation Mechanism (ETM) approach. Furthermore, we extend the ETM-based method into exploiting the marginal probability distribution of Unbalanced Optimal Transport (UOT) with KL divergence for validating its generalization. Once the marginal probabilities of UOT/SemiUOT are determined, they can be transformed into a classical Optimal Transport (OT) problem. Moreover, we propose a KKT-Multiplier regularization term combined with Multiplier Regularized Optimal Transport (MROT) to achieve more accurate matching results. We conduct several numerical experiments to demonstrate the effectiveness of our proposed methods in addressing UOT/SemiUOT problems.

Applications · Neuroscience, Cognitive Science

Xiuning Zhang, Xincheng Lu, Nihong Chen, Yuanyuan Mi

Serial dependence reflects how recent sensory history shapes current perception, producing two opposing biases: repulsion, where perception is repelled from recent stimuli, and attraction, where perception is drawn toward them. Repulsion typically occurs at the sensory perception stage, while attraction arises at the post-perception stage. To uncover the neural basis of these effects, we developed a two-layer continuous attractor neural network model incorporating synaptic short-term plasticity (STP). The lower layer, dominated by synaptic depression, models sensory processing and drives repulsion due to sustained neurotransmitter depletion. The higher layer, dominated by synaptic facilitation, models post-perception processing and drives attraction by sustained high neurotransmitter release probability. Our model successfully explains the serial dependence phenomena observed in the visual orientation judgment experiments, highlighting STP as the critical mechanism, with its time constants defining the temporal windows of repulsion and attraction. Furthermore, the model provides a neural foundation for the Bayesian interpretation of serial dependence. This study advances our understanding of how the neural system leverages STP to balance sensitivity in sensory perception with stability in post-perceptual cognition.

Theory · Everything Else

Bochen Lyu, Xiaojing Zhang, Fangyi Zheng, He Wang, Zheng Wang, Zhanxing Zhu

This paper establishes a continuous time approximation, a piece-wise continuous differential equation, for the discrete Heavy-Ball (HB) momentum method with explicit discretization error. Investigating continuous differential equations has been a promising approach for studying the discrete optimization methods. Despite the crucial role of momentum in gradient-based optimization methods, the gap between the original dynamics and the continuous time approximations due to the discretization error has not been comprehensively bridged yet. In this work, we study the HB momentum method in continuous time while putting more focus on the discretization error to provide additional theoretical tools to this area. In particular, we design a first-order piece-wise continuous differential equation, where we add a number of counter terms to account for the discretization error explicitly. As a result, we provide a continuous time model for the HB momentum method that allows the control of discretization error to arbitrary order of the learning rate. As an application, we leverage it to find a new implicit regularization of the directional smoothness and investigate the implicit bias of HB for diagonal linear networks, indicating how our results can be used in deep learning. Our theoretical findings are further supported by numerical experiments.

Optimization · Optimization and Learning under Uncertainty

Zefan Cai, Wen Xiao, Hanshi Sun, cheng Luo, Yikai Zhang, Ke Wan, Yucheng Li, Yeyang Zhou, Li-Wen Chang, Jiuxiang Gu 等

Reasoning models have demonstrated impressive performance in self-reflection and chain-of-thought reasoning. However, they often produce excessively long outputs, leading to prohibitively large key-value (KV) caches during inference. While chain-of-thought inference significantly improves performance on complex reasoning tasks, it can also lead to reasoning failures when deployed with existing KV cache compression approaches. To address this, we propose Redundancy-aware KV Cache Compression for Reasoning models (R-KV), a novel method specifically targeting redundant tokens in reasoning models. Our method preserves nearly 100% of the full KV cache performance using only 10% of the KV cache, substantially outperforming existing KV cache baselines, which reach only 60% of the performance. Remarkably, R-KV even achieves 105% of full KV cache performance with 38% of the KV cache. This KV-cache reduction also leads to a 50% memory saving and a 2x speedup over standard chain-of-thought reasoning inference. Experimental results show that R-KV consistently outperforms existing KV cache compression baselines across two mathematical reasoning datasets.

Theory · Learning Theory

Emmeran Johnson, David Martinez-Rubio, Ciara Pike-Burke, Patrick Rebeschini

We study online convex optimization on $\ell_p$-balls in $\mathbb{R}^d$ for $p > 2$. While always sub-linear, the optimal regret exhibits a shift between the high-dimensional setting ($d > T$), when the dimension $d$ is greater than the time horizon $T$ and the low-dimensional setting ($d \leq T$). We show that Follow-the-Regularised-Leader (FTRL) with time-varying regularisation which is adaptive to the dimension regime is anytime optimal for all dimension regimes. Motivated by this, we ask whether it is possible to obtain anytime optimality of FTRL with fixed non-adaptive regularisation. Our main result establishes that for separable regularisers, adaptivity in the regulariser is necessary, and that any fixed regulariser will be sub-optimal in one of the two dimension regimes. Finally, we provide lower bounds which rule out sub-linear regret bounds for the linear bandit problem in sufficiently high-dimension for all $\ell_p$-balls with $p \geq 1$.

Deep Learning · Algorithms

Chenheng Zhang, Tianqi Du, Jizhe Zhang, Mingqing Xiao, Yifei Wang, Yisen Wang, Zhouchen Lin

Conventional research on large language models (LLMs) has primarily focused on refining output distributions, while paying less attention to the decoding process that transforms these distributions into final responses. Recent advances, such as scaling the computation of inference time with reward models, have underscored the importance of decoding, but these methods often suffer from high computational costs and limited applicability. In this paper, we revisit LLM generation through the lens of recommender systems, conceptualizing the decoding process as analogous to the ranking stage in recommendation pipelines. From this perspective, we observe that both traditional decoding methods and reward models exhibit clear limitations such as redundancy. Motivated by this insight, we propose Language Ranker, a novel framework that introduces a lightweight module to rerank candidate responses using features extracted by the base model. Experiments across a wide range of tasks show that Language Ranker achieves performance comparable to large-scale reward models, while requiring only <0.5M additional parameters, significantly reducing the computational overhead during both training and inference stages. This highlights the efficiency and effectiveness of our method, showcasing its potential to fully unlock the capabilities of LLMs.

Computer Vision · Segmentation

Xiaolong Wang, Lixiang Ru, Ziyuan Huang, Kaixiang Ji, DanDan Zheng, Jingdong Chen, Jun Zhou

We propose a novel AutoRegressive Generation-based paradigm for image Segmentation (ARGenSeg), achieving multimodal understanding and pixel-level perception within a unified framework. Prior works integrating image segmentation into multimodal large language models (MLLMs) typically employ either boundary points representation or dedicated segmentation heads. These methods rely on discrete representations or semantic prompts fed into task-specific decoders, which limits the ability of the MLLM to capture fine-grained visual details. To address these challenges, we introduce a segmentation framework for MLLM based on image generation, which naturally produces dense masks for target objects. We leverage MLLM to output visual tokens and detokenize them into images using an universal VQ-VAE, making the segmentation fully dependent on the pixel-level understanding of the MLLM. To reduce inference latency, we employ a next-scale-prediction strategy to generate required visual tokens in parallel. Extensive experiments demonstrate that our method surpasses prior state-of-the-art approaches on multiple segmentation datasets with a remarkable boost in inference speed, while maintaining strong understanding capabilities.

Optimization · Neural Network Optimization

Binrui Shen, LiangYuan, Shengxin Zhu

Graph matching, usually cast as a discrete Quadratic Assignment Problem (QAP), aims to identify correspondences between nodes in two graphs. Since QAP is NP-hard, many methods its discrete constraints by projecting the discrete feasible set onto its convex hull and solving the resulting continuous problem. However, these relaxations inevitably enlarge the feasible set and introduce two errors: sensitivity to numerical scales and geometric misalignment between the relaxed and original feasible domains. To address these issues, we propose a novel relaxation framework to reformulate the projection step as a Frobenius-Regularized Linear Assignment (FRA) problem. This formulation incorporates a tunable regularization term to curb the inflation of the feasible region and ensure numerical scale invariance. To solve the FRA efficiently, we introduce a scaling algorithm for doubly stochastic normalization. Leveraging its favorable computational properties, we design a theoretically grounded, accelerated mixed-precision algorithm. Building on these components, we propose Frobenius-Regularized Assignment Matching (FRAM), which approximates the QAP solution through a sequence of FRA problems. Extensive CPU experiments show that FRAM consistently outperforms all baselines. On GPUs, with mixed precision, FRAM achieves up to a 370× speedup over its FP64 CPU implementation without sacrificing accuracy.

Computer Vision · Image and Video Generation

Sepehr Dehdashtian, Mashrur Mahmud Morshed, Jacob Seidman, Gaurav Bharaj, Vishnu Boddeti

Synthetic image detectors (SIDs) are a key defense against the risks posed by the growing realism of images from text-to-image (T2I) models. Red teaming improves SID’s effectiveness by identifying and exploiting their failure modes via misclassified synthetic images. However, existing red-teaming solutions (i) require white-box access to SIDs, which is infeasible for proprietary state-of-the-art detectors, and (ii) generate image-specific attacks through expensive online optimization. To address these limitations, we propose PolyJuice, the first black-box, image-agnostic red-teaming method for SIDs, based on an observed distribution shift in the T2I latent space between samples correctly and incorrectly classified by the SID. PolyJuice generates attacks by (i) identifying the direction of this shift through a lightweight offline process that only requires black-box access to the SID, and (ii) exploiting this direction by universally steering all generated images towards the SID’s failure modes. PolyJuice-steered T2I models are significantly more effective at deceiving SIDs (up to 84%) compared to their unsteered counterparts. We also show that the steering directions can be estimated efficiently at lower resolutions and transferred to higher resolutions using simple interpolation, reducing computational overhead. Finally, tuning SID models on PolyJuice-augmented datasets notably enhances the performance of the detectors (up to 30%).

Optimization

yuepeng zheng, Fu Luo, Zhenkun Wang, Yaoxin Wu, Yu Zhou

Multi-Task Learning (MTL) in Neural Combinatorial Optimization (NCO) is a promising approach for training a unified model capable of solving multiple Vehicle Routing Problem (VRP) variants. However, existing Reinforcement Learning (RL)-based multi-task methods can only train light decoder models on small-scale problems, exhibiting limited generalization ability when solving large-scale problems. To overcome this limitation, this work introduces a novel multi-task learning method driven by knowledge distillation (MTL-KD), which enables efficient training of heavy decoder models with strong generalization ability. The proposed MTL-KD method transfers policy knowledge from multiple distinct RL-based single-task models to a single heavy decoder model, facilitating label-free training and effectively improving the model's generalization ability across diverse tasks. In addition, we introduce a flexible inference strategy termed Random Reordering Re-Construction (R3C), which is specifically adapted for diverse VRP tasks and further boosts the performance of the multi-task model. Experimental results on 6 seen and 10 unseen VRP variants with up to 1,000 nodes indicate that our proposed method consistently achieves superior performance on both uniform and real-world benchmarks, demonstrating robust generalization abilities. The code is available at [https://github.com/CIAM-Group/MTLKD](https://github.com/CIAM-Group/MTLKD).

Applications · Robotics

Ryan Punamiya, Dhruv Patel, Patcharapong Aphiwetsa, Pranav Kuppili, Lawrence Zhu, Simar Kareer, Judy Hoffman, Danfei Xu

Egocentric human experience data presents a vast resource for scaling up end-to-end imitation learning for robotic manipulation. However, significant domain gaps in visual appearance, sensor modalities, and kinematics between human and robot impede knowledge transfer. This paper presents EgoBridge, a unified co-training framework that explicitly aligns the policy latent spaces between human and robot data using domain adaptation. Through a measure of discrepancy on the joint policy latent features and actions based on Optimal Transport (OT), we learn observation representations that not only align between the human and robot domain but also preserve the action-relevant information critical for policy learning. EgoBridge achieves a significant absolute policy success rate improvement by 44% over human-augmented cross-embodiment baselines in three real-world single-arm and bimanual manipulation tasks. EgoBridge also generalizes to new objects, scenes, and tasks seen only in human data, where baselines fail entirely. Videos and additional information can be found at https://ego-bridge.github.io/

Optimization · Optimization and Learning under Uncertainty

Yunuo Zhang, Baiting Luo, Ayan Mukhopadhyay, Gabor Karsai, Abhishek Dubey

In Partially Observable Markov Decision Processes (POMDPs), maintaining and updating belief distributions over possible underlying states provides a principled way to summarize action-observation history for effective decision-making under uncertainty. As environments grow more realistic, belief distributions develop complexity that standard mathematical models cannot accurately capture, creating a fundamental challenge in maintaining representational accuracy. Despite advances in deep learning and probabilistic modeling, existing POMDP belief approximation methods fail to accurately represent complex uncertainty structures such as high-dimensional, multi-modal belief distributions, resulting in estimation errors that lead to suboptimal agent behaviors. To address this challenge, we present ESCORT (Efficient Stein-variational and sliced Consistency-Optimized Representation for Temporal beliefs), a particle-based framework for capturing complex, multi-modal distributions in high-dimensional belief spaces. ESCORT extends SVGD with two key innovations: correlation-aware projections that model dependencies between state dimensions, and temporal consistency constraints that stabilize updates while preserving correlation structures. This approach retains SVGD's attractive-repulsive particle dynamics while enabling accurate modeling of intricate correlation patterns. Unlike particle filters prone to degeneracy or parametric methods with fixed representational capacity, ESCORT dynamically adapts to belief landscape complexity without resampling or restrictive distributional assumptions. We demonstrate ESCORT's effectiveness through extensive evaluations on both POMDP domains and synthetic multi-modal distributions of varying dimensionality, where it consistently outperforms state-of-the-art methods in terms of belief approximation accuracy and downstream decision quality.

Applications · Everything Else

Dail Kim, Joon-Hyuk Chang

Target sound extraction aims to isolate target sound sources from an input mixture using a target clue to identify the sounds of interest. To address the challenge posed by the wide variety of sounds, recent work has introduced privileged knowledge distillation (PKD), which utilizes privileged information (PI) about the target sound, available only during training. While PKD has shown promise, existing approaches often suffer from overfitting of the teacher model for the overly rich PI and ineffective knowledge transfer to the student model. In this paper, we propose Disentangled Codec Knowledge Distillation (DCKD) to mitigate these issues by regulating the amount and the flow of target sound information within the teacher model. We begin by extracting a compressed representation of the target sound using a neural audio codec to regulate the amount of PI. Disentangled representation learning is then applied to remove class information and extract fine-grained temporal information as PI. Subsequently, an n-hot vector as the class information and the class-independent PI are used to condition the early and later layers of the teacher model, respectively, forming a regulated coarse-to-fine target information flow. The resulting representation is transferred to the student model through feature-level knowledge distillation. Experimental results show that DCKD consistently improves existing methods across model architectures under the multi-target selection condition.

Social Aspects · Accountability, Transparency and Interpretability

Samuele Bortolotti, Emanuele Marconato, Paolo Morettin, Andrea Passerini, Stefano Teso

Concept-based Models are neural networks that learn a concept extractor to map inputs to high-level concepts and an inference layer to translate these into predictions. Ensuring these modules produce interpretable concepts and behave reliably in out-of-distribution is crucial, yet the conditions for achieving this remain unclear. We study this problem by establishing a novel connection between Concept-based Models and reasoning shortcuts (RSs), a common issue where models achieve high accuracy by learning low-quality concepts, even when the inference layer is fixed and provided upfront. Specifically, we extend RSs to the more complex setting of Concept-based Models and derive theoretical conditions for identifying both the concepts and the inference layer. Our empirical results highlight the impact of RSs and show that existing methods, even combined with multiple natural mitigation strategies, often fail to meet these conditions in practice.