arXiv subjects

Li Li

Publications and source records attributed to Li Li.

At least 109 records · Page 6Linked to original sources

arXiv2026

ERPO: Token-Level Entropy-Regulated Policy Optimization for Large Reasoning Models

Reinforcement learning from verifiable rewards has significantly advanced the reasoning capabilities of large language models. However, Group Relative Policy Optimization (GRPO) typically assigns a uniform, sequence-level advantage to all tokens, thereby overlooking the intrinsic information heterogeneity along reasoning chains. We show that this coarse-grained credit assignment leads to premature entropy collapse and encourages the model to generate redundant, low-quality reasoning paths. Through systematic empirical analysis, we identify Critical Decision Pivots (CDPs): transient high-entropy states where the policy's trajectory is most sensitive to perturbations. These pivots represent the "forks in the road" where effective multi-path exploration is most crucial yet often suppressed by uniform advantage signals. Building on these insights, we propose Entropy-Regulated Policy Optimization (ERPO), which transitions the optimization focus from coarse sequences to fine-grained token dynamics. ERPO introduces three synergistic components: (i) Entropy-aware Gating, which adaptively amplifies exploration at CDPs to facilitate diverse path discovery; (ii) Bucket-based Implicit Normalization, which mitigates difficulty bias by aligning token progress windows; and (iii) Result-anchored Advantage Synthesis, which re-weights token-level signals via outcome-driven anchors. Extensive experiments on competitive mathematical benchmarks demonstrate that ERPO significantly outperforms GRPO. Notably, ERPO not only boosts reasoning accuracy but also yields significantly more concise and robust derivation paths, while achieving performance comparable to large models with orders of magnitude more parameters.

cs.LG↗

arXiv2026

Accurate Precipitation Forecast by Efficiently Learning from Massive Atmospheric Variables and Unbalanced Distribution

Short-term (0-24 hours) precipitation forecasting is highly valuable to socioeconomic activities and public safety. However, the highly complex evolution patterns of precipitation events, the extreme imbalance between precipitation and non-precipitation samples, and the inability of existing models to efficiently and effectively utilize large volumes of multi-source atmospheric observation data hinder improvements in precipitation forecasting accuracy and computational efficiency. To address the above challenges, this study developed a novel forecasting model capable of effectively and efficiently utilizing massive atmospheric observations by automatically extracting and iteratively predicting the latent features strongly associated with precipitation evolution. Furthermore, this study introduces a 'WMCE' loss function, designed to accurately discriminate extremely scarce precipitation events while precisely predicting their intensity values. Extensive experiments on two datasets demonstrate that our proposed model substantially and consistently outperforms all prevalent baselines in both accuracy and efficiency. Moreover, the proposed forecasting model substantially lowers the computational cost required to obtain valuable predictions compared to existing approaches, thereby positioning it as a milestone for efficient and practical precipitation forecasting.

cs.LG↗

arXiv2026

HillInfer: Efficient Long-Context LLM Inference on the Edge with Hierarchical KV Eviction using SmartSSD

Deploying Large Language Models (LLMs) on memory-constrained AI Personal Computers (AIPCs) enables low-latency, privacy-preserving inference, but long-context generation is fundamentally bottlenecked by the linearly growing Key-Value (KV) cache. While dynamic KV eviction mitigates this memory wall, existing offloading strategies either trigger crippling PCIe I/O bottlenecks on standard SSDs or suffer from FPGA resource exhaustion by forcing compute-intensive exact attention on a single, weak Computational Storage Drive (CSD). In this paper, we propose HillInfer, a CSD-assisted KV eviction framework that introduces a paradigm shift: offloading strictly lightweight token importance evaluation to a single CSD (e.g., SmartSSD) on AIPCs. To fully capitalize on this lightweight offloading strategy, HillInfer orchestrates a Hierarchical KV Cache Manager (HKM) that leverages temporal locality and dynamic token hit rates to physically partition cache pools, thereby eliminating cross-device I/O thrashing. Additionally, we design an Adaptive Prefetch-based Pipeline (APP) that adaptively balances the evaluation workload between the host CPU and the SmartSSD, effectively masking the heterogeneous straggler effect. Finally, we introduce a CSD-based Evaluation Configuration (CEC) to enable resource-efficient near-data processing on the FPGA. Extensive experiments on a commodity AIPC demonstrate that HillInfer achieves up to an 8.56$\times$ speedup over state-of-the-art baselines, delivering low-latency, I/O-efficient long-context inference without sacrificing model accuracy.

cs.AR↗

arXiv2026

Three Creates All: You Only Sample 3 Steps

Diffusion models deliver high-fidelity generation but remain slow at inference time due to many sequential network evaluations. We find that standard timestep conditioning becomes a key bottleneck for few-step sampling. Motivated by layer-dependent denoising dynamics, we propose Multi-layer Time Embedding Optimization (MTEO), which freeze the pretrained diffusion backbone and distill a small set of step-wise, layer-wise time embeddings from reference trajectories. MTEO is plug-and-play with existing ODE solvers, adds no inference-time overhead, and trains only a tiny fraction of parameters. Extensive experiments across diverse datasets and backbones show state-of-the-art performance in the few-step sampling and substantially narrow the gap between distillation-based and lightweight methods. Code will be available.

cs.LG↗

arXiv2026

Universally Diverging Grüneisen Ratio of Holographic Quantum Criticality

Quantum criticality is a hallmark of strongly correlated electron systems, as seen in heavy-fermion materials and high-temperature superconductors. Holographic duality provides a powerful framework to investigate these systems by translating them into weakly coupled classical gravity living in one higher dimension. Here, we harness this approach to study a field-induced quantum critical point with dynamical exponent $z=3$ in Einstein-Maxwell-Chern-Simons theory. Our analysis of its thermodynamic properties reveals a new universality class. Notably, we identify a diverging Grüneisen ratio with universal scaling $\sim T^{-2/3}$, a behavior that closely mirrors recent experiments on the heavy-fermion material CeRh$_6$Ge$_4$. These findings advance our understanding of metallic quantum criticality and highlight the potential of holographic duality as a tool for studying correlated quantum matters.

cond-mat.str-el↗

arXiv2026

Demonstration of High-Fidelity Gates in a Strongly Anharmonic with Long-Coherence C-Shunt Flux Qubit

We demonstrate high-fidelity single-qubit gates on a C-shunt flux qubit that simultaneously combines a large anharmonicity ($\mathcal{A}/2π=848~\mathrm{MHz}$) with long relaxation time ($T_1 = 23~μ\text{s}$). The large anharmonicity significantly suppresses leakage to higher energy levels, enabling fast and precise microwave control. Using DRAG pulses and randomized benchmarking, the qubit achieves gate fidelities exceeding 99.9\%, highlighting the capability of C-shunt flux qubits for robust and high-performance quantum operations. These results establish them as a promising platform for scalable quantum information processing.

quant-ph↗

arXiv2026

An FPGA Implementation of Displacement Vector Search for Intra Pattern Copy in JPEG XS

Recently, progress has been made on the Intra Pattern Copy (IPC) tool for JPEG XS, an image compression standard designed for low-latency and low-complexity coding. IPC performs wavelet-domain intra compensation predictions to reduce spatial redundancy in screen content. A key module of IPC is the displacement vector (DV) search, which aims to solve the optimal prediction reference offset. However, the DV search process is computationally intensive, posing challenges for practical hardware deployment. In this paper, we propose an efficient pipelined FPGA architecture design for the DV search module to promote the practical deployment of IPC. Optimized memory organization, which leverages the IPC computational characteristics and data inherent reuse patterns, is further introduced to enhance the performance. Experimental results show that our proposed architecture achieves a throughput of 38.3 Mpixels/s with a power consumption of 277 mW, demonstrating its feasibility for practical hardware implementation in IPC and other predictive coding tools, and providing a promising foundation for ASIC deployment.

cs.AR↗

arXiv2026

Spatially-Augmented Sequence-to-Sequence Neural Diarization for Meetings

This paper proposes a Spatially-Augmented Sequence-to-Sequence Neural Diarization (SA-S2SND) framework, which integrates direction-of-arrival (DOA) cues estimated by SRP-DNN into the S2SND backbone. A two-stage training strategy is adopted: the model is first trained with single-channel audio and DOA features, and then further optimized with multi-channel inputs under DOA guidance. In addition, a simulated DOA generation scheme is introduced to alleviate dependence on matched multi-channel corpora. On the AliMeeting dataset, SA-S2SND consistently outperform the S2SND baseline, achieving a 7.4% relative DER reduction in the offline mode and over 19% improvement when combined with channel attention. These results demonstrate that spatial cues are highly complementary to cross-channel modeling, yielding good performance in both online and offline settings.

eess.AS↗

arXiv2026

TSEmbed: Unlocking Task Scaling in Universal Multimodal Embeddings

Despite the exceptional reasoning capabilities of Multimodal Large Language Models (MLLMs), their adaptation into universal embedding models is significantly impeded by task conflict. To address this, we propose TSEmbed, a universal multimodal embedding framework that synergizes Mixture-of-Experts (MoE) with Low-Rank Adaptation (LoRA) to explicitly disentangle conflicting task objectives. Moreover, we introduce Expert-Aware Negative Sampling (EANS), a novel strategy that leverages expert routing distributions as an intrinsic proxy for semantic similarity. By dynamically prioritizing informative hard negatives that share expert activation patterns with the query, EANS effectively sharpens the model's discriminative power and refines embedding boundaries. To ensure training stability, we further devise a two-stage learning paradigm that solidifies expert specialization before optimizing representations via EANS. TSEmbed achieves state-of-the-art performance on both the Massive Multimodal Embedding Benchmark (MMEB) and real-world industrial production datasets, laying a foundation for task-level scaling in universal multimodal embeddings.

cs.CL↗

arXiv2026

EchoGen: Generating Visual Echoes in Any Scene via Feed-Forward Subject-Driven Auto-Regressive Model

Subject-driven generation is a critical task in creative AI; yet current state-of-the-art methods present a stark trade-off. They either rely on computationally expensive, per-subject fine-tuning, sacrificing efficiency and zero-shot capability, or employ feed-forward architectures built on diffusion models, which are inherently plagued by slow inference speeds. Visual Auto-Regressive (VAR) models are renowned for their rapid sampling speeds and strong generative quality, making them an ideal yet underexplored foundation for resolving this tension. To bridge this gap, we introduce EchoGen, a pioneering framework that empowers VAR models with subject-driven generation capabilities. The core design of EchoGen is an effective dual-path injection strategy that disentangles a subject's high-level semantic identity from its low-level fine-grained details, enabling enhanced controllability and fidelity. We employ a semantic encoder to extract the subject's abstract identity, which is injected through decoupled cross-attention to guide the overall composition. Concurrently, a content encoder captures intricate visual details, which are integrated via a multi-modal attention mechanism to ensure high-fidelity texture and structural preservation. To the best of our knowledge, EchoGen is the first feed-forward subject-driven framework built upon VAR models. Both quantitative and qualitative results substantiate our design, demonstrating that EchoGen achieves subject fidelity and image quality comparable to state-of-the-art diffusion-based methods with significantly lower sampling latency. The code and models are publicly available at https://github.com/drx-code/EchoGen.

cs.CV↗

arXiv2026

Quantum Corrections to $η/s$ from JT Gravity

We revisit the computation of the shear viscosity to entropy ratio $η/s$ at finite chemical potential in a holographic model that takes into account the quantum fluctuations in the IR region of near-extremal black branes. Such quantum corrections can be computed from JT gravity and generate non-trivial temperature dependence for $η/s$, which deviates from the universal $1/4π$ result. In the semi-classical regime, $η/s$ attains a minimum which is below the KSS bound, generated by the presence of the quantum effects. In the quantum regime at lower temperatures, $η/s$ increases and is well above the KSS bound. We also compare the shear viscosity to the quantum-corrected absorption cross-section of near-extremal black holes, and find agreement.

hep-th↗

arXiv2026

A Survey on Federated Fine-tuning of Large Language Models

Large Language Models (LLMs) have demonstrated impressive success across various tasks. Integrating LLMs with Federated Learning (FL), a paradigm known as FedLLM, offers a promising avenue for collaborative model adaptation while preserving data privacy. This survey provides a systematic and comprehensive review of FedLLM. We begin by tracing the historical development of both LLMs and FL, summarizing relevant prior research to set the context. Subsequently, we delve into an in-depth analysis of the fundamental challenges inherent in deploying FedLLM. Addressing these challenges often requires efficient adaptation strategies; therefore, we conduct an extensive examination of existing Parameter-Efficient Fine-tuning (PEFT) methods and explore their applicability within the FL framework. To rigorously evaluate the performance of FedLLM, we undertake a thorough review of existing fine-tuning datasets and evaluation benchmarks. Furthermore, we discuss FedLLM's diverse real-world applications across multiple domains. Finally, we identify critical open challenges and outline promising research directions to foster future advancements in FedLLM. This survey aims to serve as a foundational resource for researchers and practitioners, offering valuable insights into the rapidly evolving landscape of federated fine-tuning for LLMs. It also establishes a roadmap for future innovations in privacy-preserving AI. We actively maintain a \href{https://github.com/Clin0212/Awesome-Federated-LLM-Learning}{GitHub repo} to track cutting-edge advancements in this field.

cs.LG↗

arXiv2026

Agent Banana: High-Fidelity Image Editing with Agentic Thinking and Tooling

We study instruction-based image editing under professional workflows and identify three persistent challenges: (i) editors often over-edit, modifying content beyond the user's intent; (ii) existing models are largely single-turn, while multi-turn edits can alter object faithfulness; and (iii) evaluation at around 1K resolution is misaligned with real workflows that often operate on ultra high-definition images (e.g., 4K). We propose Agent Banana, a hierarchical agentic planner-executor framework for high-fidelity, object-aware, deliberative editing. Agent Banana introduces two key mechanisms: (1) Context Folding, which compresses long interaction histories into structured memory for stable long-horizon control; and (2) Image Layer Decomposition, which performs localized layer-based edits to preserve non-target regions while enabling native-resolution outputs. To support rigorous evaluation, we build HDD-Bench, a high-definition, dialogue-based benchmark featuring verifiable stepwise targets and native 4K images (11.8M pixels) for diagnosing long-horizon failures. On HDD-Bench, Agent Banana achieves the best multi-turn consistency and background fidelity (e.g., IC 0.871, SSIM-OM 0.84, LPIPS-OM 0.12) while remaining competitive on instruction following, and also attains strong performance on standard single-turn editing benchmarks. We hope this work advances reliable, professional-grade agentic image editing and its integration into real workflows.

cs.CV↗

arXiv2026

CoAct-1: Computer-using Multi-Agent System with Coding Actions

Autonomous agents that operate computers via Graphical User Interfaces (GUIs) often struggle with efficiency and reliability on complex, long-horizon tasks. While augmenting these agents with planners can improve task decomposition, they remain constrained by the inherent limitations of performing all actions through GUI manipulation, leading to brittleness and inefficiency. In this work, we introduce a more robust and flexible paradigm: enabling agents to use coding as a enhanced action. We present CoAct-1, a novel multi-agent system that synergistically combines GUI-based control with direct programmatic execution. CoAct-1 features an Orchestrator that dynamically delegates subtasks to either a conventional GUI Operator or a specialized Programmer agent, which can write and execute Python or Bash scripts. This hybrid approach allows the agent to bypass inefficient GUI action sequences for tasks like file management and data processing, while still leveraging visual interaction when necessary. We evaluate our system on the challenging OSWorld benchmark, where CoAct-1 achieves a new state-of-the-art success rate of 60.76%, significantly outperforming prior methods. Furthermore, our approach dramatically improves efficiency, reducing the average number of steps required to complete a task to just 10.15, compared to 15 for leading GUI agents. Our results demonstrate that integrating coding as a core action provides a more powerful, efficient, and scalable path toward generalized computer automation.

cs.CL↗

arXiv2026

M3OOD: Automatic Selection of Multimodal OOD Detectors

Out-of-distribution (OOD) robustness is a critical challenge for modern machine learning systems, particularly as they increasingly operate in multimodal settings involving inputs like video, audio, and sensor data. Currently, many OOD detection methods have been proposed, each with different designs targeting various distribution shifts. A single OOD detector may not prevail across all the scenarios; therefore, how can we automatically select an ideal OOD detection model for different distribution shifts? Due to the inherent unsupervised nature of the OOD detection task, it is difficult to predict model performance and find a universally Best model. Also, systematically comparing models on the new unseen data is costly or even impractical. To address this challenge, we introduce M3OOD, a meta-learning-based framework for OOD detector selection in multimodal settings. Meta learning offers a solution by learning from historical model behaviors, enabling rapid adaptation to new data distribution shifts with minimal supervision. Our approach combines multimodal embeddings with handcrafted meta-features that capture distributional and cross-modal characteristics to represent datasets. By leveraging historical performance across diverse multimodal benchmarks, M3OOD can recommend suitable detectors for a new data distribution shift. Experimental evaluation demonstrates that M3OOD consistently outperforms 10 competitive baselines across 12 test scenarios with minimal computational overhead.

cs.LG↗

arXiv2026

AXLearn: Modular, Hardware-Agnostic Large Model Training

AXLearn is a production system which facilitates scalable and high-performance training of large deep learning models. Compared to other state-of-art deep learning systems, AXLearn has a unique focus on modularity and support for hardware-agnostic training. AXLearn's internal interfaces between software components follow strict encapsulation, allowing different components to be assembled to facilitate rapid model development and experimentation on different hardware infrastructure. AXLearn maintains constant complexity as we scale the components in the system, compared to linear or quadratic complexity in state-of-the-art training systems. This allows integrating features such as Rotary Position Embeddings (RoPE) into AXLearn across hundred of modules with just 10 lines of code, compared to hundreds as required in other systems. At the same time, AXLearn maintains equivalent performance compared to state-of-the-art training systems. Finally, we share our experience in the development and operation of AXLearn at Apple.

cs.LG↗

arXiv2026

Floe: Federated Specialization for Real-Time LLM-SLM Inference

Deploying large language models (LLMs) in real-time systems remains challenging due to their substantial computational demands and privacy concerns. We propose Floe, a hybrid federated learning framework designed for latency-sensitive, resource-constrained environments. Floe combines a cloud-based black-box LLM with lightweight small language models (SLMs) on edge devices to enable low-latency, privacy-preserving inference. Personal data and fine-tuning remain on-device, while the cloud LLM contributes general knowledge without exposing proprietary weights. A heterogeneity-aware LoRA adaptation strategy enables efficient edge deployment across diverse hardware, and a logit-level fusion mechanism enables real-time coordination between edge and cloud models. Extensive experiments demonstrate that Floe enhances user privacy and personalization. Moreover, it significantly improves model performance and reduces inference latency on edge devices under real-time constraints compared with baseline approaches.

cs.DC↗

arXiv2026

Fast Surrogate Learning for Multi-Objective UAV Placement in Motorway Intelligent Transportation System

We address multi-objective unmanned aerial vehicle (UAV) placement for motorway intelligent transportation systems, where deployments must balance coverage, link quality, and UAV count under geometric constraints. We construct a reproducible benchmark from highD motorway recordings with recording-level splits and generate Pareto-optimal labels via NSGA-II. A preference rule yields deployable targets while preserving multi-objective evaluation. We train fast surrogate models that map unordered vehicle positions to UAV count and continuous placements, using permutation-aware losses and constraint-regularized training across set-based and sequence-based architectures. The evaluation protocol combines Pareto quality metrics, success-rate curves, runtime benchmarks, and robustness studies, with uncertainty quantified by recording-level bootstrap. Results indicate that permutation-invariant set models provide the strongest coverage--SNR--count trade-off among learned predictors and approach NSGA-II quality while enabling real-time inference. Under shared budgets, they offer a more favorable success--latency trade-off than heuristic baselines. The benchmark, splits are released to support reproducible ITS deployment studies and to facilitate comparisons under shared operational budgets.

cs.NE↗