论文检索

Applications · Robotics

Dong Wang, Zilong Chen, Jirong Liu, Ziqing Qiao, Xin Xiao, Bingyi Kang, Hongtao Wu, Xiao Ma, Tao Kong, Huaping Liu

Integrating Vision-Language Models (VLMs) into robotics has facilitated the development of generalizable Vision-Language Action (VLA) policies. However, unified discrete frameworks lag behind decoupled continuous designs due to limitations in action chunking and temporal modeling. To address this, we introduce **RoboOmni**, a unified multi-modal next-token prediction framework. Challenging the assumption that continuous modeling is essential for high-performance manipulation, **RoboOmni** demonstrates that *actions are just another modality* capable of being effectively modeled discretely. At the core of our method is Multi-Token Action Prediction (MTAP), which integrates action chunking directly into the discrete tokenizer. This design resolves temporal modeling bottlenecks and significantly reduces distribution shift between training and inference. By preserving the native VLM training and inference pipeline, **RoboOmni** naturally benefits from large-scale multimodal co-training and modern decoding optimizations. Extensive evaluations on the CALVIN, SimplerEnv, and real-world platforms confirm that **RoboOmni** establishes new state-of-the-art performance, significantly outperforming diffusion-based baselines such as $\pi_0$. Notably, combining our proposed MTAP with the FAST tokenizer achieves a 94.4\% average success rate on CALVIN, while the Bin tokenizer implementation attains a 27$\times$ inference speedup compared to OpenVLA.

Zhiqi Zhang, Zhiyu Zeng, Ruohan Zhan, Dennis Zhang

While Randomized controlled trials (RCTs), or A/B tests, are the gold standard for optimizing online-platform policies, they are limited by discrete testing levels. This approach is suboptimal for continuous variables (e.g., prices and incentives), as it fails to extrapolate to untested values or account for user heterogeneity. We address this by developing Deep Learning for Policy Targeting (\textsf{DLPT}) to learn personalized continuous policies from discrete RCTs using high-dimensional features. We prove our estimators are asymptotically unbiased and consistent, achieving a $\sqrt{n}$-regret bound. In a collaboration with a leading social media platform to optimize creator incentives, we show that \textsf{DLPT} substantially outperforms existing benchmarks. In a collaboration with a leading social media platform to optimize creator incentives, we show that \textsf{DLPT} substantially outperforms existing benchmarks.

Deep Learning · Generative Models and Autoencoders

Sizhuang He, Yangtian Zhang, Shiyang Zhang, David van Dijk

The finite symmetric group $S_n$ provides a natural domain for permutations, yet learning probability distributions on $S_n$ is challenging due to its factorially growing size and discrete, non-Euclidean structure. Recent permutation diffusion methods define forward noising via shuffle-based random walks (e.g., riffle shuffles) and learn reverse transitions with Plackett–Luce (PL) variants, but the resulting trajectories can be abrupt and increasingly hard to denoise as $n$ grows. We propose *Soft-Rank Diffusion*, a discrete diffusion framework that replaces shuffle-based corruption with a structured soft-rank forward process: we lift permutations to a continuous latent representation of order by relaxing discrete ranks into soft ranks, yielding smoother and more tractable trajectories. For the reverse process, we introduce *contextualized generalized Plackett–Luce (cGPL)* denoisers that generalize prior PL-style parameterizations and improve expressivity for sequential decision structures. Experiments on sorting and combinatorial optimization benchmarks show that Soft-Rank Diffusion consistently outperforms prior diffusion baselines, with particularly strong gains in long-sequence and intrinsically sequential settings.

Deep Learning · Large Language Models

Ren-Biao Liu, Li Xin-Ye, Hui Sun, Yali Du, Jiang-Tian Xue, Ming Li

Training large language models for code generation requires selecting high-quality data from solution pools where each problem admits multiple correct implementations. Conventional studies on data selection hold that sophisticated strategies that employ various optimization objectives, such as diversity maximization or difficulty ranking, should outperform naive random sampling. However, research on *whether sophisticated selection methods truly benefit code generation* remains limited. In this work, we systematically evaluate multiple selection strategies across different representation spaces, including continuous embeddings, discrete tokens, and syntactic structures, using various base language models. We observe counterintuitive phenomena, suggesting that sophisticated methods may not yield robust benefits. Instead, simple random sampling achieves consistently competitive performance across all models, exhibiting greater stability and transferability than sophisticated methods. We attribute that an *implicit knowledge consensus* exists among diverse correct solutions, such that random selection already covers the common algorithmic knowledge required for training. Our findings provide practical insights into data selection for code generation, suggesting practitioners can adopt simple random sampling without sacrificing performance.

Jian Li, Hua Huang

While Large Language Models (LLMs) possess strong reasoning capabilities, enabling them to learn continuously from experience without parametric retraining remains an open challenge. Existing Retrieval-Augmented Generation (RAG) approaches typically treat memory as a static or append-only corpus, leading to "memory saturation''---where accumulating noise and redundant information degrade performance over time. To address this, we propose an Experience Risk Minimization (ERM) framework that formalizes the experience library as a learnable parameter under an explicit capacity budget. We introduce Textual Stochastic Gradient Descent (TSGD), a discrete optimization algorithm that refines this library via failure-driven Add, Edit, and Delete operations. TSGD estimates ``textual gradients'' through self-reflection and employs a dual-verification mechanism to ensure generalization, effectively preventing overfitting to local errors. Empirical results on MATH and AIME benchmarks demonstrate that TSGD achieves state-of-the-art performance, improving accuracy by up to 18.7\% over zero-shot baselines and significantly outperforming static RAG, all while maintaining a compact memory footprint (compressing hundreds of experiences into $\approx$30 high-utility rules).

Deep Learning · Large Language Models

Kaiyu Huang, Xingyu Wang, Mingze Kong, Zhubo Shi, Yuqian Hou, Hong Xu, Zhongxiang Dai, Minchen Yu, Qingjiang Shi

In real-world deployments of large language models (LLMs), balancing inference quality and computational cost has become a central challenge. Existing approaches tackle this trade-off along two largely independent dimensions: model routing, which switches among models of different scales to match request complexity, and test-time scaling (TTS), which adjusts inference-time compute within a fixed model for fine-grained control. However, this decoupled design introduces inherent limitations. Model routing yields coarse-grained, discrete performance changes due to the sparse set of model scales, while single-model TTS often encounters capacity ceilings and exhibits diminishing returns as compute increases. Moreover, treating the two mechanisms separately restricts adaptability in dynamic inference environments. To overcome these limitations, we introduce *Unified Inference Scaling (UIS)*, which unifies model routing and TTS in a single optimization space. Building on this formulation, we propose UniScale, an online framework that models adaptive UIS as a contextual multi-armed bandit problem and learns inference policies via LinUCB. The framework incorporates efficiency-aware learning and cost modeling to ensure stable and scalable optimization over high-dimensional action spaces. Evaluation shows that UniScale effectively exploits the synergy in the UIS space to deliver a fine-grained and consistently better quality-cost trade-off across diverse, dynamic inference scenarios.

Anders Bo Ipsen, Stratis Skoulakis

We study the revenue of approximate correlated equilibrium in discrete first price auctions - the set of allowable bids is $\mathcal{B} = \{0, 1/k, \dots, 1 - 1/k, 1\}$ for some $k \in \mathbb{N}$. We show that the revenue of any $\epsilon$-\textit{approximate} correlated equilibrium is at least $v_2 - \Theta(1/k)- \Theta(\epsilon k^2)$, where $v_2 \geq 0$ is the second-highest valuation. Our results establish the first polynomial convergence rates on the revenue generated by no-swap regret bidders in first-price auctions. For instance, if bidders admit the optimal swap regret of $\mathcal{O}(\sqrt{k T})$, then the time-averaged revenue is at least $v_2 - \Theta(1/k) - \Theta(\epsilon)$ after $\mathcal{O}(k^5/\epsilon^2)$ rounds.

Optimization · Discrete and Combinatorial Optimization

Ilay Yavlovich, Jad Agbaria, Jose Yallouz, Muhamed Mhamed, Nir Weinberger

The Linear Assignment Problem (LAP) is a fundamental combinatorial optimization task with applications ranging from computer vision to logistics. Classical exact solvers such as the Hungarian and Jonker--Volgenant (LAPJV) algorithms guarantee optimality, but their cubic time complexity $\mathcal{O}(N^3)$ becomes a bottleneck for large-scale instances. Recent learning-based approaches aim to replace these solvers with neural models, often sacrificing exactness or failing to scale due to memory constraints. We propose a *learning-augmented* framework that accelerates exact assignment solvers while maintaining optimality and worst-case guarantees. Our method predicts dual variables to warm-start a classical solver, with a fallback that prevents asymptotic runtime degradation when the learned advice is unreliable. We introduce **RowDualNet**, a lightweight row-independent architecture that avoids the $\mathcal{O}(N^2)$ memory bottleneck of graph-based models, enabling neural warm-starting at large scale ($N=16{,}384$). Feasibility is ensured via a constructive mechanism based on LP duality (namely, the *Min-Trick*), eliminating costly iterative projection. Empirically, our approach reduces the search effort of LAPJV and achieves over $2{\times}$ speedups on challenging synthetic distributions, in addition to improving over $1.25{\times}$ and $1.5{\times}$ on real-world tracking (MOT) and transportation (LPT) datasets, respectively, while strictly maintaining full optimality, effectively yielding a robust zero-shot generalization to real-world tasks.

Deep Learning · Large Language Models

Jiahao Wang, Bingyu Liang, Chenhao Hu, Longhui Zhang, Xuebo Liu, Min zhang, Jing Li, Xuelong Li

Despite remarkable performance on complex tasks, Large Reasoning Models (LRMs) often generate excessively long Chain-of-Thoughts (CoT), inflating computational costs even for simple queries. Existing efforts to mitigate this inefficiency typically rely on discrete reasoning modes or fixed budget tiers, lacking a principled criterion of when reasoning is sufficient. In this work, we introduce *Minimal Sufficient CoT* (MSC), defined as the shortest prefix of a CoT trajectory which is adequate for producing the correct answer. We empirically show that MSC not only reduces reasoning tokens, but also improves accuracy across difficulty levels. Building on MSC, we propose *Sufficiency-guided Continuous Adaptive Reasoning* (SuCo), a two-stage training framework for autonomous reasoning control along a continuous spectrum. In stage 1, *MSC-Aligned Fine-Tuning* (MFT) constructs MSC data using problem-adaptive sufficiency thresholds that naturally scale with question difficulty, then fine-tunes the model to internalize concise yet sufficient reasoning patterns. In stage 2, *Sufficiency-Aware Policy Optimization* (SAPO) further optimizes the model through reinforcement learning with dynamic complexity tracking and sufficiency-aware rewards that penalize both over- and under-thinking. Extensive experiments across mathematics, code, and science benchmarks show that SuCo consistently achieves improvements in both accuracy and reasoning efficiency.

Probabilistic Methods · Variational Inference

Sylvia Liu, Valeria Vitelli, Carlo Mannino, Arnoldo Frigessi, Ida Scheel

We propose the Pseudo-Mallows distribution over the set of all permutations of $n$ items, to approximate the posterior distribution of the Bayesian Mallows model. The Bayesian Mallows model has been successfully used for recommender systems to learn personal preferences from highly incomplete users data. However current inference algorithms do not scale, preventing its use in real-time applications. The Pseudo-Mallows distribution is a product of univariate discrete Mallows-like distributions, where the quality of the approximation depends on the order of the $n$ items in the factorization sequence. In a variational setting, we optimize the variational order parameter by minimising a marginalized KL-divergence, conjecturing a certain form of the optimal variational order that depends on the data, and proposing an approximation algorithm for this discrete optimization. Empirical evidence and some theory support our conjecture. We demonstrate on clicking data that variational inference via the Pseudo-Mallows distribution allows much faster probabilistic preference learning compared to alternative MCMC-based options.

Deep Learning · Large Language Models

Changhai Zhou, Shiyang Zhang, Yuhua Zhou, Qian Qiao, Jun Gao, Cheng Jin, KAIZHOU QIN, Weizhong Zhang

Quantization followed by parameter-efficient fine-tuning has emerged as a promising paradigm for downstream adaptation under tight GPU memory constraints. However, this sequential pipeline fails to leverage the intricate interaction between quantization bit-width and LoRA rank. Specifically, a carefully optimized quantization allocation with low quantization error does not always translate to strong fine-tuning performance, and different bit-width and rank configurations can lead to significantly varying outcomes under the same memory budget. To address this limitation, we propose AutoQRA, a joint optimization framework that simultaneously optimizes the bit-width and LoRA rank configuration for each layer during the mixed quantized fine-tuning process. To tackle the challenges posed by the large discrete search space and the high evaluation cost associated with frequent fine-tuning iterations, AutoQRA decomposes the optimization process into two stages. First, it first conducts a global multi-fidelity evolutionary search, where the initial population is warm-started by injecting layer-wise importance priors. This stage employs specific operators and a performance model to efficiently screen candidate configurations. Second, trust-region Bayesian optimization is applied to locally refine promising regions of the search space and identify optimal configurations under the given memory budget. This approach enables active compensation for quantization noise in specific layers during training. Experiments show that AutoQRA achieves performance close to full-precision fine-tuning with a memory footprint comparable to uniform 4-bit methods.

Optimization · Zero-order and Black-box Optimization

Baek Seong-Eun, Lee Jung-Mok, Kim Sung-Bin, Tae-Hyun Oh

Fine-tuning Large Language Models (LLMs) with Low-Rank Adaptation (LoRA) offers a resource-efficient way to personalize or specialize. However, LoRA is highly sensitive to hyperparameter choices, and performing an exhaustive hyperparameter search remains computationally intensive. To address these challenges, we propose a framework that integrates the domain knowledge of pre-trained LLMs into the Bayesian Optimization (BO) process to efficiently search for LoRA hyperparameters. To leverage pre-trained LLMs' knowledge, our approach repurposes them as a discrete-to-continuous mapping module to link hyperparameters and their domain knowledge to a continuous vector space, where BO is conducted. We design and control the mapping via language prompting, providing a domain-aware textual prompt that describes the relationships among hyperparameters and their respective roles. This allows us to explicitly inject domain knowledge about LoRA into the LLM in natural language. We also introduce an additional learnable token to capture residual information that is difficult to describe linguistically in the prompt. This aids BO to sample more high-performing hyperparameters. In addition, by leveraging the strong correlation observed between the performance obtained from full and subset training datasets in LoRA training regimes, we introduce proxy training and evaluation using a data subset. This significantly improves the efficiency of our method. We demonstrate that our hyperparameter, discovered with only about 30 iterations, achieves more than 20% performance improvement over standard hyperparameters found from about 45,000 combinations. Code will be released.

Deep Learning · Large Language Models

Yash Akhauri, Mohamed Abdelfattah

Efficient LLM inference research has largely focused on reducing the cost of each decoding step (e.g., using quantization, pruning, or sparse attention), typically applying a uniform computation budget to every generated token. In practice, token difficulty varies widely, so static compression can over-compute on easy steps and under-compute on hard ones. We study **dynamic budget allocation** for autoregressive decoding: learning how much computation to spend **per token** from within a single model. Self-Optimizing Language Models (SOL) pair a frozen LLM with a lightweight policy network that reads the LLM hidden state and selects a discrete **efficiency action** at each decode step. Actions can jointly control (i) token-level attention sparsity, (ii) structured activation pruning in the MLP, and (iii) activation quantization bit-width, while leaving the base model weights unchanged. We train the policy with group-relative policy optimization on teacher-forced episodes: the token sequence is fixed, while we sample multiple compute schedules (i.e., “counterfactual” schedules that vary only the efficiency actions for the same token path) and compare their likelihoods under the same supervision. Our reward trades off language‑model quality against soft penalties that encourage episode‑average budget usage to match a requested target. Across model variants and compute regimes, SOL improves quality at matched budget over static allocation and strong random schedule search, offering a complementary axis for inference‑efficiency optimization. SOL discovers a better quality-efficiency pareto-front across all our experiments and improves MMLU accuracy by upto 7.3\% over uniform budget allocation strategies.

Reinforcement Learning · Inverse

Raphaël Baur, Yannick Metz, Maria Gkoulta, Mennatallah El-Assady, Giorgia Ramponi, Thomas Kleine Buening

Reward learning typically relies on a single feedback type or combines multiple feedback types using manually weighted loss terms. Currently, it remains unclear how to jointly learn reward functions from heterogeneous feedback types such as demonstrations, comparisons, ratings, rankings, and stops that provide qualitatively different signals. We address this challenge by formulating reward learning from multiple feedback types as Bayesian inference over a shared latent reward function, where each feedback type contributes information through an explicit likelihood. We introduce a scalable amortized variational inference approach that learns a shared reward encoder and feedback-specific likelihood decoders and is trained by optimizing a single evidence lower bound. Our approach avoids reducing feedback to a common intermediate representation and eliminates the need for manual loss balancing. Across discrete and continuous-control benchmarks, we show that jointly inferred reward posteriors outperform single-type baselines, exploit complementary information across feedback types, and yield policies that are more robust to environment perturbations. The inferred reward uncertainty further provides interpretable signals for analyzing model confidence and consistency across feedback types.

Deep Learning · Large Language Models

Weiyu Huang, Pengle Zhang, Xiaolu Zhang, JUN ZHOU, Jun Zhu, Jianfei Chen

Structured pruning reduces LLM inference cost by removing low-importance architectural components. This can be viewed as learning a multiplicative gate for each component under an $\ell_0$ sparsity constraint. Due to the discreteness of the $\ell_0$ norm, prior work typically adopts stochastic hard-concrete relaxations to enable differentiable optimization; however, this stochasticity can introduce a train--test mismatch when sampled masks are discretized for deployment and restricting masks to a bounded, near-binary range. To address this, we propose Deterministic Differentiable Pruning (DDP), a mask-only optimization method that eliminates stochasticity by directly optimizing a deterministic soft surrogate of the discrete $\ell_0$ objective. Compared with prior approaches, DDP offers greater expressiveness, reduced train--test mismatch, and faster convergence. We apply our method to several dense and MoE models, including Qwen3-32B and Qwen3-30B-A3B, achieving a performance loss as small as 1\% on downstream tasks while outperforming previous methods at 20\% sparsity. We further demonstrate end-to-end inference speedups in realistic deployment settings with vLLM.

Reinforcement Learning · Deep RL

Chen Jia

We propose to model text generation as a continuous-time latent dynamical process, where token generation is formulated as a Markov Decision Process whose internal state evolves via a neural ODE. This formulation bridges discrete token sequences and continuous semantic evolution, providing a theoretically grounded approach for coherent long-range generation. The framework is optimized via reinforcement learning, maximizing a composite objective that integrates task-specific rewards with knowledge distillation from a powerful pre-trained language model. Experiments demonstrate that our method, Continuous-Time Latent Language Model (CT-LLM), outperforms discrete baselines in generation coherence and long-context performance, offering a new paradigm for fluid and controllable language generation.

Applications · Time Series

Ruituo Wu, Hongyu Zhang, Qiang Wang, Jiawei Du, Wei Cui, Ce Zhu, Bing Li

Learning from continuous-time trajectories requires modeling multivariate sensor measurements generated by underlying physical or dynamical processes. Under extreme data compression and heterogeneous sampling, directly optimizing synthetic signals as discrete sample values becomes fundamentally misaligned with the underlining \emph{continuous-time physical processes}, often producing high-frequency, non-physical artifacts that overfit specific models and break reuse across architectures and sampling rates. We propose CHESS, a \emph{function-first} synthesis framework shifts optimization from discrete samples to underlying continuous-time signal trajectories. CHESS injects physics-induced structure by jointly enforcing low-rank spatial coherence and piecewise Chebyshev polynomial temporal parameterization, constraining synthesis to a physically meaningful function manifold. We provide theoretical analysis establishing explicit smoothness and stability guaranties. Experiments on diverse sensor testbeds under the dataset distillation protocol demonstrate CHESS consistently outperform state-of-the-art methods with a compression ratios up to $133\times$ for each synthetic sample. Furthermore, CHESS exhibits strong cross-architecture generalization and enables zero-shot adaptation across different sampling resolutions.

Deep Learning · Large Language Models

Liulu He, Xuan Ang Liu, Juntao Liu, Taolue Feng, Ting Lu, Chunsheng Gan, ZHIYV PENG, Yuan Du, Li Du, Huanrui Yang 等

Existing quantization methods are fundamentally limited by rigid, integer-based bit-widths (e.g., 2, 3-bit), creating a "deployment gap" where LLMs cannot be optimally fitted to specific memory budgets. To bridge this gap, we introduce LiftQuant, a novel framework that enables continuous bit-width control for true Pareto-optimal deployment. The core innovation is a "lift-then-project" mechanism: we represent d-dimensional weight vectors by projecting a simple 1-bit lattice from a tunable D-dimensional "lifted" space. By adjusting the lifted dimension D, LiftQuant naturally yields an effective bit-width of D/d, allowing for seamless, continuous resolution adjustment rather than discrete steps. This projection generates a structured yet non-uniform codebook, capturing the expressive power of Vector Quantization. Crucially, its decoding path relies solely on linear transformations and 1-bit uniform quantizers, retaining hardware-friendly efficiency. This flexibility is transformative: LiftQuant enables a 70B LLM to be compressed to 2.4 bits to precisely fit a 24GB GPU, where its performance significantly surpasses state-of-the-art 2-bit models. With a decoding throughput up to 6.7x faster than FP16, LiftQuant redefines compression as a continuous optimization problem, paving the way for a new generation of hardware-aware LLM deployment.

Social Aspects · Robustness

Jayoung Kim, Kookjin Lee, Noseong Park, Sanghyun Hong

Implicit neural representations (INRs) encode data as continuous coordinate-based functions parameterized by neural networks, shifting downstream tasks such as image recognition to operate on functional rather than discrete representations. Despite their increasing adoption, the adversarial robustness of INR-based classification pipelines remain largely underexplored. In this work, we present the first systematic study of adversarial robustness in INR-based classifiers. A key challenge is that generating an INR requires $\text{\emph{training}}$ a neural network for each input sample, resulting in an optimization-in-the-loop forward pass that renders standard gradient-based attacks computationally prohibitive. To address this, we design surrogate models that amortizes the INR-generation process, serving as a practical proxy for attacking INR-based classifiers. We also develop speed-up techniques that substantially reduce the training cost of the surrogate. We show that in contrast to recent work, INR-based classifiers are vulnerable: under adversarial input perturbations, classification accuracy collapses to near zero. Moreover, existing countermeasures designed to operate on discrete representations offer limited protection.

Optimization · Stochastic

Alessandro Manenti, Cesare Alippi

Latent categorical variables are frequently found in deep learning architectures. They can model actions in discrete reinforcement-learning environments, represent categories in latent-variable models, or express relations in graph neural networks. Despite their widespread use, their discrete nature poses significant challenges to gradient-descent learning algorithms. While a substantial body of work has offered improved gradient estimation techniques, we take a complementary approach. Specifically, we: 1) revisit the ubiquitous *softmax* function and demonstrate its limitations from an information-geometric perspective; 2) replace the *softmax* with the *catnat* function, a function composed by a sequence of hierarchical binary splits; we prove that this choice offers significant advantages to gradient descent due to the resulting diagonal Fisher Information Matrix. A rich set of experiments - including graph structure learning, variational autoencoders, and reinforcement learning - empirically show that the proposed function improves the learning efficiency and yields models characterized by consistently higher test performance. *Catnat* is simple to implement and seamlessly integrates into existing codebases. Moreover, it remains compatible with standard training stabilization techniques and, as such, offers a better alternative to the *softmax* function.