Search arXiv⌕ Search

arXiv subjects

Liming Liu

Publications and source records attributed to Liming Liu.

At least 19 recordsLinked to original sources
arXiv2026

Decode-Branch Transformers: Decoupling the Primary Prefill Path from Additional Decode Computation

As large language models serve ever more requests, cumulative inference cost is growing relative to the one-time cost of training. In typical serving, prompt prefill runs in parallel and is compute-bound, whereas autoregressive decode is sequential and memory-traffic-bound. Conventional width or depth scaling raises both costs together, since every added layer is evaluated in both phases and enlarges the weights read at each decode step. We instead ask whether additional learned computation can be allocated to continuation prediction while preserving prompt-wide primary computation and a single KV cache. We realize this with the Decode-Branch Transformer. Its primary path alone processes the prompt and writes the KV cache; the decode branch is omitted during prefill and activated only from the final prompt position onward, adding continuation computation without writing state or affecting the primary path. The paths share attention, MLP, and output matrices, using separate token embeddings with lightweight coupling. Grouped decode reuses loaded weight tiles and the primary KV cache across both paths, so the added arithmetic does not proportionally increase dominant memory traffic or decode latency. Across matched-token comparisons, Decode-Branch achieves lower validation loss across architectures and data settings. In MoE models, the primary and branch expert fan-outs become independent knobs for trading prompt cost, decode cost, and predictive quality. We study two expert-allocation regimes, holding prefill or decode computation fixed, and expose a prefill-decode-quality trade-off enabled by phase-specific expert allocation.

cs.AI↗
arXiv2026

Understanding and Overcoming Cross-modal Fusion Bias in Multimodal Anomaly Detection From A Fisher Information Perspective

Current advancements in Multimodal Anomaly Detection (MAD) are largely driven by enhancing multimodal fusion, particularly through the integration of RGB and Depth data for richer anomaly representation. However, less attention was devoted to analyzing the role of cross-modal fusion bias, a well-known challenge in multimodal learning, in MAD. This gap motivates a key question: can we overcome this bias to break the performance bottleneck of current work? In this paper, we first analyze the impact of cross-modal fusion bias in MAD via the Fisher Information Matrix. Then, grounded in these findings, we propose UCFB, a simple yet effective plug-and-play framework designed to mitigate cross-modal fusion bias in MAD. It achieves this by jointly employing Fisher-information-guided dynamic calibration to adjust modality-specific regularization weights and canonical similarity analysis to improve inter-modal interactions. Extensive experiments on the MVTec 3D-AD and Eyecandies datasets demonstrate that UCFB achieves consistent improvements in single-class, multi-class, and few-shot settings.

cs.CV↗
arXiv2026

Diagnose Before You Compress: Prediction-Independent Bottleneck Witness Refinement for LLM Serving Traces

Production LLM serving generates millions of diverse requests, making full-trace replay across serving configurations increasingly expensive. Existing trace reduction methods mainly preserve workload distributions or representative requests, but bottleneck-revealing workloads may be rare and non-representative. Moreover, evidence for one component cannot compensate for missing evidence in another, while using predicted bottlenecks as target truth creates circular evaluation. These limitations make it necessary to preserve evidence for every bottleneck component rather than rely on workload representativeness alone. We propose Bottleneck-Preserving Witnessing (BPW), a quality-constrained framework for compact and diagnostically reliable LLM serving replay suites. BPW first performs Workload Candidate Nomination using response-blind workload features and closed source-side measurements. This stage identifies workloads that may expose scheduler, prefill, decode, or KV-cache bottlenecks. Coverage-Priority Sequence Construction then organizes multi-component proposals as reusable hyperedges and prioritizes weak and uncovered dimensions. Finally, Bottleneck Truth Verification derives prediction-independent labels solely from direct target-system measurements. The verified results determine the earliest prefix satisfying the direct two-witness requirement for every component. Experiments on BurstGPT, ServeGen, and Mooncake show that BPW reaches the verified gate with a compact workload set and outperforms 16 policies, achieving relative improvements of 2.3% and 16.3% in Mean prefix Macro-F1 and WBRC-AUC, respectively. Stage-resolved and sensitivity analyses confirm the distinct contributions and local stability of its three stages. Our code is publicly available at https://github.com/llmllmllm/BPW

cs.AI↗
arXiv2026

daVinci-kernel: Co-Evolving Skill Selection, Summarization, and Utilization via RL for GPU Kernel Optimization

GPU kernel optimization represents a paradigm where functional correctness is assumed and execution efficiency is the objective. We present daVinci-kernel, a reinforcement learning framework that couples skill discovery with skill exploitation through a dynamically evolving skill library. daVinci-kernel jointly trains three agents sharing one LLM backbone: a Skill Selection Agent that retrieves relevant techniques via BM25 and LLM reranking, a Policy Agent that generates multi-turn CUDA/Triton kernels conditioned on selected skills, and a Skill Summary Agent that distills successful rollouts into reusable skills. Candidate skills are added only after execution-based verification confirms reproducible speedups. All three agents share a single LLM backbone, are initialized via a structured SFT cold start on diversity-filtered data, and are then jointly optimized end-to-end with multi-turn REINFORCE and per-agent advantage estimation. On KernelBench, daVinci-kernel-14B achieves 37.2%, 70.6%, and 32.2% on Level 1, Level 2, and Level 3 under the Fast$_1$ threshold, outperforming the strongest prior RL-trained model, Dr\. Kernel-14B.

cs.LG↗
arXiv2026

PREF-Gate: Provenance-Constrained Relational Evidence Fusion with Validation-Gated Selection for Graph Fraud Detection

Relational fraud detection can exploit both label-free graph context and label-derived neighborhood evidence, but these two information sources obey different validity conditions. In particular, neighborhood risk becomes invalid when a queried node's own label, or any validation or test label, enters its construction. We formulate this issue as provenance-constrained relational evidence use and present PREF-Gate, an auditable decision framework with two fixed experts and a finite validation gate. The context expert uses attributes, one-hop means, feature residuals, and degree descriptors without labels. The evidence expert adds self-excluded, training-label-only neighborhood risk and empirical-Bayes summaries that expose support, uncertainty, availability, and shrinkage. Before test inference, the gate selects either expert or one of three pre-specified probability mixtures and fixes the decision threshold. On Amazon, YelpChi, and TFinance, using five identical stratified splits and 14 same-protocol methods, PREF-Gate obtains mean AUPRC values of 0.9085, 0.8104, and 0.8913. It selects the label-free expert on all Amazon and YelpChi splits and an evidence mixture on all TFinance splits. Thus, the main result is conditional rather than universal: label-derived relational evidence is useful only where held-out validation supports it. The framework couples competitive ranking performance with an explicit label-provenance contract, finite selection policy, failure accounting, and review-budget evaluation, providing an auditable knowledge-based decision pipeline for graph fraud detection.

cs.AI↗
arXiv2026

Beyond Sparse Supervision: Diffusion-Guided Learning for Few-Shot Graph Fraud Detection

Graph-based fraud detection is essential for safeguarding large-scale transaction systems, where undetected anomalies may lead to substantial financial losses and security risks. Real-world fraud graphs pose two coupled challenges: sparse and imbalanced supervision, where verified fraudulent labels are scarce and heavily skewed toward benign accounts, and representation dilution, where spatial message passing may oversmooth camouflaged anomalies while spectral filters may suppress fraud-relevant mid- and high-frequency irregularities. To address these challenges, we propose ADC-GNN, short for Attention-guided Diffusion-Contrastive Graph Neural Network, a unified framework that combines diffusion-guided feature augmentation, contrastive representation learning, and multi-hop spectral attention for few-shot graph fraud detection. The diffusion component is formulated as a feature-space denoising augmentation mechanism rather than a full topology-generative graph diffusion model: it constructs noise-perturbed node-feature views under a cosine schedule and uses contrastive learning to stabilize node representations across perturbations. The spectral attention module further adaptively emphasizes fraud-relevant hop-level and relation-level cues. We evaluate ADC-GNN primarily on three public benchmarks and additionally report a proprietary real-world telecom transaction dataset with approximately 60,000 records as a private case study. Under the 1% training setting, ADC-GNN achieves consistent improvements over original graph fraud baselines and four protocol-consistent recent graph anomaly/fraud baselines on the public benchmarks. Additional analyses on split stability, training ratios, oversampling alternatives, module-level ablations, diffusion schedules, and runtime and memory-consumption comparisons further characterize the effective operating regime of ADC-GNN.

cs.LG↗
arXiv2026

DistFlow: A Fully Distributed RL Framework for Scalable and Efficient LLM Post-Training

Effectively scaling Reinforcement Learning (RL) is crucial for enhancing the reasoning and alignment of Large Language Models. The massive data and complex execution flows inherent in these tasks require a distributed architecture capable of efficient scaling. However, to simplify programming and dependency management, mainstream frameworks often rely on a centralized architecture where a single node dispatches both control and data. This inherent coupling creates significant communication bottlenecks, severely limiting system scalability and efficiency. We present DISTFLOW, a novel, fully distributed RL framework that adopts a multi-controller paradigm. By decoupling data transmission from control dispatch, DISTFLOW establishes a parallelism-aware, decentralized Data Coordinator that leverages local caching, load balancing, and asynchronous double buffer to minimize communication overhead and mitigate straggler effects. For control logic, it introduces a task scheduler built upon Directed Acyclic Graph (DAG) that facilitates fine-grained, independent execution. Experimental results demonstrate that DISTFLOW achieves near-linear scalability up to 512 GPUs and delivers up to a 2.63x throughput improvement over state-of-the-art (SOTA) frameworks. The source code is available at: https://github.com/sii-research/siiRL.

cs.DC↗
arXiv2026

BackPlay: Head-Only Look-Back Self-Correction for Diffusion Language Models

Diffusion Language Models (DLMs) decode multiple tokens in parallel, but aggressive multi-token decoding amplifies cross-token dependency errors and can sharply degrade generation quality. We propose BackPlay, a frozen-backbone self-correction framework that trains only a lightweight correction head on a finetuned DLM without updating any backbone or adapter parameters. Because the head is trained on errors produced by the same frozen generator used at inference time, its training distribution aligns with the error patterns of the deployed model. We further introduce Look-back Correction, a training mechanism that injects predictions from earlier, more corrupted denoising states into later, richer contexts, enabling the head to leverage later context to detect mistakes made in earlier generation steps. During inference, BackPlay periodically revisits previously generated tokens through selective remasking and regeneration to limit error accumulation. Across mathematical reasoning and code generation benchmarks, BackPlay improves the speed--quality trade-off of the underlying DLM under multi-token decoding.

cs.LG↗
arXiv2026

daVinci-LLM:Towards the Science of Pretraining

The foundational pretraining phase determines a model's capability ceiling, as post-training struggles to overcome capability foundations established during pretraining, yet it remains critically under-explored. This stems from a structural paradox: organizations with computational resources operate under commercial pressures that inhibit transparent disclosure, while academic institutions possess research freedom but lack pretraining-scale computational resources. daVinci-LLM occupies this unexplored intersection, combining industrial-scale resources with full research freedom to advance the science of pretraining. We adopt a fully-open paradigm that treats openness as scientific methodology, releasing complete data processing pipelines, full training processes, and systematic exploration results. Recognizing that the field lacks systematic methodology for data processing, we employ the Data Darwinism framework, a principled L0-L9 taxonomy from filtering to synthesis. We train a 3B-parameter model from random initialization across 8T tokens using a two-stage adaptive curriculum that progressively shifts from foundational capabilities to reasoning-intensive enhancement. Through 200+ controlled ablations, we establish that: processing depth systematically enhances capabilities, establishing it as a critical dimension alongside volume scaling; different domains exhibit distinct saturation dynamics, necessitating adaptive strategies from proportion adjustments to format shifts; compositional balance enables targeted intensification while preventing performance collapse; how evaluation protocol choices shape our understanding of pretraining progress. By releasing the complete exploration process, we enable the community to build upon our findings and systematic methodologies to form accumulative scientific knowledge in pretraining.

cs.AI↗
arXiv2026

daVinci-Env: Open SWE Environment Synthesis at Scale

Training capable software engineering (SWE) agents demands large-scale, executable, and verifiable environments that provide dynamic feedback loops for iterative code editing, test execution, and solution refinement. However, existing open-source datasets remain limited in scale and repository diversity, while industrial solutions are opaque with unreleased infrastructure, creating a prohibitive barrier for most academic research groups. We present OpenSWE, the largest fully transparent framework for SWE agent training in Python, comprising 45,320 executable Docker environments spanning over 12.8k repositories, with all Dockerfiles, evaluation scripts, and infrastructure fully open-sourced for reproducibility. OpenSWE is built through a multi-agent synthesis pipeline deployed across a 64-node distributed cluster, automating repository exploration, Dockerfile construction, evaluation script generation, and iterative test analysis. Beyond scale, we propose a quality-centric filtering pipeline that characterizes the inherent difficulty of each environment, filtering out instances that are either unsolvable or insufficiently challenging and retaining only those that maximize learning efficiency. With $891K spent on environment construction and an additional $576K on trajectory sampling and difficulty-aware curation, the entire project represents a total investment of approximately $1.47 million, yielding about 13,000 curated trajectories from roughly 9,000 quality guaranteed environments. Extensive experiments validate OpenSWE's effectiveness: OpenSWE-32B and OpenSWE-72B achieve 62.4% and 66.0% on SWE-bench Verified, establishing SOTA among Qwen2.5 series. Moreover, SWE-focused training yields substantial out-of-domain improvements, including up to 12 points on mathematical reasoning and 5 points on science benchmarks, without degrading factual recall.

cs.SE↗
arXiv2026

Towards an Incremental Unified Multimodal Anomaly Detection: Augmenting Multimodal Denoising From an Information Bottleneck Perspective

The quest for incremental unified multimodal anomaly detection seeks to empower a single model with the ability to systematically detect anomalies across all categories and support incremental learning to accommodate emerging objects/categories. Central to this pursuit is resolving the catastrophic forgetting dilemma, which involves acquiring new knowledge while preserving prior learned knowledge. Despite some efforts to address this dilemma, a key oversight persists: ignoring the potential impact of spurious and redundant features on catastrophic forgetting. In this paper, we delve into the negative effect of spurious and redundant features on this dilemma in incremental unified frameworks, and reveal that under similar conditions, the multimodal framework developed by naive aggregation of unimodal architectures is more prone to forgetting. To address this issue, we introduce a novel denoising framework called IB-IUMAD, which exploits the complementary benefits of the Mamba decoder and information bottleneck fusion module: the former dedicated to disentangle inter-object feature coupling, preventing spurious feature interference between objects; the latter serves to filter out redundant features from the fused features, thus explicitly preserving discriminative information. A series of theoretical analyses and experiments on MVTec 3D-AD and Eyecandies datasets demonstrates the effectiveness and competitive performance of IB-IUMAD.

cs.CV↗
arXiv2026

$R^2$-Mesh: Reinforcement Learning Powered Mesh Reconstruction via Geometry and Appearance Refinement

Mesh reconstruction from Neural Radiance Fields (NeRF) is widely used in 3D reconstruction and has been applied across numerous domains. However, existing methods typically rely solely on the given training set images, which restricts supervision to limited observations and makes it difficult to fully constrain geometry and appearance. Moreover, the contribution of each viewpoint for training is not uniform and changes dynamically during the optimization process, which can result in suboptimal guidance for both geometric refinement and rendering quality. To address these limitations, we propose $R^2$-Mesh, a reinforcement learning framework that combines NeRF-rendered pseudo-supervision with online viewpoint selection. Our key insight is to exploit NeRF's rendering ability to synthesize additional high-quality images, enriching training with diverse viewpoint information. To ensure that supervision focuses on the most beneficial perspectives, we introduce a UCB-based strategy with a geometry-aware reward, which dynamically balances exploration and exploitation to identify informative viewpoints throughout training. Within this framework, we jointly optimize SDF geometry and view-dependent appearance under differentiable rendering, while periodically refining meshes to capture fine geometric details. Experiments demonstrate that our method achieves competitive results in both geometric accuracy and rendering quality.

cs.CV↗
arXiv2026

Exploiting Convexity of Neural Networks in Dynamic Operating Envelope Optimization for Distributed Energy Resources

The increasing penetration of distributed energy resources (DERs) brings opportunities and challenges to the operation of distribution systems. To ensure network integrity, dynamic operating envelopes (DOEs) are issued by utilities to DERs as their time-varying export/import power limits. Due to the non-convex nature of power flow equations, the optimization of DOEs faces a dilemma of solution accuracy and computation efficiency. To bridge this gap, in this paper, we facilitate DOE optimization by exploiting the convexity of input convex neural networks (ICNNs). A DOE optimization model is first presented, comprehensively considering multiple operational constraints. We propose a constraint embedding method that allows us to replace the non-convex power flow constraints with trained ICNN models and convexify the problem. To further speed up DOE optimization, we propose a linear relaxation of the ICNN-based DOE optimization problem, for which the tightness is theoretically proven. The effectiveness of the proposed method is validated with numerical case studies. Results show that the proposed ICNN-based method outperforms other benchmark methods in optimizing DOEs in terms of both solution quality and solution time.

eess.SY↗
arXiv2026

HybridPrompt: Bridging Generative Priors and Traditional Codecs for Mobile Streaming

In Video on Demand (VoD) scenarios, traditional codecs are the industry standard due to their high decoding efficiency. However, they suffer from severe quality degradation under low bandwidth conditions. While emerging generative neural codecs offer significantly higher perceptual quality, their reliance on heavy frame-by-frame generation makes real-time playback on mobile devices impractical. We ask: is it possible to combine the blazing-fast speed of traditional standards with the superior visual fidelity of neural approaches? We present HybridPrompt, the first generative-based video system capable of achieving real-time 1080p decoding at over 150 FPS on a commercial smartphone. Specifically, we employ a hybrid architecture that encodes Keyframes using a generative model while relying on traditional codecs for the remaining frames. A major challenge is that the two paradigms have conflicting objectives: the "hallucinated" details from generative models often misalign with the rigid prediction mechanisms of traditional codecs, causing bitrate inefficiency. To address this, we demonstrate that the traditional decoding process is differentiable, enabling an end-to-end optimization loop. This allows us to use subsequent frames as additional supervision, forcing the generative model to synthesize keyframes that are not only perceptually high-fidelity but also mathematically optimal references for the traditional codec. By integrating a two-stage generation strategy, our system outperforms pure neural baselines by orders of magnitude in speed while achieving an average LPIPS gain of 8% over traditional codecs at 200kbps.

eess.IV↗
arXiv2026

Artic: AI-oriented Real-time Communication for MLLM Video Assistant

AI Video Assistant emerges as a new paradigm for Real-time Communication (RTC), where one peer is a Multimodal Large Language Model (MLLM) deployed in the cloud. This makes interaction between humans and AI more intuitive, akin to chatting with a real person. However, a fundamental mismatch exists between current RTC frameworks and AI Video Assistants, stemming from the drastic shift in Quality of Experience (QoE) and more challenging networks. Measurements on our production prototype also confirm that current RTC fails, causing latency spikes and accuracy drops. To address these challenges, we propose Artic, an AI-oriented RTC framework for MLLM Video Assistants, exploring the shift from "humans watching video" to "AI understanding video." Specifically, Artic proposes: (1) Response Capability-aware Adaptive Bitrate, which utilizes MLLM accuracy saturation to proactively cap bitrate, reserving bandwidth headroom to absorb future fluctuations for latency reduction; (2) Zero-overhead Context-aware Streaming, which allocates limited bitrate to regions most important for the response, maintaining accuracy even under ultra-low bitrates; and (3) Degraded Video Understanding Benchmark, the first benchmark evaluating how RTC-induced video degradation affects MLLM accuracy. Prototype experiments using real-world uplink traces show that compared with existing methods, Artic significantly improves accuracy by 15.12% and reduces latency by 135.31 ms. We will release the benchmark and codes at https://github.com/pku-netvideo/DeViBench.

cs.NI↗
arXiv2026

Smaller is Better: Generative Models Can Power Short Video Preloading

Preloading is widely used in short video platforms to minimize playback stalls by downloading future content in advance. However, existing strategies face a tradeoff. Aggressive preloading reduces stalls but wastes bandwidth, while conservative strategies save data but increase the risk of playback stalls. This paper presents PromptPream, a computation powered preloading paradigm that breaks this tradeoff by using local computation to reduce bandwidth demand. Instead of transmitting pixel level video chunks, PromptPream sends compact semantic prompts that are decoded into high quality frames using generative models such as Stable Diffusion. We propose three core techniques to enable this paradigm: (1) a gradient based prompt inversion method that compresses frames into small sets of compact token embeddings; (2) a computation aware scheduling strategy that jointly optimizes network and compute resource usage; and (3) a scalable searching algorithm that addresses the enlarged scheduling space introduced by scheduler. Evaluations show that PromptStream reduces both stalls and bandwidth waste by over 31%, and improves Quality of Experience (QoE) by 45%, compared to traditional strategies.

eess.IV↗
arXiv2026

Camel: Frame-Level Bandwidth Estimation for Low-Latency Live Streaming under Video Bitrate Undershooting

Low-latency live streaming (LLS) has emerged as a popular web application, with many platforms adopting real-time protocols such as WebRTC to minimize end-to-end latency. However, we observe a counter-intuitive phenomenon: even when the actual encoded bitrate does not fully utilize the available bandwidth, stalling events remain frequent. This insufficient bandwidth utilization arises from the intrinsic temporal variations of real-time video encoding, which cause conventional packet-level congestion control algorithms to misestimate available bandwidth. When a high-bitrate frame is suddenly produced, sending at the wrong rate can either trigger packet loss or increase queueing delay, resulting in playback stalls. To address these issues, we present Camel, a novel frame-level congestion control algorithm (CCA) tailored for LLS. Our insight is to use frame-level network feedback to capture the true network capacity, immune to the irregular sending pattern caused by encoding. Camel comprises three key modules: the Bandwidth and Delay Estimator and the Congestion Detector, which jointly determine the average sending rate, and the Bursting Length Controller, which governs the emission pattern to prevent packet loss. We evaluate Camel on both large-scale real-world deployments and controlled simulations. In the real-world platform with 250M users and 2B sessions across 150+ countries, Camel achieves up to a 70.8% increase in 1080P resolution ratio, a 14.4% increase in media bitrate, and up to a 14.1% reduction in stalling ratio. In simulations under undershooting, shallow buffers, and network jitter, Camel outperforms existing congestion control algorithms, with up to 19.8% higher bitrate, 93.0% lower stalling ratio, and 23.9% improvement in bandwidth estimation accuracy.

eess.IV↗
arXiv2026

R-Meshfusion: Reinforcement Learning Powered Sparse-View Mesh Reconstruction with Diffusion Priors

Mesh reconstruction from multi-view images is a fundamental problem in computer vision, but its performance degrades significantly under sparse-view conditions, especially in unseen regions where no ground-truth observations are available. While recent advances in diffusion models have demonstrated strong capabilities in synthesizing novel views from limited inputs, their outputs often suffer from visual artifacts and lack 3D consistency, posing challenges for reliable mesh optimization. In this paper, we propose a novel framework that leverages diffusion models to enhance sparse-view mesh reconstruction in a principled and reliable manner. To address the instability of diffusion outputs, we propose a Consensus Diffusion Module that filters unreliable generations via interquartile range (IQR) analysis and performs variance-aware image fusion to produce robust pseudo-supervision. Building on this, we design an online reinforcement learning strategy based on the Upper Confidence Bound (UCB) to adaptively select the most informative viewpoints for enhancement, guided by diffusion loss. Finally, the fused images are used to jointly supervise a NeRF-based model alongside sparse-view ground truth, ensuring consistency across both geometry and appearance. Extensive experiments demonstrate that our method achieves significant improvements in both geometric quality and rendering quality.

cs.CV↗