论文检索

Deep Learning · Large Language Models

Emmy Liu, Graham Neubig, Chenyan Xiong

Midtraining, the practice of mixing specialized data with more general pretraining data in an intermediate training phase, has become widespread in language model development, yet there is little understanding of what makes it effective. We propose that midtraining functions as distributional bridging by providing better initialization for posttraining. We conduct controlled pretraining experiments, and find that midtraining benefits are largest for domains distant from general pretraining data, such as code and math, and scale with the proximity advantage the midtraining data provides toward the target distribution. In these domains, midtraining consistently outperforms continued pretraining on specialized data alone both in-domain and in terms of mitigating forgetting. We further conduct an investigation on the starting time and mixture weight of midtraining data, using code as a case study, and find that time of introduction and mixture weight interact strongly such that early introduction of specialized data is amenable to high mixture weights, while late introduction requires lower ones. This suggests that late introduction of specialized data outside a plasticity window cannot be compensated for by increasing data mixtures later in training. Beyond midtraining itself, this suggests that distributional transitions between any training phases may benefit from similar bridging strategies.

Deep Learning · Generative Models and Autoencoders

Zhongwei Zhang, Fuchen Long, Wei Li, Zhaofan Qiu, Wu Liu, Ting Yao, Tao Mei

The In-context generation paradigm has demonstrated strong power in instructional image editing for better synthesis quality. Nevertheless, shaping such in-context learning for instructional video editing is not trivial. Without specifying editing regions, the results can suffer from the issue of inaccurate editing regions and the token interference between different areas. To address these, we present ReCo, a new instructional video editing paradigm that novelly delves into **Re**gional **Co**nstraint modeling between editing and non-editing areas. Technically, ReCo width-wise concatenates source and target video for joint denoising. In model training, ReCo formulates regional constraints with two regularization terms, i.e., latent and attention regularization, on one-step backward denoised latents and attention maps, respectively. The former increases the latent discrepancy of the editing region between source and target videos while reducing that of non-editing areas, emphasizing editing area modification and alleviating unexpected content generation. The latter suppresses the attention of tokens in the editing region to the tokens in counterpart of the source video, thereby mitigating their interference during novel object generation in target video. Furthermore, we propose a large-scale, high-quality video editing dataset, i.e., ReCo-Data, comprising 500K instruction-video pairs. Extensive experiments conducted on four major instruction-based video editing tasks verify the superiority of ReCo.

Applications · Everything Else

Hanyang Wang, Xiwei Wu, Qinxiang Cao

Formal verification of system software with complex heap manipulations remains challenging. Standard automated solvers frequently fail to discharge separation logic verification conditions even when correct specifications like loop invariants are provided, forcing verification engineers to manually construct proofs. While large language models (LLMs) have shown promise in proof synthesis, specialized approaches for separation logic remain unexplored. To bridge this gap, we introduce SL-VC (Separation Logic Verification Conditions), a benchmark of 208 verification conditions derived from real-world C programs, including data structures and algorithms, the LiteOS kernel's linked list library, and the mini-gmp library. Our evaluation reveals that general-purpose LLMs and existing LLM-based Rocq provers struggle to effectively discharge these verification conditions. To address this challenge, we propose SPLIT (Split spatial and pure Proving with LLM-frIendly Tactics), a novel framework that enables predictable proof state transitions through an LLM-friendly tactic library, combined with a two-stage workflow that separates spatial and pure reasoning to align with separation logic semantics. Experimental results on SL-VC demonstrate that SPLIT consistently outperforms existing approaches, showing that LLM-assisted proof synthesis is a promising solution for separation logic verification of real-world system software.

Applications · Computer Vision

Jingxuan Wu, Zhenglin Wan, Xingrui Yu, Yuzhe YANG, Bo An, Ivor Tsang, Yang You

Flow-based text-to-image models follow deterministic trajectories, making it costly to explore diverse modes under limited sampling budgets. Existing approaches to improving diversity often rely on retraining or degrade image fidelity. To address this limitation, we present a training-free, inference-time control mechanism that makes the flow itself diversity-aware. Our core insight is to encourage diversity through guidance that is geometrically decoupled from the model’s quality-seeking direction. Our method simultaneously encourages lateral spread among trajectories via a feature-space objective and reintroduces uncertainty through a time-scheduled stochastic perturbation. Crucially, this perturbation is projected to be orthogonal to the generation flow, a geometric constraint that allows it to boost variation without degrading image details or prompt fidelity. Theoretically, we show that this design monotonically increases a volume surrogate while approximately preserving the marginal distribution, providing a principled explanation for the robustness of generation quality. Empirically, across multiple text-to-image settings under fixed sampling budgets, our method consistently improves diversity metrics such as the Vendi Score and Brisque over strong baselines, while upholding image quality and alignment.

Reinforcement Learning · Everything Else

Zhenglin Wan, Jingxuan Wu, Xingrui Yu, Chubin Zhang, Mingcong Lei, Bo An, Ivor Tsang, Yang You

Flow Matching (FM) has shown remarkable ability in modeling complex distributions and achieves strong performance in offline imitation learning for cloning expert behaviors. However, despite its behavioral cloning expressiveness, FM-based policies are inherently limited by their lack of environmental interaction and exploration. This leads to poor generalization in unseen scenarios beyond the expert demonstrations, underscoring the necessity of online interaction with environment. Unfortunately, optimizing FM policies via online interaction is challenging and inefficient due to instability in gradient computation and high inference costs. To address these issues, we propose to let a "student'' policy with simple MLP structure to explore the environment and be online updated via RL algorithm with a reward model. This reward model is associated with a "teacher'' FM model, containing rich information of expert data distribution. Furthermore, the same "teacher'' FM model is utilized to regularize the "student'' policy's behavior to stabilize policy learning. Due to the student’s simple architecture, we avoid the gradient instability of FM policies and enable efficient online exploration, while still leveraging the expressiveness of the teacher FM model. Extensive experiments shows that our approach significantly enhances learning efficiency, generalization, and robustness, especially when learning from suboptimal expert data.

Social Aspects · Fairness

Nika Haghtalab, Ariel Procaccia, Han Shao, Serena Wang, Kunhe Yang

Recent leaderboard-based evaluations of large language models aggregate user feedback by fitting a Bradley--Terry model to pairwise comparisons, producing a single global ranking based on a latent quality score. While appealing for its simplicity, this approach is incompatible with heterogeneous preferences: when LLMs are used across diverse tasks and use cases, users who favor fundamentally different model behaviors can be systematically misrepresented when collapsed into a single quality score. To address this issue, we study *pluralistic leaderboards* that aim to remain *stable* with respect to heterogeneous user populations. Drawing on ideas from social choice theory, we adapt the notion of *local stability*, which requires that no model outside the top-$k$ positions is collectively preferred to the top-$k$ set by more than $O(1/k)$ fraction of users. Building on techniques from the social choice literature, we design an alternative leaderboard mechanism that satisfies local stability while eliciting only $\widetilde{O}(k)$ pairwise comparisons per user, where $k$ is the size of the prefix for which stability is guaranteed. Using data from LMArena, we show that standard Bradley--Terry aggregation can violate local stability in practice, whereas our method provides substantially stronger stability guarantees.

Reinforcement Learning · Online

Qingzhang Zeng, Peixi Peng, hang li, Luntong Li, Yonghong Tian

Sample efficiency remains one of the central challenges in modern deep reinforcement learning (DRL). In recent years, world model approaches have significantly improved the sample efficiency of model-based reinforcement learning (MBRL) by enabling policy learning in imagination, thereby reducing the need for direct interaction with the real environment. However, most existing world model methods are trained independently for each task or perform multi-task learning in offline settings, failing to fully exploit the latent relationships among tasks in online interactive scenarios. To address this limitation, we propose TaskLoom, a knowledge-sharing world model architecture for online reinforcement learning. TaskLoom adopts a grouped two-stage training paradigm: in the first stage, fine-grained knowledge is shared among tasks within each group, while in the second stage, coarse-grained knowledge is exchanged across groups, enabling hierarchical knowledge transfer and reuse. Experimental results show that TaskLoom outperforms baseline methods on widely used benchmarks such as Proprio Control and Visual Control, validating the effectiveness of the proposed knowledge-sharing mechanism for both low-dimensional state and high-dimensional visual inputs.

Deep Learning · Robustness

Zheng Wang, Yi Zhang, Siddartha Khastgir, carsten maple, Xingyu Zhao

Deep learning (DL) models, despite their remarkable success, remain vulnerable to small input perturbations that can cause erroneous outputs, motivating probabilistic robustness (PR) as a complementary notion to adversarial robustness (AR) for stochastic reliability assessment. However, existing PR formulations assume a fixed, known perturbation distribution, which is often unavailable or misspecified in practice. To address this limitation, we propose non-parametric probabilistic robustness (NPPR), a more conservative PR estimator over an admissible family of perturbation distributions. We instantiate NPPR with a tractable estimator (GMM-based) that supports four dependency structures (independent, label-, input-, and joint-dependent perturbations). We show that NPPR provably interpolates between AR and PR. Experiments on CIFAR-10/100 and TinyImageNet across multiple architectures show that NPPR yields consistently lower (more conservative) PR estimates than PR computed under common assumed distributions (e.g., Gaussian/Uniform), with up to 40\% reduction in representative settings.

Deep Learning · Large Language Models

Tianyi MA, Yiyue Qian, Yiyang Li, Zehong Wang, Yifan Ding, Zheyuan Zhang, Yan Liang, Chuxu Zhang, Yanfang Ye

Autoregressive models generate sequences monotonically, where any sampled token, even if erroneous or sub-optimal, becomes a permanent condition for all subsequent steps. This structural limitation means that autoregressive models cannot revisit or revise earlier decisions, i.e., a capability essential for complex generation tasks where exploration and correction are necessary. To this end, we propose N-MARS, a Non-Monotonic AutoregRessive Sequence modeling framework that enables models to generate, evaluate, and revise tokens within a single forward pass, effectively allowing exploration before commitment. We operationalize this framework through a learned erase token that retracts the previous token, enabling on-the-fly revision within standard autoregressive decoding. To train the model, we introduce a sequence augmentation method that constructs error-correction trajectories from model-generated deviations paired with ground-truth references. We then propose masked supervised fine-tuning (mSFT) that exposes the model to errors as context for learning when to revise, without optimizing their likelihood. Finally, we refine the model with group relative policy optimization (GRPO) that incentivizes judicious usage, i.e., rewarding effective corrections while penalizing unsuccessful ones. We conduct comprehensive theoretical and empirical analysis to validate the effectiveness of N-MARS, demonstrating a robust foundation for non-monotonic sequence modeling.

Reinforcement Learning · Multi-agent

Haibo Jin, Peng Kuang, Ye Yu, Xiaopeng Yuan, Haohan Wang

While existing multi-agent systems (MAS) can handle complex problems by enabling collaboration among multiple agents, they are often highly task-specific, relying on manually crafted agent roles and interaction prompts, which leads to increased architectural complexity and limited reusability across tasks. Moreover, most MAS communicate primarily through natural language, making them vulnerable to error accumulation and instability in long-context, multi-stage interactions within internal agent histories. In this work, we propose \textbf{Agent Primitives}, a set of reusable latent building blocks for LLM-based MAS. Inspired by neural network design, where complex models are built from reusable components, we observe that many existing MAS architectures can be decomposed into a small number of recurring internal computation patterns. Based on this observation, we instantiate three primitives: Review, Voting and Selection, and Planning and Execution. All primitives communicate internally via key–value (KV) cache, which improves both robustness and efficiency by mitigating information degradation across multi-stage interactions. To enable automatic system construction, an Organizer agent selects and composes primitives for each query, guided by a lightweight knowledge pool of previously successful configurations, forming a primitive-based MAS. Experiments show that primitives-based MAS improve average accuracy by 12.0–16.5\% over single-agent baselines, reduce token usage and inference latency by approximately 3$\times$–4$\times$ compared to text-based MAS, while incurring only 1.3$\times$–1.6$\times$ overhead relative to single-agent inference and providing more stable performance across model backbones.

Deep Learning · Foundation Models

Tengyao Tu, Yulin Li, Huiling Zhen, Libo Qin, Zhoujun Wei, Jinghua Piao, Zhuotao Tian, Yong Li, Min zhang

Recent advances in Large Reasoning Models (LRMs) demonstrate remarkable performance improvements by iteratively reflecting, exploring, and executing complex tasks, yet suffer from inefficiencies due to redundant reasoning, known as "overthinking". Existing methods to mitigate this issue either rely on static difficulty estimates or require task-specific training, and thus fail to adapt to the dynamic complexity during reasoning. In this work, we empirically show that the problem difficulty evolves dynamically throughout the reasoning process and is linearly encoded in the LRM’s step-level embeddings. Building on this insight, we propose DyCon, a training-free framework that leverages latent step-level representations to explicitly model the evolving task difficulty, enabling the dynamic control of reasoning depth to mitigate the overthinking issue. Extensive experiments conducted on four models ranging from 4B to 32B, and across twelve benchmarks in math reasoning, general question answering, and coding tasks demonstrate that DyCon significantly enhances reasoning efficiency by reducing redundant steps without sacrificing accuracy or generalization. Code and models will be made publicly available.

Theory · Domain Adaptation and Transfer Learning

Jiaxuan Cheng

Plasticity—a neural network's ability to adapt to new tasks—is critical for continual and transfer learning. Existing measures, such as effective rank, dead neuron fraction, and weight norm, lack theoretical grounding and correlate poorly with performance on new tasks. We introduce *local redundancy*, an information-theoretic measure derived from universal compression theory. We define local redundancy as the worst-case redundancy of a local model family—parameters in an infinitesimal neighborhood along gradient directions—and show this is a principled measure of plasticity. Although local redundancy is intractable to compute exactly, we prove that the expected squared gradient norm on a synthetic memorization task provides an efficiently computable lower bound. Experiments on continual image classification and time series transfer learning demonstrate that local redundancy predicts downstream performance better than existing measures and enables pretraining checkpoint selection where validation loss plateaus.

Applications · Everything Else

Cheng Huang, Jinghua Piao, Wang Ranran, Yong Li

Financial reasoning requires precise execution. While Large Language Model (LLM) agents have shown encouraging progress in financial reasoning, their effectiveness in realistic financial workflows is severely hindered by the lack of holistic benchmarks and the fragility of unstructured reasoning. To evaluate these capabilities, we introduce ProFinR, the first Professional Finance Reasoning benchmark that covers four types of financial tasks, comprising 528 expert-designed tasks. To solve these complex financial reasoning questions, we construct Financial Tool Universe, a tool library containing 53 domain-specific tools organized into 13 categories. Building on the tool library, we introduce ProFinAgents, a structured agent framework based on Directed Acyclic Graph (DAG) and Case-Based Memory (CBM). Compared with strictly sequential workflows, ProFinAgent coordinates tool execution through DAG. This allows for parallel execution and reduces latency compared to serial pipelines. Furthermore, the CBM component refines decision-making over time by retrieving prior cases to mitigate reasoning failures. Experimental results demonstrate that ProFinAgent achieves a 49.81% performance gain over state-of-the-art baselines with a 47.1% reduction in inference latency.

Reinforcement Learning · Multi-agent

Simin Li, Zihao Mao, Zheng Yuwei, Linhao Wang, Ruixiao Xu, Chengdong Ma, Zhiqian Liu, Xin Yu, Yuqing Ma, Xin Wang 等

Partial agent failure becomes inevitable when systems scale up, making it crucial to identify the subset of agents whose failure causes worst-case system performance degradations. We study this Vulnerable Agent Identification (VAI) problem in large-scale multi-agent reinforcement learning (MARL). We frame VAI as a Hierarchical Adversarial Decentralized Mean Field Control (HAD-MFC), where where the upper level selects vulnerable agents as an NP-hard task and the lower level learns their worst-case adversarial policies via mean-field MARL. The two problems are coupled together, making HAD-MFC difficult to solve. To handle this, we first decouple the hierarchical process by Fenchel-Rockafellar transform, resulting a regularized mean-field Bellman operator for upper level that enables independent learning at each level, thus reducing computational complexity. We next reformulate the upper-level NP-hard problem as an MDP with dense rewards, allowing sequential identification of vulnerable agents via greedy and RL algorithms. This decomposition provably preserves the optimal solution. Experiments show our method effectively identifies more vulnerable agents in large-scale MARL and the rule-based system, fooling system into worse failures, and reveals the vulnerability of each agent in large systems. Code available at \url{https://anonymous.4open.science/r/VAI-5F61/}.

Applications · Computer Vision

Jingcheng Xie, Yinda Chen, Xiaoyu Liu, Haoyuan Shi, Zhiwei Xiong

Segment Anything Models (SAMs) often struggle with certain specialized tasks. A common approach is to fine-tune models with specific task labels, but this often leads to overfitting, introduces model bias and significantly degrades their generalization ability. To overcome these challenges, we propose SHERPA, a novel framework that leverages a smaller SAM to guide the fine-tuning of a larger SAM via task-relevant features. Specifically, we first leverage the Fisher Ratio Separation (FRS) module to separate high task-relevant features and preserve the ability of the large SAM to perform other general tasks. Then, the Guiding Feature Extraction (GFE) module is used to extract representative guiding features from the fine-tuned small SAMs. We leverage small SAMs tailored for specific tasks (including natural image segmentation, biomedical image segmentation, and video object segmentation) as guidance and then evaluate the SHERPA scheme to fine-tune larger SAM series models. Our experiments demonstrate that SHERPA enhances the retention of generalization ability across those diverse tasks, by up to 11.1\%, and improves specific task performance by up to 2.2\%.

Applications · Social Sciences

Jinghua Piao, zhentao liu, Cheng Huang, Huang Jiarui, Songwei Li, Wang Ranran, Yong Li

Stock markets are one of the most complex systems in the modern world, where prices emerge from billions of decentralized interactions among heterogeneous participants in an ever-evolving information landscape. While high-fidelity simulation is important for understanding market dynamics, existing approaches face a persistent trade-off between structural and behavioral fidelity. To this end, we propose MarketSim, a large-scale stock market simulation framework with generative agents. MarketSim introduces a hierarchical multi-agent architecture that decouples strategic reasoning from high-frequency execution, enabling LLM agents to operate in a nanosecond-resolution, NASDAQ-like continuous double auction market. Building on this, we simulate over 15,000 heterogeneous market participants whose interactions shape and are shaped by an evolving market environment grounded in more than 12k real-world news articles, policy documents, and earnings reports. To evaluate our proposed MarketSim, we develop a comprehensive benchmark that includes stocks from 8 GICS sectors and 3 representative real-world scenarios, along with 5 stylized facts for market complexity and 5 price-related statistical metrics. Extensive experiments demonstrate that MarketSim not only captures key complexity properties of real-world markets, but also outperforms state-of-the-art baselines in tracking high-frequency price dynamics with an average MAPE of 3.48%, providing a scalable testbed for market analysis.

General Machine Learning · Representation Learning

Seonglae Cho, Zekun Wu, Adriano Koshiyama

Sparse Autoencoders (SAEs) decompose LLM activations into interpretable features, yet existing SAE-based steering methods require contrastive datasets or large activation stores. We introduce CorrSteer, which selects steering features by correlating task outcomes with SAE activations computed during generation, then validates these selections through intervention. This two-stage approach treats correlation as a selection heuristic and intervention as the causal test: features that both correlate with success and improve performance when amplified are retained. Coefficients derive from mean activations on correct samples, yielding a fully automated pipeline without task-specific tuning. On Gemma-2 2B and LLaMA-3.1 8B, CorrSteer achieves +3.3% on MMLU (4k samples) and +27.2% on HarmBench (108 samples), with lower side-effect ratios than fine-tuning despite comparable accuracy. Selected features cluster into interpretable categories: structured-output features for multiple-choice tasks, refusal features for safety, and domain-specific semantics for specialized benchmarks. The method scales to $10^5$ SAE features via streaming correlation ($O(1)$ in dataset size), requiring no backward passes or activation storage.

Deep Learning · Large Language Models

Ezra Winston, Zico Kolter

Language models acquire syntax and world knowledge together, entangling the two in ways that limit efficiency and controllability. We show that syntax can be learned while suppressing semantic plausibility and world‑knowledge cues, yielding more efficient and controllable models. We train tiny LMs on grammatical nonsense — syntactically well-formed text with semantic content ablated via constrained relexicalization (SAMBAL). Models trained on this data perform comparably to standard pretraining on syntactic benchmarks (BLiMP, SyntaxGym) while scoring at chance on world knowledge probes (EWoK). On targeted grammar-plausibility conflict probes, content-neutral models prefer grammaticality where standard models prefer plausibility, and their representations show more syntactic vs lexical alignment. On efficiency, disentanglement yields substantial sample and parameter gains: in low‑resource regimes, a 5M‑parameter model matches a 30M‑parameter baseline at the same data budget. On controllability, content-neutral models adapt rapidly to a new domain with minimal exposure, suggesting the feasibility of modular post‑hoc knowledge specialization.

Reinforcement Learning · Batch/Offline

Pengcheng Wang, Qinghang Liu, Haotian Lin, Yiheng Li, Guojian Zhan, Masayoshi Tomizuka, Yixiao Wang

Learning domain adaptive policies that can generalize to unseen transition dynamics, remains a fundamental challenge in learning-based control. Substantial progress has been made through domain representation learning to capture domain-specific information, thus enabling domain-aware decision making. We analyze the process of learning domain representations through dynamical prediction and find that selecting contexts adjacent to the current step causes the learned representations to entangle static domain information with varying dynamical properties. Such mixture can confuse the conditioned policy, thereby constraining zero-shot adaptation. To tackle the challenge, we propose DADP (Domain Adaptive Diffusion Policy), which achieves robust adaptation through unsupervised disentanglement and domain-aware diffusion injection. First, we introduce Lagged Context Dynamical Prediction, a strategy that conditions future state estimation on a historical offset context; by increasing this temporal gap, we unsupervisedly disentangle static domain representations by filtering out transient properties. Second, we integrate the learned domain representations directly into the generative process by biasing the prior distribution and reformulating the diffusion target. Extensive experiments on challenging benchmarks across locomotion and manipulation demonstrate the superior performance, and the generalizability of DADP over prior methods. More visualization results are available on the https://anonymous.4open.science/w/dadp-webpage-D841.

Applications · Everything Else

Zhongju Yuan, Geraint Wiggins, Dick Botteldooren

Audio provides critical situational cues, yet current Audio Language Models (ALMs) face an attention bottleneck in long-form recordings where dominant background patterns can dilute rare, salient events. We introduce NAACA, a training-free NeuroAuditory Attentive Cognitive Architecture that reframes attention allocation as an auditory salience filtering problem. At its core is OWM, a neuro-inspired Oscillatory Working Memory that maintains stable attractor-like states and triggers higher-cognition ALM processing only when adaptive energy fluctuations signal perceptual salience, triggering higher-level reasoning. On XD-Violence, NAACA improves AudioQwen’s average precision (AP) from 53.50\% to 70.60\% while reducing unnecessary ALM invocations. Furthermore, qualitative case studies on the Urban Soundscapes of the World (USoW) dataset show that OWM captures novel events and subcategory shifts while remaining robust to transient pauses and ambient urban noise.