Inference
-
One-Sided Expert-Parallel All-to-All for Mixture-of-Experts Inference over NVLink
2026-11-09 20:06:38Bo Li, Chulian Zhang, Or Biri, Anthony Chang, Anton Korzh, Noam Duvdevani, XingFei Xi, Daniel Stokes, Zhenhuan Chen, Ludwig Schneider, Artem Y. Polyakov, Julien Demouth, Chandler ZhouAbstract
Expert parallelism (EP) shards the experts of a Mixture-of-Experts (MoE) layer across GPUs, so every MoE layer requires two all-to-all communications, namely dispatch and combine. Compared with training, inference processes relatively few tokens per GPU at a time, and at small batch sizes fixed costs that do not grow with the number of tokens, rather than data movement, dominate the latency. We present NVLinkOneSided, a one-sided EP all-to-all for GPUs within a single NVLink domain. Each source rank pushes its tokens, without first exchanging metadata, into rank-major buffers on the target GPUs, and the grouped GEMMs read them in place instead of from a per-expert copy. We provide two protocols for the synchronization that the communication requires, namely Fence and Sentinel, for high-throughput and low-latency scenarios respectively. On Blackwell GPUs, NVLinkOneSided completes dispatch and combine 1.2-3.0$\times$ faster than DeepEP, with the largest speedups at small batch sizes. The resulting MoE layer is up to 2.4$\times$ faster than DeepEP with DeepGEMM. It even outperforms the fused DeepGEMM MegaMoE at up to 128 tokens per GPU, making it still competitive for decoding. We expose NVLinkOneSided through a modular interface, so that it is agnostic to the activation quantization format, decoupled from the MoE compute backend and easy to integrate into inference frameworks.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13188
PDF
-
Continuous Angular Control in Text-to-Image Generation via Geometric Token Encoding
2026-11-09 19:55:34Furkan G\"uzelant, Basavaraj Sunagad, Aysegul Dundar, Adam KortylewskiAbstract
Text-to-image diffusion models can render a photorealistic analog clock but cannot control the time it displays. Across four pretrained models, fewer than 1% of generated clocks place both hands within 6 degrees of the requested time; the models instead collapse onto visually dominant configurations such as the canonical 10:10 pose. The same failure appears when a specific fraction of a circular object is requested, where models return complete or conventionally sliced objects. We observe that these two failures share a geometric structure: both target quantities can be parameterized by angular variables. We introduce a geometric token encoding framework that conditions generation directly on these angles. Each angular component is encoded as a periodic sine-cosine feature and projected into a role-specific token embedding that is injected into the prompt before a frozen text encoder. To spatially ground these tokens without requiring localization input at inference, we introduce Stochastic Localization Relaxation (SLR), which stochastically relaxes attention supervision from component-level to object-level to unconstrained localization during training. For clocks, we additionally introduce a hand-swap preference objective that discourages assigning the correct angles to the wrong hand roles. On analog clocks our method reaches 63-78% Acc@6 degrees across evaluation methods, compared with below 1% for the baselines. On circular-fraction generation, it achieves 89.5% Acc@0.10, while maintaining 88.8% on unseen object categories and 84.9% on unseen target values, compared with roughly 20% for pretrained baselines.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13172
PDF
-
Which Successes Make Good References? Action-Sequence Alignment for Multi-Turn On-Policy Distillation
2026-11-09 19:18:25Mufei Li, Nikki Lijing Kuang, Zhuoer Wang, Samson Koelle, Jie Yuan, Manqing Mao, Hong Wang, Yanjun Lin, James Feng, Daniel Edmiston, Pan Li, Wei NiuAbstract
On-policy distillation (OPD) reduces train-inference distribution mismatch by supervising a student on its own trajectories. For multi-turn tool-use agents, conditioning the teacher on a validated same-task successful trajectory as a reference can provide useful hindsight, but different successes may reach incompatible interaction states and offer misleading supervision. We cast successful-reference selection as a free-target optimal transport problem, which reduces to assigning each failed trajectory its most compatible successful reference. Because latent state paths are hidden, we use action-sequence alignment as an observable proxy for compatibility and instantiate it with longest common subsequence (LCS), which identifies a shared ordered backbone of actions while allowing recoverable detours. This perspective makes trajectory compatibility the central criterion for reference selection in multi-turn OPD. Experiments on ALFWorld and the airline domain of $\tau^3$-bench show that LCS is the only evaluated reference selector to improve over reference-free OPD on both ALFWorld overall scores and all three airline metrics. Without further optimization, the airline-trained models also transfer to the retail domain, where LCS improves all three metrics.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13135
PDF
-
Catch It Where It Happens: Real Time Error Correction During LLM Reasoning
2026-11-09 19:00:40Rotem Elimelech, Michael Shalyt, Ido KaminerAbstract
Large language models (LLMs) are increasingly trusted with mathematical, scientific and engineering work, yet their mistakes are often confidently written and hard to spot. A single wrong step tends to propagate, making the model waste inference tokens following a false premise. Existing safeguards either check only the final answer and restart from scratch, or rely on a verifier LLM that competes for GPU capacity and slows down response time. We propose online, deterministic verification: many kinds of reasoning steps can be validated by code, so a step can be verified in real time, as it is written, without any model call or GPU use. Whenever a check fails, we point out the mistake and let the model continue from the same spot. We demonstrate this approach on symbolic mathematics, where derivations already contain explicit expressions that a computer algebra system can check in milliseconds. Tested on the ASyMoB benchmark with three open-weights models, our online validation improves accuracy by 1.4-5.6 percentage points, corresponding to a 1.8-14.9% net reduction in incorrect answers, while increasing token usage by only 6-7%. Because the verifier is independent of the model, the approach complements progress in LLMs rather than competing with it: every new model can benefit immediately, and the benefit grows as deterministic checkers extend to more kinds of reasoning steps.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13107
PDF
-
One Temperature to Rule Them All?
2026-11-09 19:00:01Aviv Orly (School of Physics and Astronomy, Tel Aviv University, Israel), Ori Shem-Ur (School of Physics and Astronomy, Tel Aviv University, Israel), Yaron Oz (School of Physics and Astronomy, Tel Aviv University, Israel)Abstract
Inference-time sampling is a central mechanism for improving large language models, with temperature controlling the tradeoff between reliability and diversity. While recent work suggests that varying temperature across samples can improve multi-sample inference, simple fixed-temperature schemes remain the standard in practice. We investigate this discrepancy in the canonical pass@K setting, asking when temperature diversity is useful, and when a single tuned temperature suffices. To this end, we analyze per-question success probabilities across multiple models and datasets, tracking how they vary with sampling temperature. We identify several consistent properties of these profiles, including that harder questions are less likely to share efficient sampling temperatures, although this trend is only logarithmic in hardness. We then derive an efficient method for evaluating the theoretically optimal multi-temperature allocation scheme, and show that although such a scheme improves performance, its gains over single-temperature baselines are typically modest when temperature is tuned as a function of the sampling budget K. We explain this phenomenon through mathematical mechanisms that we call dulling effects, and show how they ensure the robustness of temperature choices across different question distributions. Our results clarify when multi-temperature sampling is useful, and how temperatures should ideally be chosen.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13077
PDF
-
APS-VLA: Action-Guided and Future-Aware Token Pruning for Efficient Vision-Language-Action Models
2026-11-09 09:28:08Xiaoyang Xing, Chun-Mei Feng, Anthony Cohn, Xiaoliang Gong, Asoke NandiAbstract
Vision-language-action (VLA) policies have demonstrated strong generalization in robotic manipulation, yet processing dense visual tokens often incurs high inference latency. Visual token pruning offers a practical way to reduce this cost, but identifying a compact set of tokens that preserves reliable action generation remains challenging. Existing methods largely rely on static or immediate heuristic cues to estimate token importance. We introduce APS-VLA, a framework that learns visual token selection from robot demonstrations while keeping the pretrained VLA backbone frozen. A lightweight Decision Influence Estimator scores tokens conditioned on current visual observations, language instructions, and robot proprioception. Its training combines two complementary signals: action supervision matches both the full-input policy predictions and demonstrated actions, while an auxiliary visual transition objective uses demonstrated actions to predict subsequent visual features from the selected representation. At deployment, the estimator performs a single hard selection and packing step before the policy's native action-generation procedure; the auxiliary predictor is removed. Experiments with OpenVLA-OFT on LIBERO and with $\pi_{0.5}$ and XR-1 on RoboCasa365 demonstrate inference acceleration with modest changes in task success. Across five real-world tasks on two dual-arm platforms, APS-VLA reduces mean policy inference latency from 137.25 to 93.25 ms, achieving a $1.47\times$ speedup with an average success rate of 61%, compared with 60% for the unpruned baseline. Together, these results support APS-VLA as a modular, plug-in framework for efficient robotic control across VLA architectures, preserving the underlying policy parameters and native action-generation procedure.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13030
PDF
-
khammoon: From a Linguistic Guideline to a Thai Morpheme Segmenter
2026-11-09 03:19:53Warit Sirichotedumrong, Adisai Na-Thalang, Chanakan Wittayasakpan, Kritsadha Phatcharoen, Kunat PipatanakulAbstract
Thai is written without spaces between words, making segmentation a foundational step in most Thai NLP pipelines. Existing tools primarily reproduce the word-level conventions of corpora such as BEST; however, because the notion of a "word" in Thai lacks a unified linguistic consensus, these conventions struggle with productive compounding, leading to arbitrary boundaries and ambiguity in downstream representations. Instead, we propose operating at the morphological level, focusing on khammoon, the smallest atomic unit that carries meaning. Segmenting at the morphemic level provides an explicit, consistent morphological representation that eliminates compounding ambiguity. Crucially, this representation benefits downstream learning: models trained on morphemes perform on par with or exceed word-level baselines across part-of-speech tagging, named-entity recognition, and sentiment classification, demonstrating that linguistic consistency improves representation quality without downstream performance penalty. To operationalize this representation, we translate linguistic principles into a formal decision guideline with near-perfect human agreement ($\kappa \approx 0.99$), scale annotation with an LLM, and correct systematic errors with human labels. Finally, we train khammoon, a lightweight 1M-parameter convolutional BiLSTM segmenter designed for efficient CPU inference, achieving a boundary F1 of 0.992 on out-of-domain text. The guideline, datasets, model, and library are all publicly released.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13012
PDF
-
G-STAR: Group-Scaled Token Capacity for Deeply Compressed Autoregressive Generation
2026-11-08 16:59:18Xu Huang, Ye Huang, Xiaotong Li, Zhengrong Yue, Jialv Zou, Minhao Jing, Haodong Li, Wenhao Yan, Yaxin Luo, Duomin Wang, Daquan ZhouAbstract
Discrete autoregressive models offer a scalable framework for image generation. However, they face a fundamental trade-off: long token sequences preserve representation quality but slow generation, while aggressive spatial compression reduces representation capacity. Based on product quantization, we show that, under a fixed nominal information budget, group size should scale quadratically with the per-axis spatial compression factor. Building on this principle, we propose GSTAR, a spatial--group co-scaling framework that enables \(32\times\) and even effective \(64\times\) compression while preserving strong reconstruction, understanding, and generation performance. Since larger groups introduce intra-group dependencies that must be modeled efficiently, we analyze their impact on inference latency and generation quality and introduce a semi-autoregressive decoder that combines parallel feature computation with lightweight causal modeling. With these designs, our highly compressed models perform competitively with leading models across five representative text-to-image benchmarks (GenEval, DPG-Bench, PRISM, LongText-Bench, and Qwen-Image-Bench). At \(1024\times1024\) resolution, our model delivers a \(20\times\) speedup over prior discrete autoregressive models, generating each image in 2.5 seconds on a single NVIDIA H200. Together, these results show that highly compressed discrete representations can enable high-quality autoregressive generation at low latency.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13004
PDF
-
ZeroPoint: Lightweight Energy-Based Decision Model
Abstract
Energy-based models (EBMs) assign a scalar energy to every joint value of the observed and predicted variables. In decision models, the prediction is chosen from a finite list of options provided with each input. Inference then reduces to comparing the energies of the options, and training lowers the energy of the correct option relative to the others. A complete decision must also account for the possibility that no option is acceptable. However, the energies learned by a standard EBM define only relative compatibility: adding the same input-dependent quantity to all energies leaves the predictions, the Gibbs distribution, and the usual training losses unchanged. The model therefore has no learned reference level against which to decide whether any option fits. We propose ZeroPoint, an EBM that makes the empty answer a state of the model with energy zero for every input. This fixed state provides an explicit reference for the energies of all listed options, whose non-negative depths measure their compatibility with the input relative to the empty answer. The options and the empty answer share one distribution, in which an option at zero depth receives no probability. The decision to select an option or to abstain is thus part of ordinary inference, trained with the negative log-likelihood loss, rather than a separate rejection rule. The energy function is a small multilingual encoder that reads the input together with all the options and is invariant to their order. Ablations show that most of the benefit comes from training with examples in which no option fits. With the same encoder and data, the fixed zero separates inputs with and without an acceptable option slightly better than a learned rejection class and clearly better than a thresholded softmax, while leaving the choice among options unchanged.
Abstract (translated)
URL
https://arxiv.org/abs/2611.12992
PDF
-
Partition, Refine, and Detect: Preserving Novel Category Capacity in Open-Vocabulary Object Detection
2026-11-08 04:13:25Wei Zhang, Zhu Teng, Zhixiang Chi, Junnan Zou, Yi Liu, Jianping FanAbstract
Open-vocabulary object detection (OVD) expects the model trained on base classes to recognize both base and novel classes at inference. Typical methods rely on CLIP-style vision-language embeddings, yet the learned space is often biased toward base categories, with limited effective capacity preserved for novel concepts. Such a base-dominated representation can compress novel semantics, leading to semantic collapse and mis-calibrated confidence at inference. To address this issue, we present PR-OVD, a partition-and-refine framework that explicitly reserves capacity for novel categories and strengthens category separation, while being seamlessly integrated into different detector architectures. Specifically, PR-OVD first performs coarse partitioning to build global divisions for base, novel, and an explicit background partition in both visual and textual modalities, preventing object-irrelevant embeddings from leaking into category-specific representations. It then applies category-level refinement within each partition to sharpen decision boundaries and alleviate the lack of fine-grained region-level pretraining in CLIP. To further provide reliable visual anchors for novel categories, we design a cognitively inspired Structure Prompt (SP) that encodes both egocentric attributes and allocentric context, facilitating stable partition-and-refine learning. Extensive experiments on two OVD benchmarks show that PR-OVD delivers an average 6.5\% proportional improvement on novel categories over the state-of-the-art methods. Notably, PR-OVD also benefits other open-vocabulary tasks, including segmentation and classification, further confirming its effectiveness and generality.
Abstract (translated)
URL
https://arxiv.org/abs/2611.12976
PDF
-
ADIndex: Layer-Adaptive Sparse Attention Indexing via Group-Level Distillation
2026-11-07 06:35:09Jin-Fei Qi, Yiduo Li, Jiang-Xin Shi, Kan Liu, Tao Lan, Hanlin Tang, Yu-Feng LiAbstract
DeepSeek Sparse Attention (DSA) reduces the computational cost of attention through token-level selection. However, its indexer still scores the entire causal prefix for each query, making indexing a dominant source of latency in long-context inference. Notably, the indexer alone can account for more than 50% of total prefill GPU time at long context lengths. To address this bottleneck, we propose ADIndex, an Adaptive and Distillation-driven Indexing method for token-level sparse attention. ADIndex integrates three key components: (i) a layer-wise calibration strategy that adaptively accommodates cross-query retrieval overlap, motivated by our observation of pronounced inter-layer heterogeneity; (ii) a group-level distillation scheme that leverages this heterogeneity to efficiently train adaptive layer-wise group indexers, capturing salient retrieval signals from individual queries; and (iii) a low-latency inference strategy that reduces repeated full-prefix scans through token bank sharing within query groups while retaining query-specific refinement. Experiments on DeepSeek-V3.2 and GLM-5.1 demonstrate that ADIndex preserves near-lossless task performance on LongBench, LongBench v2, and RULER while delivering substantial efficiency gains over standard DSA, including prefill and decoding kernel speedups of up to 6.22x and 3.40x, respectively, at a context length of 256K across the layer-wise reuse configurations, as well as an end-to-end serving speedup of 1.54x at a context length of 128K.
Abstract (translated)
URL
https://arxiv.org/abs/2611.12956
PDF
-
Video-Coder: Reinforcing Agentic Coding for Video Generation
2026-11-06 18:59:57Chaoyang Wang, Sicheng Gao, Junan Hu, Kunyang Li, Dongyang Chen, Yiyang Zhou, Huaxiu Yao, Yong Jae Lee, Radu Timofte, Yuzhang ShangAbstract
Recent years have witnessed remarkable progress in video generation in terms of visual quality and temporal consistency. However, generating videos with precise spatial structure, coherent motion dynamics, and controllable scene composition remains challenging. In this paper, we introduce Video-Coder, the first attempt to train a code-driven video generation agent, which autonomously writes and executes code to construct a dynamic 3D scene (Draft Video) that provides structure and motion signals to downstream video generators. To support this paradigm, we design a dedicated data pipeline and curate two high-quality datasets, Video-Coder-SFT-6k and Video-Coder-RL-2k, which contain diverse prompts and ground-truth reference videos. We further develop the Video- Coder Harness, a unified interactive environment integrating code execution, 3D scene rendering, and execution feedback for agent training and inference. We then train Video-Coder with SFT fol- lowed by harness-native agentic reinforcement learning with dual reward feedback, which combines execution-based and video-based rewards. Additionally, we design an annealing training strategy that gradually reduces the execution reward over time, allowing the model to rely more on the video reward in the later training stages. Experiments show that Video-Coder achieves significant gains, improving Minimax-H3 by approximately 7 points on PhyGenBench and 24 points on VBench-2.0. We hope this work can serve as an open foundation for coding agents in video generation, and we fully open-source our data, models, and code
Abstract (translated)
URL
https://arxiv.org/abs/2611.09894
PDF
-
CodeQuant2: Enhanced Outlier Mitigation and Adaptive Clustering for Low-Precision LRMs
2026-11-06 18:48:04Xingyu Liu, Xiangyang Yin, Tianhua Xia, Haiyu Wang, Sai Qian ZhangAbstract
Large reasoning models (LRMs) improve task performance by generating longer intermediate reasoning traces, but this substantially increases inference costs and makes low-precision quantization increasingly important. However, aggressively quantized models often perform poorly on reasoning tasks. We observe larger activation fluctuations during the first decoding steps of reasoning. Calibration that omits these early states may miss their activation fluctuations. To address these reasoning-specific quantization challenges, we propose \textbf{CodeQuant2}, a clustering-based LUT quantization framework for low-precision large reasoning models. CodeQuant2 mitigates activation outliers with rotation and scaling, uses clustering to represent non-uniform weight distributions, and adapts codebook sizes to the quantization difficulty of each layer. Guided by our observation of reasoning-time activation behavior, CodeQuant2 further improves calibration sample selection to better match actual decoding statistics. Combined with our customized LUT CUDA kernel, CodeQuant2 delivers substantial accuracy improvements on reasoning benchmarks while achieving up to $2.85\times$ speedup over BF16.
Abstract (translated)
URL
https://arxiv.org/abs/2611.09868
PDF
-
Grasp-TAGS: Task-Aware Grasp Selection via Critic Distillation for Prehensile Loco-Manipulation
2026-11-06 18:44:33Xun Tu, Minghao Zou, Karthik DesinghAbstract
Loco-manipulation, which requires mobile manipulators to coordinate locomotion and manipulation over long horizons, is a key capability for large-object rearrangement in applications such as service robotics and warehouse automation. Despite recent progress, reliably executing these tasks remains challenging because performance depends on complex robot-object and object-environment interaction dynamics. Existing approaches address this challenge either through model-based methods, which often require accurate dynamics models, or through learning-based methods, which reduce this requirement but typically assume a fixed or predefined initial grasp. However, different geometrically stable grasps can lead to substantially different downstream manipulation performance. Therefore, we propose Grasp-TAGS, a task-aware grasp selection framework for goal-conditioned large-object rearrangement. We train a unified PPO-based reinforcement learning policy across multiple initial grasp configurations and distill the learned critic into a lightweight grasp utility network that ranks candidate grasps using only the candidate grasp configuration and goal information. At inference time, we select the grasp with the highest predicted downstream utility for the downstream policy to complete the task, before even making a contact. Experiments demonstrate that Grasp-TAGS improves rearrangement success over both fixed single-grasp policies and random grasp selection, achieving an average success rate of 89.84% compared with 79.87% for the best-performing single-grasp policies, an improvement of 9.97 percentage points. Grasp-TAGS also reduces final orientation errors across all evaluated objects. For more qualitative results, refer to the supplementary webpage: this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2611.09864
PDF
-
Flow-Guided Target-Space Alignment via Path Consistency
2026-11-06 18:44:08Ruizhi Yuan, Zeqiu Yu, Wei Gao, Wei Chen, Xiwei Tang, Lu TangAbstract
Many multimodal alignment problems require mapping a source modality into a fixed space defined by a target modality, where the target space is not merely an intermediate embedding but is used directly for downstream analysis. In this setting, successful alignment should preserve both the global structure of the target distribution and the instance-level correspondence observed in paired source--target training data. Existing objectives typically address only part of this goal: contrastive losses encourage relative similarity, distribution-matching losses improve marginal agreement, and pointwise regression fits each source sample to its paired target without explicitly modeling target-space path geometry. We propose FlowAlign, a flow-guided target-space alignment method that repurposes conditional flow matching as a training-time alignment signal rather than an inference-time transport model. FlowAlign uses the learned flow field to guide the alignment of source-side representations with the fixed target space, encouraging them to preserve paired correspondence while remaining compatible with the global target-space structure. We further provide theoretical analysis showing how FlowAlign helps improve paired correspondence by controlling endpoint accuracy and capturing local target-space geometry. Experiments on synthetic data, paired single-cell multiome benchmarks, and image--text alignment demonstrate that FlowAlign improves instance level correspondence while preserving target-space structure across different modality settings.
Abstract (translated)
URL
https://arxiv.org/abs/2611.09863
PDF
-
Seeing What Matters: Benchmarking Task-Conditioned Actionable Spatial Belief for Manipulation
2026-11-06 17:49:56Yina Jian, Mingkun XuAbstract
When should a robot look at the cup, and when at the portafilter? Spatial-reasoning benchmarks evaluate whether a model can describe a scene, but manipulation needs something more specific: the ability to acquire the particular evidence that makes the current command executable. A view that is informative for one task can be useless for another, even at the same workstation. We introduce Seeing What Matters, a benchmark for task-conditioned actionable spatial belief in simulated tabletop manipulation. Across three domains--coffee workstation, drawer/cabinet, and microwave/appliance--the benchmark provides 860 frozen evaluation cases; expanding across methods, budgets, and inference repeats yields more than 115,000 evaluator episodes under two protocols: closed-loop actionability and same-object functional out-of-distribution generalization. Each case pairs a set of candidate RGB views with a manipulation command, shared scripted skills, and hidden postconditions, so the benchmark directly measures whether a view-selection policy acquires the observations its command actually needs. We also provide a diagnostic reference pipeline that maintains a spatial-functional belief state and ranks views by missing evidence. At primary budget settings, this pipeline reaches 98.8% closed-loop and 98.9% functional-OOD success; ablating the functional or command-binding components collapses performance to near zero. Seven vision-language models, evaluated on the same cases, reach only 26-43%. Replacing the reference pipeline's coverage metadata with frozen RGB-derived scores preserves 98.5%/98.9% success; shuffling the visual scores collapses it to 15%/11%, confirming that the result depends on image content. These gaps suggest that spatial belief for manipulation should be evaluated not by scene-level completeness, but by whether it acquires the evidence the current command requires.
Abstract (translated)
URL
https://arxiv.org/abs/2611.09794
PDF
-
ROSE: Risk-Aware Orthogonal Subspace Navigation for Lifelong Knowledge Editing in Multimodal Large Language Models
2026-11-06 17:32:49Lingyun Song, Ziyao Chen, Kang Pan, Xiaolin Han, Yifei Zhang, Xiaoqi Wang, Yudai Pan, Xuequn ShangAbstract
Multimodal Large Language Models (MLLMs) require continuous factual updates to stay current, yet lifelong knowledge editing remains a daunting challenge. The primary obstacle is the severe Locality Erosion and Catastrophic Forgetting, where sequential edits inevitably interfere with pre-trained capabilities and prior updates. In this paper, we propose \textbf{ROSE}, a \textbf{R}isk-aware \textbf{O}rthogonal \textbf{S}ubspace navigation framework for lifelong knowledge \textbf{E}diting. ROSE addresses forgetting by aiming to restrict parameter updates to the orthogonal complement of previously edited subspaces. Crucially, it introduces a risk-guided mechanism that utilizes a dynamically computed risk mask to safeguard specific factual parameter subspaces essential for model stability. During inference, ROSE employs a multi-granularity knowledge integration strategy, featuring prototype-anchored semantic gating and perturbation-driven subspace fusion to precisely activate edited knowledge for relevant queries while limiting interference on unrelated inputs. Comprehensive experiments on two prominent multimodal benchmarks show that ROSE is competitive with state-of-the-art approaches across five key metrics, while maintaining outstanding stability over $1,000$ sequential updates. Our code is available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2611.09770
PDF
-
A lightweight convolutional neural network for target detection in an edge environment
2026-11-06 17:22:37Hughes Perreault, Ola Ahmad, Daniel CzubokaAbstract
This paper proposes a novel, lightweight convolutional neural network (CNN) designed for edge device deployment. The CNN is developed to recognize and classify threats or targets of interest. The model is equipped with a real-time hotspot detection algorithm, which enables it to process information quickly and accurately, resulting in improved inference time of up to 20 fps and a detection range of more than 400 meters with typical thermal situational awareness sensors. The proposed CNN has been rigorously tested and validated to demonstrate its high accuracy, sensitivity, and low false positive rate, providing a reliable solution for edge devices to detect potential targets. Additionally, the lightweight design of the CNN allows for easy deployment, making it an ideal solution for extending the security bubble for edge devices. The proposed model can be used for a variety of applications, including surveillance, security, and object recognition.
Abstract (translated)
URL
https://arxiv.org/abs/2611.09756
PDF
-
PoseRieFlow: Riemannian Flow Matching for Whole-Body Text-to-Pose Generation
2026-11-06 16:40:03Jan Spindler, Bardia Ardakanian, Patrick Stotko, Reinhard KleinAbstract
We present PoseRieFlow, a compact flow-matching framework for text-conditioned generation of whole-body 3D human poses. Prior text-to-pose methods either neglect articulated hands or model rotations in Euclidean space despite their non-Euclidean geometry. PoseRieFlow instead formulates generation directly on the product manifold of SMPL-H joint rotations, learning a text-conditioned vector field that transports samples from an isotropic $\mathrm{SO}(3)$ Gaussian prior to the data distribution along Riemannian geodesics. A part-tokenized Diffusion Transformer aggregates the 52 joints into eight body-part tokens, allocating a dedicated token and nonlinear projector to each hand, and restores per-joint resolution through part-specific decoders, shortening the sequence while improving kinematic fidelity over per-joint tokenization. With only 21.6M trainable parameters, PoseRieFlow supports expressive SMPL-H generation with detailed hand articulation at a fraction of the scale of recent LLM-based generators. On PoseScript, it attains the lowest whole-body kinematic error among all hand-producing methods, with consistent kinematic gains from the Riemannian formulation over its Euclidean counterpart at every practical inference budget, while distributional and retrieval metrics remain comparable between them.
Abstract (translated)
URL
https://arxiv.org/abs/2611.09692
PDF
-
Decoupled Mode Connectivity for Base-to-Novel Generalization in Vision-Language Models
2026-11-06 16:04:41Imad Eddine Marouf, Khalid Oublal, Enzo Tartaglione, St\'ephane Lathuili\`ereAbstract
Prompt learning adapts vision-language models such as CLIP by optimizing a short learnable prompt. The cross-entropy objective pulls this prompt toward base-class specialization, whereas generalization to unseen classes benefits from staying close to CLIP's zero-shot feature space. Because both objectives act on the same prompt, their gradients conflict at convergence, and in our experiments no single-prompt loss design clearly improves on the cosine baseline. We propose Decoupled Mode Connectivity (DMC), which gives each objective its own prompt. Starting from a base-trained prompt, DMC trains a specialization prompt and a generalization prompt, and merges their text features with a fixed coefficient at test time, at the inference cost of a single prompt. A linear mode connectivity (LMC) corridor in text-feature space keeps the classification loss low for every classifier interpolated between the two endpoints, and a Visual Anchor KL term preserves CLIP's zero-shot class-similarity distribution of each image during specialization. We further argue that such regularizers should satisfy class-permutation invariance (CPI), and show via Fano's inequality, in an idealized model, that violating CPI lower-bounds the drop in novel accuracy by a term proportional to the mutual information between the prompt and the base-class labels. Across 11 datasets, DMC improves the average harmonic mean of base and novel accuracy over MergeTune for both CoOp and KgCoOp, and its interpolation curves pass beyond the single-prompt operating point, shifting the base-novel frontier rather than trading along it. Code is available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2611.09655
PDF
