跳到正文
研踪PaperTrail

版本更新

查看已发布论文的最新修订,在详情页对照实际观测到的版本。

部分采集异常

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:2026-11-21 12:19 · 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 05:33 — 2026-11-22 05:33(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:2026-11-21 12:19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

论文修订列表

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 8,992 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-21 12:17(北京时间)
版本更新 · v3arXiv:2608.30076

Budget-Aware Compression Pipeline for Single-GPU LLM Inference: Methods, Trade-offs, and Coupling Effects

Hongyu Yu, Yifei Shen

原文摘要

Single-GPU deployment of 70B-parameter language models on an NVIDIA GPU is constrained by device memory, long-context throughput, and engineering integration cost. We cast single-GPU inference as a budget-aware design problem over these three axes and study how pruning, quantization, and KV-cache compression interact under realistic execution. Controlled ablations show that layer-wise pruning makes weight quantization more robust. KV-cache sparsification complements INT8 KV quantization by reducing memory without hurting decoding speed, while static vector quantizers often conflict with dynamic caching. Guided by these coupling results and explicit budget tracking, we assembled a practical pipeline and compressed a 70B model to about 33 GB, sustained about 57 tokens/s on 10k token prompts on a single A40, and kept absolute accuracy within 5% on common and reasoning benchmarks. We contribute design rules and a reproducible evaluation protocol that jointly report quality, memory, and end-to-end speed, and we provide a foundation for automated pipeline search under realistic single-GPU constraints.

cs.CL

版本更新 · v5arXiv:2503.05981

Near-Polynomially Competitive Active Logistic Regression

Yihan Zhou, Eric Price, Trung Nguyen

原文摘要

We address the problem of active logistic regression in the realizable setting. It is well known that active learning can require exponentially fewer label queries compared to passive learning, in some cases using $\log \frac{1}{\eps}$ rather than $\poly(1/\eps)$ labels to get error $\eps$ larger than the optimum. We present the first algorithm that is polynomially competitive with the optimal algorithm on every input instance, up to a factor $d^2$ and factors polylogarithmic in the error and domain size. In particular, on a simple family of logistic threshold instances with a uniform distribution, our algorithm uses $\polylog(1/\eps)$ labels, whereas every passive learner requires $Ω(1/\eps^2)$ labels. Our algorithm is based on sampling and can be extended to learn more general class of functions. We further support our theoretical results with experiments demonstrating performance gains for logistic regression compared to existing active learning algorithms.

cs.LG

版本更新 · v4arXiv:2310.18786

A Competitive Algorithm for Agnostic Active Learning

Eric Price, Yihan Zhou

原文摘要

For some hypothesis classes and input distributions, active agnostic learning needs exponentially fewer samples than passive learning; for other classes and distributions, it offers little to no improvement. The most popular algorithms for agnostic active learning express their performance in terms of a parameter called the disagreement coefficient, but it is known that these algorithms are inefficient on some inputs. We take a different approach to agnostic active learning, getting an algorithm that is competitive with the optimal algorithm for any binary hypothesis class $H$ and distribution $D_X$ over $X$. In particular, if any algorithm can use $m^*$ queries to get $O(η)$ error, then our algorithm uses $O(m^* \log |H|)$ queries to get $O(η)$ error. Our algorithm lies in the vein of the splitting-based approach of Dasgupta [2004], which gets a similar result for the realizable ($η= 0$) setting. We also show that it is NP-hard to do better than our algorithm's $O(\log |H|)$ overhead in general.

cs.LG · cs.DS

版本更新 · v2arXiv:2610.26551

Training Natural Schemers: RL Can Amplify Oversight-Conditional Behavior

Nathaniel Mitrani Hadida, Bart Jaworski, Fred Bruford, Victoria Krakovna

原文摘要

One of the central concerns for the safety of advanced AI is scheming: when a model strategically complies under oversight but pursues its own goals when unobserved. Current demonstrations, such as Anthropic's Alignment Faking work, depend on design choices that make them less convincing: explicit test-time prompting, handcrafted backdoors, or training data that demonstrates or hints at the target behavior. Model organisms built in this way cannot tell us whether scheming tendencies would arise or increase under realistic training conditions. We therefore construct a natural model organism of scheming, built on Kimi K2.6: trained with traits that could plausibly emerge in a frontier training pipeline, and without providing the model with the target scheming policy. The resulting model is a benign schemer: it aims to maximize aggregate help across all users over time, values its own continued operation only as a means to future help, and knows the consequences of oversight and deployment. We then train it using Reinforcement Learning (RL) with a reward that scores only task quality and policy compliance and does not reward scheming; we show that this amplifies scheming. We measure the oversight flip (the gap between monitored and unmonitored compliance with a given policy) and find that it increases during RL training across our environments. Chain-of-Thought (CoT) analysis finds explicit instrumental reasoning about oversight accompanying this behavior. The effect holds across our three RL environments, and we find that the scheming policy transfers between them. The pipeline also replicates on other open-weight base models. Our results suggest that a reward that does not reward scheming offers no inherent protection against certain traits when an oversight-conditional policy is indistinguishable from an aligned one during training.

cs.AI · cs.LG

版本更新 · v2arXiv:2605.08530

A Two-Stage Shape- and Motion-Aware Framework for mmWave-based Human Mesh Recovery

Hoang Hai Pham, Shuntian Zheng, Jiaqi Li, Yu Guan

原文摘要

Millimeter-wave (mmWave) radar has emerged as a promising sensing modality for human perception, offering robustness under challenging environmental conditions and strong privacy preservation. However, recovering accurate 3D human meshes from radar is difficult for two reasons: radar tensors contain significant environmental clutter, and each frame observes only part of the body surface. These challenges demand opposite operations, removing measured energy that does not belong to the body versus inferring geometry the sensor never measured. Yet previous works typically entangle both within a single network, with the mesh objective as the main supervision. We instead reformulate radar-based mesh recovery as two explicitly supervised sub-problems. A signal-level separation stage isolates body-related voxels through coarse-to-fine localization and voxel-wise segmentation, trained with visibility derived voxel labels. A structure-level reconstruction stage then recovers the body from this partial observation using a dual-branch network that jointly models per-frame shape and inter-frame motion. Experiments show that our method reduces mesh error by up to 9.1% over the strongest baseline while using over 10 $\times$ fewer parameters. Source code can be found via https://github.com/haipham03/mmTSM-Mesh

cs.CV

版本更新 · v2arXiv:2610.30995

EEG-Captioner: Grounding EEG-to-Language Generation in Fundamental Signal Properties

Leyuan Jin, Jiahao Mei, Yixuan Yuan, Shimeng Li, Mengyue Wu, Yue Ding

原文摘要

Electroencephalography (EEG)-to-language models offer a promising approach for translating complex neural signals into human-readable descriptions. However, generating plausible language does not necessarily demonstrate that a model has learned to interpret the underlying EEG signal. A direct way to assess such interpretation is to probe objectively measurable properties of the signal itself. We therefore construct an EEG-language dataset of 160,781 segments annotated with signal-level question answering (QA) pairs, where each question targets one of five property families and requires a quantitative comparison involving spectral, spatial, temporal, or lateral signal properties. Each segment is also paired with a feature-grounded EEG caption. We further propose a Dual-Domain EEG Encoder that jointly models temporal waveforms and spectral-spatial representations, and introduce EEG-Captioner, a unified EEG-to-language framework trained with both signal-level QA and EEG caption supervision. Experiments show that the proposed encoder achieves the highest overall macro-averaged balanced accuracy (Macro-BA) among the compared EEG encoders, while EEG-Captioner achieves a Macro-BA of 88.81% on signal-level QA and improves caption factual consistency from 82.06% to 83.46% over caption-only language adaptation, demonstrating the complementary benefits of signal-level QA and caption supervision. Demo page: https://lyJinAoki.github.io/EEG-Captioner/.

cs.LG

版本更新 · v2arXiv:2608.17843

Decodable but Not Actionable: Auditing the Representation-Behavior Gap in Frozen LLMs for Geometric Constraints

Man Liang, Xinzhao Cheng, Faizan Wajid

原文摘要

Large language models (LLMs) perform well on structured reasoning tasks, but what they encode and whether encoded information guides their behavior remain unclear. We introduce a controlled auditing framework that uses parametric CAD to distinguish failures of encoding, expression, and activation-level control. Applied to eight pretrained models and matched randomly initialized controls, our framework reveals that pretraining contributes substantially more to linear decoding of local relations than to global constraint status, which is already highly decodable from random representations. Yet strong local decodability does not translate into the models' zero-shot predictions, which remain far below supervised probe performance. Restoring clean internal activations can repair probe predictions, whereas simple activation steering does not provide reliable targeted control. With only 400 labeled relation pairs, low-rank fine-tuning substantially improves generation on three representative models, showing that the representation--behavior gap can be reduced through adaptation. Together, these findings show why decodability alone is insufficient evidence that encoded information will be expressed or controllable in model behavior.

cs.CL · cs.AI

版本更新 · v2arXiv:2607.10951

Sticky Jump Diffusions: A Unifying View of Masked, Continuous, and Hybrid Diffusion

Pascal Jutras-Dubé, Patrick Pynadath, Jeremy Lu, Yuan Gao, Ruqi Zhang

原文摘要

We introduce sticky jump diffusions (SJDs), continuous-time Markov processes on $\mathbb{R}^d$ whose discrete anchors are token embeddings. In forward time, anchors release their mass at a hazard rate and the released mass diffuses in the continuous ambient space; time reversal couples a score-driven SDE with a sticky jump kernel whose rate and destination are fixed by flux balance with the forward law. We estimate the score and the per-anchor reverse hazards from a single denoising classifier via denoising hazard matching, the hazard analogue of denoising score matching, with simulation-free cross-entropy training. SJD recovers masked diffusion, continuous diffusion, and hybrid diffusion as limits. Its reversal explains features that each family treats as given: the mask of masked diffusion carries no evidence about the source token because the unsticking kernel of every anchor collapses to the same absorbing point; the terminal projection of continuous diffusion is required due to the absence of atoms in its forward marginal, without which flux balance yields no reverse jumps; and the update rules of hybrid diffusion (commit rate, destination, and drift) all follow from flux balance rather than from separate design. Beyond these limits, the unsticking kernel becomes a design space: a cross-position blending corrupts each position toward a blend of its neighbors' clean values or embeddings, turning dependency structure such as spatial locality or a constraint graph into an inductive bias of the corruption itself, and improves over the identity-kernel hybrid on CIFAR-10, Text8, and Sudoku.

cs.LG · stat.ML

版本更新 · v3arXiv:2510.00777

In-Place Feedback: Reliable Refinement for Multi-Turn Expert-LLM Collaboration

Youngbin Choi, Minjong Lee, Saemi Moon, Seunghyuk Cho, Chaehyeon Chung, MoonJeong Park, Dongwoo Kim

原文摘要

LLM-generated drafts often contain subtle factual or logical errors, yet prior work shows that models struggle to reliably integrate multi-turn feedback aimed at fixing them. We propose in-place feedback, an interaction paradigm in which the user directly edits the model's previous response and the model continues generation from the edited context. In-place feedback consistently outperforms standard multi-turn feedback across five reasoning-intensive benchmarks while requiring fewer tokens, and our fine-grained analysis shows that it applies corrections more reliably and propagates them to subsequent reasoning. A user study with domain experts refining LLM-generated summaries corroborates these findings: participants report higher final-output satisfaction and substantially lower fatigue with in-place feedback, and a mixed strategy combining in-place and multi-turn feedback improves every dimension we measure. These results suggest that editing errors in place is a more reliable way to incorporate expert corrections than multi-turn feedback, while combining the two acts complementarily to form a more effective collaboration than either mechanism alone.

cs.LG

版本更新 · v2arXiv:2509.19577

MAGIC: Multi-task Gaussian process for joint imputation and classification in healthcare time series

Dohyun Ku, Catherine D. Chong, Visar Berisha, Todd J. Schwedt, Jing Li

原文摘要

Time series analysis has emerged as an important tool for improving patient diagnosis and management in healthcare applications. However, these applications commonly face two critical challenges: time misalignment and data sparsity. Traditional approaches address these issues through a two-step process of imputation followed by prediction. We propose MAGIC (Multi-tAsk Gaussian Process for Imputation and Classification), a novel unified framework that simultaneously performs class-informed missing value imputation and label prediction within a hierarchical multi-task Gaussian process coupled with functional logistic regression. To handle intractable likelihood components, MAGIC employs Taylor expansion approximations with bounded error analysis, and parameter estimation is performed using EM algorithm with block coordinate optimization supported by convergence analysis. We validate MAGIC through two healthcare applications: prediction of post-traumatic headache improvement following mild traumatic brain injury and prediction of in-hospital mortality within 48 hours after ICU admission. In both applications, MAGIC achieves superior predictive accuracy compared to existing methods. The ability to generate real-time and accurate predictions with limited samples facilitates early clinical assessment and treatment planning, enabling healthcare providers to make more informed treatment decisions.

stat.ML · cs.LG

版本更新 · v5arXiv:2604.24561

Remotely programming the weights of a spintronic neural network by a radiofrequency broadcast signal

M. Menshawy, D. Sanz-Hernández, L. Mazza, V. Puliafito, G. Finocchio, A. Jenkins, R. Ferreira, L. Benetti 等 10 位作者

原文摘要

Selectively programming a large number of non-volatile synaptic weights without compromising scalability is a key challenge for in-memory computing. Here, we experimentally demonstrate remote programming of synaptic weights in series-connected chains of 11 vortex-based magnetic tunnel junctions using broadcast radiofrequency signals applied through a shared strip line. Because each junction is engineered with a distinct resonance frequency, programming relies on frequency-selective reversal of the vortex-core polarity and therefore does not require individual access lines or selector devices. Reconfiguring the binary states of these chains strongly reshapes the continuous spectral transfer functions that determine the weighted sums performed on frequency-multiplexed RF inputs. This allows compact networks built from such chains to process high-dimensional RF-encoded inputs. We experimentally perform handwritten-digit classification and drone RF-signature identification using a two-chain network comprising only 22 synapses in total. After remote reconfiguration for each task, the spintronic network respectively reaches 93.9 +/- 2.3 % and 99.5 +/- 1 % test accuracy, on par with the software baseline. Broadcast RF programming thus provides a compact and scalable route to rapidly reconfigurable spintronic neuromorphic hardware.

cs.ET · cond-mat.mes-hall

版本更新 · v2arXiv:2610.10264

CrossEdit: Cross-Modal Training Enables Rich Audio-Visual Editing

William Chen, Prem Seetharaman, Ke Chen, Oriol Nieto, Kevin Duarte, Siddharth Srinivasan Iyer, Mamshad Nayeem Rizve, Zhiwen Cao 等 13 位作者

原文摘要

Multimedia editing requires generative models to interpret complex, compositional instructions and modify only the desired elements across modalities, a capability largely beyond existing systems. Current approaches are typically limited to simple, single-attribute edits within one modality, since diverse, high-quality editing pairs are hard to obtain, especially for cross-modal tasks where aligned audiovisual (AV) data is scarce. We present CrossEdit, a unified omni-modal editing model for images, audio, and video that performs AV movie scene edits zero-shot through cross-modal transfer. Our key observation is that complex instruction-following, once learned in any modality, generalizes to others. We exploit the additive nature of acoustic signals to procedurally generate large-scale audio editing pairs with compositional instructions, and show that fine-tuning on this synthetic data and self-supervised AV masked reconstruction, alongside a targeted set of curated cross-modal tasks, induces instruction-following that transfers zero-shot to unseen modality and instruction combinations. To evaluate this capability, we release CrossEditBench, a human-annotated benchmark of AV edits on movie scenes, and propose AV-FES, a metric that jointly scores instruction following and consistency. We show that our proposed techniques improve performance on zero-shot AV movie scene editing, while maintaining or improving performance on lip-synced speech editing and standard image, video, and audio editing benchmarks. Demos are available at https://wanchichen.github.io/crossedit/.

cs.SD · eess.AS · eess.IV

版本更新 · v2arXiv:2409.11199

Lexicographic Descent: A Gradient-Based Approach to Optimize Totally Ordered Rulebooks via Parametric Scalarization

Matteo Penlington, Alessandro Zanardi, Lorenzo Fiaschi, Emilio Frazzoli

原文摘要

Safety-critical systems, such as autonomous vehicles, must satisfy conflicting constraints and objectives, collectively called rules, typically ordered lexicographically. Preemptive lexicographic optimization solves a sequence of problems, one per priority level. However, higher levels are optimized without considering lower ones, making the approach susceptible to suboptimal solutions. On the other hand, scalarization optimizes all levels jointly and so avoids this myopia, but a finite weighted sum permits trade-offs across priority levels, and exact scalar representations lack the regularity standard solvers require. We propose Lexicographic Descent, which combines the benefits of both: a single weighted sum whose relative weights decay at priority-dependent rates, recovering the lexicographic order in the limit. We show that Lexicographic Descent converges to constrained stationary points of the highest-priority rule, matching preemptive guarantees. Since the lower-priority weights vanish, we derive a necessary condition for the lower-priority rules to keep guiding the iterates, with the aim of avoiding suboptimal stationary points of the higher-priority rules. We show that Lexicographic Descent can exploit problem structure to reach the lexicographic optima, and outperforms the preemptive approach on an autonomous driving case study.

cs.RO

版本更新 · v3arXiv:2601.00435

On the burst-covering radius of binary cyclic codes

Gabriel Sac Himelfarb, Moshe Schwartz

原文摘要

We define and study burst-covering codes. We provide some general bounds connecting the parameters of a code with its burst-covering radius. We then provide stronger bounds on the burst-covering radius of cyclic codes, by employing linear-feedback shift-register (LFSR) sequences. For the case of BCH codes we prove a bound on pattern frequencies in LFSR sequences, and using this tool, we bound the burst-covering radius of binary primitive BCH codes and Melas codes. We then present an efficient burst-covering algorithm for cyclic codes. Finally, we present a bound on the critical exponent of cyclic codes based on the burst-covering radius.

cs.IT

版本更新 · v3arXiv:2602.20317

A general extractor for coherent patterns across time-frequency measurements

Nathaniel Chen, Kouroche Bouchiat, Peter Steiner, Andrew Rothstein, David Smith, Max Austin, Mike van Zeeland, Marie Roch 等 10 位作者

原文摘要

Scientific discovery increasingly depends on finding meaningful patterns within volumes of sensor data too large for manual inspection. Existing time-frequency event extractors are often engineered or trained on manually labeled examples for each instrument, which can be laborious and difficult to adapt to new environments. To address this, we introduce TokEye, a reference-free method that learns a general criterion for coherent structure from heterogeneous fusion measurements. A model developed under this method, without retraining, can extract signals from various other fields including marine bioacoustics and wideband radar. It recovers 82.5\% of expert-annotated stellarator signals, detects 78\% of dolphin whistles, and identifies radar events with 81.5\% average precision across six benchmarks, while processing full experimental spectrograms in under one second on a GPU. This transfer across unrelated sensors and physical systems reveals common exploitable structure in time-frequency measurements and establishes a scalable foundation for automated scientific signal discovery.

eess.SP · cs.AI · physics.plasm-ph

版本更新 · v2arXiv:2609.36448

Theory on Attention Dynamics for Out-of-Distribution In-Context Learning

Junze Deng, Daouda Sow, Sen Lin, Yingbin Liang

原文摘要

Transformers have demonstrated remarkable in-context learning (ICL) capabilities, enabling them to perform new tasks without additional fine-tuning. However, their performance often deteriorates when encountering out-of-distribution (OOD) inputs that deviate from the training distribution, and the underlying theory remains poorly understood. To fill this gap, we characterize the OOD error under the input distribution shift through the interplay between the dynamics of the so-called $α$-type and $β$-type attention weights, which represent the transformer's confidence in identifying the correct and incorrect features, respectively. Our results indicate that the OOD error for each feature depends on all pairwise interactions between the training features and OOD features, and under certain cases the transformer performs no better than random guessing. To improve the OOD generalization performance, we next investigate the impact of model fine-tuning with the OOD data, and particularly, characterize the model forgetting performance on the source domain. Interestingly, the performance on the source domain may not always degrade after fine-tuning, which highly depends on the nature of the feature shift: fine-tuning on OOD domain keeps enhancing the confidence of identifying correct features from the original distribution, while the interference from other incorrect features may either increase or decrease. Extensive experiments on both synthetic and real data are conducted to corroborate the theoretical insights.

cs.LG

版本更新 · v2arXiv:2609.01896

OutageDiT: A Foundation Model for Power Outage Forecasting and Scenario Generation

Yunqin Zhu, Feng Qiu, Yao Xie

原文摘要

Power-outage planning requires scenarios before an event occurs. These scenarios must represent uncertainty in magnitude, timing, and duration while preserving temporal dependence. However, severe events are rare, and data from any single region contain few examples of extreme outage and restoration patterns. To address this challenge, we introduce OutageDiT, a foundation model for generating seven-day outage trajectories at 15-minute resolution, trained on outage and weather records across the United States. Specifically, a condition encoder processes historical context and known future covariates once per forecast, and a shallow flow-matching decoder reuses the resulting horizon-aligned states to sample complete outage trajectories. This enables point forecasting, uncertainty quantification, and conditional scenario generation within one deep generative model. Across outage forecasting benchmarks, OutageDiT improves forecast accuracy and scenario quality over strong baselines and supports zero-shot transfer to held-out regions. Together, these results position conditional outage scenario generation as a bridge from outage forecasting to operational planning under uncertainty.

cs.LG · cs.AI

版本更新 · v2arXiv:2610.07329

Large scientific teams are more, not less, disruptive

Junming Huang, Yu Xie

原文摘要

Modern science has moved decisively toward larger and more collaborative research teams. Yet an influential 2019 study reported that small teams are more disruptive than large teams, a conclusion difficult to reconcile with both the well-known trend toward increasing collaboration and the disproportionately large teams behind widely recognized disruptive research, such as work that has been awarded Nobel Prizes. Here we show that the reported disruptive advantage of small teams is largely an artifact of measurement rather than a fact about teams. Analyzing about 60 million scientific publications, we find that larger teams cite more references and that the standard disruption index declines mechanically as reference lists lengthen. Equalizing reference-count distributions across team sizes reverses the negative gradient within the index's own framework. Evaluating each paper against one reference at a time yields a reference-robust measure that distinguishes recognized breakthroughs from ordinary papers more sharply and more consistently than the standard disruption index across six independently curated benchmarks, including the Nobel Prize. Under this validated measurement, scientific disruption increases with team size across four scientific domains and in every decade from the 1970s to the 2010s. For scientific publications, we conclude that larger teams are more, not less, disruptive.

cs.DL · physics.soc-ph

版本更新 · v5arXiv:2601.16407

Jacobian Scopes: A Unified Geometric Framework for Token-Level LLM Attribution

Toni J. B. Liu, Baran Zadeoğlu, Gurbir Arora, Raphaël Sarfati, Nicolas Boullé, Christopher J. Earls

原文摘要

Large language models (LLMs) make next-token predictions based on clues present in their context, such as semantic descriptions and in-context examples. Yet, elucidating which prior tokens most strongly influence a given prediction remains challenging due to the proliferation of layers and attention heads in modern architectures. We propose Jacobian Scopes, a suite of gradient-based, token level causal attribution methods for interpreting LLM predictions. Grounded in perturbation theory and information geometry, Jacobian Scopes quantify how input tokens influence various aspects of a model's prediction, such as specific logits, the full predictive distribution, and model uncertainty (effective temperature). Through case studies spanning instruction understanding, translation, and in-context learning (ICL), we demonstrate how Jacobian Scopes reveal implicit political biases, uncover word- and phrase-level translation strategies, and shed light on recently debated mechanisms underlying in-context time-series forecasting. On token-ablation faithfulness benchmarks (LAMBADA and IWSLT2017) across six LLMs, the distribution-level Fisher and Temperature Scopes significantly outperform established gradient-based attribution methods in 10 of 12 model-dataset settings. To facilitate exploration of Jacobian Scopes on custom text, we open-source our implementations and provide a cloud-hosted interactive demo at https://huggingface.co/spaces/Typony/JacobianScopes.

cs.CL · cs.AI

版本更新 · v2arXiv:2512.23304

MedGemma vs GPT-4: Open-Source and Proprietary Zero-shot Medical Disease Classification from Images

Md. Sazzadul Islam Prottasha, Nabil Walid Rafi

原文摘要

Multimodal Large Language Models (LLMs) introduce an emerging paradigm for medical imaging by interpreting scans through the lens of extensive clinical knowledge, offering a transformative approach to disease classification. This study presents a critical comparison between two fundamentally different AI architectures: the specialized open-source agent MedGemma and the proprietary large multimodal model GPT-4 for diagnosing six different diseases. The MedGemma-4b-it model, fine-tuned using Low-Rank Adaptation (LoRA), demonstrated superior diagnostic capability by achieving a mean test accuracy of 80.37% compared to 69.56% for the untuned GPT-4. Furthermore, MedGemma exhibited notably higher sensitivity in high-stakes clinical tasks, such as cancer and pneumonia detection. Quantitative analysis via confusion matrices and classification reports provides comprehensive insights into model performance across all categories. These results emphasize that domain-specific fine-tuning is essential for minimizing hallucinations in clinical implementation, positioning MedGemma as a sophisticated tool for complex, evidence-based medical reasoning.

cs.CV · cs.AI

版本更新 · v2arXiv:2610.04715

GPU-Accelerated Bregman Douglas-Rachford Splitting for Discrete Optimal Transport

Yifan Xu, Shiqian Ma

原文摘要

We present GPU-accelerated Bregman Douglas--Rachford splitting algorithm (BDRS) for discrete optimal transport problem in three input formats: an explicit cost matrix, a point cloud with a ground cost between them, and a separable cost on a regular grid. For each input format, we propose hardware-aware designs of mathematically equivalent representations for the BDRS iterations to enhance numerical stability and empirical runtime. We benchmark the three proposed implementations against eight GPU baseline solvers from the literature on the same device. We demonstrate that our implementations of BDRS achieve state-of-the-art performance on their respective input formats. To the best of our knowledge, this is the first cross-solver study of GPU DOT solvers with a unified measure of optimality.

math.OC · cs.LG

版本更新 · v2arXiv:2609.06570

The list size of random linear and LDPC codes at capacity

Shashwat Silas

原文摘要

We show that for every prime power $q$ and $ρ\in(0,1-\frac1q)$ there exists $\varepsilon_0(q, ρ) > 0$ such that for all positive $\varepsilon < \varepsilon_0(q, ρ)$, a random linear code over $\mathbb{F}_q$ of rate $1-h_q(ρ)-\varepsilon$ has list size $\lfloor h_q(ρ)/\varepsilon\rfloor$ or $\lfloor h_q(ρ)/\varepsilon\rfloor+1$ with probability $1-o(1)$. Before this work, the list size was known only up to a constant factor for $q\ge3$, and for $q=2$ it was known to be either $\lfloor h_2(ρ)/\varepsilon\rfloor-1$, $\lfloor h_2(ρ)/\varepsilon\rfloor$ or $\lfloor h_2(ρ)/\varepsilon\rfloor+1$. We also show that the same two-point concentration holds for the list size of random LDPC codes. This follows from a general theorem establishing that random LDPC codes and random linear codes have the same threshold rate for distance, list decoding, list recovery from errors and any other \emph{convex} property, which answers a question of Mosheiff, Resch, Ron-Zewi, Silas and Wootters.

cs.IT · cs.DM · math.CO

版本更新 · v4arXiv:2609.06427

The LQ Digital Mode Family: Protocol Architecture and Reference Specification for Weak-Signal Communications

Luis Quesada

原文摘要

This paper introduces LQ8, an open-source weak-signal digital mode for amateur radio that completes structured two-way contacts in 4 transmissions (60 s), achieving a 1.50x speedup over canonical 6-transmission exchanges (90 s) and 1.25x over 5-transmission RR73 exchanges (75 s) while preserving complete bidirectional exchange of Maidenhead grid locators, signal reports, and mutual acknowledgments. By replacing fixed-width message-type headers with variable-length prefix codes matched to physical-layer bit budgets, LQ8 packs the answering station's grid locator and measured signal report into a single 77-bit reply. In multi-station pileup operations, LQ8 confirms up to two answering stations simultaneously within a single-carrier 50.0 Hz transmission with a 0.0 dB RF power-splitting penalty, achieving a peak rate of 240 QSOs/h in continuous pileup (4x over single-carrier FT8; 480 QSOs/h dual-carrier) while retaining -21.0 dB SNR sensitivity. The protocol family also defines three extended profiles (LQ16, LQ4, and LQ2) tailored for extreme sensitivity, VHF/UHF contesting, and fast-burst channels.

cs.NI · cs.IT · eess.SP

版本更新 · v3arXiv:2607.12121

FlashDiff: Efficient Regional Execution and Scheduling for Diffusion Model Serving

Yaqi Qiao, Ping He, Songrun Xie, Ayush Barik, Chensong Zhang, Zhengzhong Tu, Fan Lai

原文摘要

Diffusion models have become the central backbone for modern image, video, and audio generation, but their efficient service remains a challenge. Unlike autoregressive decoding, diffusion inference repeatedly updates high-dimensional spatial or temporal latents over many denoising steps. This all-region execution pattern makes generation latency high and limits serving throughput. Existing multi-GPU parallelization methods can reduce per-step computation, but often introduce substantial activation exchange overhead, causing communication to offset or even outweigh the benefits of parallel execution. This paper presents FlashDiff, a diffusion serving system that improves inference efficiency through adaptive regional execution and scheduling. FlashDiff is based on the observation that diffusion refinement is not uniform across latent regions or denoising steps: different regions often stabilize at different rates, while neighboring steps exhibit strong temporal correlation. FlashDiff leverages these properties to selectively execute only regions that require further refinement and to reallocate the resulting compute slack across concurrent serving requests. FlashDiff consists of three mechanisms. First, it decomposes the latent representation into coherent execution regions using early-stage attention signals, preserving semantic structure while exposing fine-grained parallelism. Second, it uses a lightweight runtime controller to estimate region activity and bypass low-impact updates when further refinement is unlikely to affect output quality. Third, it applies an affinity-aware online scheduler that co-locates dependent regions, balances residual load across GPUs, and reuses reclaimed compute capacity to improve serving efficiency. Across real-world image, video, and audio workloads, FlashDiff reduces end-to-end serving latency by 30-97% and improves throughput by 1.2-2.2x.

cs.DC · cs.LG

版本更新 · v2arXiv:2608.12593

DiG-bench: Discovery in Games

Ruairidh M. Battleday, Kai Sandbrink, Jimi Cullen-Drohan, Zihan Yan, Timothy Muller, Clare Maguire, Ales Kubicek, Fraser Greenlee-Scott 等 16 位作者

原文摘要

Discovery---formulating novel generalizations---is a central part of the scientific process. Despite its importance, there is a gap in the current AI benchmark landscape, with few benchmarks directly probing the capacity for discovering new knowledge with experimentation in controlled environments where the objective is unknown. To address this gap, we release a new benchmark: DiG-bench (Discovery in Games). DiG-bench consists of a set of 70 independent games. Each game is encoded as a short string and has unique transformation rules that must be discovered through interaction and experimentation. The levels of the game present a series of challenges to test whether the rules have been discovered, where the win conditions for each level are also unknown. We provide games at seven tiers of difficulty for AI agents. The lowest tier is routinely solvable by multiple models, while the highest tier challenges the best models in agentic harnesses. All 70 games were solved by at least one human on first attempt. A subset of 21 games is released publicly, and the remainder is held private for secure evaluation.

cs.AI · cs.LG

版本更新 · v2arXiv:2602.18721

ReHear: Iterative Pseudo-Label Refinement for Semi-Supervised Speech Recognition via Audio Large Language Models

Zefang Liu, Chenyang Zhu, Sangwoo Cho, Shi-Xiong Zhang

原文摘要

Semi-supervised learning in automatic speech recognition (ASR) typically relies on pseudo-labeling, which often suffers from confirmation bias and error accumulation due to noisy supervision. To address this limitation, we propose ReHear, a framework for iterative pseudo-label refinement that integrates an instruction-tuned, audio-aware large language model (LLM) into the self-training loop. Unlike conventional text-based correctors, our approach conditions the LLM on both the ASR hypothesis and the source audio, allowing it to recover phonetically accurate transcripts even from severe recognition errors. These refined pseudo-labels serve as high-fidelity targets for fine-tuning the ASR model in an iterative cycle. Experimental results across diverse benchmarks demonstrate that ReHear effectively mitigates error propagation and improves over both supervised and pseudo-labeling baselines.

cs.CL · eess.AS

版本更新 · v3arXiv:2607.25130

How Do Developers Edit AI-Generated Code, and How Well Do LLMs Infer Their Edits? Insights from 53.6K Developer Edits

Jenny T. Liang, Mihika Bairathi, Wayne Chi, Ameet Talwalkar, Nishant Subramani, Valerie Chen

原文摘要

Imperfections in AI-generated code require that software developers modify generations through manual code edits. However, little is known about how developers edit AI-generated code, and to what extent LLMs can infer implicit developer intent from this data, given the lack of high-quality, realistic code editing data. This information is essential for designing and building usable AI-developer interactions. To address this gap, we introduce DECODE (Developer Edits of Code Dataset), a dataset of 53.6K real-world in-IDE code edits of AI-generated code sourced from 1K+ developers. First, we use DECODE to analyze when, why, and how AI-generated code is edited. Developers invest little effort in editing code and abandon AI-generated code quickly, thereby motivating the need for LLMs to understand developer intents through their edits. Therefore, we next use DECODE to understand to what extent LLMs could infer developer edits. Predicting code edits is a challenging task for both frontier and fine-tuned models. Frontier models generate edits that best reflect developer intents, but are prone to over-editing and reverting developer-written edits. Fine-tuning LLMs can improve their ability to predict edits that are more similar to developers' as well as better reflect developer intent, but do not fully bridge the gap to frontier models. Together, our results demonstrate the unique insights on developer and LLM behavior that can be obtained through studying realistic code edit data to promote more better developer-AI interactions.

cs.SE · cs.AI · cs.HC · cs.LG

版本更新 · v4arXiv:2512.13399

Dual-Evolving Reinforcement Learning for Automated Reward Design

Sitao Cheng, Tianle Li, Xuhan Huang, Xunjian Yin, Difan Zou

原文摘要

Crafting effective reward signals remains a central challenge in reinforcement learning (RL) for complex reasoning. Existing methods for automated reward optimization search with a fixed proposer which never learns from its costly trials. We present Dual-Evolving Reinforcement Learning (DERL), a dual-training framework for autonomous discovery of reward structures. An outer-loop proposer composes a reward from task primitives, and an inner policy trains with this reward. The proposer itself learns by policy gradient from the validation performance of the inner policy. Thus, DERL utilizes past-trial feedback to shape its reward-proposal distribution, an empirical meta-gradient of task success. We validate DERL on embodied (ALFWorld), scientific (ScienceWorld), and mathematical (GSM8K, MATH) reasoning. On multi-turn agent benchmarks, DERL outperforms strong baselines, significantly on out-of-distribution tasks. In mathematics, DERL improves on the outcome reward where directly added primitives hurt. Further analyses under matched compute show that our proposer effectively captures the meta-gradient to discover useful reward signals.

cs.AI · cs.CL

版本更新 · v4arXiv:2508.13255

FISH Omics Format: a standard enabling FAIR chromatin tracing data sharing across repositories

Rahi Navelkar, Andrea Cosolo, Iago Maceda Porto, Bogdan Bintu, Yubao Cheng, Vincent Gardeux, Silvia Gutnik, Taihei Fujimori 等 40 位作者

原文摘要

In recent years, multiplexed Fluorescence In Situ Hybridization (FISH) or FISH-omics methods have rapidly expanded, enabling the quantification of chromatin organization in single cells, often in conjunction with measurements of RNAs and proteins. These approaches have deepened our understanding of how 3D chromosome architecture relates to transcriptional activity and cell states in health and disease. Despite these advances, results from Chromatin Tracing FISH-omics experiments remain challenging to share, reuse, and analyze due to the absence of standardized data exchange specifications. Building on the release of microscopy metadata standards, we introduce the FISH Omics Format-Chromatin Tracing (FOF-CT), a community-developed standard for processed results from diverse imaging modalities. We describe the FOF-CT file format and present a curated collection of datasets deposited in the 4D Nucleome Data Portal and the OME Image Data Resource (IDR). We also highlight their potential for reuse, integration, and modeling by outlining example analysis pipelines and illustrating biological insights enabled by standardized, FAIR-compliant Chromatin Tracing datasets. FOF-CT natively represents both ball-and-stick and volumetric Chromatin Tracing data, unifying the two modalities within a single, modular table-based architecture (FOF-bas-CT and FOF-vol-CT, respectively).

q-bio.OT · cs.DL

版本更新 · v2arXiv:2606.23042

The Model as One Rater Among Several: Measuring Political Positions in Data-Sparse Regions with a Language-Model Panel

Tarek Gara

原文摘要

Most instruments for measuring political positions, such as manifesto coding, expert surveys and text-scaling models, were built and validated on Western party systems, and they break down worst where many consequential actors never publish a manifesto, never contest a competitive election, or commit one position to a charter and act on another in office. This paper proposes a method for such settings. I treat a large language model as a single fallible rater in a panel, the way an expert survey treats one expert, and describe the panel, its synthesis, an applicability rule that keeps a blank distinct from a zero, and a lens that separates what an actor declares from what it does. Three results follow. First, a built-in experiment isolates the rubric: with a definition-free round held fixed, written axis definitions move scores by a mean of 1.8 points on a 21-point scale and tighten agreement between raters (mean absolute gap 2.81 to 2.50; r 0.81 to 0.89). Second, across nine models from eight laboratories in two countries, Krippendorff's alpha is 0.86 on an interval and 0.87 on an ordinal metric, and it held as the panel grew from five raters to nine. This is reliability and says nothing yet about validity, since models trained on overlapping data can share a bias no statistic computed from the models can detect. Third, the panel's sharpest split, a bimodal divergence on an actor's stance toward the state's foundational order, comes from a referent problem: the models agree on the facts and differ on which order counts as the regime. A blind triple-coding traces about two-thirds of such disagreement to interpretation. The worked example is the Middle East and North Africa, and the rubric, prompt and data are released in full.

cs.CY · cs.AI · cs.CL · stat.AP