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

+0新增论文
+0其中 AI 相关
+2已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-15 02:35 · 全站最近入库：2026-11-14 15:09各信源进度 →

统计口径与数据新鲜度

统计 2026-11-14 03:31 — 2026-11-15 03:31（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：2026-11-15 02:35最近发现新增版本：2026-11-14 15:09库内论文最新更新时间：2026-11-13 02:59

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。

23,348 已收录论文19,580 AI 相关6,729 有修订112 HF 社区发现

浏览论文库

研究日报 / 周报 →按信源浏览 →

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 88 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-15 02:34（北京时间）
版本更新 · v2arXiv:2601.22275

VMonarch：基于结构化注意力的高效视频扩散Transformer

VMonarch: Efficient Video Diffusion Transformers with Structured Attention

Cheng Liang, Haoxian Chen, Liang Hou, Qi Fan, Gangshan Wu, Xin Tao, Limin Wang

AI 速读 基于摘要

作者提出VMonarch，用Monarch结构化矩阵表示视频DiT中的稀疏时空注意力，以降低注意力计算复杂度并加速长视频生成。

cs.CV · cs.AI

arXiv ↗PDF ↗
查看原文摘要

The quadratic complexity of the attention mechanism severely limits the context scalability of Video Diffusion Transformers (DiTs). We find that the highly sparse spatio-temporal attention patterns exhibited in Video DiTs can be naturally represented by the Monarch matrix. It is a class of structured matrices with flexible sparsity, enabling sub-quadratic attention via an alternating minimization algorithm. Accordingly, we propose VMonarch, a novel attention mechanism for Video DiTs that enables efficient computation over the dynamic sparse patterns with structured Monarch matrices. First, we adapt spatio-temporal Monarch factorization to explicitly capture the intra-frame and inter-frame correlations of the video data. Second, we introduce a recomputation strategy to mitigate artifacts arising from instabilities during alternating minimization of Monarch matrices. Third, we propose a novel online entropy algorithm fused into FlashAttention, enabling fast Monarch matrix updates for long sequences. Extensive experiments demonstrate that VMonarch achieves comparable or superior generation quality to full attention on VBench after minimal tuning. It overcomes the attention bottleneck in Video DiTs, reduces attention FLOPs by a factor of 17.5, and achieves a speedup of over 5x in attention computation for long videos, surpassing state-of-the-art sparse attention methods at 90% sparsity. Code is available at https://github.com/EricLina/VMonarch.

新论文 · v1arXiv:2611.16795

KuaiRecV2: Benchmarking Large-Scale Continual Learning for Diversified and Multi-task Recommendation

Chenxu Li, Shuchang Liu, Hantao Shu, Wei Yuan, Zhe Xu, Guoqing Hu, Yan Wang, Biao Yang 等 18 位作者

原文摘要

Recommender systems, which retrieve items of interest to users based on their interaction histories, have recently witnessed several key technological advancements, benefiting from the semantic representation space and model-data scaling. Nevertheless, practitioners and researchers have frequently encountered critical inconsistencies between simple offline evaluations and complex online results, and this gap still exists in the era of large models. In this work, we address several key factors that contribute to this inconsistency: 1) the \textit{large-scale continual learning challenge} that assumes a practical demand to maintain long-term model effectiveness under dynamic item pools and distributional shifts; 2) the \textit{combinatorial diversity challenge} that requires the generation strategy to solve the accuracy-diversity trade-off in both the finer token level and the holistic item level; and 3) the \textit{multi-task reward balancing challenge}, which is particularly critical for modern generative models that can accurately model the sequence distribution but are less controllable for the multi-task demands. To address these challenges and further pave the way for the development of more realistic recommendation solutions, we introduce KuaiRecV2, a deliberately constructed benchmark with industrial-level datasets from Kuaishou. Specifically, we provide a large-scale interaction dataset with million-scale multi-modal item information and long-term interaction records across 30 days, with a dynamic video candidate pool. Additionally, we provide a rigorous benchmark that evaluates retrieval accuracy, diversity, and multi-task performance in one rubric system, where an effective advancement occurs if and only if the new method improves all metrics. The code is available at https://github.com/cxli2333/KuaiRecV2

cs.IR

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.16589

STVLA: Spatial-aware and Temporal-memory forVision-Language-Action Models

Zhuo Han, Yeqiang Wang, Xinran Wang, Hongwei Han, Yi Xiong, Zunian Wan, Jianbo Mao, Huan Peng 等 11 位作者

原文摘要

Robotic manipulation often requires both spatial scene understanding and task history. However, many vision-language-action (VLA) policies rely primarily on current observations, without explicitly integrating spatial structure and task history. Appearance-based memories can be sensitive to occlusion and viewpoint changes, while spatial representations of the current scene alone do not preserve how the scene has changed during task execution. We introduce STVLA, a framework that incorporates features from a pretrained visual geometry foundation model into temporal memory alongside visual representations. This allows the policy to recall past scene structure as context for current action decisions. To select informative history, STVLA uses a mathematically grounded scoring rule that balances relevance to the current observation with complementary evidence from memory. Across evaluations on LIBERO, LIBERO-Mem, MIKASA-Robo, and SimplerEnv Bridge, STVLA achieves the highest average success rate among the compared methods on each benchmark. On real-world block stacking, object organization, and phone packaging, it improves success over $π_{0.5}$ by $58.33$, $30.00$, and $35.00$ percentage points, respectively. These results support integrating scene geometry into temporal memory to improve manipulation requiring both spatial precision and task history. Project website: \href{https://stvla.github.io/}{https://stvla.github.io/}

cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.22321

AVC-Reward: Event-Centric Audio-Video Consistency Evaluation

Yuruo Wu, Kaihang Pan, Kai Shen, Jiacheng Li, Shuyuan Tu, Li Tuoyu, Juncheng Li, Siliang Tang

原文摘要

As joint audio-video generation models become more capable, they can generate audio-visual scenes exhibiting three forms of complexity: multi-event composition, multi-layer audio mixture, and asymmetric cross-modal relations. However, current evaluations largely reuse metrics from earlier simple and unidirectional tasks. They typically evaluate an audio-video clip as a whole and rely on simple predefined relations, making them unsuitable for these increasingly complex scenes. We propose an event-centric evaluation framework that disentangles audio and visual events at the event level, analyzes each event independently, cross-validates relations across modalities, and aggregates the judgments into a coarse-to-fine, three-level consistency metric. The metric assesses global narrative harmony, event-level temporal synchronization, and fine-grained attribute alignment, providing a general formulation for audio-video consistency evaluation across diverse generation settings and cross-modal relations. We instantiate the metric as AVC-Reward, a trainable audio-video reward model. To evaluate the agreement between automatic judgments and human perception across generation paradigms and relation complexities, we construct AVC-Benchmark, which is primarily built from diverse positive and negative samples generated by various types of audio-video generation models. AVC-Reward achieves state-of-the-art alignment with human judgment among automatic methods, provides interpretable rationales that localize failures, and improves the audio-video consistency of a joint generation model when used as an online reinforcement-learning signal. Project page: https://tsuki0512.github.io/AVC-Reward.github.io/

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.16224

Disentangled Embedding Optimization for Diffusion-Based Dataset Distillation

Hongcheng Li, Yucan Zhou, Yibin Wang, Xiaoyan Gu, Bo Li, Weiping Wang

原文摘要

Dataset distillation aims to compress large-scale datasets into compact synthetic subsets that enable efficient model training with comparable performance to the original data. Recent advances show that diffusion models are highly effective for this task due to their strong generative capability. Existing diffusion-based approaches typically perform class-wise clustering and use the obtained cluster prototypes as diffusion inputs to generate distilled samples. However, the uniformly distributed original samples in a category inevitably lead to entanglement near cluster boundaries, pulling the cluster prototypes closer and reducing the inter-cluster separation. Consequently, the diffusion model receives overly similar latent inputs, resulting in synthetic data with limited diversity. To address this issue, we propose Disentangled Embedding Optimization (DEO), a simple yet effective framework that replaces static clustering with trainable embeddings. These embeddings are optimized to match the Gaussian statistics (i.e., mean and covariance) of the original data distribution, while a diversity regularizer is designed to encourage separation among these embeddings to improve the intra-class variability. Extensive experiments demonstrate that our proposed DEO can generate more diversified and representative distilled data than existing cluster-based diffusion methods, achieving state-of-the-art performance and superior generalization across multiple datasets.

cs.CV

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2604.08410

BLaFA: Bridging Language to 3D Functional Affordances for Dexterous Robotic Grasping in Tool Use

Fan Yang, Wenrui Chen, Guorun Yan, Ruize Liao, Wanjun Jia, Dongsheng Luo, Jianwen Hu, Jiacheng Lin 等 11 位作者

原文摘要

In industrial assembly, equipment maintenance, and everyday tool-use scenarios, robots need not only to grasp objects stably but also to establish task-appropriate contact configurations according to task intentions and tool functionality. However, existing methods lack explicit mappings from open-vocabulary instructions to 3D functional contact structures and dexterous grasp configurations, limiting robots' adaptability to diverse tools and functional tasks. To address this limitation, we propose BLaFA (Bridging Language to 3D Functional Affordances for Dexterous Robotic Grasping), a structured framework for language-driven functional dexterous grasping. First, a Knowledge-guided Language Parsing (KLP) module leverages structured knowledge to transform natural-language instructions into semantic sextuples of manipulation constraints. Second, a 3D Gaussian Splatting based hierarchical semantic representation and a Triangular Functional Point Localization (TriLocation) module are developed to localize 3D functional contact keypoints under task-tool topology constraints. Finally, a 3D Keypoint Grasp Matrix Transformation (KGT3D+) module maps the localized keypoints and semantic constraints to executable wrist poses and finger-level joint and force commands. Evaluations on 100 open-vocabulary language instructions, 14 task-tool combinations covering four functional intents, and real-world robotic experiments in representative scenes demonstrate that BLaFA can effectively translate natural-language instructions into task-aligned 3D functional contact structures and corresponding dexterous grasps, thereby enabling language-driven robotic tool use across diverse tasks. The source code will be made publicly available. Code will be publicly available at https://github.com/PopeyePxx/BLaFA.

cs.CV · cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2605.02171

QuIVer: Efficient Graph Indexing with Two-Bit Construction and Navigation

Wenxuan Xiao, Peidong Zhu, Chengcheng Li, Xu Cao

原文摘要

High-dimensional vector indices must balance construction time, memory, and query throughput. Two-bit comparisons reduce computation and storage, but preserving useful graph connections and search candidates requires scores suited to each stage. We present QuIVer, a graph index that uses a training-free sign-magnitude representation for construction and navigation while reserving original vectors for bounded final reranking. Construction discovers candidates with Hamming scores, orders them with an integer proxy, and defers reverse-edge pruning across parallel insertions. Search uses reconstruction norms and bitwise inner products to order both the frontier and retained candidates, without query calibration or an exact anchor. Experiments cover 21 datasets and compare 30 external baseline modes on three core workloads. QuIVer's validation-selected 95% configurations lie on the measured Pareto frontiers of achieved recall, throughput, index preparation time, and memory, whether memory counts navigation state alone or also the original vectors. On Cohere-5M, QuIVer reaches 95.90% Recall@10 at 3,075 single-thread queries/s, $11.74\times$ the throughput of Lucene BBQ with FP32 reranking at 95.83% recall. Against SymphonyQG at 95.87%, it retains 81% of single-thread and 97% of 32-thread throughput while reducing index preparation time by 54%, total logical memory by 47%, and navigation storage by 87%. An SSD mode with INT8 reranking serves Cohere-5M at about 5 GiB of steady-state process RSS. Controlled crossovers and candidate-coverage diagnostics connect these results to construction and retention quality. Across 21 datasets, high recall holds for most text and image embeddings, while non-negative descriptors, cross-modal queries, and random data expose the limits of the representation.

cs.DB

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.15490

ActionCodec2: A Streaming Action Tokenizer

Zibin Dong, Yicheng Liu, Shiduo Zhang, Tianyuan Yuan, Baijun Ye, Jinqiu Wang, Hang Zhao

原文摘要

Most action tokenizers for autoregressive vision language action (VLA) models compress motion into chunks that must be decoded in full before their actions become executable. We introduce ActionCodec2, a non-neural streaming tokenizer that releases actions incrementally while keeping trajectory reconstruction error within a prescribed tolerance. Its encoder quantizes each step relative to the decoder's reconstructed state, so the selected motion primitive corrects residual error rather than allowing errors from independently quantized increments to accumulate. Finding the smallest such primitive set is a geometric covering problem; we derive the exact minimum for translation and constant-factor-tight bounds for rotation. Set-BPE then losslessly merges co-occurring primitive symbols across coordinates and neighboring steps, without a fixed coordinate ordering. A universal version fitted on 20k hours of data from 24 embodiments averages 8.8 tokens per second for 15 Hz end-effector motion on this corpus. With the same VLA backbone, ActionCodec2 outperforms the compared tokenizers on three simulation benchmarks and in zero-shot real-robot evaluation. Its streaming representation allows policies to incorporate state feedback within an action chunk, substantially improving performance over open-loop execution under randomized dynamics, external disturbances, and moving objects. The same tokens also can serve as a direct control interface for VLM agents, which compose successful action sequences into reusable macros. These agent-built macros reduce control decisions and thinking-token usage across attempts and transfer to a held-out task. More results are available at https://zibindong.github.io/actioncodec2-page/

cs.RO

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.15234

Schedule-Calibrated Commitment Order in Blockwise Masked Diffusion Language Models

Gaurav Sarkar, Umang Pandey, Pratik Prabhanjan Brahma, Zicheng Liu, Emad Barsoum

原文摘要

Sequential block schedules can make token commitment in masked diffusion language models appear strongly left-to-right even when within-block order is random or reversed. For equal contiguous blocks processed without interleaving in positional order, with untied irreversible commitments, we use a Spearman-rank identity to express global position-commitment correlation as a deterministic schedule reference plus a weighted mean within-block correlation. In a reported paired study of LLaDA-8B-Instruct on 500 math and code prompts with 256 generated positions and updates, increasing block size from 64 to 128 lowers mean global correlation from 0.9845 to 0.9149. Uniform-random within-block references are already 0.9375 and 0.7500. After calibration, mean within-block correlations are 0.751 and 0.660, so the larger raw residual at block size 128 does not indicate stronger local positional ordering. We give sharp bounds and derive calibrated intervals from reported summaries, while documenting the limitations of the aggregate-only empirical release. The analysis supports schedule-aware interpretation of commitment order; its empirical scope is one checkpoint and two schedules, without evaluation of output quality or inference speed.

cs.LG · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.15225

Can MLLMs Reason over Surround-View Driving Videos? A Multi-View Autonomous Driving QA Benchmark

Shunming Liu, Jianyi Peng, Cheng Li, Lu Hou, Pei Huang, Yajing Sun, Huaxiao Liu

原文摘要

Surround-view video understanding in autonomous driving is a crucial capability for Multimodal Large Language Models (MLLMs) to achieve reliable driving reasoning. Real-world driving often requires information beyond the front-view camera. Existing MLLMs primarily focus on front-view information, while their spatio-temporal understanding and reasoning abilities remain unknown. To address this gap, we propose MVDriveBench, a high-information-density video QA benchmark for autonomous driving, characterized by synchronized surround-view inputs, spatio-temporal understanding, and reasoning-chain-oriented evaluation. We design 12 categories of QA tasks that cover the complete reasoning pipeline from low-level visual perception to high-level driving decision-making, enabling a systematic evaluation of MLLMs surround-view spatio-temporal understanding and reasoning capabilities in complex driving environments. We conduct a comprehensive evaluation of 23 open-source, closed-source, and expert MLLMs on MVDriveBench. The results show that current MLLMs still exhibit insufficient surround-view spatio-temporal reasoning ability. Meanwhile, closed-loop evaluations of specialist models in AlpaSim reveal a correlation between surround-view spatio-temporal reasoning capabilities and closed-loop driving performance, highlighting the importance of comprehensive multi-view understanding for reliable vehicle control.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.15049

Science Prediction Universe

Zengji Tu, Xiang Liu, Jiarui Sun, Ruoyu Wu, Zhihui Qi, Chengda Wu, Yutao Zhou, Bo Yang 等 36 位作者

原文摘要

Humanity has long sought to explain natural phenomena. Reliable prediction of experimental outcomes is a central test of our understanding of the processes that produce them. If artificial general intelligence (AGI) is to surpass human scientific capabilities, it must also predict experimental outcomes more reliably than humans. Such capabilities could accelerate scientific discovery across disciplines, help address humanity's most pressing challenges, and improve human well-being. We introduce Science Prediction Universe (SPU), a benchmark of hidden experimental outcomes with 100 public verified tasks from chemistry, materials science, biology and medicine, and 16 private tasks from unpublished wet-lab experiments. Given an experimental setup and candidate conditions, an agent works with any resources it needs to rank the candidates by a measured outcome. We evaluate thirteen agent configurations spanning twelve frontier models across two harnesses (Codex and Claude Code). Current agents remain far from saturating wet-lab outcome prediction, with substantial performance disparities across models: on the public benchmark, configurations separate into two distinct tiers, with task-adjusted Spearman correlations ranging from 0.227 for the lowest-performing configuration to 0.478 for the best. Analysis of 1,155 scored public trajectories (227,103 tool calls) suggests that the bottleneck lies in scientific judgment rather than tool execution: agents rely on shared monotone priors, agreeing far more with one another than with experimental reality (0.52 vs. 0.33), and frequently compute trivial physical or chemical proxies that do not influence the final outcome.

cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.14741

Reconstructing Fine-grained Memory Reuse Structure in GPU Kernels

Yanbo Zhao, Yueming Hao, Zecheng Li, Mao Lin, Shuyin Jiao, Keren Zhou, Xu Liu, Jiajia Li

原文摘要

Optimizing GPU data movement requires knowing which instructions access the same data and across which execution scopes reuse occurs. Existing profilers can attribute memory behavior to individual instructions but do not expose the reuse relations among them, while aliasing, dynamic indexing, and runtime parameters prevent static analysis from establishing these relations in general. Cycle-level simulation can reconstruct such relations, but at costs prohibitive for production workloads. We introduce two abstractions for representing this structure: Topological Distance, which labels each dynamic memory access with the smallest execution scope (thread, warp, block, or grid) containing its most recent reuse, and PCDepGraph, which organizes these labels into a PC-level graph of the kernel's memory reuse structure. Defined over logical execution scopes rather than transient hardware state, the representation remains stable across runs and GPU targets while directly indicating the sharing scope that candidate optimization mechanisms must cover. Our tool, cuVein, reconstructs PCDepGraph directly from optimized binaries, without sampling, at generally under 10x overhead. Across HPC and AI workloads on NVIDIA RTX 4090 and H100, optimizations guided by PCDepGraph eliminate up to 90.18% of memory instructions and yield up to 2.84x kernel speedup; reuse-scope distributions remain consistent across platforms, supporting that the representation captures program structure rather than transient hardware behavior.

cs.DC · cs.PF · cs.PL

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.22588

SAG2: Embedding-Free Multi-Hop Retrieval with BM25 and Decision Models

Yuchao Wu, XingCheng Liang, Guanxian Li

原文摘要

Multi-hop retrieval must find passages that become useful only after intermediate evidence is retrieved. We present SAG2, an embedding-free retriever whose only persistent structure is a SQLite lexical index, built without an LLM. Where SAG and HippoRAG 2 rely on a generative LLM for graph extraction and selection, SAG2 uses a decision model. Given the question and a shared context of selected retrieved passages, the decision model scores whether each candidate supplies an answer-chain fact. Terms activated from the top-scoring passages retrieve unvisited passages through the lexical index, which holds a passage-term hypergraph implicitly. SAG2 reaches Recall@10 of 86.62%, 97.43%, and 99.00% on 1,000-question subsets of MuSiQue, 2WikiMultiHopQA, and HotpotQA, the highest among the compared systems. With a decision model priced at about a fifth of SAG's LLM, indexing and retrieval on MuSiQue cost USD 4.33 against USD 14.85 for SAG. On EnterpriseRAG-Bench, SAG2 reaches 81.12% document Recall@10. As the benchmark's nested corpus grows 447.5-fold, SAG2's Recall@5 lead over BM25 alone widens from 7.28 to 14.76 points, at a nearly constant model cost per question. Because indexing makes no model call, SAG2 suits corpora that change often or must be searchable as soon as they are indexed.

cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.09672

ODEAction: From Predicting Chunks to Evolving Actions

Qishen Wang, Pengxiang Ding, Shifeng Bao, Mingyang Sun, Hengtao Li, Yi Gao, Baohua Yin, Yang Liu 等 17 位作者

原文摘要

Vision-language-action (VLA) models typically generate actions in chunks, dividing continuous robot execution into separately predicted action chunks, which can introduce discontinuities at chunk boundaries. We propose ODEAction, a chunk-free action generation paradigm that replaces action chunk prediction with Continuous Action Evolution. We use an ordinary differential equation (ODE) to describe the continuous evolution of actions, and define the vector field of this ODE as the Action Transition Field to characterize how actions evolve with respect to evolution time. During inference, the model takes the current action as the initial condition and numerically solves the ODE defined by the learned Action Transition Field, thereby recursively generating subsequent actions with explicit dependencies. ODEAction is implemented as a modular action head that can be integrated with different vision-language backbones. Experiments on LIBERO, EBench, RoboTwin, and real-robot tasks show that, compared with action-chunk-based baselines, ODEAction improves both task performance and trajectory quality, with particularly pronounced gains on fine manipulation tasks.

cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.11399

RoCoMark: Robust Code Watermarking Against LLM-Based Functionality-Preserving Rewriting Attacks

Hongjie Zhang, Xingcheng Li, Hanzhou Wu, Zhihua Xia, Xinpeng Zhang

原文摘要

Large Language Models (LLMs) present significant challenges to source tracing of generated code, as LLM rewriting can substantially alter the syntactic and structural characteristics of watermarked code. However, existing watermarking methods often suffer severe performance degradation under LLM-based functionality-preserving rewriting attacks. We attribute this vulnerability not merely to the intensity of rewriting, but to a fundamental distribution mismatch between the perturbations encountered during watermark training and the global, non-deterministic rewriting behaviors induced by LLMs. In order to deal with this challenge, we propose RoCoMark, a Robust Code waterMarking framework that explicitly learns and adapts to LLM rewriting behaviors. RoCoMark is a multi-round minimax game between the watermark embedder and LLM rewriter. In each round, the embedder first learns threat profiling from LLM-rewritten code by optimizing the watermark encoder through contrastive learning and knowledge distillation while mitigating catastrophic representation drift. Then the embedder improves strategies by optimizing the watermark encoder and extractor simultaneously. Finally, an adaptive attack-evolution mechanism based on Reflexion and short- and long-term memory pools is applied, enabling continuous interaction between the watermark embedder and evolving rewriting strategies. This iterative attacker-defender process progressively improves the robustness of the watermark against diverse and previously unseen LLM rewriting behaviors. Extensive experiments on four typical benchmarks, under attacks generated by five LLMs, show the effectiveness of RoCoMark. RoCoMark achieves an average detection AUROC of 74.2% and a Rewrite BitACC of 75.3%, while maintaining a code functionality pass rate above 99.7%.

cs.CR

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.13044

Cyclically covering subspaces and additive perfect codes

Yangcheng Li, Pingzhi Yuan

原文摘要

Let $h_q(n)$ be the largest codimension of a subspace of $\mathbb{F}_q^n$ whose cyclic shifts cover the whole space. For $r>0$, we characterize the inequality $h_q(n)\ge r$ in terms of cyclic $\mathbb{F}_q$-linear codes over vector alphabets with covering radius one, no nonzero word of block weight one, and a scalar dual generated by the shifts of one word. For $k,m\ge2$, let $I_q(j)$ count the monic irreducible polynomials of degree $j$ over $\mathbb{F}_q$. If $k$ is prime or $m\ge2\sum_{j=1}^{k-1}I_q(j)+2$, we prove that a cyclic $\mathbb{F}_q$ linear perfect code of length $\frac{q^{km}-1}{q^k-1}$ over $\mathbb{F}_q^k$ exists if and only if $\gcd(m,q-1)=1$. When this condition holds, we also obtain \(h_q\!\left(\frac{q^{km}-1}{q^k-1}\right)=k(m-1).\) Restriction of scalars gives the same criterion for every $m\ge2$ when $q$ is odd, $q+1$ is a power of two, and $k=2p$ with $p$ prime. We also show that $h_q(q^k+1)=k$ exactly when $q$ is even, for every $k\ge1$.

cs.IT · math.CO · math.NT

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2606.21023

Demystifying Numerical Instability in LLM Inference: Achieving Reproducible Inference for Mission-Critical Tasks with HEAL

Zhenting Zhu, Lucas Thai, Shan Yu, Yicheng Liu, Yifan Qiao, Chenxi Wang, Harry Xu, Junyi Shu

原文摘要

As Large Language Models (LLMs) deploy into mission-critical domains (e.g., finance, medicine, and law), output reproducibility has become a strict system requirement. While practitioners use greedy decoding to eliminate algorithmic stochasticity, empirical deployments with 16-bit precisions still exhibit catastrophic output divergence across heterogeneous GPUs. Through SASS-level profiling, we reveal that this inconsistency is fundamentally driven by truncation errors introduced during downcasting at kernel boundaries. However, achieving reproducibility via a global FP32 pipeline incurs prohibitive system penalties: bypassing 16-bit hardware accelerators hurts compute efficiency, while upcasting the KV cache doubles memory overhead. To bridge this gap, we propose Hybrid Error ALleviation (HEAL), a targeted intervention that approximates FP32 precision while resolving hardware constraints through two targeted mechanisms. First, recognizing that floating-point formats underutilize their bit-width for Q, K, V tensors, HEAL applies INT16 quantization that preserves numerical stability without expanding the KV cache footprint. Second, HEAL synthesizes high-precision matrix multiplications via an algebraic error compensation strategy, executing entirely on high-throughput 16-bit Tensor Cores. To evaluate our approach practically, we introduce MCR-Bench, a benchmark targeting reproducibility in mission-critical tasks. HEAL achieves the same level of reproducibility on downstream tasks as the FP32 baseline while reducing the performance overhead by up to 6.0$\times$.

cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.10846

Efficient SC-Flip Decoding for Polar Codes through Decision-Boundary Normalization

Gulinisha Yiming, Zhongjun Yang, Zhicheng Liu, Zhiming Ma, Xumin Ni, Yuting Liu

原文摘要

Accurate first erroneous position (FEP) identification is key to improving the error-correction performance of successive cancellation flip (SCF) decoding. However, FEP identification remains challenging with a limited number of re-decoding attempts. To address this issue, we propose decision-boundary normalization (DBN) to improve FEP identification and develop a novel DBN-SCF decoding algorithm. DBN normalizes the magnitude of each decision log-likelihood ratio (LLR) by the standard deviation predicted by the Gaussian approximation (GA) for the corresponding bit-channel. This allows a single normalized threshold to be shared across bit-channels within each code configuration, giving it a consistent statistical meaning and avoiding per-channel calibration. DBN-SCF decoding selects candidate flip positions from the information set in the natural successive cancellation (SC) decoding order using a threshold, without sorting. Moreover, DBN-SCF decoding achieves better error-correction performance with a limited number of re-decoding attempts. It also requires lower average decoding complexity than existing SCF decoding methods at a comparable level of error-correction performance.

cs.IT

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.10719

No Second Run: Does RSI Stop at Frontier Pre-training?

Xiaojie Cai, Jiayi Chen, Junwei Chen, Yu Chen, Sikai Cheng, Jinquan Dai, Siyuan Feng, Bin Fu 等 59 位作者

原文摘要

Pre-training is where every foundation model begins; bringing recursive self-improvement (RSI) there would let AI improve the process that creates AI. Yet public RSI work has so far targeted only post-training, which starts from an existing base model, runs trials costing GPU-hours, and returns verifiable scores. Pre-training breaks all three conditions at orders-of-magnitude higher cost: a run consumes hundreds of thousands of GPU-hours over months and cannot be repeated; its objective, post-training headroom, cannot be read off any single score; and what must improve is not a model but the whole production system, from data pipelines and training infrastructure to the teams that run them. To our knowledge, we are the first to apply RSI to an entire frontier pre-training run from scratch and to report it in detail. RSI does not stop at frontier pre-training, but it changes form: since no end-to-end loop can close over a single unrepeatable run, we decompose it into smaller loops that can, each repairing a step the full run breaks. The result is 801-30B-A3B-Base, a 30B-A3B MoE trained from scratch on 11.8T tokens in three months by a newly formed team. We release it as a base model; the lightweight SFT reported here (6B tokens, no RL) only probes post-training headroom. As a base model, 801-30B-A3B outperforms Qwen3-30B-A3B and Nemotron-3.5-Lightning-30B-A3B, trailing only the larger Qwen3.5-35B-A3B, and keeps this position under identical SFT; with lightweight SFT alone, it achieves the highest average on four agentic benchmarks against fully post-trained models, including Qwen3.6-35B-A3B. Since post-training RSI presupposes a strong base model, bringing RSI into pre-training acts on the foundation every later stage builds on. We open-source 801-30B-A3B-Base and the lightweight SFT data at https://github.com/GAIR-NLP/801-30B .

cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2610.17623

RELAY: Selective Self-Distillation with Adaptive Hand-off for Rubric-Based Reinforcement Learning

Kunwang Zheng, Xinyu Sun, Yu Cao, Shuocheng Li, Lei Cheng, Ying Liu, Qilong Zhang, Jinlong Chen 等 10 位作者

原文摘要

Rubric-based reinforcement learning grades a response against many written criteria but collapses them into one reward, which discards the detail the rubric carries and gives no GRPO reward gradient when a group's responses tie. A frozen teacher that reads the rubric as privileged context provides token-level supervision, but we observe an early plateau in the gains over GRPO. RELAY uses advantages and rewards to select positive-advantage and low-scoring zero-advantage rollouts, and monitors the distillation loss to decide when to retire the teacher. Stage 1 combines selective distillation from a frozen teacher with GRPO; a plateau detector then removes the teacher, and Stage 2 continues with GRPO alone. Calibrated on one model-domain pair and reused unchanged in three others, the plateau detector fires between steps 122 and 171. RELAY scores highest in-domain among the compared methods in all four settings, ahead of GRPO on RaR-Medicine by 4.62 points with Qwen3-1.7B and 6.02 with Qwen3-4B. On Qwen3-1.7B/Medicine, removing the teacher from a shared checkpoint outperforms keeping, weakening or refreshing it. Distilling selected rollouts covering about half the response tokens also outperforms distilling all rollouts at a matched average coefficient.

cs.LG · cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.10634

Coverage-Verified Sparse Attention: Closed-Loop Quality Control for Long-Context LLM Decoding

Yan-Cheng Li, Chun-Yi Lee

原文摘要

Sparse attention accelerates long-context decoding by reading a subset of the KV cache, but existing methods are open-loop: the kernel commits to its selection with no signal indicating whether enough attention mass was preserved and no recovery path when it was not. The needed signal already exists inside the kernel. The fraction of softmax mass on the selected tokens, which we call coverage, satisfies an exact algebraic equality with per-step output error and serves a dual role as the acceptance criterion and the interpolation weight for recovery. Coverage-Verified Sparse Attention (CVSA) turns coverage into a training-free, closed-loop verify-then-correct decode loop that accepts high-coverage drafts and recovers only the heads that fall short, with no sampling, no draft model, and no speculative-decoding infrastructure. Quality varies by only $0.14$ points across the threshold sweep; budget and scheduling ablations characterize the deployed operating point. On LongBench at 7B and 70B, CVSA comes within $1.13$ points of dense attention, statistically matching it at 70B, and it loses $0.8$ points on RULER; measured median decode-step speedup reaches $6.21\times$ at 128K context.

cs.LG · cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.32069

DriveMind: Mind-Evolving Belief Tracking for Closed-Loop Autonomous Driving

Yan Shi, Kehua Chen, Jiarui Qi, Yihong Tang, Haicheng Liao, Yinhai Wang

原文摘要

Recent multimodal large language models (MLLMs) can perceive a traffic scene, reason about it, and propose a driving action within a single model. However, existing MLLM-based drivers reason from scratch at each frame and commit to a single explanation, discarding earlier hypotheses about the scene. When the scene is ambiguous, the agent locks in on the wrong interpretation and acts on it before evidence accumulates. In this paper, we propose DriveMind, an MLLM-based driving framework that keeps several hypotheses about the scene alive at the same time and updates their weights as new frames arrive. Inspired by particle filtering, DriveMind propagates each hypothesis through a VLM, scores it against the current observation, and resamples to maintain diversity. On Bench2Drive, adding belief tracking to the same driving pipeline improves the Driving Score (DS) from 68.50 to 81.37 and the Success Rate (SR) from 41.36% to 54.09%. The largest gains occur on ambiguous scenarios such as unsignalized intersections (+20.0 DS) and oncoming-traffic interactions (+20.5 DS). Beyond the overall metrics, DriveMind also improves fine-grained interactive driving abilities such as merging, emergency braking, and give-way.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.08989

LEF-JEPA: Denoising-Trained Latent Energy Fields for Spatial Representation Learning

Zhiyuan Liu, Zhihao Xu, Sicheng Liu, Huanhuan Huang

原文摘要

A joint-embedding predictive architecture (JEPA) that applies prediction supervision only at masked positions struggles to clearly capture object shapes and boundaries in its latent representations. We further observe that some existing methods improve spatial structure while also retaining rich variations in texture and color. However, preserving more low-level visual information does not necessarily make representations better suited to prediction and action. This raises a key question: how can we strengthen the spatial structure in latent representations that supports prediction and action without over-encoding these details? We introduce LEF-JEPA, which learns spatial structure through target-guided latent recovery and transfers this supervision to the visual encoder through a relative-energy constraint; the energy module is used only during pretraining and adds no inference overhead. To reduce the cost of energy optimization during pretraining, we further construct a conditional Ising Hamiltonian that allows low-energy-state search to be performed by specialized Ising solvers. On physical hardware, we validate the feasibility of achieving lower forward latency with fewer local updates. Under a unified CIFAR-10 pretraining and frozen-encoder evaluation protocol, both realizations outperform V-JEPA2.1 in image classification, foreground segmentation, and visual planning.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2601.19912

Analysis of LLM Vulnerability to GPU Soft Errors: An Instruction-Level Fault Injection Study

Duo Chai, Zizhen Liu, Shuhuai Wang, Songwei Pei, Cheng Liu, Huawei Li, Shangguang Wang

原文摘要

Large language models (LLMs) are highly compute- and memory-intensive, posing significant demands on high-performance GPUs. At the same time, advances in GPU technology driven by shrinking transistor sizes and lower operating voltages have made these devices increasingly susceptible to soft errors, particularly in radiation-prone environments such as satellite computing scenarios. While prior work has examined GPU soft errors, most studies have focused on general-purpose applications or conventional neural networks primarily used for vision tasks such as classification and detection. However, vulnerability analysis of modern LLMs remains limited, despite their rapid adoption in diverse application scenarios. Given the unique characteristics of LLMs, their resilience to soft errors may differ substantially from earlier models. To bridge this gap, we conduct the first instruction-level fault injection study of LLM inference to analyze the factors affecting LLM vulnerability. Our approach reveals reliability characteristics from multiple perspectives, highlighting the effects of model architecture, parameter scale, and task complexity. These findings provide new insights into LLM reliability and inform the design of more effective fault tolerance mechanisms.

cs.AR · cs.AI · cs.CR

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.07896

AffordJEV: Affordance Grounding by Confirming Candidate Regions

Jincheng Li, Yutian Zhang, Jianyu Zhang, Xinshun Wang, Mengyuan Liu

原文摘要

Affordance grounding asks which visible part of an object performs a requested operation, a basic capability for embodied manipulation and human--robot interaction. Most existing methods follow a System Two paradigm: a multimodal language model first writes a chain of thought and then turns its conclusion into a box or points for a segmenter. This incurs considerable inference cost, gives the semantic decision little direct control over the extent of the final region, and makes mistakes hard to correct. Yet locating an affordance is essentially a perceptual judgment that need not be argued out in words. We therefore propose, in which frozen segmenters propose candidate regions and a System One model decides directly which of them perform the operation, generating no text and reading each decision from a single forward pass. Our contributions are twofold. (1) An inference framework that confirms candidate regions level by level, from object to part to sub-part, with evidence passed between the levels in both directions so that the levels correct one another. (2) A System One model trained for this framework: existing System One models are built for general-purpose decisions, judge one candidate per request and are ill-suited to the framework's region questions, so ours learns directly from the requests the framework itself produces and judges all candidates of a level at once. Trained only on InstructPart and evaluated with definition-aligned operation words, outperforms chain-of-thought methods in domain and in zero-shot transfer to UMD and IIT-AFF, is better overall than existing System One models run in the same framework, and needs fewer judge requests. To our knowledge, is the first approach to bring System One models to affordance grounding.

cs.RO

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.07605

The Gating Mechanism of KV Cache Eviction

Hao Gu, Xintong Yang, Zhiwei Li, Lei Zhu, Jian Yang, Qiyuan Zhu, Dezhi Li, Jingyuan Zhang 等 13 位作者

原文摘要

KV cache eviction must decide which entries carry information that the rest of the cache can no longer supply. A natural way to estimate this is a learnable gate: give each cached entry a gate on its attention weight and train the gate on the model's own objective. The gate belongs inside the normalizing softmax, where suppressing one entry hands its mass to the others rather than discarding it. That same normalizer absorbs any per-query constant, so a constant gate cancels and reproduces the frozen model exactly. Trained on the language-modeling loss the gate stays near a constant score while its loss falls throughout: optimization succeeds, but a frozen backbone was fit for dense attention, so its loss prefers the no-op, and a near-constant score ranks nothing. The no-op is not a floor the gate improves from but what its objective asks for. A second obstacle survives once the gate moves, since the same gradient charges the gate for errors the frozen model already makes, which no reweighting of positions separates out. Much of the machinery in learnable eviction, from sparsity penalties to relaxations of the discrete decision, works around the first obstacle rather than removing it; both are removed by putting dense attention beyond the gate's reach and making the dense model's predictions its target. Under these two conditions capacity is never the missing ingredient: neither a larger scorer, nor explicit access to the past, nor a trainable backbone helps. The same holds for what eviction unavoidably discards, which is recovered by parameter-free centroids read inside the same softmax far better than by any memory read outside it, since the mass they must return is a share of that normalizer and available nowhere else. At 75\% KV cache compression, this gate reaches 90.14 on RULER at 8K and 46.03 on LongBench with Qwen3-8B, 3.72 and 1.36 points below full-cache inference.

cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.07542

FlashOffloading: Low-Latency KV Cache Offloading for Long-Context LLM Serving

Xuan Wang, Zhuanhao Wu, Tingzhu Fang, Xushi Li, Sicheng Zhang, Zhenyu Yang, Feng Liu, Zhaohui Fan 等 20 位作者

原文摘要

In large language model (LLM) serving systems, the key-value (KV) cache grows linearly with sequence length and concurrency, so long-context inference easily hits the high-bandwidth memory (HBM) capacity wall and results in low concurrency. Offloading the KV cache to host dynamic random-access memory (DRAM) frees HBM space and raises concurrency, but the additional cache management and host-to-device (H2D) load overhead inflates inference latency, higher concurrency fails to translate into effective throughput. We present FlashOffloading, a host-DRAM-centric inference architecture that fully offloads the KV cache in both the prefill and decode phases. FlashOffloading builds rolling HBM cache for prefill, a hot cache for decode, a reverse sparse index with zero extra miss-lookup overhead, and a compute--load overlapped kernel, optimizing KV offloading latency across the entire pipeline and delivering throughput gains under a time-per-output-token (TPOT) service-level objective (SLO). In experiments deploying GLM-5.2 with multi-token prediction (MTP3) speculative decoding on an Ascend910C prefill--decode (PD)-disaggregated platform, FlashOffloading improves prefill throughput by up to 1.38\(\times\) and decode throughput by up to 1.91\(\times\) under a 30 ms TPOT constraint. FlashOffloading is integrated into an open-source serving framework: https://github.com/vllm-project/vllm-ascend/tree/releases/v0.30.0.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.07534

Route Before Response: Fail-Closed Routing for LLM-Enabled Customer-Service Software

Yuchen Cao, Zhenyu Mao, Jacky Wai Keung, Yifei Wang, Guancheng Lin, Jingyu Zhang

原文摘要

LLM-enabled customer-service software in financial institutions must choose a downstream service before responding. Classifiers and LLM routers can identify a relevant label or destination, but a composed application must still verify dispatch permission and entity registration. We study this integration problem as boundary-aware pre-response routing. We present RouteBR, a controller in which one model call proposes a primary service action and registered object, then deterministic code checks identifiers and versioned rules before dispatch or a specified non-dispatch outcome. Lower-ranked candidates cannot replace a blocked primary route. We construct 600 privacy-preserving scenarios from recurring confidential-request patterns at a real financial institution; three engineers jointly reviewed the scenarios and labels. On the 200-case CONFIRM MAIN split, RouteBR reaches 97.5% final-path exact match. A matched one-call baseline given the same rules reaches 95.5%, whereas a diagnostic baseline without the rule program obtains 94.0% action accuracy but only 41.0% final-path exact match. Across 166 non-dispatch MAIN and FLIP cases, RouteBR produces no false dispatch. The artifact pairs all 1,960 prediction rows with raw receipts while excluding gold and reviewer fields from model requests. These results support separating model proposals from code-owned, replayable execution. Our implementation is available at https://github.com/YgcCoder/RouteBR.

cs.SE

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2606.00734

EMA: Approximate Nearest Neighbor Search with General Attribute Filtering and Dynamic Updates

Mocheng Li, Xiao Yan, Baotong Lu, James Cheng, Chenhao Ma

原文摘要

Filtering approximate nearest neighbor search (Filtering ANNS) retrieves approximate nearest neighbors subject to attribute constraints in applications such as recommendation, retrieval-augmented generation (RAG), and LLM agent memory. However, predicate-specialized methods target either range or label filtering and can incur substantial index construction time, whereas predicate-agnostic methods support arbitrary predicates but struggle to sustain throughput and recall across a wide range of selectivities. In this paper, we propose EMA, a Filtering ANNS algorithm that supports multi-predicate queries over mixed numerical and categorical attributes as well as efficient dynamic updates. EMA attaches compact summaries, called Markers, to graph edges to record the attributes of candidates removed by geometric pruning. During query processing, EMA performs Marker-guided joint search with bounded edge recovery, enabling efficient filtering while preserving graph navigability. On multi-predicate workloads, EMA achieves 1.20x-28.18x the query throughput of the best competing baseline at comparable recall. In our complete-record replacement experiment, EMA retains approximately 98% of its initial query throughput without rebuilding the graph.

cs.DB

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06670

Polynomial-Time Differentially Private Synthetic Graph Release with Nearly-Linear Additive Error

Marek Eliáš, Gautam Kamath, Janardhan Kulkarni, George Z. Li, Jingcheng Liu, Sepideh Mahabadi, Pan Peng, Jakub Tarnawski 等 10 位作者

原文摘要

We give a polynomial-time $(\varepsilon, δ)$-edge-differentially private algorithm that releases a synthetic graph approximately preserving all cut values of a nonnegative weighted input graph $G$. Our algorithm outputs a nonnegative weighted graph $H$ such that, with probability at least $1-β$, simultaneously for all $S\subseteq V$, \[ \left|w_G(S,V\setminus S)-w_H(S,V\setminus S)\right| \leq γw_G(S,V\setminus S) + O\left(\frac{n}{γ\varepsilon} \log^7 \frac{n}{δβγ\varepsilon}\right). \] For fixed $γ, \varepsilon$ and $δ= β= n^{-10}$, the additive error is $n \operatorname{polylog} n$, improving upon the previous efficient bound of $n^{13/12 + o(1)}$. Furthermore, we establish an $Ω(n/γ\varepsilon)$ lower bound for the relevant parameter regime, showing that our result is tight up to polylogarithmic factors. Our algorithm uses a log-det-regularized SDP to generate fractional queries for multiplicative-weights updates and certify accuracy for all cuts. The privacy analysis charges privacy loss only fractionally per mechanism call, based on the amount of mass on edge coordinates retired from subsequent iterations. Our synthetic graph release yields improved error bounds for several downstream tasks in private graph analysis including correlation clustering, Max-Cut, Max-$k$-Cut, constrained cut problems and hierarchical clustering. Most notably, for unweighted complete-graph correlation clustering, we achieve the optimal $\widetilde O(n)$ additive error while reducing the best previously known approximation ratio for this additive error from several thousand to $1.387$. We also provide a formalization of our results in Lean.

cs.DS · math.ST

查看论文
arXiv ↗PDF ↗