跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+1085新增论文
+884其中 AI 相关
+130已有论文修订
+94新增 AI 导读

arXiv 最近完成扫描:2026-11-19 06:59 · 全站最近入库:2026-11-19 06:59各信源进度 →

统计口径与数据新鲜度

统计 2026-11-18 09:24 — 2026-11-19 09:24(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

最近完成检查:2026-11-19 06:59最近发现新增版本:2026-11-19 06:59库内论文最新更新时间:2026-11-18 08:00

每分钟刷新已保存的数据。空检查不代表采集异常;新收录可能包含历史论文。AI 导读按新增版本计数。

26,871 已收录论文22,375 AI 相关8,156 有修订112 HF 社区发现

浏览论文库

时间范围跟随排序:版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常,查看状态 →

找到 16 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-19 06:57(北京时间)
新论文 · v1arXiv:2611.22086

HyCoInfer: Cloud-Edge Collaborative Inference for Large Language Models on Heterogeneous GPU-FPGA Architectures

Zewen Huang, Xiangwen Zhuge, Yingkun Hu, Zheng Yang, Qiang Ma

原文摘要

Large language model (LLM) serving relies predominantly on cloud GPUs, yet autoregressive decoding is memory-bound and often leaves their arithmetic units underutilized. Edge FPGAs offer attractive decoding efficiency, but are ill suited to sustaining a high-concurrency serving workload alone. This paper presents HyCoInfer, a heterogeneous cloud-edge serving system that assigns prefill to cloud GPUs and distributes decoding across cloud GPUs and edge FPGAs. The design addresses three barriers to practical offloading: capacity balancing across heterogeneous accelerators, the large KV cache crossing a constrained wide-area link, and the latency added by heterogeneous execution. HyCoInfer incorporates FPGA decode capacity into GPU prefill/decode role partitioning and places requests according to queue load, prompt length, and network state. It further combines layer-wise pipelined transfer, decoder-side precomputation, and bandwidth-adaptive KV cache compression based on residual vector quantization. We implement HyCoInfer using two NVIDIA V100 GPUs and an AMD Xilinx Alveo U280 and evaluate it over 100-1000 Mbps links. At 500 Mbps, HyCoInfer delivers up to 4.11 times the request goodput of cloud-only vLLM and reduces mean end-to-end latency by up to 61.1%. Across different network conditions and workloads, the system improves overall serving capacity, with the largest gains on communication-intensive tasks, and reduces per-request energy.

cs.OS · cs.AI

新论文 · v1arXiv:2611.21312

PointZip: Statistically Lossless Compression for Climate Data with Pointwise Error Bounds

Jiayong Li, Pratyai Mazumder, Yuejiang Yu, Langwen Huang, Alexandru Calotoiu, Torsten Hoefler

原文摘要

High-resolution climate simulations generate data volumes that increasingly strain storage, I/O, and data dissemination. Error-bounded lossy compressors can reduce these costs, but their tolerances are usually user-chosen and difficult to standardize across heterogeneous climate datasets. We address this limitation with statistically lossless compression, where the pointwise compression error is bounded by the pointwise data uncertainty. We propose PointZip, a two-stage compressor that combines a rate-distortion-efficient base layer with error-bound-aware residual coding to enforce pointwise guarantees. On outputs from the widely used climate modeling framework ICON, PointZip makes 1km simulations practical by reducing storage bandwidth requirements by 72x. On the ERA5 climate dataset, PointZip reaches 69.3x on the most representative variables, reducing the commonly used subset to 4 TB. In downstream evaluations, PointZip preserves data spectra, extreme events, and air parcel trajectories, while increasing neural weather forecast RMSE by less than 1%, with negligible impact on the Fractions Skill Score (FSS).

cs.CE

版本更新 · v3arXiv:2506.16250

Graph-Cover-based Characterization of the Bethe Partition Function of Double-Edge Factor Graphs

Yuwen Huang, Pascal O. Vontobel

原文摘要

For standard factor graphs (S-FGs), whose local functions are nonnegative-real valued, a popular approximation of the partition function is given by the Bethe approximation, which is also known as the Bethe partition function. Vontobel gave a combinatorial characterization of the Bethe partition function with the help of graph covers. The proof of this characterization heavily relied on the method of types. In this paper, we study double-edge factor graphs (DE-FGs), whose local functions are complex valued and satisfy certain positive-semidefinite constraints. DE-FGs and their partition functions are particularly relevant to quantum information processing. Approximating the partition function of a DE-FG is more difficult than approximating the partition function of an S-FG because the DE-FG partition function is a sum of complex values rather than nonnegative-real values. We develop a Bethe approximation of the partition function of a DE-FG based on sum-product-algorithm (SPA) fixed points. However, one cannot directly apply the method of types to prove a similar graph-cover-based combinatorial characterization of the Bethe partition function as in the case of S-FGs. Despite this, we establish such a combinatorial characterization for a class of DE-FGs that satisfy an easily checkable condition. Towards proving this characterization, we apply a suitable loop-calculus transform (LCT) to DE-FGs. Originally, the LCT was introduced by Chertkov and Chernyak as a special linear transform for S-FGs and later extended by Mori. Our proposed LCT is applicable to both S-FGs and DE-FGs and generalizes prior versions by handling zero-valued SPA fixed-point message components, which are common in DE-FGs. Supported by numerical results, we conjecture that the graph-cover-based combinatorial characterization of the Bethe partition function holds more broadly for DE-FGs.

quant-ph · cs.IT

版本更新 · v3arXiv:2605.13276

D-VLA: A High-Concurrency Distributed Asynchronous Reinforcement Learning Framework for Vision-Language-Action Models

Yucheng Guo, Yongjian Guo, Zhong Guan, Wen Huang, Haodong Yue, Shuai Di, Junwu Xiong, Yicheng Gong 等 9 位作者

原文摘要

The rapid evolution of Embodied AI has enabled Vision-Language-Action (VLA) models to excel in multimodal perception and task execution. However, applying Reinforcement Learning (RL) to these massive models in large-scale distributed environments faces severe systemic bottlenecks, primarily due to the resource conflict between high-fidelity physical simulation and the intensive VRAM/bandwidth demands of deep learning. This conflict often leaves overall throughput constrained by execution-phase inefficiencies. To address these challenges, we propose D-VLA, a high-concurrency, low-latency distributed RL framework for large-scale embodied foundation models. D-VLA introduces "Plane Decoupling," physically isolating high-frequency training data from low-frequency weight control to eliminate interference between simulation and optimization. We further design a four-thread asynchronous "Swimlane" pipeline, enabling full parallel overlap of sampling, inference, gradient computation, and parameter distribution. Additionally, a dual-pool VRAM management model and topology-aware replication resolve memory fragmentation and optimize communication efficiency. Experiments on benchmarks like LIBERO show that D-VLA significantly outperforms mainstream RL frameworks in throughput and sampling efficiency for billion-parameter VLA models. In trillion-parameter scalability tests, our framework maintains exceptional stability and linear speedup, providing a robust system for high-performance general-purpose embodied agents.

cs.AI · cs.RO

新论文 · v1arXiv:2611.15459

Fear of Memorizing Objects: End-to-End Zero-Shot Model-Based Object Detection

Leon Greiner, Wout Boerdijk, Junwen Huang, Benjamin Busam, Rudolph Triebel, Maximilian Durner

原文摘要

This paper presents FOMO (Fear of Memorizing Objects), an end-to-end trainable transformer-based architecture for zero-shot model-based object detection. Existing approaches typically follow a proposal-and-match pipeline: target-agnostic mask proposals are generated and then matched to template features in a set of loosely coupled stages, making them susceptible to fragmented proposals in clutter and degraded matching under occlusions. FOMO elegantly replaces this paradigm by injecting object template information early in the detection pipeline: Bidirectional attention between learnable latent tokens, multi-view CAD templates, and query image features produces object-aware image features and conditioned queries which are subsequently decoded into bounding-box and confidence predictions. We train FOMO on Clutter6D, a novel large-scale synthetic dataset with over 500k images featuring more than 10k diverse object models rendered in highly cluttered scenes with multiple instances and varying occlusion levels, which addresses the limited object diversity and lack of realistic dense clutter in existing datasets for training zero-shot multi-instance algorithms. In true zero-shot fashion, FOMO achieves 54.2 mAP on BOP-Classic-Core at 0.23 s per image, ranking second overall with 3.8 pp difference to the top method while being more than 7x faster. Throughout comprehensive ablation studies we investigate and verify architectural design choices.

cs.CV

新论文 · v1arXiv:2611.13413

Placing LLM Lookup Tables by Per-Layer Prefetch Deadlines

Wen Huang, Jack Liu

原文摘要

DeepSeek-V4.1's 203 GB of Engram n-gram tables sit in host DRAM today, taking 190 of the 223 GiB a 4-GPU serving process uses. Row indices depend only on input tokens, so reads can start with each step and overlap earlier layers. V4.1's two Engram layers have windows that differ by 30x (0.3 vs. 9.3 ms). We show that this deadline, together with session reuse and contention on a shared device, decides which tier each table can use and which layers need cache, and we give a model that predicts the resulting slowdown from measured read times. Agent and chat traffic reuse rows within sessions, and static hot sets transfer poorly. We build SlackTier, a table tier for SGLang that issues all lookups at step start. It reads the short-window layer first and splits a small GPU row cache across layers. The path runs inside the decode CUDA graph and is bit-exact with the original gather. Layer 14's table can go to NVMe uncached with no measurable decode cost. With both tables on four NVMe disks, mean time per output token on agent and chat replays rises at most 1.4% over unmodified SGLang, and host memory falls from 212 to 34 GiB. Both tables need no cache when read over RDMA (remote direct memory access) from another node's memory; one copy serves two instances. On a KV-constrained agent workload, using the freed memory as host KV cache at the same memory budget raises median throughput by 36% and cuts median p90 time to first token from 38 to 24 s.

cs.DC · cs.AI · cs.LG

新论文 · v1arXiv:2611.13412

Bursts That Outlast Themselves: Persistent Degradation in Prefix-Cached Agent Serving

Wen Huang, Jack Liu

原文摘要

LLM agents drive inference engines with closed-loop sessions whose context grows every turn, and serving them economically depends on prefix caching: a hit costs an incremental prefill of about a thousand tokens, a miss recomputes a 10--40K-token context. We show that this makes prefix-cached agent serving metastable. With an identical base workload, a few minutes of extra arrivals leaves a vLLM server in a degraded regime--a ~10% hit ratio, about half the throughput, and a backlog that grows for the remaining 70 minutes--while its no-burst twin stays healthy (9 of 12 pairs; none in the reverse direction). The regime appears at loads the cache-warm server sustains, is absent without prefix caching, and also appears with Qwen3-32B on SWE-bench agent trace shapes and on SGLang, which tolerates larger triggers. Slower service keeps more sessions in the system, their working set outgrows the KV pool, and every turn becomes a recomputation; a first-order workload-balance condition, calibrated independently, brackets when the regime can drain. Getting out is the harder problem: throttling new admissions recovered 1 of 7 matched collapsed runs, while suspending admitted sessions against a capacity-based limit recovered all of them within minutes. A CPU offload tier delays the collapse but does not prevent it. A budgeter that bounds the resident KV demand of active sessions to the pool, with thresholds that need no per-workload tuning, prevented the failure in every run we tested.

cs.DC · cs.AI · cs.LG

新论文 · v1arXiv:2611.12509

UniHam: A Large-Scale SOC-Complete Dataset and Benchmark for Hamiltonian Learning in Materials

Yuewen Huang, Pin Chen, Yutong Lu

原文摘要

Accurate prediction of electronic Hamiltonians would enable broad property inference while avoiding the high computational cost of Density Functional Theory (DFT). However, progress toward general-purpose materials foundation models is limited by a data bottleneck: existing Hamiltonian datasets are typically small, lack structural diversity, and often omit essential relativistic physics such as spin--orbit coupling (SOC). We therefore construct UniHam, a large-scale Hamiltonian dataset and benchmark suite comprising 100,000+ DFT-computed complex-valued Hermitian Hamiltonians with full SOC, covering 72 elements and a wide range of crystal geometries and symmetries (spanning diverse lattice types and space-group families). Building on UniHam, we benchmark two representative state-of-the-art models under a standardized protocol and introduce complementary evaluation metrics that jointly assess three dimensions: (i) Hamiltonian reconstruction accuracy, (ii) out-of-distribution (OOD) generalization across composition/symmetry shifts, and (iii) the ability to support downstream property prediction from the predicted Hamiltonians. Experiments on UniHam demonstrate that the proposed benchmark and metrics effectively differentiate model capabilities, revealing intrinsic SOC- and element-dependent failure modes, large variations in compositional OOD robustness, and the necessity of spectral-level evaluation to assess whether Hamiltonian predictions reliably support downstream electronic-structure properties. Overall, UniHam provides a reproducible, SOC-complete benchmark that can sharpen model comparisons and accelerate the development of next-generation foundation models for quantum materials.

cs.LG · cond-mat.mtrl-sci

版本更新 · v4arXiv:2505.23277

Sentinel: Decoding Context Utilization via Attention Probing for Efficient LLM Context Compression

Yong Zhang, Heng Li, Yanwen Huang, Ning Cheng, Yang Guo, Yun Zhu, Yanmeng Wang, Shaojun Wang 等 9 位作者

原文摘要

Retrieved contexts are often long, noisy, and redundant, motivating query-aware compression before answer generation. Native attention provides signals for context selection, but heterogeneous head behaviors complicate their direct aggregation into sentence-utility scores. We introduce Sentinel, a sentence-level compressor that learns to combine these signals through a signed linear readout across layers and heads. Only the readout is trained using existing QA evidence annotations, while the backbone remains frozen and sentence scoring requires no autoregressive generation. Across twelve LongBench tasks, Sentinel with a 0.5B backbone outperforms Full Context on multiple tasks under a 2,000-token budget with each of three readers: Qwen2.5-7B-Instruct, Llama-3.1-8B-Instruct, and DeepSeek V4.1 Flash. Despite English-only QA supervision, the same readout transfers to Chinese tasks without task-specific retraining. Length-stratified profiling further shows lower compression latency and GPU memory usage than LLMLingua-2 and CPC.

cs.CL · cs.AI

新论文 · v1arXiv:2611.10356

SepLRM: A Foundation Model Paradigm for Recommendation

Di Chen, Xijun Xiao, Qin Ren, Weijiang Lai, Zhijian Pan, Zhentao Pan, Pengfei Nie, Cheng Yuan 等 26 位作者

原文摘要

Large recommendation models (LRMs) improve recommendation performance by scaling behavior sequence length, model capacity, and training compute. However, the return on investment (ROI) of continued scaling deteriorates as marginal gains diminish and long-sequence encoding costs rise, especially when this computation is repeatedly executed by distinct models for diverse recommendation stages and business scenarios. To address this issue, we propose SepLRM, a recommendation foundation model (RFM) paradigm that identifies reusable sequence encoding and reorganizes multiple end-to-end (E2E) LRMs into one RFM and multiple independently optimized vertical models (VMs). The RFM adopts an encoder-decoder architecture and exports its final-layer encoder state as the shared key-value (KV) cache that is independent of downstream task-offered candidates. Unlike candidate-specific logits or aggregated user embeddings, the KV cache preserves fine-grained behavioral information. Separation at this native KV interface also allows each VM to construct candidate-conditioned queries while retaining the same query-KV interaction as its E2E counterpart, enabling high-fidelity capability transfer. To ensure both transfer effectiveness and system efficiency, we co-design cache compression, the training and serving workflows, and LG-Store, a distributed storage system for KV caches. Experiments on three large-scale datasets show that SepLRM achieves state-of-the-art offline performance with a transfer rate of approximately 100%. It also demonstrates strong transferability across model iterations, recommendation stages, and business scenarios. Deployment in two large-scale industrial scenarios substantially improves the ROI of sequence scaling and delivers consistent gains in online A/B tests at ByteDance.

cs.IR

新论文 · v1arXiv:2611.02306

Lie Routing Vision Language Models for View-Robust 3D Understanding

Jingjun Yi, Chen Hu, Qi Bi, Hao Zheng, Huimin Huang, Yongkang, Wei Ji, Yawen Huang 等 10 位作者

原文摘要

Multimodal Large Language Models (MLLMs) have made substantial progress in 3D scene understanding from multi-view RGB observations. Yet a fixed object and query can elicit inconsistent target-related responses across camera viewpoints, a phenomenon we characterize as viewpoint-induced semantic drift. We introduce LieVLM, a Lie routing framework for view-robust 3D understanding. Motivated by the rotational component of viewpoint change, Lie routing adapts visual representations by applying local $\mathrm{SO}(3)$ transformations to projected features within learned three-dimensional subspaces. Geometry tokens extracted by VGGT predict local rotation parameters and channel-wise routing coefficients. A closed-form Padé--Cayley map transforms projected visual features, whose identity-relative corrections are projected back and gated into the original visual and geometry features. This design retains the pretrained semantic and direct spatial context while learning view-dependent feature adjustments. Experiments on 3D captioning, grounding, detection, and spatial reasoning show consistent improvements over additive fusion across training scales and backbones. Controlled ablations examine the roles of geometric conditioning and transformation structure. Viewpoint analyses further reveal systematic variation in the predicted routing states alongside more localized target responses and stronger task performance across viewpoint changes.

cs.CV · cs.LG

新论文 · v1arXiv:2610.38775

Skillful and realistic precipitation forecasting with reinforced diffusion

Ozan Sener, Anand Avati, David Hafner, Jierui Lin, Laura Manduchi, Xiaoying Pang, Langwen Huang, Torsten Hoefler 等 10 位作者

原文摘要

Accurate precipitation forecasting remains elusive as physics-based models struggle to resolve mesoscale convection and machine learning often yields physically inconsistent, blurred results that consistently underestimate extreme events. We present FitzRoy, the first system that simultaneously achieves state-of-the-art predictive skill and physical realism over the continental United States. Through a reinforced diffusion framework, where a conditional discrete diffusion model is refined via reinforcement learning, FitzRoy effectively resolves the long-standing trade-off between pointwise accuracy and structural fidelity. Across lead times up to 24 hours, FitzRoy consistently matches or exceeds state-of-the-art models while preserving structural realism and capturing extreme events. Its low-latency inference further enables deployment in real-time operational settings. In case studies such as the 2025 Texas Hill Country floods and the 2026 Missouri floods, FitzRoy signals record-breaking accumulations that existing models largely missed. By bridging the gap between realism and skill, FitzRoy establishes a new frontier for high-fidelity precipitation forecasting.

cs.LG · physics.ao-ph

版本更新 · v2arXiv:2606.19350

Pruning via Causal Attribution Preserves Reasoning Performance in Large Language Models

Amogh Sheth, Andrew Lin, Biruk Assefa, Yi Wen Huang, Yuhao Ge

原文摘要

Large language models (LLMs) excel at multi-step reasoning but incur substantial inference cost. We introduce Causal Attribution Pruning (CAP), a training-free method that identifies critical attention heads by measuring their causal impact on reasoning tasks and uses these head-level scores to guide fine-grained weight pruning. For each attention head, CAP estimates the expected performance degradation when the head is masked during forward passes on a small calibration set of reasoning problems. These causal scores are then converted into weight-level importance values for the corresponding projection matrices. Unlike magnitude-only or activation-based criteria, CAP's interventional measurement directly captures each head's functional contribution, yielding relative accuracy gains of up to 61% over Wanda on ARC-Challenge at 20% sparsity. We evaluate CAP on GSM8K, StrategyQA, and ARC-Challenge using Llama-3-8B-Instruct and Mistral-7B-Instruct at 10%, 20%, and 50% sparsity. At moderate sparsity (10-20%), CAP improves over Wanda in most model-benchmark configurations. with especially large gains on ARC-Challenge for Llama-3. Our results suggest that attention-head-level causal attribution can better preserve reasoning performance on downstream benchmarks than correlational pruning criteria at equivalent sparsity, while remaining limited by coarse MLP attribution at 50% sparsity.

cs.CL

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1055ACL Anthology · ACL 2026

CoEvolve: Training LLM Agents via Agent-Data Mutual Evolution

Shidong Yang, Ziyu Ma Hunan University, Tongwen Huang Alibaba Group, Yiming Hu, Yong Wang, Xiangxiang Chu Alibaba Group

原文摘要

Reinforcement learning for LLM agents is typically conducted on a static data distribution, which fails to adapt to the agent’s evolving behavior and leads to poor coverage of complex environment interactions. To address these challenges, we propose CoEvolve, an agent-data mutual evolution framework that enables LLM agents to improve through closed-loop, interaction-driven training. Specifically, CoEvolve extracts feedback signals such as forgetting and uncertainty from rollout trajectories to identify failure-prone interaction patterns, and utilizes them to guide LLM-based task synthesis. The synthesized tasks are validated through environment interaction and utilized to update the data distribution, enabling joint adaptation of the agent and its data. Extensive experiments on AppWorld and BFCL across Qwen2.5-7B, Qwen3-4B, and Qwen3-30B-A3B demonstrate consistent and significant improvements over strong base models, yielding absolute gains of 19.43%, 15.58%, and 18.14%, respectively.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1792ACL Anthology · ACL 2026

Trait Activation in Silicon: A Situation-Aware Framework for Psychologically Grounded Role-Playing

Zuolong Li University of Science and Technology of China, Pingyu Wu University of Science and Technology of China, Xianwen Huang, Tianyi Wei Microsoft, Wenbo Zhou University of Science and Technology of China

原文摘要

Role-playing agents (RPAs) have made significant strides in mimicking static character identities. However, their personality simulations remain superficial, lacking a profound understanding of complex human psychological mechanisms. We identify a critical bottleneck termed “Personality Inertia”—a behavioral rigidity where RLHF-induced alignment bias traps models in a sanitized, “helpful assistant” persona. This inertia prevents models from adapting to diverse social contexts or expressing essential but negative traits under pressure. To bridge this gap, we propose PD-LLM, a situation-aware framework grounded in Trait Activation Theory. PD-LLM introduces Bipolar Latent Decomposition, which decouples personality traits into bidirectional LoRA adapters. These adapters are dynamically modulated by a situation-aware module based on the DIAMONDS taxonomy, allowing for precise behavioral regulation. Empirical results show that while baseline methods fail to synchronize multidimensional traits under pressure, PD-LLM achieves superior performance in both static fidelity and dynamic adaptability. By advancing from prompt engineering to intrinsic parameter control, PD-LLM effectively overcomes personality rigidity, facilitating the creation of vivid and psychologically consistent agents.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.796ACL Anthology · ACL 2026

A Data-Centric Approach to Generalizable Speech Deepfake Detection

Wen Huang, Yuchen Mao Shanghai Jiaotong University, Yanmin Qian

原文摘要

Achieving robust generalization in speech deepfake detection (SDD) remains a primary challenge, as models often fail to detect unseen forgery methods. While research has focused on model-centric and algorithm-centric solutions, the impact of data composition is often underexplored. This paper proposes a data-centric approach, analyzing the SDD data landscape from two practical perspectives: constructing a single dataset and aggregating multiple datasets. To address the first perspective, we conduct a large-scale empirical study to characterize the data scaling laws for SDD, quantifying the impact of source and generator diversity. To address the second, we propose the Diversity-Optimized Sampling Strategy (DOSS), a principled framework for mixing heterogeneous data with two implementations: DOSS-Select (pruning) and DOSS-Weight (re-weighting). Our experiments show that DOSS-Select outperforms the naive aggregation baseline while using only 3% of the total available data. Furthermore, our final model, trained on a 12k-hour curated data pool using the optimal DOSS-Weight strategy, achieves state-of-the-art performance, outperforming large-scale baselines with greater data and model efficiency on both public benchmarks and a new challenge set of various commercial APIs.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表