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

+1095新增论文
+868其中 AI 相关
+92已有论文修订
+95新增 AI 导读

arXiv 最近完成扫描:2026-11-17 04:58 · 全站最近入库:2026-11-17 13:07各信源进度 →

统计口径与数据新鲜度

统计 2026-11-16 13:22 — 2026-11-17 13:22(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:2026-11-17 04:58最近发现新增版本:2026-11-17 13:07库内论文最新更新时间:2026-11-16 17:16

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。

24,849 已收录论文20,754 AI 相关7,213 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 24,849 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-17 04:57(北京时间)
版本更新 · v3arXiv:2606.09864

Alignment Collapse Under KV Cache Quantization: Diagnosis and Mitigation

Bruce Changlong Xu, Adarsh Kumarappan, Mu Zhou

原文摘要

Key-value (KV) cache quantization is widely used to reduce LLM inference memory, yet existing evaluations focus only on perplexity and accuracy without assessing the safety impact. In this study, we explore alignment preservation under KV cache quantization. Across eleven instruction-tuned models (3.8B-72B) and five benchmarks (1,894 prompts), we find that low-bit quantization can silently destroy safety alignment: Mistral-7B loses 15.2% of its refusals at only 1.03x perplexity, and no universal safe bit-width exists, with sharp model-specific phase transitions invisible to standard metrics. We identify that the root cause is geometric: safety features occupy a low-dimensional activation subspace $10^2$-$10^3\times$ more vulnerable to quantization noise than the full representation space perplexity averages over. We propose Per-Channel Reduction (PCR), a diagnostic that classifies each model into one of three mechanistic failure modes: outlier-crushes-safety, where safety lives in non-outlier channels collaterally damaged by outlier-driven scale factors; outlier-as-safety, where safety overlaps outlier channels and finer granularity cannot rescue it; and multi-layer dilution, where safety is distributed across many layers and per-layer fixes fail. PCR predicts the correct mitigation direction on all nine primary models and one held-out model from an independent family using 20 calibration prompts. PCR generalizes across unseen prompts, models, and quantizer families, including production quantizers (KIVI: up to 97.2% recovery) and rotation-based quantization (QuaRot-style Hadamard), succeeding where attention-based allocation methods fail. The resulting training-free protocol, requiring ~35 GPU-minutes, recovers up to 97% of lost alignment at under 1% latency and minimal memory overhead, addressing vulnerabilities confirmed in production vLLM serving with FP8 KV cache on NVIDIA GPUs.

cs.LG · cs.AI · cs.ET

版本更新 · v2arXiv:2611.06559

From Fixed Defaults to Calibrated Activation Policies for Long-Term Time-Series Forecasting

Shijie Xu, Boyan Tang, Kai He, Jianghua Wu

原文摘要

Long-term time-series forecasting supports decisions in energy, transportation, and other applications where reliable predictions are needed beyond the observed data. Although forecasting research has advanced model architectures, fixed activation configurations provide limited guidance on how activation choices should vary across datasets and prediction horizons. This study aims to establish when task-specific activation choices improve forecasting accuracy and how previous experiments can guide the selection of suitable configurations for new tasks. We develop a calibrated selection policy that learns a reliable default from prior experiments, adapts the activation function and its location to each task, and retains the default when evidence for a change is insufficient. In an evaluation supported by 1,225 Informer training runs across 17 dataset variants and 51 forecasting tasks, the policy reduces mean squared error by 7.86% on average relative to the conventional fixed-activation baseline and improves 72.55% of tasks. These findings establish activation choice and placement as practical opportunities for improving existing forecasting models and show how accumulated experimental evidence can guide configuration decisions without repeating an exhaustive search for every task.

cs.LG

新论文 · v1arXiv:2611.07463

FREE-LIVO: Physically Decoupled Yet Tightly Coupled LiDAR-Inertial-Visual Odometry

Reon Tabata, Kenji Koide, Shuji Oishi, Taku Okawara, Mizuho Aoki, Aoki Takanose, Masashi Yokozuka, Jun Miura

原文摘要

LiDAR-Inertial-Visual Odometry (LIVO) integrates complementary LiDAR, camera, and IMU measurements to achieve robust state estimation. However, most existing methods assume a fixed relative pose and accurate time synchronization between sensors. Toward modular and heterogeneous robotic systems, this paper presents FREE-LIVO, a tightly coupled odometry method that requires neither a rigid LiDAR-camera rig nor LiDAR-camera time synchronization. The proposed method uses a pretrained image feature matcher to establish correspondences between a camera image and a dense LiDAR intensity image generated from a single LiDAR scan using a generative model. From the resulting 3D point-to-pixel correspondences, we construct reprojection factors that directly constrain LiDAR and camera poses at different acquisition times. These factors are jointly optimized in a single factor graph together with constraints from point cloud registration, visual feature tracking, and the IMU measurements associated with each sensor. On six sequences from the public M3DGR dataset, FREE-LIVO achieves an average absolute trajectory error (ATE) RMSE of 0.45 m without provided LiDAR-camera extrinsic parameters and exhibits minimal accuracy degradation under time synchronization errors of up to 0.5 s. Furthermore, experiments with the LiDAR and camera mounted on separate robots demonstrate that the method can jointly estimate the trajectories of independently moving sensors.

cs.RO · cs.CV

新论文 · v1arXiv:2611.07462

Information-Spectrum General Formulas for the Capacity Region of Classical-Quantum Multiple Access Channels

Takuya Kubo, Hiroshi Nagaoka

原文摘要

For classical-quantum multiple access channels (cq-MACs) with two senders, we give several general expressions for the capacity region, valid for arbitrary sequences of channels. In the case of asymptotically vanishing error ($\varepsilon=0$), the capacity region is expressed in terms of spectral inf-mutual information rates. For every $0\le\varepsilon<1$, the $\varepsilon$-capacity region is expressed in terms of a quantity defined through a linear combination of three alternative hypotheses. In particular, the outer bound of Kubo and Nagaoka (2015) is tight. Our results include the quantum counterpart of the general formula of Han (1998) for the classical MAC. The most important ingredient is the one-shot inner bound of Sen (2021), which we apply to each member of an arbitrary channel sequence.

cs.IT · quant-ph

版本更新 · v3arXiv:2611.00177

AuroraWorld: Structured 4D World Modeling for Exo-to-Ego Video Generation in Robot Manipulation

Hao Xue, Peng-Fei Zhang, Ying Cheng, Lanyun Zhu, Lei Zhu, Heng Tao Shen

原文摘要

Egocentric observations provide a robot-centric view of manipulation processes, capturing contact interactions and object-state changes that are important for visuomotor policy learning but difficult to observe from external cameras. However, collecting such data at scale remains expensive due to additional hardware requirements and platform-specific deployment costs. Synthesizing egocentric observations from large-scale exocentric manipulation videos provides a scalable alternative. However, existing methods struggle to synthesize physically consistent and policy-relevant egocentric videos under substantial viewpoint transformations, severe occlusions, complex object interactions, and inherently sparse observations from available exocentric viewpoints. Code will be made publicly available at https://github.com/xueha0/AuroraWorld

cs.AI · cs.CV

新论文 · v1arXiv:2611.07461

The Last Mile to the Real World: Agent-Native Primitives as Reusable Building Blocks for Solving Real-World Tasks

Haibo Jin, Peng Kuang, Xiaopeng Yuan, Lucheng Fu, Jerry Wang, Xucheng Yu, Yaoning Yu, Dehao Wu 等 14 位作者

原文摘要

How can an agent handle a real-world task end to end? Frontier models with their own harnesses still fall short on real-world benchmarks: on Terminal-Bench 4.0, GPT-5.6 Sol with Codex resolves 37.3% of the tasks and Claude Opus 5 with Claude Code 51.8%. These tasks fail at the step from reaching a resource to applying it, which we call the last mile to the real world. Existing approaches put the resource within reach but leave the competence to apply it with the caller, so each harness rebuilds that competence. We introduce agent-native primitives, reusable units that equip a resource with a resident LLM that applies it in context and communicates with other primitives, and HEELO (Harness Engineering around Evolving Agent-Native Primitives for the Last Mile to Real-WOrld Tasks), a harness engineering framework that builds each task's harness from a library of such primitives. HEELO composes the primitives a task needs into a task-specific graph, verifies the result against evidence from the environment, revises the primitives a failure implicates, and admits new or improved primitives after validation. We instantiate HEELO with eight primitive families, each built for one kind of task, and evaluate them on more than twenty benchmarks. Dev-Primitives raise Terminal-Bench 4.0 resolution from 37.3% under Codex to 55.5% on the same backbone, Harness Primitives composed per task raise Terminal-Bench 2.0 Pass@1 from 60.0% for the seed harness to 72.2%, Tool Primitives raise ToolBench pass rate from 55.8% to 75.1%, and Navigation Primitives raise GAIA-Text-103 accuracy from 45.6% to 58.3%. On the 46 real-world tasks of our heelo.dev platform, Claude Opus 5.5 with Claude Code and GPT-6.1 Sol with Codex complete 9 each on their own and 39 and 35 when connected to HEELO through our MCP server, with 1.9$\times$ and 2.3$\times$ less time and 1.55$\times$ lower cost per completed task.

cs.CL · cs.AI · cs.LG · cs.MA · cs.SE

新论文 · v1arXiv:2611.07458

Physics-informed latent conditional flow matching for generative modeling of granular flows on inclined planes

Ziqi Han, Kuang Shi, Xuyang Li, Teng Man, Yimin Lu

原文摘要

The internal kinematic and mechanical states of granular flows are difficult to observe or measure, limiting the interpretation of internal particle interactions and their role in flow evolution. In this study, we propose a physics-informed latent conditional flow matching (PI-LCFM) framework for reconstructing complete multivariable flow sequences from partial boundary observations. A variational autoencoder and temporal flow-matching model learn a reusable latent prior over velocity, granular temperature, stress, and solid volume fraction. During inference, boundary-velocity or basal-stress observations and mass- and momentum-conservation constraints guide the latent trajectory through the decoded physical fields, without retraining either model. The results demonstrate that the model accurately reconstructs full spatiotemporal fields from observations containing only 0.3\% of the complete field information, extrapolates beyond the training ranges of inclination and particle number, and recovers key trends in runout scaling, energy evolution, and stress profiles. Compared with a frame-by-frame conditional flow-matching baseline, PI-LCFM reduces the mean velocity-component root-mean-square error by 45.2\%, while observation-guided reconstruction of 500-frame sequences achieves an approximately 44-fold speedup. Physics guidance further improves conservation consistency, supporting the broader potential of this framework for inferring hidden states in particulate systems, with potential applications envisioned in geophysical hazard assessment and industrial process monitoring.

cs.CE

版本更新 · v6arXiv:2603.20957

Alignment Whack-a-Mole : Finetuning Activates Verbatim Recall of Copyrighted Books in Large Language Models

Xinyue Liu, Niloofar Mireshghallah, Jane C. Ginsburg, Tuhin Chakrabarty

原文摘要

Frontier LLM companies have repeatedly assured courts and regulators that their models do not store copies of training data. They further rely on safety alignment strategies via RLHF, system prompts, and output filters to block verbatim regurgitation of copyrighted works, and have cited the efficacy of these measures in their legal defenses against copyright infringement claims. We show that finetuning bypasses these protections: by training models to expand plot summaries into full text, a task naturally suited for commercial writing assistants, we cause GPT-4o, Gemini-2.5-Pro, and DeepSeek-V3.1 to reproduce up to 85-90\% of held-out copyrighted books, with single verbatim spans exceeding 460 words, using only semantic descriptions as prompts and no actual book text. This extraction generalizes across authors: finetuning exclusively on Haruki Murakami's novels unlocks verbatim recall of copyrighted books from over 30 unrelated authors. The effect is not specific to any training author or corpus: random author pairs and public-domain finetuning data produce comparable extraction, while finetuning on synthetic text yields far little extraction, indicating that finetuning on individual authors' works reactivates latent memorization from pretraining. Three models from different providers memorize the same books in the same regions ($r \ge 0.90$), pointing to an industry-wide vulnerability. Our findings offer compelling evidence that model weights store copies of copyrighted works and that the security failures that manifest after finetuning on individual authors' works undermine a key premise of recent fair use rulings, where courts have conditioned favorable outcomes on the adequacy of measures preventing reproduction of protected expression.

cs.CL · cs.AI · cs.CY

版本更新 · v2arXiv:2608.06886

HazeSpikeMamba: Coupling Spiking-Inspired and State-Space Features for Self-Supervised Real-World Dehazing

Haoran Liu, Li Xu, Mingzhe Liu, Huibin Li, Peng Li, Guibin Zan, Simon X. Yang

原文摘要

Dehazing networks are commonly trained on synthetic hazy-clear pairs, but their performance often drops on real photographs. Synthetic haze generated using the atmospheric scattering model does not fully capture the variability of real haze, and paired real hazy-clear images are scarce. In this work, we propose HazeSpikeMamba, a compact dehazing framework that combines a spiking-inspired local path and an attentive state-space global path in a multi-scale U-Net. The local path uses TPCNNSpike, a new spike-emission scheme inspired by the neighborhood coupling of Pulse-Coupled Neural Network (PCNN). Unlike grouped directional scanning, TPCNNSpike updates all neurons in parallel using the previous firing states of their Gaussian-weighted neighborhoods. The global path adapts the Attentive State-Space Module of MambaIRv2, retaining semantic prompting and sequence reordering while removing the window self-attention branch. Its state-space processing models long-range dependencies with complexity linear in sequence length. For target-domain adaptation, a frozen degradation network, pretrained on paired NH-HAZE data, re-synthesizes haze from the dehazed prediction. The reconstruction error updates only the final restoration layers of HazeSpikeMamba without haze-free labels during adaptation. A shared checkpoint is adapted once on each complete unlabeled target set, making the evaluation dataset-level and transductive rather than zero-shot or per-image optimization. The forward network contains 2.02M active parameters and requires 13.27G nominal MACs (measured with thop at 256x256 input). This adaptation consistently improves BRISQUE and NIMA on RTTS, URHI, and HSTS. On RTTS, BRISQUE decreases from 30.13 to 27.72 and NIMA increases from 4.13 to 4.87. Under this transductive protocol, the adapted model also achieves the best BRISQUE and NIMA on URHI and HSTS among the compared methods.

cs.CV

版本更新 · v2arXiv:2608.02663

Variable-aware representation, not relation typing, explains the performance of an event-graph model for ICU mortality prediction: temporal and external validation

Mohammad Nasir Uddin, Eklachur Rahman Bhuiyan, Rahnuma Tabassum Orpita

原文摘要

Event-graph neural networks represent intensive care unit (ICU) records as typed entities linked by timestamped observations, and their gains are often attributed to relation-specific message passing. We tested which components matter once leakage is removed and validation is temporal and external. In 31,142 first ICU stays of at least 48 hours in the Medical Information Mart for Intensive Care IV (MIMIC-IV), models predicting post-48-hour in-hospital mortality were developed on 2008-2016 stays, tested on held-out 2008-2016 stays and on 2017-2019 and 2020-2022 stays, and applied without retraining to 54,696 stays from 205 hospitals in the eICU Collaborative Research Database. A leak in a previous version, total ICU length of stay used as an input, was removed. A variable-aware representation (a learned embedding of each node's variable plus per-variable normalisation) raised the mean single-model area under the receiver operating characteristic curve (AUROC) from 0.846 to 0.896 in 2020-2022 and from 0.699 to 0.809 externally. With it, relation-specific and shared (relation-collapsed) message passing differed by less than 0.005 AUROC in every test set, within a +/-0.01 equivalence margin, although relation typing had appeared beneficial without it. The revised model's AUROC was 0.878-0.900 across MIMIC-IV periods, 0.15-0.19 above three severity scores and 0.007-0.011 below gradient boosting. Externally, the five-model average reached AUROC 0.825 versus 0.788 for the Acute Physiology and Chronic Health Evaluation (APACHE) IVa score on the 48,184 stays with an APACHE prediction. In this model, representation mattered more than message-passing architecture.

cs.LG · cs.AI

版本更新 · v2arXiv:2611.01345

Unified dynamics of training, learning, inference, and reinforcement in neural networks

Mian Wang

原文摘要

Through nine nanoGPT experiments, we established training-learning-inference relations: realized updates elicit state-dependent finite-amplitude responses; learning persistently reorganizes distributed functional support, with target-specific readout-boundary crossings determining capability transitions; inference recruits learned support without changing the learned state. Factors identified in the fifth experiment yielded a direct predictor of post-update capability transitions, achieving 91.43% accuracy and 91.49% four-way macro recall across 5,088 targets. We tested these relations and prediction across five additional neural architectures. In a seventh system, concentrated correct feedback eroded previously mastered skills; rebalanced feedback restored their accuracy from 70.31% to 99.48%. Linking inference consequences to subsequent training closes these relations into unified training-learning-inference-reinforcement dynamics. We subsequently established a unified system of dynamical equations, used it to reconcile positive and negative findings across existing research, and derived mathematical relationships in other neural-network theories using only the assumptions of the original studies.

cs.LG

新论文 · v1arXiv:2611.07453

Latency-Aware Adaptive Best-of-$N$

Alon Jacobson, Qian Xie, Peter I. Frazier

原文摘要

Test-time compute (TTC) improves language model responses by spending additional computation at inference time, and the right amount of compute depends on the prompt. Adaptive TTC methods allocate compute based on signals of response quality such as observed rewards, often by stopping once the signal exceeds a fixed threshold, and generally measure cost by the number of generations or tokens. But in online serving, latency is also a cost: it grows with server load, so a method using a fixed threshold spends too little compute when generation is fast and too much when it is slow. For Best-of-$N$ sampling, a widely used TTC method, we propose a two-stage adaptive algorithm that draws an initial batch, then uses the observed rewards and the batch's latency to decide whether to draw a second batch and at what size by comparing expected improvement against predicted cost: it draws one when the first batch's best reward is below a threshold that decreases with the first batch's latency, without explicitly measuring server load. We map each reward to its value under the cumulative distribution function (CDF) of the prompt's reward distribution, so the same tradeoff parameters can be reused across tasks; a regression fitted offline on other prompts predicts a distribution for the CDF value of the first batch's best reward, giving the expected improvement in closed form. We evaluate the method by simulation on a preference-based alignment benchmark, measuring generations-equivalent cost: generations plus latency, converted to generations. The method reaches Best-of-$N$'s reward at lower generations-equivalent cost, and conditioning on the measured first-batch latency saves at least as much of that cost as fixed-threshold (latency-unaware) baselines, and more when latency is highly variable, when it grows with batch size, and when the second batch's latency is highly predictable from the first's.

cs.CL · cs.LG

版本更新 · v2arXiv:2605.19256

Distribution Matching Distillation without Fake Score Network

Youngjoong Kim, Deokyeong Lee, Jaesik Park

原文摘要

Distribution Matching Distillation (DMD) provides an effective distribution-level correction for few-step generation, while relying on an auxiliary fake-score network to track the evolving generative distribution. Recent work combines DMD with a flow-map objective to exploit the benefits of both methods: mode-seeking and mode-covering behaviors. Since this fake-score network increases memory and update overhead, we study whether this explicit tracker can be avoided when the generator has a flow-map structure. In this work, we propose Fake-Score-network-Free DMD (FSF-DMD), a DMD-style correction for flow-map generators, replacing the auxiliary fake-score estimator with a generator-induced surrogate. We observe that this surrogate can be derived from the definition of the fake velocity under a certain perturbation assumption. Building on this insight, we derive a DMD-style objective without a fake-side network, extend it to backward simulation for flow maps, and to a self-teacher variant for training from scratch. In our ImageNet-1K $256\times 256$ experiments, FSF-DMD is competitive with the studied DMD baselines when initialized from pretrained networks and when trained from scratch.

cs.CV

新论文 · v1arXiv:2611.07452

Understanding Self-Supervised Representation Learning for Encrypted Traffic

Zongxi Liu, Shuai Mao, Jie Chuai, Hongyang Du

原文摘要

SSL has been adopted for encrypted traffic classification to learn representations from unlabeled traces and reduce reliance on manual feature engineering. However, recent studies show that data leakage can inflate reported performance and that pretrained models can underperform shallow classifiers once leakage is controlled. These findings raise questions about what information SSL can learn from encrypted traffic and how different pretraining mechanisms exploit this information. In this paper, we systematically investigate these questions by examining the sources of learnable information, the effectiveness of contextual pretraining, alternative self-supervision mechanisms, and the limits of representation learning. We first find that visible protocol fields retain contextual dependencies that are strongest between adjacent bytes, while payload interventions establish the classification value of retained plaintext structure. Guided by these insights, we investigate byte-level contextual learning through ByteBERT and TrafficDCNet, showing that aligning prediction granularity with local dependencies produces more discriminative representations. We further examine cross-view contrastive learning through AlignTraffic and find that complementary supervision improves representation quality beyond contextual prediction alone. Across four public classification tasks, AlignTraffic outperforms shallow learning by 5.97% to 30.70%, with additional evaluations on industrial traffic demonstrating its effectiveness in broader settings. Finally, evaluations on Tor traffic and individual packets reveal limitations when application information is concealed or unevenly distributed. These findings provide a systematic understanding of how observable information and self-supervision mechanisms shape the effectiveness and scope of representation learning for encrypted traffic.

cs.NI

版本更新 · v3arXiv:2502.16800

The w-value: Market Implementation and gamma-Core Stability in Pollution-Control Games

Juanjuan Fan, Ying Wang

原文摘要

This paper proposes a novel solution concept, called the w-value, for cooperative pollution-control games. The w-value is axiomatically founded on three principles: Pareto Optimality (PO), Market Equilibrium (ME), and Fiscal Balance (FB), which together yield an efficient and implementable welfare distribution. We establish its existence and uniqueness under standard concavity and convexity assumptions, and provide a market mechanism that implements it through price constraints and an initial permit allocation. A key theoretical advantage is its strong stability: the w-value belongs to the gamma-core, which makes it resistant to coalitional deviations. Moreover, the w-value can be obtained through a simple step-by-step procedure, which facilitates its application to complex real-world problems. A numerical illustration demonstrates the step-by-step calculation process and confirms the gamma-core membership numerically. Finally, we apply the concept to international climate negotiations, showing how it provides a theoretical rationale for the shift from rigid mandatory emission reduction systems toward the flexible structure of Nationally Determined Contributions.

econ.TH · cs.GT

版本更新 · v2arXiv:2611.06157

Neural Stochastic Forecasting for Lateral Molecular Diffusion Simulations

Jacob Ede Levine, Mario Hidalgo-Soria, Chen Yun Wen, Yun Lyna Luo, Sai Chandra Kosaraju

原文摘要

Predicting the long-time diffusion of membrane proteins is challenging because molecular dynamics (MD) simulations are computationally expensive, while experimental single-particle tracking is often limited by finite observation times. Here, we introduce a neural stochastic (NS) model for forecasting the collective diffusion of membrane proteins from coarse-grained MD trajectories. The model combines a transformer-based representation of protein configurations, recurrent networks operating at multiple temporal scales, and a stochastic latent process with learned memory. A cross-attention decoder then generates correlated protein displacements while accounting for the instantaneous system geometry and periodic boundary conditions. We evaluate the model in two membrane systems: seven interacting PIEZO channels and a mixed system containing seven PIEZO and ten TREK channels. For the seven PIEZO system, the NS model reproduces the anomalous diffusion exponent measured in MD simulations $α\sim 1$, substantially improving on a previously developed variational autoregressive model that exhibits superdiffusive behavior. The model also generates trajectories extending to 40,000 frames and captures the distinct diffusion characteristics of PIEZO and TREK populations in the mixed system. Comparisons of time-averaged mean-square displacements, Van Hove displacement distributions, and non-Gaussian parameters show that the model preserves key statistical features of the simulated dynamics. These results establish neural stochastic modeling as a promising approach for generating long trajectories of multi-protein membrane diffusion and extending the statistical analysis of molecular dynamics simulations.

cs.LG

版本更新 · v2arXiv:2607.03236

TACG: Trajectory-Aware Commit Gating for Diffusion Language Model Decoding

Chengcheng Wang, Tingzhang Luo, Wenhao Li, Jianyuan Guo, Chang Xu

原文摘要

Diffusion language models predict many masked positions at each denoising step, but their efficiency and output quality depend on which predictions enter the evolving context. Confidence is informative but incomplete for this decision: at matched confidence, proposals with different identity and support histories exhibit different one-step revision rates. We propose Trajectory-Aware Commit Gating (TACG), a training-free decoder that separates the current token proposal from the decision to commit it. TACG scores temporal support for the current proposal by contrasting its logits with an exponential moving average of earlier logits, requires the proposal identity to persist through a History Gate, and promotes a capped number of additional positions per step. The gate decides which positions to commit and when; every written token remains the contemporaneous base proposal. On complete four-task LLaDA evaluations under the stated single-seed protocol, measured task-specific operating points reach 51.02\% mean accuracy with 67.82 mean model evaluations, compared with 48.00\% and 76.59 for Native Confidence. Under a matched per-event commit-count replay, history-guided selection exceeds confidence ranking by 8.3 accuracy points at identical per-sample NFE. History-reference interventions alter commitment timing while preserving the base-proposal write rule. Experiments on Dream and LLaDA2-Mini characterize quality--cost behavior across backbones and generation lengths. The code is publicly available at https://github.com/Clarence-CV/TACG-DLLM.

cs.CL

版本更新 · v2arXiv:2611.01844

OpenNWM: A Generalist Navigation World Model with Latent Action Pretraining

Dujun Nie, Ruiqi Song, Zixuan Wang, Hanqing Wang, Yifan Han, Baiyong Ding, Wenzhao Zheng, Hangbin Wu 等 9 位作者

原文摘要

Navigation world models predict future observations conditioned on navigation actions to support planning, but their generalization across real-world environments remains constrained by the limited coverage of action-labeled data. Action-free navigation videos provide diverse visual experience, yet lack the recorded actions required for action-conditioned training. We introduce OpenNWM, a generalist navigation world model with latent action pretraining. It learns universal latent actions from large-scale action-free videos through visual prediction pretraining, and further aligns them with physical motions via action-conditioned post-training. To support video pretraining, we curate NavAnywhere, a large-scale action-free navigation video dataset comprising 17.5 million visual observations from 15 heterogeneous sources. It captures navigation by ground robots, humans, and drones, with diverse viewpoints and motion patterns across five broad scene types: residential areas, public and commercial spaces, urban settings, parks, and natural landscapes. Experiments demonstrate improved visual prediction on in-domain and out-of-domain benchmarks, together with improved navigation performance. These results show that latent-action pretraining enables diverse, action-free videos to improve dynamics learning and generalization for open-world navigation.

cs.RO · cs.AI

版本更新 · v2arXiv:2605.17204

Event-Grounded Sparse Autoencoders for Vision-Language-Action Policies

Xinchen Jin, Aditya Chatterjee, Pranav Kumar, Rohan Paleja

原文摘要

Vision-Language-Action (VLA) policies translate language and visual inputs into robot actions, where their hidden representations directly shape closed-loop behavior. However, mechanistic interpretability tools from language and vision-language models do not transfer cleanly to VLAs: outputs are robot actions rather than human-readable tokens, and validating internal changes requires expensive closed-loop rollouts. We propose an event-grounded interpretability pipeline that anchors SAE feature analysis to behavioral events rather than text contexts. We extract keyframes from end-effector trajectories and cluster them within each task using visual, robot-state, and temporal cues. The resulting events guide SAE feature ranking, with optional VLM annotations providing descriptions of the events. To our knowledge, our pipeline is among the first to ground SAE-based VLA analysis in closed-loop behavioral events. Across two simulation architectures and a real-robot study, event-grounded ranking yields the strongest causal effects on OpenVLA and transfers to the continuous action chunks of $π_{0.5}$. SAE is a sparse but imperfect intervention basis: usability varies with architecture and intervention site, and aggressive intervention reveals safety and interpretability limits. Overall, event-grounded SAE analysis emerges as a practical starting point for behavior-anchored VLA interpretability, motivating future work on SAE analysis beyond end-effector motion, finer-grained closed-loop evaluation, and safe interventions for high-stakes VLA deployments.

cs.RO · cs.AI

版本更新 · v2arXiv:2611.05793

ZipDrift: Generating Multimodal Short-Horizon MPC Costs with One-Step Drifting Models

Ke Zhang

原文摘要

Short-horizon model predictive control (MPC) is fast but myopic. ZipMPC compensates by learning a context-dependent correction of the short-horizon cost that imitates a long-horizon expert. We observe that the optimal correction is often multimodal: when a slower opponent can be passed on either side, good corrections of opposite sign coexist, and the expert's choice between them hinges on tiny cost differences. Deterministic regression then averages modes or inherits inconsistent expert tie-breaking. We propose ZipDrift, which trains a one-step drifting generator over MPC cost corrections, tilted toward low closed-loop cost and learned only from cost evaluations of candidate corrections, without differentiating through the optimiser. A few generated candidates are ranked by a learned critic, and the model-based MPC then enforces dynamics and constraints. In a kinematic overtaking benchmark derived from ZipMPC, more than half of the contexts are bimodal and the manual short-horizon MPC crashes in 28% of them. With an eight-dimensional stagewise correction, evaluated on 1000 fresh contexts and five training seeds, ZipDrift with eight candidates crashes in 0.74% of episodes, compared with 3.4% for an idealised ZipMPC-style regressor and 2.1% for critic-ranked random candidates with the same budget. A mixture density network (MDN) with the same selection is equally good (0.76%). The benefit of drifting lies in sample quality: in a pre-specified test on further unseen contexts, a single drifting sample crashes in 8.7% of episodes against 13.2% for the MDN, and ZipDrift needs four candidates instead of eight to fall below 1%. With many candidates, all methods degrade as selection exploits errors of the critic.

cs.RO · cs.LG · eess.SY

新论文 · v1arXiv:2611.07444

Two-stage residual learning for calibrated uncertainty in solar irradiance forecasts

RR Flanagan, E Kasha, PW Talbot, AS Epiny, MR Deinert

原文摘要

Accurate forecasting of renewable generation requires not only point predictions but also reliable estimates of the uncertainty attached to them, particularly at extended horizons. Machine learning models are increasingly used to predict solar output, but they typically report a single value and capture the associated uncertainty poorly. The standard remedy is to train a model to predict the parameters of an assumed probability distribution directly. Here we develop an alternative two-stage method, motivated by synthetic solar signal generation, in which the forecast error is modeled separately from the expected irradiance. A first long short-term memory (LSTM) network predicts the expected global horizontal irradiance, and a second network models the distribution of the forecast error, which is added back to the mean to give both the forecast and its uncertainty. Using historical weather data from the National Solar Radiation Database and a probability integral transform evaluation, we show that explicitly modeling the error brings the predicted standard deviation markedly closer to the historical value, reducing the deviation between the two by 34.6 percent for a standalone LSTM and by 56.5 percent when the LSTM is coupled to Gaussian process regression, the most reliable of the configurations tested. Because the method models the total forecast error directly, it captures the full predictive uncertainty that matters for decisions, rather than only the model contribution that dropout-based approaches estimate.

cs.LG · eess.SY

版本更新 · v2arXiv:2609.34915

Muon Sublates the Edge of Stability in LLM Pretraining

Yanzhe Chen, Qifang Zhao, Xiaoxiao Xu, Fanghui Liu

原文摘要

Muon is increasingly used for language-model pretraining, yet its large-step dynamics are not captured by the classical edge-of-stability (EoS) picture of gradient descent (GD). In GD, loss neutrality, equal-magnitude update reversal, and marginal stability meet at a single learning-rate-dependent edge. We show that Muon breaks this coupling. For stochastic no-momentum Muon, we derive a coherence-corrected conditional loss-neutral boundary $2ρ_b/η$, while temporal alignment follows a separate geometry. Controlled experiments show that loss balance and temporal alignment respond differently to learning rate and batch size. Across our language model experiments, the 130M Llama-like LLM runs exhibit loss-boundary tracking with weak negative alignment, whereas the studied 1B LLM configuration shows stronger partial cancellation; in both settings, directions remain far from coherent reversal while training continues to improve. These results support a split EoS picture for Muon: a stochastic loss-neutral edge survives, but it is not accompanied by a universal temporal-direction signature. The source code for reproducing the experiments can be found in https://github.com/cyzebra/Muon-Sublates-the-Edge-of-Stability-in-LLM-Pretraining

cs.LG

新论文 · v1arXiv:2611.07443

VIPA: View-Invariant Projector-Based Adversarial Attack

Jiyu Han, Qingyue Deng, Bingyao Huang

原文摘要

Projector-based adversarial attack aims to physically manipulate real-world scenes by projecting adversarial patterns, thereby causing deep image classifiers to produce incorrect predictions. However, existing stealthy projector-based adversarial attack methods model the project-and-capture process in a single and static view, limiting their view-invariant capability and practical applicability in dynamic environments. In this paper, we introduce View-Invariant Projector-Based Adversarial Attack (VIPA), a novel method designed to overcome these limitations by achieving both view-invariant and classifier-agnostic adversarial attacks in the physical world. We first leverage a view-invariant projector-camera system simulation method to model the physical interactions between projected patterns and real-world surfaces. To ensure robustness and stealthiness of the attack across different views, VIPA optimizes adversarial projections by aggregating simulated attack losses from multiple views. This joint optimization across diverse views helps maintain robustness regardless of the viewpoints. Finally, the optimized patterns are physically projected into real-world scenes, where they successfully fool various classifiers from different viewpoints, thereby enabling robust and practical view-invariant adversarial attacks. Our experiments in both targeted and untargeted attacks demonstrate that VIPA consistently achieves higher attack success rates than existing methods, while also enhancing stealthiness and ensuring minimal perceptual degradation across all tested views.

cs.CV

版本更新 · v4arXiv:2508.03829

Majority Bit-Aware Watermarking For Large Language Models

Jiahao Xu, Rui Hu, Zikai Zhang, Olivera Kotevska

原文摘要

The growing deployment of Large Language Models (LLMs) has raised concerns about their misuse in generating harmful or deceptive content. To address this issue, watermarking methods have been proposed to embed identifiable multi-bit messages into generated text for misuse tracing. However, existing methods often suffer from a fundamental trade-off between text quality and decoding accuracy. In particular, they have to restrict the size of the preferred token set (i.e., green list) during encoding to maintain a detectable watermark signal for decoding, which inevitably degrades generation quality. To improve this trade-off, we propose a novel message encoding paradigm called \textit{majority bit-aware encoding}, which relaxes the watermark signal strength from the green list size. This strategy allows for a strong watermark signal to be preserved in generated texts even when using a large green list. We introduce two instantiations of this paradigm: MajorMark and MajorMark$^{+}$, where the latter is specifically optimized for long messages. Extensive experiments on state-of-the-art LLMs demonstrate that our methods achieve higher decoding accuracy and superior text quality compared to prior baselines. The code is at https://github.com/JiiahaoXU/MajorMark.

cs.CL · cs.CR

新论文 · v1arXiv:2611.07442

Req2Rust: A Requirements-Guided Agent Harness for Project-Level C++-to-Rust Migration

Shengbo Wang, Mingwei Liu, Yuwen Chen, Yanlin Wang

原文摘要

Memory-safety concerns make Rust an attractive target for modernizing legacy C/C++ systems, yet project-level C++ migration must balance semantic fidelity with Rust-oriented reconstruction. Source-structure-preserving migration can track low-level behavior, but it can also carry source-language organization, unsafe idioms, and project assumptions into Rust. Direct agent-based translation gives the model more freedom to construct a Rust project, but semantic coverage depends on the model identifying the source behaviors that must be preserved. This problem is especially acute for C++, where source behaviors are often hidden in object models, destructor-driven lifetime protocols, and other hard-to-localize C++ semantics. This paper presents Req2Rust, the first project-level LLM-based C++-to-Rust migration harness. Req2Rust first derives an evidence-grounded Migration Requirements Specification (MRS) that records the behaviors and constraints that must be preserved. It then synthesizes a Rust Design Specification (RDS) and realizes it as an initial Rust project. Because real projects contain implementation details and idioms that only become visible during build and validation, Req2Rust treats the initial Rust project as an executable refinement state. Compiler, semantic, and unsafe-reduction validators produce validation reports for improving buildability, behavioral alignment, and unsafe usage. We evaluate Req2Rust on general C++ projects and modules from the large-scale Openharmony codebase against a ClaudeCode baseline. Req2Rust produces buildable Rust crates in both settings and improves test pass rate from 69.5% to 91.9% on general projects and from 45.4% to 87.9% on Openharmony modules.

cs.SE

新论文 · v1arXiv:2611.07441

Closed-Loop Physics-Based Manipulation Evaluation at a Commodity Price: What 4.4 L4 GPU hours at $3.50 Buys You

Rohun Kshirsagar, Diana Rypkema

原文摘要

Running evaluations of contact-rich robotic manipulation at statistical scale has largely been possible only at well-resourced labs, highlighting the need for affordable and efficient options for broader audiences. We present a reproducible evaluation harness wrapper based on a previous study at 200-rollout scale on serverless commodity GPUs, at a cost of approximately $3.50 USD and 38 minutes of wall-clock per replication (4.4 L4-GPU-hours across eight parallel shards). Using the harness, we report a partially controlled negative result on action chunking with sparse data: three independently trained chunked checkpoints produce zero genuine successes across 780 rollouts, while three single-step checkpoints range from 2.6% to 11.8% success rate. A per-step-replanning control partially isolates chunk prediction, rather than open-loop chunk execution as the mechanism, with the failure localized to the training formulation and multi-step prediction. This work further democratizes physics-based policy development and evaluation through our reusable harness.

cs.RO

版本更新 · v2arXiv:2603.15221

Preference-Guided Adversarial Policy Optimization for Long-Tail Robust Driving

Tong Nie, Yihong Tang, Junlin He, Yuewen Mei, Jie Sun, Lijun Sun, Wei Ma, Jian Sun

原文摘要

Deploying autonomous driving systems requires robustness against long-tail scenarios that are rare but safety-critical. While adversarial training offers a promising solution, existing methods typically decouple scenario generation from policy optimization and rely on heuristic surrogates. This leads to objective misalignment and fails to capture the shifting failure modes of evolving policies. This paper presents ADV-0, a closed-loop min-max optimization framework that treats the interaction between driving policy (defender) and adversarial agent (attacker) as a zero-sum Markov game. By aligning the attacker's utility directly with the defender's objective, we reveal the optimal adversary distribution. To make this tractable, we cast dynamic adversary evolution as iterative preference learning, efficiently approximating this optimum and offering an algorithm-agnostic solution to the game. Theoretically, ADV-0 converges to a Nash Equilibrium and maximizes a certified lower bound on real-world performance. Experiments indicate that it effectively exposes diverse safety-critical failures and greatly enhances the generalizability of both learned policies and motion planners against unseen long-tail risks.

cs.LG · cs.AI

新论文 · v1arXiv:2611.07437

IMUDINO: Self-distilled Foundation Model for Inertial Sensing

Zhoutong Ye, Chengwen Zhang, Zixuan Wang, Xufeng Jian, Zhe He, Qingyang Wan, Xutong Wang, Jiaqi Liu 等 12 位作者

原文摘要

We present IMUDINO, the first foundation model (FM) for inertial sensing to achieve competitive performance across diverse sensing tasks including human activity recognition (HAR), health, motion capture, and odometry. Prior inertial FMs often focus on one subfield of inertial sensing and sometimes underperform state-of-the-art supervised methods (e.g., in HAR), making their scope considerably narrower than FMs for vision and natural language. Moreover, they often require fixed input length and sampling rate, and generalize poorly to unseen sensor locations. To build a general-purpose foundation model for inertial sensing, we adapt the DINO self-supervised learning (SSL) paradigm to inertial data through a set of semantics-preserving motion transformations that simulate variations in motion pattern and sensor setup. For model architecture, we use a plain transformer encoder with frequency-aware positional encoding that natively supports varying input length and sampling rate. This simple setup proved remarkably effective for learning global and dense features generalizable across users, sensor setups, tasks, and, in the case of robot odometry, embodiments. We hope IMUDINO can serve as a backbone for inertial sensing and kickstart the development of general-purpose inertial FMs.

cs.LG · cs.AI · cs.CV

版本更新 · v2arXiv:2607.10630

World Models as Adversaries: Multi-Agent Self-Play Fine-Tuning for Robust Motion Planning

Tong Nie, Yuewen Mei, Junlin He, Yihong Tang, Jian Sun, Wei Ma

原文摘要

Robust motion planning in dense traffic requires autonomous vehicles to interact in rare and safety-critical scenarios that are underrepresented in naturalistic driving data. Although adversarial training offers a feasible solution, existing methods often rely on external scenario generators, heuristic perturbations, or simulator-heavy rollouts, which makes them difficult to integrate with modern autoregressive planners. Here, we cast adversarially robust planner learning as a constrained min-max game and propose Adversarial World Modeling (AWM), a theoretically grounded multi-agent self-play fine-tuning framework. Since solving the exact game is intractable, AWM introduces a principled decoupled solver. In the inner minimization, the planner's predictive world model is converted into a role-conditioned adversary that learns sparse, scene-adaptive attack coalitions via counterfactual credit assignment. In the outer maximization, the ego planner optimizes a regret-aware robust best response against the frozen AWM, utilizing tail-risk weighting and reference-anchored trust regions to improve hard-case recovery while preserving nominal driving behavior. Experiments on the nuPlan and InterPlan benchmarks demonstrate that our method generates transferable adversarial interactions and yields a robust planner that achieves competitive closed-loop performance in both nominal and highly interactive long-tail scenarios. Theoretical analysis justifies the decoupled solver and the main optimization components.

cs.RO · cs.AI

版本更新 · v2arXiv:2611.06623

Amortized Optimization for Calibrating Stochastic Social Simulators

Muyao Zhong, Peng Yang

原文摘要

Stochastic simulators need suitable parameter values to reproduce observed behaviour in complex systems. Searching for these values through repeated simulations makes frequent recalibration computationally expensive. We develop an amortized optimizer that uses simulations collected in advance to learn how to choose parameters that make simulated outcomes agree with observations. Once fitted, it maps new observations to feasible parameters without further simulator calls. The method retains alternative parameter settings that can explain similar observations and compares their responses, avoiding the conflicting labels that arise when training a single parameter regressor. Historical observations also provide supervision by showing which simulated responses agree with the data, even when the underlying parameters are unknown. Experiments across four simulators compare calibration accuracy and query time. Dedicated cost analyses quantify traffic preparation and the recorded financial reuse costs. On new traffic targets, local correction more than halves query time relative to empirical and surrogate selection; a learned inverse is faster still, with similar mean error. In a retrospective comparison on real financial data, our method returns parameters about $1.07\times10^{7}$ times faster than a 10-minute Gaussian-process Bayesian optimization search, with mean calibration error 3.06% higher. Reusing simulation work across observations supports repeated parameter updates when the available time is too short for a new search.

cs.CE · cs.CY