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

+197新增论文
+151其中 AI 相关
+16已有论文修订
+91新增 AI 导读

arXiv 最近完成扫描:2026-11-22 06:41 · 全站最近入库:2026-11-21 12:19各信源进度 →

统计口径与数据新鲜度

统计 2026-11-21 08:29 — 2026-11-22 08:29(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:2026-11-22 06:41最近发现新增版本:2026-11-21 12:19库内论文最新更新时间:2026-11-20 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。

29,158 已收录论文24,181 AI 相关8,992 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 21,490 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-22 06:39(北京时间)
新论文 · v1arXiv:2611.20094

FRAAC: A French Multimodal Large Language Model and Dataset forPictogram-Based Augmentative and Alternative Communication

Alexandre Audibert, François Portet, Benoit Favre, Maja Jonck Hjuler

原文摘要

Pictograms are a core modality in augmentative and alternative communication (AAC), yet recent multimodal large language models (MLLMs) cannot reliably interpret or produce pictogram sequences. Two obstacles explain this gap: the absence of a standardized pictogram grammar and the scarcity of parallel text-pictogram data. We introduce Retrieval-Augmented Pictogram Selection (RAPS), an annotation method that refines a rule-based transcription system's output with LLM-based concept extraction and retrieval over a pictogram vocabulary, handling cases where hand-written rules fail (named entities, numbers, polysemous words) without requiring new rules. Applied to French, RAPS yields a synthetic corpus of 1.4 million parallel pairs, of which 300k are publicly released. In a human ranking study where annotators order four pictogram transcriptions of the same sentence, RAPS outperforms the rule-based state of the art. Moreover, we show that standard reference-based metrics separate systems of clearly different quality but fail to rank strong systems against each other. Finally, we build an instruction-tuned model that supports interaction in several AAC modalities: it accepts pictogram or text input and answers in pictograms, speech, or text.

cs.CL · cs.AI

版本更新 · v2arXiv:2611.07940

CardioLM: A multilingual suite of small language models for the cardiology domain

Manuela-Daniela Danu, Miguel Rodríguez-Ortega, Alberto Becerra-Tome, Pablo Arancibia, Fernando Gallego-Donoso, Lorenzo Pratesi, Xi Bai, Andrei-Damian Dinescu 等 40 位作者

原文摘要

Background: Cardiovascular care generates large volumes of clinical free-text documentation in local languages. However, clinical natural language processing (NLP) resources remain concentrated in English, and no existing benchmark is cardiology-specific, multilingual, and broad in entity coverage. Methods: We present two contributions. The first is CardioCCC, a corpus of 508 cardiology case reports originally published in Spanish and extended to Czech, Dutch, English, Italian, Romanian, and Swedish through machine translation, annotation projection, and clinician review, with annotations covering diseases, symptoms, procedures, and medications. The second is CardioLM, comprising CardioBERTa, a set of language-specific encoders obtained by continued pretraining on cardiology and biomedical corpora, and CardioNER, a set of named entity recognition (NER) models fine-tuned on CardioCCC. Results: Cardiology-specific pretraining reduced pseudo-perplexity by up to a factor of eight, yet changed strict span-level F1 by at most 0.04, with benefits concentrated in languages whose backbones lacked biomedical coverage. The jointly trained multilingual XLM-RoBERTa-large models achieved the best strict F1 in nearly every language and entity type, exceeding the strongest monolingual system by up to 0.13 F1. A large share of residual errors concerned boundary detection: relaxed matching recovered 0.10-0.23 F1 for disease, procedure, and symptom, but only 0.01-0.07 for medication. Clinicians rated CardioNER well above a rule- and dictionary-based baseline, with the mean rating across entity types rising from 26% to 73%. Conclusions: Specialty-specific pretraining is not the decisive factor for downstream entity recognition, whereas multilingual joint training yields the strongest results in nearly all settings and a simpler deployment path for multi-country clinical information extraction.

cs.CL · cs.AI

新论文 · v1arXiv:2611.20089

See the Lives of Stars in a Handful of Light

Shuhong Liu, Gengjia Chang, Quanfeng Xu, Ziteng Cui, Qianyu Zhou, Xiaoyu Dong, Lin Gu, Tatsuya Harada

原文摘要

Galaxies evolve over hundreds of millions of years, and no telescope has ever watched one move. Cosmological simulations make this evolution visible as a sequence of multi-view images, inviting the question of whether a galaxy can be recovered in four dimensions from its images. This setting defeats existing dynamic scene methods, since millions of stars span only a few pixels each, faint populations blend into diffuse light, and free deformations reproduce the images without obeying gravity. We present Cosmos4D, which rests on two components. Gravitational 4D Gaussian Splatting represents each star as a self-luminous Gaussian whose motion is integrated through a gravitational potential and a low-dimensional force field shared by all stars. Luminosity-stratified dynamics give bright stars individual orbits supervised across image scales and carry faint populations with a velocity field estimated from optical flow. We further build a 4D dataset of eight galaxies from four cosmological simulations, on which Cosmos4D outperforms dynamic Gaussian baselines in both photometric quality and physical accuracy. With these advancements, we glimpse eight hundred million years of a galaxy's life in ninety-six frames.

cs.CV

新论文 · v1arXiv:2611.20088

Optimal Pattern-Avoidance Bounds for Binary Search Trees

Anita Dürr, Arturo Merino, Karol Węgrzycki, Sorrachai Yingchareonthawornchai

原文摘要

Self-adjusting binary search trees (BSTs) perform better than balanced BSTs if the access sequence has additional structure. For example, splay trees serve a sequential access sequence in linear time. Pattern avoidance captures regularity in an access sequence and unifies several classic examples of efficient BST access, such as deque and traversal sequences. A central question is how well a BST algorithm can exploit a pattern-avoiding access sequence. We present an online BST algorithm that serves an access sequence avoiding a permutation pattern $π$ in linear time, matching the information-theoretic lower bound. The algorithm does not need to know that pattern $π$ a priori. Our results answer an open problem asked by [Kozma 2016, Problem 44, strong variant]. Previously, the best known offline bound was optimal only for constant-size patterns [Berendsohn, Kozma, and Opler, STOC'24], and the best known online bound was superlinear [Chalermsook, Pettie, Yingchareonthawornchai, SODA'24] even for a constant-size pattern. As a corollary, our BST algorithm yields an optimal sorting algorithm for a pattern-avoiding sequence that supports online insertions. Previously, an optimal sorting algorithm was achieved using a mergesort-like approach [Opler, FOCS'24] in a batch setting. A key technique is to apply recursive split decomposition [Chalermsook, Chuzhoy, Saranurak, ToC'23] jointly in time and space, along with the Marcus-Tardos proof of the Stanley-Wilf conjecture. We show that the framework can be implemented online using $k$-finger BSTs, which is of independent interest. Finally, this framework also yields a constant-factor-approximation FPT algorithm in the offline setting.

cs.DS

版本更新 · v2arXiv:2603.06950

How Private Are DNA Embeddings? Inverting Foundation Model Representations of Genomic Sequences

Jules Kreuer, Sofiane Ouaari, Nico Pfeifer

原文摘要

Genomic foundation models are increasingly used to share dense representations rather than raw DNA sequences. This is particularly pursued for mean-pooled representations that are later used for downstream prediction tasks. Whether this practice protects the privacy of the underlying sequence is largely unknown. We evaluated model-inversions against zero-shot embeddings from DNABERT-2, Evo 2, and Nucleotide Transformer v2 (NTv2). Substantial reconstruction remains possible: a single multi-length Length-Conditioned Query Decoder reconstructed NTv2 sequences with mean Levenshtein similarity from 0.98 at 10 nucleotides to 0.68 at 100, and Evo 2 sequences from nearly 1.00 at 15 nucleotides to 0.67 at 100. DNABERT-2 showed higher resistance (0.56 to 0.51), although it remained above the random baseline at short lengths. Therefore, we further evaluated, whether a linkage attack similar to Homer's attack is possible based on the mean-pooled representations of human DNA sequences. At SNP-enriched length-25 loci and known coordinates, Evo 2 and NTv2 reached 99\% identification at 32 loci and DNABERT-2 at 256 loci. With alignment-based coordinate recovery, the corresponding 99\% budgets were 64 and 128 loci for Evo 2 and NTv2, while DNABERT-2 did not reach the reported identification thresholds. Random loci required larger budgets. These findings show that mean-pooled embeddings can leak sequence and identity, with risk depending on sequence length, representation, locus informativeness, and coordinate information.

q-bio.GN · cs.AI · cs.LG

新论文 · v1arXiv:2611.20087

Does Learning Order Explain Forgetting? The Lie Bracket as an Early-Warning Signal for Catastrophic Forgetting Without Old Data

Ephraim Kofi Ayi Kwapong

原文摘要

A network trained on task A and then task B ends at different weights than one trained on B and then A. To leading order this gap is set by the Lie bracket of the two tasks' gradient fields, and Dherin (2023) conjectured that a non-vanishing bracket may be linked to catastrophic forgetting. I test this conjecture empirically. Across Split MNIST, Split Fashion-MNIST and Split CIFAR-10, the forgetting of A is almost exactly the difference in A's loss between the two training orders (Spearman rho = 0.96-1.00). I then ask whether the bracket can predict that forgetting early in training on B and without any of A's data, using only a small summary stored when A finishes: its gradient and its top 20 Hessian eigenpairs. When the old task is well learned and forgetting is slow (task-incremental), this stored bracket ranks forgetting with rho = 0.70-0.81, against 0.33-0.58 for the penalty of Elastic Weight Consolidation, and matches or exceeds re-evaluating A on its own data. When forgetting is fast (domain-incremental), the stored summary goes stale and the EWC penalty is as good or better. The bracket fails outright when the old task is only half learned; I trace this to a residual gradient on A an order of magnitude larger, which makes first-order (slope) damage matter. The bracket is therefore a useful early-warning signal for the quiet kind of forgetting, under a clear condition.

cs.LG · stat.ML

版本更新 · v3arXiv:2603.27910

GAAMA: Graph Augmented Associative Memory for Agents

Swarna Kamal Paul, Shubhendu Sharma, Nitin Sareen

原文摘要

AI agents that talk with users across many sessions need long-term memory to stay coherent and personalized. Existing approaches use either flat retrieval-augmented generation (RAG) that ignores how memories connect, or entity-centric knowledge graphs. Entity-centric graphs split each message into separate triples, which loses its context, and in conversations the speakers become hubs linked to almost every message, so propagation through them tells retrieval little. We present GAAMA, a graph-augmented associative memory that organizes conversations around topics. It keeps verbatim episodes, uses an LLM to distill them into atomic facts and topic concepts, and then synthesizes higher-order reflections. These four node types are linked by five kinds of structural edges, and topic concepts supply the cross-cutting paths, so mega-hubs do not form. Retrieval blends nearest-neighbor similarity search with Personalized PageRank whose transitions depend on edge type. We also introduce GRAFT (Graph Repair by Augmenting Facts & Topology), an offline repair step: for questions whose retrieved evidence fell short, a bounded ReAct agent searches memory again and records the grounded answer as a fresh fact, so later questions find it directly without extra query-time cost. On LoCoMo10, GAAMA achieves the highest average score of all compared methods (79.1%) and is competitive in every question category; GRAFT adds a small further gain. In a controlled comparison, replacing GAAMA's concept-linked facts with entityrelation triples lowers reward by 3.0 pp. On MemoryArena, GAAMA also has the best average across the three datasets.

cs.AI · cs.IR · cs.MA

新论文 · v1arXiv:2611.20083

WL-Headset Law: Projecting Future Headset Evolution from a Decade of Display and Mass Data

Jialin Wang, Hai-Ning Liang

原文摘要

Tracking a decade of headset display and mass changes can inform forecasts. However, comparing headset resolution requires pixels per degree (PPD), which is often unreported and estimated using unclear assumptions. Therefore, we compile the open-source, evidence-linked HMD Resolution and Mass Database (HRMD) with 241 configurations from 2016--2026 and introduce structure- and axis-aware robust calibration (SARC) to estimate missing central PPD. Compared with official PPD, SARC reduces PPD estimation error from 14.8% for traditional geometric estimation to 3.2%. Based on HRMD, we introduce the Weight-Linked Headset Law (WL-Headset Law), which models PPD by year, mass, and category. The law reveals 7.4% annual PPD growth at fixed mass and category. Meanwhile, head-worn mass fell by 6.5% annually after category adjustment. At an eyeglass-like mass of 30g, the law projects average central PPD of 60 (normal visual acuity) around 2035 and 120 (higher acuity) around 2044 if trends persist.

cs.HC

版本更新 · v2arXiv:2601.15549

VIOLA: Towards Video In-Context Learning with Minimal Annotations

Ryo Fujii, Hideo Saito, Ryo Hachiuma

原文摘要

Generalizing Multimodal Large Language Models (MLLMs) to novel video domains is essential for real-world deployment but remains challenging due to the scarcity of labeled data. While In-Context Learning (ICL) offers a training-free adaptation path, standard methods rely on large annotated pools, which are often impractical in specialized environments like industrial or surgical settings since they require the experts' annotations. To bridge this gap, we introduce VIOLA (Video In-cOntext Learning with minimal Annotation), a label-efficient framework that synergizes minimal expert supervision with abundant unlabeled data. First, to maximize the efficiency of a strict annotation budget, we propose density-uncertainty-weighted sampling. Unlike standard diversity or uncertainty strategies that risk selecting visual outliers, our method leverages density estimation to identify samples that are simultaneously diverse, representative, and informative. Second, to utilize the remaining unlabeled data without noise propagation, we construct a hybrid pool and introduce confidence-aware retrieval and confidence-aware prompting. These mechanisms explicitly model label reliability, retrieving demonstrations based on a composite score of similarity and confidence while enabling the MLLM to adaptively distinguish between verified ground truths and noisy pseudo-labels. Extensive experiments across nine diverse benchmarks using three MLLMs demonstrate that our framework significantly outperforms various baselines in low-resource settings, achieving robust adaptation with minimal annotation costs. Project page: https://fujiry0.github.io/VIOLA-project-page/

cs.CV · cs.AI

版本更新 · v2arXiv:2611.06044

Do Search-Augmented Recommenders Use the Latest Search Episode?

Jiandong Ding

原文摘要

When a user returns to a recommendation feed after searching, does the latest search episode help rank the next item? Aggregate ranking gains leave open whether search-augmented recommenders use the latest search at prediction time or benefit from broader search state and representation learning. FocalSwap is a counterfactual audit that masks or replaces the latest episode or the full search history while holding the target, candidates, recommendation history, and trained scorer fixed. A planted-signal test and GenSAR's native search-item prediction task verify that FocalSwap detects known search dependence. In proxy analyses, KuaiSAR's routing gain falls below the shuffled threshold, and Qilin's clearest matched-search advantage appears at high taxonomy overlap. The primary comparisons for SESRec and UniSAR do not establish a positive ranking gain from the latest search episode alone. For the full search history, the estimated contribution depends on the control. On 66,139 shared instances across three UniSAR checkpoints, neutral masking consistently lowers mean reciprocal rank (RR). Replacing the history with observed searches matched on category and activity profile changes 44.1 to 44.8 percent of target ranks, yet the original history has little average advantage. In the primary assignment, matched replacements yield mean RR 0.002770 to 0.006604 higher than neutral masking across checkpoints. The adjusted intervals for this difference lie above zero under the primary assignment and all five alternative assignments per checkpoint. Search audits should specify which inputs changed and what replaced them, and report both how often ranks move and how much the original search helps.

cs.IR · cs.LG

新论文 · v1arXiv:2611.20078

Infrastructure-Based Multi-Agent Reinforcement Learning with Attention for Highway Ramp Merging

Chenhao Tong, Steffen Müller

原文摘要

On-ramp merging is a typical safety and throughput problem on highways. Without proper coordination, human drivers and self-interested autonomous vehicles may make risky manoeuvres and cause accidents or congestion. With recent advances in connected autonomous vehicles (CAVs) and multi-agent reinforcement learning (MARL), vehicles can be trained to coordinate for safe and efficient merging. However, in dense traffic, forming an effective observation of the traffic state is challenging. For example, observing only the neighbouring vehicles may lead to locally optimal decisions, while observing the entire traffic can introduce much irrelevant information. In this paper, we propose a transformer-based MARL vehicle control model to address these problems. A roadside infrastructure controller receives the states of all CAVs via vehicle-to-infrastructure (V2I) communication and sends control commands back to them. An attention-based observation encoder is designed where each vehicle can filter the relevant vehicle information from the global traffic state. The proposed model is evaluated in SUMO against three baselines: a model that observes only the neighbouring vehicles, a model that encodes the global traffic information, and the rule-based SUMO model. The results show that our model achieves the highest traffic flow, safety and smoothness. Furthermore, in a four-hour simulation with dynamic traffic demand, our model successfully avoids traffic jams, while other baseline models fail at high traffic flow. We also show that our model remains effective when human-driven vehicles are present. We further show that our model, trained in ideal communication conditions, can still remain collision-free and maintain throughput under realistic V2I communication errors. Finally, we visualise the relevance weights that an ego vehicle assigns to other vehicles in multiple traffic situations.

cs.LG · cs.MA · eess.SY

版本更新 · v2arXiv:2606.12585

Revisiting the ABCs of Working with AI: Evidence from Radiologists

Daniel Martin

原文摘要

I examine the relationship between the benefits of working with AI and two baseline traits - ability and belief calibration - among all 227 radiologists available in Collab-CXR, a dataset from an experiment where AI assistance was provided during X-ray diagnosis. When analyzed jointly, both lower unassisted accuracy and smaller miscalibration, measured by the absolute gap between average confidence and accuracy, are associated with greater improvements in accuracy with AI. In addition, the association between unassisted accuracy and improvement with AI becomes more negative and statistically significant when controlling for calibration.

econ.GN · cs.HC

新论文 · v1arXiv:2611.20076

SGIP: Learning Scene-Grounded Interaction Priors for Scene-Aware Human Motion Prediction

Donghyeon Soon, Jiil Yoon, Hyeondong Kim, Daehee Park

原文摘要

Human motion prediction (HMP) aims to forecast future movement from an observed motion sequence, and recent methods have incorporated 3D scene geometry and object semantics to improve long-term plausibility. However, the destination of future motion remains ambiguous in cluttered scenes: the same observed motion can lead to different interaction targets or navigation modes. Existing methods either handle this ambiguity implicitly or rely on auxiliary signals like gaze that are hard to obtain in practice. To address this, we propose SGIP, which explicitly predicts where in the scene future motion will be grounded across both objects and open navigation regions, and uses it to condition future trajectory and full-body pose prediction. This scene-grounded interaction prior is learned via intent distillation: at training time only, a teacher model scores candidate scene regions by combining an LLM prompted with a textual description of the person's intent with geometric motion-scene features. A student model learns to reproduce these scores from only the observed motion and 3D scene, making intent and LLM unnecessary at test time. Across multiple benchmark datasets, SGIP outperforms prior scene-aware baselines on both trajectory and pose accuracy in a realistic deployment setting where neither gaze nor intent is available at inference, demonstrating the effectiveness of the proposed framework. The project page is available at https://sgip2026.github.io.

cs.CV

新论文 · v1arXiv:2611.20075

Prompt Re-Specification and Target-Mask Availability in Mask-Conditioned Visual Reinforcement Learning for Moving-Platform Landing

Ho Tim Tsui, Jongyun Kim

原文摘要

Visual reinforcement learning (RL) policies that land a drone on a moving platform can fail when the landing pad looks different from the one used in training. Using binary target masks as the visual input removes most pixel-level appearance information while preserving the spatial representation of the target, and lets a text prompt specify the target. In simulation, we test whether a frozen (fixed-weight) mask-based DrQ-v2 policy can handle a new pad when only the prompt of its fine-tuned SAM 3 detector changes. A banana-themed pad lowered landing success from 75.6% to 12.0%, and SAM 3 rarely detected it under the original prompt. A revised prompt, chosen after testing several candidates, restored success to 68.4% with no weight updates. In a separate experiment with a policy trained on ground-truth masks, filling in masks that were missing while the pad was visible helped landing far more than the tested mask-centroid and area corrections. Tracking the pad over time supplied masks more often and also kept producing them after the pad was confirmed absent from the rendered image, but it did not establish a reliable improvement in landing success. Whether a usable mask reaches the policy at all thus appears to be an important limiting factor for this approach.

cs.RO · cs.CV · cs.LG

新论文 · v1arXiv:2611.20068

Kolmogorov-Arnold Lookup Table Demappers for Amplitude Phase-Shift Keying over Nonlinear Satellite Links

Kürşat Tekbıyık, Güneş Karabulut Kurt, Antoine Lesage-Landry

原文摘要

Next-generation high-throughput satellites increasingly demodulate and decode on board, so the per-symbol cost of the receiver constrains payload power and throughput. Soft demappers for nonlinear satellite links must therefore produce accurate bit log-likelihood ratios (LLRs) at low cost. We introduce a KAN-LUT demapper, a two-layer Kolmogorov--Arnold network (KAN) whose learned univariate edge functions are stored as 128-entry lookup tables (LUTs), so that inference needs only table reads, linear interpolation, and additions. We bound the LLR error introduced by the tables and evaluate the demapper on amplitude phase-shift keying (APSK) links with low-density parity-check (LDPC) coding and a nonlinear traveling-wave-tube amplifier. For 16-, 32-, and 64-APSK, KAN-LUT stays within 0.2 dB of a per-cluster Gaussian log-MAP (LogMAP-CG) demapper, while its cost relative to LogMAP-CG falls from 87% to 27% as the order grows. It matches or slightly exceeds multilayer perceptrons (MLPs) at about one fifth of their cost, remains competitive under operating-point mismatch, and needs only tens of kilobytes of tables. KAN-LUT is therefore a low-cost candidate demapper for regenerative payloads.

eess.SP · cs.LG

新论文 · v1arXiv:2611.20067

World-Model-Guided Plan Search for Cooperative Mixed-Traffic Control at Bottlenecks

Jinghan Peng, Ning Liu, Xiangrui Kong, Sen Shen, Dongkun Han

原文摘要

Cooperative speed control by connected and automated vehicles can reduce congestion at lane drop bottlenecks. However, effective coordination depends on vehicle roles and execution timing, while initial holding costs may be offset by later traffic benefits. This paper proposes an offline plan search framework that combines world model guidance with simulator verification. A traffic world model incorporating physical dynamics guides evolutionary search over vehicle roles, target speeds, holding durations, and start times. Paired SUMO simulations screen candidates for total delay reduction and limits on additional mean delay across vehicle groups. Accepted plans form a reusable library whose actions temporarily override a frozen MAPPO policy. Across 192 paired test scenarios at 60% penetration and a demand of 2,800 vehicles per hour, the searched library reduces mean total delay by 14.3% relative to MAPPO and by 3.8% relative to a handwritten plan library. The largest increase in group mean delay is 6.2 s, below the prescribed 10 s limit. In a separate screening experiment, the guided search achieves a 30.0% pass rate compared with 2.6% for random sampling. These results support the use of predictive search and independent verification for discovering cooperative plans under group delay constraints.

cs.MA

新论文 · v1arXiv:2611.20065

MARLIN: Multiscale And Directional Sparse Representation Learning for Imbalanced SAR Ship Classification

Benyamin Hosseiny, Kamirul Kamirul, Odysseas Pappas, Alin Achim

原文摘要

Despite recent progress in SAR maritime monitoring, ship classification remains challenging due to limited labeled data, class imbalance, large intra-class variability, fine-grained inter-class similarity, and variations in target scale, orientation, and backscattering structure. To address these challenges, this paper proposes MARLIN, a class-frequency-aware framework with directional sparse and hierarchical multi-scale feature extraction for ship type classification in SAR images. The proposed architecture first constructs two complementary representations of each input image, where each views are processed by parallel encoder branches. Each branch begins with a directional sparse convolutional stem designed to capture elongated and orientation-dependent ship scattering patterns, followed by customized hierarchical multiscale feature extraction blocks. To improve learning under imbalanced class distributions, we introduce a class-frequency aware objective. The proposed objective combines margin-based classification with class-frequency-weighted prototype compactness and class-pair frequency weighted margin separation. This formulation shapes the embedding space by encouraging samples from the same class to form compact clusters while enforcing stronger separation for class pairs involving underrepresented ship categories. Extensive experiments on three complementary datasets demonstrate that in addition to being lightweight and parameter efficient, MARLIN improves both overall accuracy and class-balanced metrics, confirming its effectiveness for robust SAR ship-type classification under limited and imbalanced training conditions. The codes are available at github.com/benyaminhosseiny/marlin.

cs.CV · cs.LG

版本更新 · v2arXiv:2603.13324

The Challenge of Out-Of-Distribution Detection in Motor Imagery BCIs

Merlijn Quincent Mulder, Matias Valdenegro-Toro, Andreea Ioana Sburlea, Ivo Pascal de Jong

原文摘要

Machine Learning classifiers used in Brain-Computer Interfaces make classifications based on the distribution of examples on which they were trained. When exposed to EEG from unfamiliar classes, they can only make blind guesses. Instead of allowing such guesses, these Out-of-Distribution (OOD) samples should be detected and rejected to improve the robustness of BCIs against unfamiliar cognitive states. We study OOD detection in Motor Imagery BCIs by training a model on some classes and observing whether an unfamiliar movement class can be detected based on increased uncertainty. We tested seven different OOD detection methods and one more method that has been claimed to boost the quality of OOD detection. For many BCI users, the uncertainty for the familiar in-distribution classes can still be higher than for the out-of-distribution classes, due to the high intrinsic variability inherent in EEG signals. As a result, many OOD detection methods that have shown good performance in other machine learning domains prove to be ineffective at identifying unfamiliar motor imagery patterns in BCIs. However, we also found that OOD detection performance is correlated with on-task performance, and that Deep Ensemble models and MC-Dropout models were able to achieve on-task AUROC > 0.9, and OOD detection ability up to 0.7. This shows that for models and subjects where task performance is high, rejecting unfamiliar cognitive states becomes feasible. Our research demonstrates a Leave-One-Class-Out OOD detection setup as an experimental paradigm for studying whether models are robust against unfamiliar motor imagery actions that were not seen in the training data. This is the first benchmark in Motor Imagery BCI that evaluates this type of OOD data. Our results demonstrate how to improve the overall safety and reliability of BCIs by preventing erroneous actions.

cs.LG

新论文 · v1arXiv:2611.20064

Decode Early, Verify Late: Streaming Offline Speech Recognizers Without Accuracy Loss

Doyeob Kim

原文摘要

Voice agents need accurate transcripts soon after a user stops speaking. We propose Decode Early, Verify Late (DEVL), a training-free framework for streaming inference with offline ASR models. The server keeps partial transcripts as drafts and verifies the latest against the complete audio in one forward pass at turn end. Under stated input and decoding conditions, the final equals the model's greedy offline output in exact arithmetic. We adapt DEVL to four model families. In our five-recognizer comparison, 95-100% of final transcripts are byte-identical to the offline transcripts, WER stays within 0.13 points of offline, and single-stream time to final falls 1.5-7.3x. Serving Qwen3-ASR-1.7B on one A100, DEVL returns finals a median 22-26 ms after turn end at up to 64 streams, 2.7-3.9x sooner than the official streaming mode. It is the only system we tested that meets a 100 ms p99 target, up to 80 streams on Pipecat-STT and 64 on WildASR.

cs.LG

版本更新 · v3arXiv:2509.03206

Autonomous Learning From Success and Failure: Goal-Conditioned Supervised Learning with Negative Feedback

Zeqiang Zhang, Fabian Wurzberger, Gerrit Schmid, Sebastian Gottwald, Daniel A. Braun

原文摘要

Learning from reward functions and imitation learning of demonstrations are the two principal approaches for training autonomous systems that interact with an environment through action and observation. Both, however, require human specification for each behaviour to be acquired, a problem for long-lived self-adaptive systems whose goals and operating conditions cannot be fully anticipated at design time. Recently, Goal-Conditioned Supervised Learning (GCSL) through self-imitation has been proposed as a self-supervised alternative: by strategically relabelling goals, agents can derive policy insights from their own experiences. Despite its successes, this framework presents two notable limitations: (1) learning exclusively from self-generated experiences can exacerbate the agents' inherent biases; (2) the relabelling strategy allows agents to focus solely on successful outcomes, precluding them from learning from their mistakes. To address these issues, we propose GCSL with Negative Feedback (GCSL-NF), which evaluates each trajectory twice: positively with respect to relabelled goals, and correctively with respect to the goal originally intended. The corrective target comes from a similarity function learned contrastively from trajectory-induced neighbourhood relations, so that neither a reward function nor a geometric distance needs to be specified. Our experiments show that GCSL-NF overcomes limitations imposed by agents' initial biases, increasingly benefits from negative feedback as learning progresses, and matches or surpasses GCSL- and HER-based methods. By reducing reliance on prespecified reward functions, the proposed approach is particularly relevant for self-adaptive autonomous systems, where adaptation objectives may be diverse, changing, or difficult to engineer.

cs.LG · cs.AI

版本更新 · v4arXiv:2607.00578

Caption Bottleneck Models

Seref Baris Cagliyan, Umut Ozdemir, Merve Tapli, Emre Akbas

原文摘要

Concept Bottleneck Models (CBMs) provide interpretability by routing predictions through a layer of human-understandable concepts. However, defining an optimal concept set for a specific dataset remains an open challenge. Existing approaches rely on expensive expert annotations or LLM-generated lists based solely on class names. Even "open-vocabulary" variants typically depend on static concept sets, which restrict discovery and introduce label bias. Furthermore, traditional CBMs often suffer from information leakage, where unmodeled visual features bypass the bottleneck and compromise the integrity of the explanations. To overcome these limitations, we propose Caption Bottleneck Models (CaBM), a framework that circumvents the need for predefined concept sets by replacing rigid concept layers with free-form natural language. By representing images via LMM-generated captions and training a classifier strictly on this text, CaBM ensures a leakage-free architecture by construction. Additionally, by analyzing the text classifier post-training, CaBM autonomously discovers high-quality, dataset-specific concepts. Our results across fine- and coarse-grained benchmarks demonstrate that CaBM achieves competitive accuracy while preserving interpretability without the constraints of external dictionaries or manual labeling.

cs.CV

新论文 · v1arXiv:2611.20062

Mandarin Wh-Indefinites Follow Nonveridicality in Multilingual LLMs, Unlike English any

Guofang Tang

原文摘要

Multilingual LLMs learn mostly from English. Do they carry English polarity licensing over to other languages? English any is licensed by downward-entailing (DE) contexts, Mandarin wh-indefinites such as shenme 'what/something' by nonveridical ones, and the two classes come apart: perhaps licenses shenme but not any, while only licenses any but entails existence. We test this dissociation in a pre-registered study of twelve open LLMs (0.5B-9B, four families), using 48 matched bilingual frames in 21 contexts and a log-probability difference-in-differences normalised by negation. All 11 models that pass the measurement gate license Mandarin wh-indefinites under possibility, necessity and evidential adverbs, at 0.42-2.33 times the strength of negation, while English any gets nothing in the same contexts. An exploratory extension to Qwen2.5-32B shows the same. The English side is weaker: any under only, first and regret is licensed but at a fraction of negation, so only 3 models show the full double dissociation in sentence probabilities. When prompted to choose between sentences, most gated models show it. In exploratory directional ablations, a direction that carries English licensing, held-out classes included, does not reach Mandarin licensing in most models. The models do not simply apply the English rule to Mandarin.

cs.CL

新论文 · v1arXiv:2611.20061

Turning Errors into Successes: Designing LLM Tutoring Around Scaffolded Error Correction

Michael W. Asher, Meng Cao, Paulo F. Carvalho

原文摘要

Large language models (LLMs) could make personalized tutoring available at scale, but it remains unclear how they can best support the instructional processes that make one-on-one instruction effective. We compared two ways of using LLMs after students make errors: providing personalized explanations of those errors, or engaging students in additional cycles of scaffolded practice and feedback, with a Socratic tutor guiding students toward the correct answer. In a preregistered online experiment, 400 adults practiced setting up constraint equations and, after each attempt, received the correct answer, a generic explanation, a personalized LLM explanation of their error, or Socratic tutoring. Socratic tutoring helped: participants resolved 64% of their errors through the dialogue and scored 0.24 SD higher on practice-aligned posttest questions than those who received generic or personalized explanations. Socratic participants also found the session more interesting, less frustrating, and less costly, even though it took nearly twice as long. In contrast, we found no evidence of a benefit of personalized explanations over generic ones, and neither type of explanation reliably outperformed simply seeing the correct answer. These findings suggest that LLM tutors may add the most value not by explaining errors more precisely but by helping students correct those errors themselves through scaffolded practice. This pattern fits a mastery-experience account of tutoring, in which an effective tutor notices when a student is struggling, steps in with guidance, and works with the student until the struggle resolves into success. AI tutors may be most effective when designed around this principle.

cs.HC

新论文 · v1arXiv:2611.20058

SWAN-Rx: Efficient and Bandwidth-Scalable Neural Reception for MIMO-OFDM

Zhoubin Kou, Cong Shen

原文摘要

We present a Scalable Windowed Axial-attention Neural Receiver (SWAN-Rx) for multiple-input multiple-output orthogonal frequency-division multiplexing (MIMO-OFDM) systems. The receiver improves reception accuracy and reduces forward-pass computation while supporting frozen-weight reuse across bandwidth allocations and channel conditions. It combines a common resource-element representation of received observations and pilot context with alternating time-axis and local frequency-axis attention. Shared parameters and relative positional information permit soft detection across allocation lengths without inference-time parameter updates. At fixed subcarrier spacing, model configuration, and symbol count, bounded frequency neighborhoods give linear growth of forward-pass computation with bandwidth. Evaluations compatible with the 5G New Radio physical layer reuse weights trained on four physical resource blocks (PRBs) at up to 273 PRBs and on unseen channels. At the training bandwidth, SWAN-Rx achieves lower pre- and post-decoding bit error rates than the implemented three-step neural receiver (NRX) and DeepRx-MIMO baselines, with 16.9% and 39.7% fewer parameters and 56.3% and 35.7% fewer forward-pass operations, respectively. The same weights achieve lower pre-decoding error rates than these baselines throughout the evaluated bandwidth-extrapolation and 16-PRB channel-transfer tests, with an estimated 0.4-0.9 dB signal-to-noise ratio reduction relative to NRX at a pre-decoding error rate of $10^{-2}$ in the 16-PRB tests.

eess.SP · cs.IT · cs.LG

版本更新 · v2arXiv:2609.01126

How Evaluation Choices Shape the Measured Benefit of Online Adaptation in Time-Series Forecasting: Warmup, Learning-Rate Selection, and Adaptation-State Memory

Takumi Fujimoto, Hiroaki Nishi

原文摘要

Online adaptation can help edge time-series forecasting under distribution drift, but its measured benefit is sensitive to evaluation choices. We study six public multivariate streams, including building-sensor and smart-meter data, under a leakage-free streaming protocol. We identify two sources of comparison bias. First, the warmup budget of the static baseline has a two-sided effect on seed-mean curves: insufficient warmup undertrains the baseline, whereas excessive warmup can degrade its pre-drift generalization. Across six dataset-backbone settings, the estimated adaptation benefit changes by 3.0 to 18.8 percentage points (pp) over the 1,000-20,000-step warmup range. Second, comparing SGD with momentum (SGD+m) and Adam at a shared default learning rate conflates optimizer quality with rate sensitivity. We select both the warmup budget and each optimizer's online rate using a held-out pre-drift validation slice without accessing test data. Under this procedure, Adam outperforms SGD+m in 310 of 360 cells, and only 4 of its 360 cells fall below the static baseline. Third, we characterize accuracy against adaptation-state memory and A100-measured per-update latency for full, head-only, and calibration-based adaptation. In the evaluated PatchTST frontier settings, several parameter-efficient variants are nondominated on the adaptation-state-memory axis. These findings support a validation-only commissioning procedure. Code, data, and all reported numbers: https://github.com/keiotakmin/tsf-edge-adaptation

cs.LG

新论文 · v1arXiv:2611.20050

Clinician Assessment of Laryngeal Disease with Auditory AI before Laryngoscopy: A Paired Reader Study

Siyuan Hou, Shuting Yang, Zengrui Jin, Liyu Cheng, Siyin Wang, Yunhan Shi, Xueyan Li, Rong Hu 等 16 位作者

原文摘要

Background: Before laryngoscopy, clinicians assess the likely cause of hoarseness and whether to suspect a precancerous or malignant lesion. Whether voice-based artificial intelligence (AI) supports pre-laryngoscopic categorization is unknown. Methods: We developed an auditory large language model that generates a patient-level assessment card (provisional category, voice-quality ratings, and rationale) and a prespecified malignancy alert from voice recordings, patient information, and acoustic measures obtained before laryngoscopy at one center. We evaluated both in a held-out test set and a later cohort. In a sequential reader study, 32 otolaryngology clinicians categorized held-out cases without AI and, 1 to 2 weeks later, reviewed their initial categories with the card (without the alert). Results: Across 800 paired assessments, reader accuracy was 49.8% before and 56.8% after review (difference, 7.0 percentage points; 95% CI, 1.7 to 13.5; P < 0.001); 99 categories changed from incorrect to correct and 43 from correct to incorrect (ratio, 2.30; 95% CI, 1.22 to 5.31). Accuracy rose 14.7 points when the model was correct and fell 8.7 when incorrect. Provisional-category accuracy was 72% in held-out testing and 67% in temporal validation. The malignancy alert flagged 85.7% and 86.8% of precancerous or malignant lesions. Sensitivity for precancerous or malignant lesions increased from 53.9% to 75.0% after AI-assisted review (difference, 21.1 percentage points; 95% CI, 4.3 to 42.0). Conclusions: In this single-center study, clinicians' pre-laryngoscopic categorization accuracy was higher after reviewing the AI assessment card, although the sequential design cannot fully separate AI assistance from reassessment; gains depended on model correctness without clearly selective adoption. Randomized and external evaluations, including of the malignancy alert, are needed.

eess.AS · cs.AI

新论文 · v1arXiv:2611.20049

The Frontier Must Grow - Self-proposed, interpreter-labelled training roughly doubles a 4B model's accuracy on a fresh exam as its verified frontier grows

Martin Schroder

原文摘要

A model can learn from problems it poses itself when an executable oracle supplies the answers. Qwen3-4B proposed Lua 5.1 programs; those it predicted wrong at least once in four samples joined a pool labelled by the interpreter, each round adding twice as many programs as the last (300, 900, 2100 in total), and each round trained a fresh adapter from the base on the pool. On the original exam the procedure met its registered criterion, an item-level test of each adapter against its predecessor. On a 2340-item exam drawn afterwards, accuracy rose from 8.3% to 16.7%; that follow-up's registered criteria were not met because round 1 gained nothing, while rounds 2 and 3 added 5.0 and 2.9 points on average, both intervals above zero. One fit on the same amount of base-proposed frontier data reached 14.6%, so the rounds' 2.1-point advantage is inconclusive. Three earlier registered series met no per-lineage criterion, and the same loop showed no gain on repository tasks or for a 27B agent in those configurations.

cs.AI

版本更新 · v2arXiv:2604.02097

LatentUM: Unleashing the Potential of Interleaved Cross-Modal Reasoning via a Latent-Space Unified Model

Jiachun Jin, Zetong Zhou, Xiao Yang, Hao Zhang, Pengfei Liu, Jun Zhu, Zhijie Deng

原文摘要

Unified models (UMs) hold promise for their ability to understand and generate content across heterogeneous modalities. Compared to merely generating visual content, the use of UMs for interleaved cross-modal reasoning is more promising and valuable, e.g., for solving understanding problems that require dense visual thinking, improving visual generation through self-reflection, or modeling visual dynamics of the physical world guided by stepwise action interventions. However, existing UMs typically rely on pixel decoding as a bridge when their visual representations for understanding and generation are disjoint, which is both ineffective and computationally cumbersome. In this paper, we introduce LatentUM, a novel unified model that represents all modalities within a shared semantic latent space, eliminating pixel-space mediation for model-internal reasoning over generated visual states. By making generated visual tokens directly interpretable to the model itself, LatentUM enables flexible interleaved cross-modal reasoning and generation. Empirically, LatentUM achieves state-of-the-art performance on the Visual Spatial Planning benchmark, pushes the limits of visual generation through self-reflection and demonstrates world modeling capability by predicting future visual states in the shared semantic latent space. Code and models are available at https://github.com/SJTU-DENG-Lab/LatentUM.

cs.CV · cs.LG

版本更新 · v2arXiv:2611.16869

Raysemble: Metric 3D Bone Reconstruction from a Single X-ray

Xiaoyang Li, Xinghao Yan, Yuan Chai

原文摘要

Raysemble reconstructs eight lumbar and pelvic bones from a single X-ray by separating shape and placement. Each 3D query samples regional and global features at its image projection. A shared decoder predicts normalized bone shapes. Predicted depth and camera geometry determine 3D centers and scales for resampling these shapes onto a common grid in millimeters. Evaluation uses 23 held-out patients with 43 CT scans, each generating one simulated anteroposterior radiograph. Without rigid alignment to CT,Raysemble leads 15 evaluated methods, averaging 65.24% Dice, 9.09 mm HD95, and 3.32 mm ASSD over three random seeds. With identical inputs, it improves Dice over VolumeNeRF by 5.86 percentage points.

cs.CV

新论文 · v1arXiv:2611.20047

CleanRoute: Reconciling DRC Compliance, Post-Layout Performance, and Runtime in Analog Routing

Bingyang Liu, Ziming Wei, Xiaohan Gao, Yang Hsu, David Z. Pan

原文摘要

Automated analog routing remains a long-standing challenge because design-rule checking (DRC) compliance, post-layout performance preservation, and runtime reduction are difficult to achieve together. Existing analog routers have limited coverage of joined-metal design rules, lack control over topology-induced mismatch, and face a trade-off between DRC violations and runtime on uniform routing grids. In this paper, we introduce CleanRoute, an analog router that reconciles these three targets. Hierarchical rule enforcement assigns DRC checks and repairs to stages where the required geometry is complete, covering rules both between separate metal and where metal joins. IR-drop isolation constructs supply trees from current-mirror relationships and device currents to reduce topology-induced mismatch that geometric matching cannot eliminate and voltage drops along high-current paths. A flush-line grid adds net-specific lines aligned with device-metal edges to a coarse routing grid, providing DRC-compliant connections without uniform refinement. Experiments on ten operational transconductance amplifiers (OTAs) in a commercial 40nm process show that CleanRoute produces DRC-clean and functional layouts that pass layout versus schematic (LVS) checking on all benchmarks, with the strongest overall performance preservation and a median routing runtime of 8.82s, whereas the baselines produce no DRC-clean layouts.

cs.AR