Searcharxiv⌕ Search

arXiv subjects

Hao Li

Publications and source records attributed to Hao Li.

At least 253 records · Page 14Linked to original sources
arXiv2026

LogitSpec: Accelerating Retrieval-based Speculative Decoding via Next Next Token Speculation

Speculative decoding (SD), where a small draft model is employed to propose draft tokens in advance and then the target model validates them in parallel, has emerged as a promising technique for LLM inference acceleration. Many endeavors to improve SD are to eliminate the need for a draft model and generate draft tokens in a retrieval-based manner in order to further alleviate the drafting overhead and significantly reduce the difficulty in deployment and applications. However, retrieval-based SD relies on a matching paradigm to retrieval the most relevant reference as the draft tokens, where these methods often fail to find matched and accurate draft tokens. To address this challenge, we propose LogitSpec to effectively expand the retrieval range and find the most relevant reference as drafts. Our LogitSpec is motivated by the observation that the logit of the last token can not only predict the next token, but also speculate the next next token. Specifically, LogitSpec generates draft tokens in two steps: (1) utilizing the last logit to speculate the next next token; (2) retrieving relevant reference for both the next token and the next next token. LogitSpec is training-free and plug-and-play, which can be easily integrated into existing LLM inference frameworks. Extensive experiments on a wide range of text generation benchmarks demonstrate that LogitSpec can achieve up to 2.61 $\times$ speedup and 3.28 mean accepted tokens per decoding step. Our code is available at https://github.com/smart-lty/LogitSpec.

cs.CL↗
arXiv2026

AutoResearchBench: Benchmarking AI Agents on Complex Scientific Literature Discovery

Autonomous scientific research is significantly advanced thanks to the development of AI agents. One key step in this process is finding the right scientific literature, whether to explore existing knowledge for a research problem, or to acquire evidence for verifying assumptions and supporting claims. To assess AI agents' capability in driving this process, we present AutoResearchBench, a dedicated benchmark for autonomous scientific literature discovery. AutoResearchBench consists of two complementary task types: (1) Deep Research, which requires tracking down a specific target paper through a progressive, multi-step probing process, and (2) Wide Research, which requires comprehensively collecting a set of papers satisfying given conditions. Compared to previous benchmarks on agentic web browsing, AutoResearchBench is distinguished along three dimensions: it is research-oriented, calling for in-depth comprehension of scientific concepts; literature-focused, demanding fine-grained utilization of detailed information; and open-ended, involving an unknown number of qualified papers and thus requiring deliberate reasoning and search throughout. These properties make AutoResearchBench uniquely suited for evaluating autonomous research capabilities, and extraordinarily challenging. Even the most powerful LLMs, despite having largely conquered general agentic web-browsing benchmarks such as BrowseComp, achieve only 9.39% accuracy on Deep Research and 9.31% IoU on Wide Research, while many other strong baselines fall below 5%. We publicly release the dataset and evaluation pipeline to facilitate future research in this direction. We publicly release the dataset, evaluation pipeline, and code at https://github.com/CherYou/AutoResearchBench.

cs.AI↗
arXiv2026

Golden RPG: Confidence-Adaptive Region-Aware Noise for Compositional Text-to-Image Generation

Compositional text-to-image (T2I) generation requires a model to honour multiple sub-prompts that describe distinct image regions. Recent work shows that the \emph{starting noise} of a diffusion model carries significant semantic information: ``golden'' noise predicted from text can substantially raise prompt fidelity. We observe that this noise prediction is, however, fundamentally global: the same network is asked to summarise a long, multi-region prompt with a single text embedding, which becomes the bottleneck whenever the prompt describes scenes with spatially-separated entities. We introduce \textbf{Golden RPG}, a region-aware noise predictor that extends a frozen NPNet with two trainable additions: (i) a per-region \textbf{FiLM adapter} that reshapes the predicted noise according to each sub-prompt; and (ii) a \textbf{Region Cross-Attention} layer injected between two stages of the Swin backbone, allowing different spatial locations to attend to different sub-prompt tokens. To prevent the regional conditioning from degrading samples whose prompts are already easy, we further propose a \textbf{Confidence-Adaptive Blending} head that dynamically predicts, per sample, how strongly the regional signal should override the global signal. We evaluate on the original RPG benchmark (20 prompts, 100 samples) and on four multi-region categories of T2I-CompBench (1{,}200 images, six competing methods). Golden RPG achieves the highest Cross-Region-Coherence score on every category, while matching the strongest baselines on absolute CLIP-Score and CLIP-IQA. A paired user study further shows a $\boldsymbol{\sim}$67\% preference over the strongest baseline. The adapter contains $\sim$2M trainable parameters and adds only $0.6$\,s of inference overhead on top of SDXL.

cs.CV↗
arXiv2026

Parameterized Post-Newtonian Analysis of Quadratic Gravity and Solar System Constraints

This work systematically investigates the post-Newtonian behavior of general quadratic gravity in the weak-field regime. By extending the Einstein-Hilbert action to include quadratic curvature terms as $\mathcal{L}\propto R-λC^2+μR^2$, the theory introduces two massive modes: a scalar mode and a ghost tensor mode. Using the post-Newtonian expansion method, we derive the explicit expressions for the metric for a general source up to 1.5PN order. Furthermore, for a point-mass source, we extend the solution to 2PN order and evaluate the effective Parameterized Post-Newtonian parameters $γ(r)$ and $β(r)$. The results show that deviations from General Relativity are exponentially suppressed. The theory has the feature $γ(r)\equiv 1$ when $m_R=m_W$, and to ensure that gravity remains attractive, we have $m_W>m_R/4$. The leading correction to $β(r)$ exhibiting a characteristic $\mathcal{O}(r \ln (r)e^{-mr})$ dependence. Based on the Solar System experiments, we derive preliminary constraints on the theory's parameters: $m_R,m_W\gtrsim23~\mathrm{AU}^{-1}$, corresponding to $λ\lesssim2.1\times10^{19}~\mathrm{m}^2$ and $μ\lesssim 7.1\times 10^{18}~\mathrm{m}^2$. This study provides a theoretical foundation for future tests of quadratic gravity using pulsar timing arrays, gravitational-wave observations, and laboratory-scale short-range gravity experiments.

gr-qc↗
arXiv2026

CrossGuard: Safeguarding MLLMs against Joint-Modal Implicit Malicious Attacks

Multimodal Large Language Models (MLLMs) achieve strong reasoning and perception capabilities but are increasingly vulnerable to jailbreak attacks. While existing work focuses on explicit attacks, where malicious content resides in a single modality, recent studies reveal implicit attacks, in which benign text and image inputs jointly express unsafe intent. Such joint-modal threats are difficult to detect and remain underexplored, largely due to the scarcity of high-quality implicit data. We propose ImpForge, an automated red-teaming pipeline that leverages reinforcement learning with tailored reward modules to generate diverse implicit samples across 14 domains. Building on this dataset, we further develop CrossGuard, an intent-aware safeguard providing robust and comprehensive defense against both explicit and implicit threats. Extensive experiments across safe and unsafe benchmarks, implicit and explicit attacks, and multiple out-of-domain settings demonstrate that CrossGuard significantly outperforms existing defenses, including advanced MLLMs and guardrails, achieving stronger security while maintaining high utility. This offers a balanced and practical solution for enhancing MLLM robustness against real-world multimodal threats. Our code is released: https://github.com/ZhangXu0963/CrossGuard.

cs.CR↗
arXiv2026

DualGuard: Dual-stream Large Language Model Watermarking Defense against Paraphrase and Spoofing Attack

With the rapid development of cloud-based services, large language models have become increasingly accessible through various web platforms. However, this accessibility has also led to growing risks of model abuse. LLM watermarking has emerged as an effective approach to mitigate such misuse and protect intellectual property. Existing watermarking algorithms, however, primarily focus on defending against paraphrase attacks while overlooking piggyback spoofing attacks, which can inject harmful content, compromise watermark reliability, and undermine trust in attribution. To address this limitation, we propose DualGuard, the first watermarking algorithm capable of defending against both paraphrase and spoofing attacks. DualGuard employs the adaptive dual-stream watermarking mechanism, in which two complementary watermark signals are dynamically injected based on the semantic content. This design enables DualGuard not only to detect but also to trace spoofing attacks, thereby ensuring reliable and trustworthy watermark detection. Extensive experiments conducted across multiple datasets and language models demonstrate that DualGuard achieves excellent detectability, robustness, traceability, and text quality, effectively advancing the state of LLM watermarking for real-world applications.

cs.CR↗
arXiv2026

MTRouter: Cost-Aware Multi-Turn LLM Routing with History-Model Joint Embeddings

Multi-turn, long-horizon tasks are increasingly common for large language models (LLMs), but solving them typically requires many sequential model invocations, accumulating substantial inference costs. Here, we study cost-aware multi-turn LLM routing: selecting which model to invoke at each turn from a model pool, given a fixed cost budget. We propose MTRouter, which encodes the interaction history and candidate models into joint history-model embeddings, and learns an outcome estimator from logged trajectories to predict turn-level model utility. Experiments show that MTRouter improves the performance-cost trade-off: on ScienceWorld, it surpasses GPT-5 while reducing total cost by 58.7%; on Humanity's Last Exam (HLE), it achieves competitive accuracy while reducing total cost by 43.4% relative to GPT-5, and these gains even carry over to held-out tasks. Further analyses reveal several mechanisms underlying its effectiveness: relative to prior multi-turn routers, MTRouter makes fewer model switches, is more tolerant to transient errors, and exhibits emergent specialization across models. Code: https://github.com/ZhangYiqun018/MTRouter

cs.CL↗
arXiv2026

DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence

We present a preview version of DeepSeek-V4 series, including two strong Mixture-of-Experts (MoE) language models -- DeepSeek-V4-Pro with 1.6T parameters (49B activated) and DeepSeek-V4-Flash with 284B parameters (13B activated) -- both supporting a context length of one million tokens. DeepSeek-V4 series incorporate several key upgrades in architecture and optimization: (1) a hybrid attention architecture that combines Compressed Sparse Attention (CSA) and Heavily Compressed Attention (HCA) to improve long-context efficiency; (2) Manifold-Constrained Hyper-Connections (mHC) that enhance conventional residual connections; (3) and the Muon optimizer for faster convergence and greater training stability. We pre-train both models on more than 32T diverse and high-quality tokens, followed by a comprehensive post-training pipeline that unlocks and further enhances their capabilities. DeepSeek-V4-Pro-Max, the maximum reasoning effort mode of DeepSeek-V4-Pro, redefines the state-of-the-art for open models, outperforming its predecessors in core tasks. Meanwhile, DeepSeek-V4 series are highly efficient in long-context scenarios. In the one-million-token context setting, DeepSeek-V4-Pro requires only 27% of single-token inference FLOPs and 10% of KV cache compared with DeepSeek-V3.2. This enables us to routinely support one-million-token contexts, thereby making long-horizon tasks and further test-time scaling more feasible. The model checkpoints are available at https://huggingface.co/collections/deepseek-ai/deepseek-v4.

cs.CL↗
arXiv2026

Toward Robust and Efficient ML-Based GPU Caching for Modern Inference

In modern GPU inference, cache efficiency remains a major bottleneck, and heuristic policies such as \textsc{LRU} can perform far worse than the offline optimum. Existing learning-based caching systems improve hit rates mainly through predictor design, but often follow learned predictions blindly, making performance unreliable when predictions are inaccurate. In contrast, emerging learning-augmented caching algorithms~\cite{pmlr-v80-lykouris18a,mitzenmacher2022algorithms} provide performance guarantees by carefully integrating predictions into caching policies, achieving both \emph{consistency} (near-optimality under perfect predictions) and \emph{robustness} (bounded worst-case performance under prediction errors). However, deployment remains challenging. A practical algorithm should satisfy strict time and space efficiency constraints, which some theoretical work overlooks, while also incurring low deployment overhead. We propose learning-augmented LRU, a deployment-oriented learning-augmented caching algorithm that guarantees \emph{1-consistency} and \emph{$O(k)$-robustness}, incurs low time and space overhead, and maintains strong compatibility. We further build a GPU cache, called \textsc{LCR}, on top of learning-augmented LRU to benefit from its theoretical guarantees and translate them into practical performance. In experiments, \textsc{LCR} reduces P99 time-to-first-token (TTFT) by up to 28.3\% on LLM workloads and increases throughput by up to 24.2\% on deep learning recommendation (DLRM) workloads. Even with poor predictions, performance degrades gracefully and remains close to \textsc{LRU}, demonstrating robustness with practical value.

cs.LG↗
arXiv2026

How is a gas sensor poisoned by volatile methylsiloxanes?

Volatile methyl siloxanes (VMSs), widely present in consumer and industrial products, have attracted increasing concerns due to their persistence, bioaccumulation behavior, and adverse health effects. Beyond their environmental implications, VMSs also pose operational challenges for sensing technologies because they readily decompose on sensing materials to form silicon-based compounds (e.g., silica and silane) that irreversibly impair sensing performance, a phenomenon commonly known as siloxane poisoning. Despite its prevalence, the mechanistic basis of this deactivation remains poorly understood. Herein, we present the first comprehensive theoretical study of siloxane-induced poisoning in catalytic gas sensors. Guided by our self-developed AI Agent, Digital Sensor Platform (DigSen), we first identify siloxane poisoning as a previously overlooked yet high-impact research direction. Using hexamethyldisiloxane (HMDS) as a model compound, we then conducted first-principles calculations to uncover decomposition pathways across noble metal surfaces. Strikingly, a descriptor-based microkinetic volcano model is developed to capture the trade-off between sensing activity and resistance to poisoning, enabling predictive identification of anti-poisoning candidates. These insights not only elucidate the origin of siloxane poisoning but also demonstrate how AI-driven discovery, mechanistic theory, and experiments can be integrated into a closed-loop framework for catalytic sensor design. More broadly, this AI-guided paradigm represents a generalizable strategy for materials digital discovery, offering a transferable methodology that extends well beyond siloxane systems to diverse classes of materials challenges.

physics.chem-ph↗
arXiv2026

LoViF 2026 Challenge on Real-World All-in-One Image Restoration: Methods and Results

This paper presents a review for the LoViF Challenge on Real-World All-in-One Image Restoration. The challenge aimed to advance research on real-world all-in-one image restoration under diverse real-world degradation conditions, including blur, low-light, haze, rain, and snow. It provided a unified benchmark to evaluate the robustness and generalization ability of restoration models across multiple degradation categories within a common framework. The competition attracted 124 registered participants and received 9 valid final submissions with corresponding fact sheets, significantly contributing to the progress of real-world all-in-one image restoration. This report provides a detailed analysis of the submitted methods and corresponding results, emphasizing recent progress in unified real-world image restoration. The analysis highlights effective approaches and establishes a benchmark for future research in real-world low-level vision.

cs.CV↗
arXiv2026

Rethinking Jailbreak Detection of Large Vision Language Models with Representational Contrastive Scoring

Large Vision-Language Models (LVLMs) are vulnerable to a growing array of multimodal jailbreak attacks, necessitating defenses that are both generalizable to novel threats and efficient for practical deployment. Many current strategies fall short, either targeting specific attack patterns, which limits generalization, or imposing high computational overhead. While lightweight anomaly-detection methods offer a promising direction, we find that their common one-class design tends to confuse unseen benign inputs with malicious ones, leading to unreliable over-rejection. To address this, we propose Representational Contrastive Scoring (RCS), a framework built on a key insight: the most potent safety signals reside within the LVLM's own internal representations. Our approach inspects the internal geometry of these representations, learning a lightweight projection to maximally separate benign and malicious inputs in safety-critical layers. This enables a simple yet powerful contrastive score that differentiates true malicious intent from mere distribution shift. Our instantiations, MCD (Mahalanobis Contrastive Detection) and KCD (K-nearest Contrastive Detection), achieve state-of-the-art performance on a challenging evaluation protocol designed to test generalization to unseen attack types. This work demonstrates that effective jailbreak detection can be achieved by applying simple, interpretable statistical methods to the internal representations, offering a practical path towards safer LVLM deployment.

cs.CR↗
arXiv2026

Spin State versus Potential of Zero Charge as Predictors of Density-Dependent Oxygen Reduction in M-N-C Electrocatalysts

Metal-site density strongly influences oxygen reduction activity and selectivity in M-N-C electrocatalysts, but the descriptors that predict these trends remain under debate. Here, we compare spin state and the potential of zero charge as predictors of density-dependent oxygen reduction behavior in Fe-N-C and Co-N-C catalysts. Using constrained-magnetization calculations combined with Landau analysis, we find that the ground-state magnetic moments vary only weakly across a broad range of metal-site densities, suggesting that magnetic descriptors alone cannot account for the pronounced performance changes. In contrast, explicit-solvent simulations reveal systematic density-dependent shifts in PZC, which alter the interfacial electric field and thereby modulate field-sensitive adsorption energetics of ORR intermediates. Incorporating these PZC shifts into a pH-field-coupled microkinetic model captures the density-dependent activity trends and reproduces the experimentally observed increase in two-electron selectivity at lower site densities under acidic conditions. Experimental PZC measurements further support the predicted trend. Together, these results show that PZC is a more effective predictor than spin state for density-dependent oxygen reduction activity and selectivity in M-N-C electrocatalysts.

cond-mat.mtrl-sci↗
arXiv2026

FuXi-TC: A generative framework integrating deep learning and physics-based models for improved tropical cyclone forecasts

Tropical cyclones (TCs) are among the most devastating natural hazards, yet their intensity remains notoriously difficult to predict. NWP models are constrained by both computational demands and intrinsic predictability, while state-of-the-art deep learning-based weather forecasting models tend to underestimate TC intensity due to biases in reanalysis-based training data. Here, we present FuXi-TC, a diffusion-based generative forecasting framework that combines the track prediction strength of the FuXi model with the intensity representation of NWP simulations. By conditioning a diffusion model on the large-scale forecasts of the global FuXi model, FuXi-TC effectively downscales and delivers higher-accuracy forecasts of fine-grained variable fields such as wind speed and precipitation. In evaluations across the 2024 Western North Pacific, our approach matches the TC intensity forecast skill of the operational ECMWF deterministic model while delivering superior precipitation forecasts. Meanwhile this is achieved with significantly higher inference speeds and lower computational costs. Moreover, FuXi-TC demonstrates robust zero-shot generalization directly when applied to North Atlantic hurricanes without any fine-tuning. When applied to the FuXi ensemble model, this framework effectively yields well-dispersed probabilistic forecasts and refines the ensemble intensity predictions.

physics.ao-ph↗
arXiv2026

Learning Sewing Patterns via Latent Flow Matching of Implicit Fields

Sewing patterns define the structural foundation of garments and are essential for applications such as fashion design, fabrication, and physical simulation. Despite progress in automated pattern generation, accurately modeling sewing patterns remains difficult due to the broad variability in panel geometry and seam arrangements. In this work, we introduce a sewing pattern modeling method based on an implicit representation. We represent each panel using a signed distance field that defines its boundary and an unsigned distance field that identifies seam endpoints, and encode these fields into a continuous latent space that enables differentiable meshing. A latent flow matching model learns distributions over panel combinations in this representation, and a stitching prediction module recovers seam relations from extracted edge segments. This formulation allows accurate modeling and generation of sewing patterns with complex structures. We further show that it can be used to estimate sewing patterns from images with improved accuracy relative to existing approaches, and supports applications such as pattern completion and refitting, providing a practical tool for digital fashion design.

cs.CV↗
arXiv2026

Enabling Ultra-Fast Cardiovascular Imaging Across Heterogeneous Clinical Environments with A Generalist Foundation Model and Multimodal Database

Multimodal cardiovascular magnetic resonance (CMR) imaging provides comprehensive and non-invasive insights into cardiovascular disease (CVD) diagnosis and underlying mechanisms. Despite decades of advancements, its widespread clinical adoption remains constrained by prolonged scan times, inconsistent image quality, and heterogeneity across medical environments. This underscores the urgent need for a generalist reconstruction foundation model for ultra-fast CMR imaging, one formulated for physics-constrained inverse problems in the sensor (k-space) domain, capable of adapting across diverse imaging scenarios and serving as the essential substrate for all downstream analyses. To enable this goal, we curate MMCMR-427K, the largest and most comprehensive multimodal CMR k-space database to date, comprising 427,465 multi-coil k-space data paired with structured metadata across 13 international centers, 12 CMR modalities, 15 scanners spanning four field strengths, and 17 CVD categories in populations across three continents. Building on this unprecedented resource, we introduce CardioMM, a generalist reconstruction foundation model capable of dynamically adapting to heterogeneous fast CMR imaging scenarios. CardioMM unifies semantic contextual understanding with physics-informed data consistency to deliver robust reconstructions across varied scanners, protocols, and patient presentations. Comprehensive evaluations demonstrate that CardioMM achieves state-of-the-art performance across internal centers and exhibits strong zero-shot generalization to unseen external settings. Importantly, CardioMM supports acceleration up to 24x, providing the first evidence that such extreme acquisition speed can preserve key cardiac phenotypes, quantitative myocardial biomarkers, and diagnostic image quality without compromising clinical integrity.

eess.IV↗
arXiv2026

Hamiltonian Constraints on Spontaneous Lorentz Symmetry Breaking in the Bumblebee Model

This study demonstrates that the common practice of determining spontaneous Lorentz violation via the minimum of a Lagrangian potential is generally incorrect. By analyzing the Hamiltonian structure and constraints of vector fields, we show that the true vacuum must be derived from the Hamiltonian density. We prove that the standard quadratic potential cannot consistently generate a vacuum expectation value (VEV), identifying a cubic potential as the simplest viable alternative. Furthermore, we prove that smooth potentials only support stable timelike or lightlike VEVs. These conclusions extend to higher-rank tensor fields and impose rigorous consistency constraints on higher-rank tensor fields and Lorentz-violating effective field theories.

hep-ph↗
arXiv2026

CROP: Conservative Reward for Model-based Offline Policy Optimization

Offline reinforcement learning (RL) aims to optimize a policy using collected data without online interactions. Model-based approaches are particularly appealing for addressing offline RL challenges because of their capability to mitigate the limitations of data coverage through data generation using models. Nonetheless, a prevalent issue in offline RL is the overestimation caused by distribution shift. This study proposes a novel model-based offline RL algorithm named Conservative Reward for model-based Offline Policy optimization (CROP). CROP introduces a streamlined objective that concurrently minimizes estimation error and the rewards of random actions, thereby yielding a robustly conservative reward estimator. Theoretical analysis shows that the designed conservative reward mechanism leads to a conservative policy evaluation and mitigates distribution shift. Experiments showcase that with the simple modification to reward estimation, CROP can conservatively estimate the reward and achieve competitive performance with existing methods. The source code will be available after acceptance.

cs.LG↗