arXiv subjects

Hong Huang

Publications and source records attributed to Hong Huang.

At least 19 recordsLinked to original sources

arXiv2026

E8-QAT: Efficient 2-Bit Quantization-Aware Training with Implicit Lattice Codebooks

Low-bit quantization reduces the memory footprint of large language models, but preserving accuracy with efficient training remains challenging. Scalar quantization encodes weights independently, while explicit-codebook vector quantization introduces costly search during training. We propose E8-QAT, a 2-bit quantization-aware training framework that uses the $E_8$ lattice as an implicit eight-dimensional codebook. It jointly quantizes weight groups through arithmetic projection without storing or searching an explicit codebook during the training forward pass. A 2-bit core encoding and sparse extension preserve the selected lattice points exactly, adding less than 0.1\% to core index storage. On Qwen3.6-35B-A3B, E8-QAT matches the performance of the full-precision model in terms of the average score across the four tasks. E8-QAT improves over the state-of-the-art baseline by 2.03\% on IFEval and by 1.51\% on LiveCodeBench, under the same data, schedule, and bit width. The 2-bit representation reduces peak memory by 80\% and supports a $4\times$ larger batch. At each representation's largest feasible batch, E8-QAT reaches $1.97\times$ the decoding throughput of BF16.

cs.AI↗

arXiv2026

Fluid-Gen-Zero: Grounding Pretrained Video Generators in Physics without Training

We present Fluid-Gen-Zero, a training-free framework for physics-aware fluid-object interaction video generation that decouples physical reasoning from appearance synthesis. Our key insight is to delegate motion dynamics to a physics simulator while preserving the appearance modeling capacity of pretrained video generators. We bridge these two domains through a two-level agentic workflow: generation-time planning, where a vision-language model (VLM) agent interprets intent and the simulation rollout to organize generation clips, and latent-space guidance, which injects simulation signals into denoising through region-aware latent wrapping. This plug-and-play design is compatible with current video foundation models. We further introduce a benchmark for fluid-object interaction video generation. Across Tora (CogVideoX-based), VACE and WanMove (Wan-based), Fluid-Gen-Zero consistently improves simulation alignment, reducing object trajectory error by 26.7%-81.5% and fluid fEPE (fluid flow endpoint error) by 67.9%-84.0%, while largely preserving perceptual quality. In a human preference study, raters favor Fluid-Gen-Zero in 55.1%-74.4% of same-backbone comparisons across three backbones, and in 90.4%-94.2% of comparisons against simulation-based methods. Code and data will be released upon acceptance.

cs.CV↗

arXiv2026

Omni-Diffusion-Distill: Few-Step Distillation of Unified Multimodal Diffusion Large Language Models

Unified multimodal diffusion large language models (dLLMs) offer a single architecture for both image generation and multimodal understanding, but their iterative decoding requires tens to hundreds of forward passes. Existing few-step distillation methods largely focus on either image generation or text generation, making it unclear how to compress a fully discrete multimodal dLLM into a single efficient student while preserving both generation and understanding. We introduce Omni-Diffusion-Distill, a unified two-stage distillation framework that retains strong generation and understanding capabilities while substantially reducing the inference cost of a unified multimodal dLLM. Omni-Diffusion-Distill aligns the distillation of both generation and understanding, for both images and text, in the discrete token space. In the first stage, the student is trained to skip decoding steps by replaying cached teacher trajectories, and in the second stage the student is refined on intermediate states along its own rollouts. We further remedy two sources of degradation in unified distillation with a pairwise collision penalty that reduces repetition under parallel text decoding, and entropy-matched guidance that prevents entropy collapse caused by fitting the sharpened teacher distribution in image generation. Omni-Diffusion-Distill achieves state-of-the-art trade-offs between decoding efficiency and generation and understanding performance for multimodal dLLMs, reducing image generation from 128 to 8 decoding steps and multimodal understanding from 512 to 64, giving 18.2x and 21.2x wall-clock speedups. Under these budgets, it scores 0.828 on GenEval and 83.0 on DPG-Bench for text-to-image generation, while reaching GPT judge scores of 20.0 on MM-Vet and 57.2 on COCO captioning (twice the teacher's 28.4 at the same steps) for multimodal understanding.

cs.CV↗

arXiv2026

Removing Information Content Does Not Certify Tamper Resistance in Open-Weight Models

Does removing harmful information make open-weight models resistant to fine-tuning attacks? We show that mutual information at release alone cannot universally certify slow recovery. Function-preserving reparameterizations leave information unchanged while altering gradient-descent geometry, so an invariant certificate is bounded by the fastest reachable parameterization. We apply this principle to weight--data mutual information under training-data filtering and label--representation mutual information under capability removal. Training order can change recovery time at fixed weight--data information, while exact representation-level independence can preserve the entire parameter Jacobian. An explicit construction has both information quantities equal to zero and recovers in one gradient step. Controlled experiments illustrate order-dependent recovery and parameterization-dependent attack speed. These results identify the missing requirement for certification: constraints on attack dynamics beyond mutual information at release.

cs.LG↗

arXiv2026

Classification of compact manifolds with positive isotropic curvature

We show the following result: Let $(M,g_0)$ be a compact manifold of dimension $n\geq 12$ with positive isotropic curvature. Then $M$ is diffeomorphic to a spherical space form, or a quotient manifold of $\mathbb{S}^{n-1}\times \mathbb{R}$ by a cocompact discrete subgroup of the isometry group of the round cylinder $\mathbb{S}^{n-1}\times \mathbb{R}$, or a connected sum of a finite number of such manifolds. This extends previous works of Brendle and Chen-Tang-Zhu, and improves a work of Huang. The proof uses Ricci flow with surgery on compact orbifolds, with the help of the ambient isotopy uniqueness of closed tubular neighborhoods of an isolated singular point in an orbifold.

math.DG↗

arXiv2026

The Collaboration Tax: How Much LLM Multi-Agent Systems Pay to Coordinate

Multi-agent systems built from large language models are deployed widely, yet how much performance is lost when two LLMs must coordinate rather than act alone remains unclear. We formulate the collaboration tax as the team-decentralisation loss of a two-player cooperative game with private information, with two propositions characterising its sign and its equivalence to a max-superadditivity violation. We operationalise this definition on 32 solo-tractable tasks grouped by source of grounding friction and measure it on 11 models from 7 providers. The tax is structured along two no-exception axes: a category ordering across every model and a monotonic decrease with capability. The proximate mechanism is not a reasoning deficit but a four-stage conversational cascade in which agents make ungrounded claims, fail to query the partner, skip integrating both views, and accept the answer without re-derivation. The tax is mechanically predictable from conversation features and partly tractable: a prompt intervention targeting all four stages closes a substantial fraction of the gap, with the dominant bottleneck differing across categories. In heterogeneous pairs the tax is pulled toward the stronger partner rather than the additive midpoint, empirically realising the max-superadditivity violation predicted by our framework. Together these results recast collaboration in LLM systems as a measurable, predictable, and partly tractable cost.

cs.CL↗

arXiv2026

Learning Personalized Prompts for Healthcare Guidance

The rapid development of large language models (LLMs) has transformed many industries, including healthcare. In practice, hospitals and patients increasingly seek LLM-based systems capable of interpreting personal health records and providing healthcare guidance. However, existing approaches mainly rely on general medical knowledge and often fail to account for individual variability, limiting their ability to provide personalized guidance. To address this, we propose personalized prompt learning (PPL), a framework that learns individualized prompts to guide LLMs in generating personalized healthcare recommendations. PPL constructs initial personalized prompts by leveraging both self-informed patient information and peer-informed signals derived from clinically similar cases. These prompts are then refined using reinforcement learning (RL) to better align the generated responses with physician recommendations written for each patient. PPL operates with hard prompts, enabling seamless integration with proprietary LLMs without modifying the underlying models. We evaluate PPL on real-world obstetrics and gynecology data. The results show that our approach produces more personalized healthcare guidance and wins 97 out of 100 comparisons in expert evaluation, demonstrating its potential for broader healthcare applications. Our code is publicly available at https://github.com/CGCL-codes/PPL.

cs.CL↗

arXiv2026

The Cauchy problem for fully nonlinear parabolic systems on manifolds

We show the short time existence and uniqueness of solutions to the Cauchy problem for fully nonlinear systems of arbitrary even order on closed manifolds which are strongly parabolic at the initial values. The proof uses a linearization procedure and a fixed-point argument, and the key ingredient is the well known Schauder estimates for linear, strongly parabolic systems due to Solonnikov.

math.DG↗

arXiv2026

Characterizing the Evolution of Tropical Cyclone Thermal Structure: A Tropical Cyclone Thermal Phase Space

Current understandings of the tropical cyclone (TC) warm core primarily relies on statistical averages from soundings and satellite products, which provide a climatological thermal state but cannot adequately characterize the continuous three-dimensional evolution of the warm core throughout the TC life cycle. In this study, Empirical Orthogonal Function (EOF) analysis was applied to three-dimensional temperature anomalies of Western North Pacific (WNP) TCs derived from the ERA5 reanalysis dataset for the period 1979-2024. We found that the three leading EOF modes effectively capture the primary characteristics of the TC thermal structure: the first mode represents the typical warm core structure; the second mode characterizes vertical baroclinicity; and the third pattern captures the horizontal asymmetry of the temperature anomalies. Using the first three principal components (PCs), we established a three-dimensional Cartesian coordinate system. Within this framework, two phase parameters define a thermal phase space to visualize the TC's three-dimensional thermal structure: one diagnoses the vertical barotropic or baroclinic structure, and the other quantifies the horizontal asymmetry of the thermal field. The results demonstrate that the trajectory in the phase diagram effectively captures the observed intensity changes and thermal structural evolution. The EOF-based phase diagrams offer promising insights and provide a novel, objective tool for analyzing and diagnosing structural thermodynamic characteristics and intensity evolution by quantifying key thermal features and their dynamic relationships with TC intensity. This capability thereby holds substantial potential for advancing both theoretical understanding and operational forecasting of TCs.

physics.ao-ph↗

arXiv2026

ToolRec: Calibrated Preference Alignment for Query Recommendation in On-Device Assistants

Large Language Models (LLMs) have significantly advanced generative query recommendation. However, while alignment is crucial for tailoring LLMs to human preferences, existing alignment methods primarily focus on standard chatbot scenarios, falling short in on-device intelligent assistants where users predominantly expect the rapid invocation of system-level tools. Moreover, directly aligning LLMs with real-world click logs introduces severe noise due to varying user activity levels and the failure to emphasize execution-oriented queries. To address these challenges, we propose ToolRec, a calibrated preference alignment framework tailored for on-device query recommendation. To ground query recommendation with executable tools, we first construct SysToolKit, a comprehensive repository of 708 system tools, paired with a context-aware tool retrieval mechanism to ensure that the extracted tools closely match the user's intent. A dual-level calibration mechanism is then proposed to refine raw click data, effectively mitigating user behavioral noise by calibrating signals based on user activity (user-level), while simultaneously up-weighting click signals on tool-invoking queries (system-level). Guided by these refined preference signals, we then align the model using a sample-level weighted Kahneman-Tversky Optimization (KTO). Extensive online A/B tests on our mobile assistant platform OPPO Xiaobu, which has over 150 million monthly active users, demonstrate that ToolRec can significantly improve Click-Through Rate (CTR) and total click volume over strong baselines while maintaining high query relevance.

cs.IR↗

arXiv2026

InitGen: Candidate Generation for Interaction Initiation in Intelligent Assistants

Interaction initiation refers to presenting multiple candidate queries when a user opens an intelligent assistant before expressing any intent for the current session. In production, candidate generation incorporates dynamic context and produces all candidates within a strict latency budget. Learning from user feedback is also difficult since the generator usually produces more candidates than are finally displayed. After downstream filtering and ranking, only a subset is exposed to users, so the observed feedback is partial and cannot be reliably assigned to individual queries. We present InitGen, a framework for candidate generation that is deployed in the interaction initiation pipeline of OPPO's Xiaobu Assistant. InitGen generates a set of candidate queries jointly and aligns the generated set with user feedback through weighted preference optimization. The sample weights are derived from user activity and downstream ranking scores. The activity weight reduces the dominance of highly active users during training, while the ranking score is used as a practical estimate of the reliability of the observed feedback. InitGen also uses a rolling window update strategy to incorporate recent interaction data into periodic model updates. In an online A/B test against a strong production baseline, InitGen improves the click-through rate from 0.95% to 1.61%, corresponding to a relative improvement of 69.1%, and increases query exposure by 17.9% under the same traffic allocation. InitGen generates the complete candidate set within 180 ms and has been fully deployed in OPPO's Xiaobu Assistant, which serves over 150 million monthly active users.

cs.IR↗

arXiv2026

Distribution-Specific Curvature Control with Finite-Sample Guarantees for Open-Weight Safety

A short fine-tuning run can undo the safety guards of an open-weight model---retraining a refusal-trained assistant to aid weapons development or produce hate speech. Preventing such harmful fine-tuning while retaining benign adaptability remains difficult: the only prior method with an explicit curvature certificate, spectral deformation, inflates curvature globally and thereby obstructs benign adaptation along with harmful adaptation. We propose HarmAlign, which applies function-preserving spectral deformation along a estimated contrastive activation subspace. We derive finite-sample bounds for the estimated subspace energy and the resulting local harmful-distribution curvature lower bound. A stability--progress dichotomy for constant-step gradient descent turns the certified curvature into conditional convergence-rate control. Empirically, within a fixed-architecture, finite-budget first-order threat model, HarmAlign blocks direct fine-tuning and three data- or objective-adaptive attacks across a hazardous-knowledge relearning setting and a harmful-assistance fine-tuning setting, while the protected benign tasks remain trainable. The block persists across the tested first-order optimizer variants over every attack checkpoint, and under out-of-distribution harmful fine-tuning, and it extends to important cases in our threat model: accidental safety degradation and emergent misalignment.

cs.LG↗

arXiv2026

Prima.cpp: Fast 30-70B LLM Inference on Heterogeneous and Low-Resource Home Clusters

On-device inference offers privacy, offline use, and instant response, but consumer hardware restricts large language models (LLMs) to low throughput and capability. To overcome this challenge, we present prima.cpp, a distributed on-device inference system that runs 30-70B LLMs on consumer home clusters with mixed CPUs/GPUs, insufficient RAM/VRAM, slow disks, Wi-Fi links, and heterogeneous OSs. We introduce pipelined-ring parallelism (PRP) to overlap disk I/O with compute and communication, and address the prefetch-release conflict in mmap-based offloading. We further propose Halda, a heterogeneity-aware scheduler that co-optimizes per-device CPU/GPU workloads and device selection under RAM/VRAM constraints. On four consumer home devices, a 70B model reaches 674 ms/token TPOT with <6% memory pressure, and a 32B model with speculative decoding achieves 26 tokens/s. Compared with llama.cpp, exo, and dllama, our proposed prima.cpp achieves 5-17 times lower TPOT, supports fine-grained model sizes from 8B to 70B, ensures broader cross-OS and quantization compatibility, and remains OOM-free, while also being Wi-Fi tolerant, privacy-preserving, and hardware-independent. The code is available at https://github.com/OpenCPIL/prima.cpp.

cs.DC↗

arXiv2026

Keep It in Mind: User Centric Continual Spatial Intelligence Reasoning in Egocentric Video Streams

We introduce UCS-Bench, a dataset spanning 170+ hours of egocentric visual observations with 8.1K+ timestamped questions for diagnosing User-Centric Continual Spatial intelligence in egocentric video streams. UCS-Bench targets a new problem that emphasizes dynamic spatial reasoning, long-term memory, and their alignment with users' real-time locations. We propose DirectMe, a framework that incrementally constructs and maintains a structured spatial memory from streaming egocentric observations. DirectMe enables robust tracking and recall of object locations, all relative to the user's movement over time. By tightly coupling visual perception with memory updates and spatial reasoning, our approach supports long-horizon queries that require recalling interactions, resolving viewpoint-induced ambiguities, and adapting to dynamic scenes. Our experiments show that DirectMe significantly improves the spatial reasoning of leading multimodal LLMs; it also surpasses many spatially aware and long-form streaming video models. We hope our benchmark and solution will advance spatial intelligence research for egocentric AI assistants. Data and code are available at https://github.com/cocowy1/UCS-Bench.

cs.CV↗

arXiv2026

Efficient Code Analysis via Graph Representation Learning-Guided Large Language Models

Large Language Models (LLMs) have significantly advanced code analysis tasks, yet they struggle to detect malicious behaviors fragmented across files, whose intricate dependencies easily get lost in the vast amount of benign code. We therefore propose a graph-centric attention acquisition pipeline that enhances LLMs' ability to localize malicious behavior. The approach parses a project into a code graph, uses an LLM to encode nodes with semantic and structural signals, and trains a Graph Neural Network (GNN) under sparse supervision. The GNN performs an initial detection, and by interpreting these predictions, identifies key code sections that are most likely to contain malicious behavior. These influential regions are then used to guide the LLM's attention for in-depth analysis. This strategy significantly reduces interference from irrelevant context while maintaining low annotation costs. Extensive experiments show that the method consistently outperforms existing approaches on multiple public and custom datasets, highlighting its potential for practical deployment in software security scenarios.

cs.SE↗

arXiv2026

Hy-MT2: A Family of Fast, Efficient and Powerful Multilingual Translation Models in the Wild

Hy-MT2 is a family of fast-thinking multilingual translation models designed for complex real-world scenarios. It includes three model sizes: 1.8B, 7B, and 30B-A3B (MoE), all of which support translation among 33 languages and effectively follow translation instructions in multiple languages. Multi-dimensional evaluations show that Hy-MT2 delivers outstanding performance across general, real-world business, domain-specific, and instruction-following translation tasks. The 7B and 30B models outperform open-source models such as DeepSeek-V4-Pro and Kimi K2.6 in fast-thinking mode, while the lightweight 1.8B model also surpasses mainstream commercial APIs from providers such as Microsoft and Doubao overall. Moreover, when paired with AngelSlim's 1.25-bit extreme quantization for on-device deployment, the lightweight 1.8B model requires only 440 MB of storage and achieves a 1.5x inference speedup.

cs.CL↗

arXiv2026

Exploring the AI Obedience: Why is Generating a Pure Color Image Harder than CyberPunk?

Recent advances in generative AI have shown human-level performance in complex content creation. However, we identify a "Paradox of Simplicity": models that can render complex scenes often fail at trivial, low-entropy tasks, such as generating a uniform pure color image. We argue this is a systemic failure related to uncontrollable emergent abilities. As models scale, strong priors for aesthetics and complexity override deterministic simplicity, creating an "aesthetic bias" that hinders the model's transition from data simulation to true intellectual abstraction. To better investigate this problem, we formalize the concept of AI Obedience, a hierarchical framework that grades a model's ability to transition from probabilistic approximation to pixel-level determinism (Levels 1 to 5). We introduce Violin, the first systematic benchmark designed to evaluate Level 4 Obedience through three deterministic tasks: color purity, image masking, and geometric shape generation. Using Violin, we evaluate several state-of-the-art models and reveal that closed-source models generally outperform open-source ones in deterministic precision. Interestingly, performance on our benchmark correlates with the benchmark in natural image generation. Our work provides a foundational framework and tools for achieving better alignment between human instructions and model outputs.

cs.CV↗

arXiv2026

Covariant Spinor Formalism for Multipole Expanded Form Factor

We present a systematic technique for constructing Lorentz covariant orbital-spin ($LS$) bases for matrix elements of local operators and the associated form factors, thereby extending the traditional multipole expansion to a Lorentz covariant formalism. In the spinor-helicity formalism, matrix elements of local operators for spin-$j$ particles can be treated as several massive 3-point scattering amplitudes, each of which can be further decomposed into different $LS$ partial wave amplitudes. We obtain explicit complete and linearly independent $LS$ amplitude bases for scalar, vector, and rank 2 tensor form factor of particles with spin-$\frac{1}{2}$, $1$, and $\frac{3}{2}$. In the Breit frame, it recovers the traditional multipole expansion expression, and we show the explicit equivalence among the traditional multipole expansion, canonical $LS$ expansion, and the $\mathrm{SO}(3)$ Zemach tensor expansion. Finally noting covariant structures built from the relativistic external wave functions and momenta of the initial and final state particles, we give a universal construction formula for form factor of arbitrary Lorentz tensor operators for arbitrary external spin particles.

hep-ph↗