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

+729新增论文
+575其中 AI 相关
+82已有论文修订
+95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 2026-11-20 16:31 — 2026-11-21 16:31(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

最近完成检查:2026-11-21 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

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 6,976 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)
新论文 · v1arXiv:2611.21937

SLED: Sensor SLicing from Edge to Datacenter

Alan Subedi, Avhishek Biswas, Apala Pramanik, Amlan Balabantaray, Mikhail Titov, Santosh Pitla, Ozgur Kilic, Mehmet C. Vuran

原文摘要

Modern sensing systems generate large volumes of heterogeneous data streams from multiple sensor modalities, increasing reliance on datacenters for computation, artificial intelligence/superintelligence (AI/SI) inference, and AI/SI training. Each sensor stream may have different network and compute requirements depending on its type and data volume. However, current wireless systems typically allocate resources at the device level, leading to contention among co-located sensor streams, where bulk data transfers may starve latency-critical telemetry. To this end, this paper introduces Sensor Slicing (SL). SL provides deterministic, medium access control (MAC)-level isolation for individual sensors within a single user equipment (UE) in open radio access networks (O-RAN). Using agentic coding techniques, SL is implemented in two O-RAN-compliant packages, reducing development time by 400 times (2 h 40 min [OCUDU] vs. 6.5 weeks [OAI]), 228-fold fewer human prompts (4 vs. 913), and less than half the tokens (0.64B vs. 1.43B). An associated end-to-end framework, SLED, is implemented to establish per-sensor radio and network isolation across edge, campus, and remote high-performance computing (HPC) tiers. Over-the-air wireless experiments demonstrate that while conventional network slicing causes critical delays to increase more than 2,300-fold under high traffic load, SLED maintains stable latency. Scaling evaluations on the two RAN stacks show that throughput remains within 0.9% and 1.4% of expected allocations even as the number of sensor slices increases. The results demonstrate that SLED preserves latency-critical telemetry while effectively utilizing resources across the sensor-to-HPC pipeline. The SLED software and evaluation datasets are publicly available at https://github.com/UNL-CPN-Lab/SLED-Sensor-SLicing-from-Edge-to-Datacenter.

cs.AI · cs.CE · cs.DC · cs.ET · cs.NI

新论文 · v1arXiv:2611.21932

V2V Communications Using Blockchain-Enabled 6G Technology and Federated Learning

Tahir H. Ahmed, Jun Jiat Tiang, Azwan Mahmud, Dinh-Thuan Do, Truong Tran, Shahid Mumtaz

原文摘要

This study proposes an interesting approach for vehicle-to-vehicle (V2V) communication, which integrates blockchain technology, federated learning (FL), and allocation optimization of latency and resources. The research evaluates the proposed system using various performance metrics such as packet delivery ratio (PDR), model accuracy, and latency and demonstrates its superiority over existing techniques. Furthermore, the system provides enhanced security through consensus optimization and k-anonymity for data privacy. Overall, the proposed system is a promising solution for efficient and secure V2V communication in the era of connected and autonomous vehicles. Moreover, the proposed approach achieves higher reliability, lower latency, and better resource utilization compared to traditional 5G.

cs.NI · cs.AI · cs.CR · cs.LG

版本更新 · v2arXiv:2611.02380

Malware Clustering in the Wild: Taxonomy, Challenges, and Evaluation Metrics

Raguvir S, Bojing Li, Prajna Bhandary, Charles Nicholas

原文摘要

As malware continues to grow in volume, diversity, and sophistication, clustering has emerged as a critical technique for analyzing large malware corpora. Clustering enables the discovery of structural relationships, behavioral patterns, and evolutionary trends, and is widely used to group unknown samples, infer malware family boundaries, and identify novel variants. However, existing research remains fragmented, employing a wide range of clustering techniques with no unified perspective that systematically organizes prior work or clearly identifies gaps in the literature. This survey addresses this gap by organizing previous malware clustering research into seven categories: centroid-based, hierarchical-based, density-based, distribution-based, mixed- type, fuzzy, and AI-driven approaches. In addition, this survey summarizes publicly available malware datasets commonly used in clustering research and examines the feature representations derived from these datasets. Our goal is to clarify the current landscape and outline promising avenues for future research on malware clustering, including how AI-driven methods may shape large-scale and reproducible analysis.

cs.CR · cs.AI

新论文 · v1arXiv:2611.21925

DNeo: Your Parallel Draft Model Needs Better Target Feature Encoding

Jameson Sandler, Shane Bergsma, Mostafa Elhoushi, Bin Claire Zhang, Gavia Gray, Joel Hestness, Nolan Dey, Ferdinando Fioretto

原文摘要

Speculative-diffusion decoding uses a small parallel drafter model to generate b draft tokens in one forward pass, which then undergo parallel verification via a larger target model. Faster inference can be achieved when multiple draft tokens pass verification simultaneously. Recent parallel draft architectures (DFlash, DSpark) improve draft quality by conditioning on rich features computed by the target, through a mechanism known as feature encoding. However, standard feature encoding mechanisms have three limitations: (1) they are lossy, applying encoding projections that are destructive to the underlying target feature-space via a discovered effect known as residual mixing, (2) they are memory-less, with no record of previously processed proposals, and (3) they utilize features sampled uniformly from across the target layers, omitting important hidden states. To address these issues, we propose Mixture-of-Features Attention (MoFA), a lightweight feature-encoding approach equipped with a working memory that encodes previously rejected proposals. Leveraging MoFA, rejection memory, and a set of further enhancements, including 2-tap causal convolutions, prefix-only cross-attention, and non-uniform target layer selection, the final architecture, DNeo, achieves 1.3x faster generation speed than SoTA (DFlash-2) without impacting drafter footprint or generation quality. DNeo marks a new state of the art in high quality, fast LLM inference.

cs.AI · cs.CL

版本更新 · v2arXiv:2608.27754

Towards Efficient Auto-Interpretability of Boltz-1 Sparse Autoencoders

Piotr Jedryszek, Oliver M. Crook

原文摘要

Sparse autoencoders (SAEs) could turn AI models in Biology and other fields into engines of scientific discovery by explaining the superhuman capabilities of those models. We postulate that a latent is useful if we know three things: whether it is coherent, whether it can be described, and whether that description has predictive power. These questions are routinely conflated. We assemble them into a single pipeline and report some practical innovations. First, cross-seed dictionary stability prioritises which latents are worth spending resources to investigate. Second, an intruder-detection task asks whether a latent's activating examples share a recognizable pattern. Third, a separate pass proposes a candidate biological description which we convert into falsifiable predictions which can be tested in silico. Deployed on the Boltz-1 Pairformer trunk, stability prioritisation finds interpretable latents using about 4.4 times fewer latent evaluations each, and at 5.2 times lower measured LLM cost, while recovering over half of them. However, we notice that in our setup the LLM judge is heavily biased towards consistent residue identity and a simple filter that selects for features with high amino-acid identity precision outperforms the stability thresholding. This simpler method has the added benefit of requiring no extra SAE seeds but it remains uncertain if it reflects a genuinely useful heuristic or an limitation of the LLM judge.

q-bio.QM · cs.AI

新论文 · v1arXiv:2611.21919

ALOHA Lightning: Learning Fast and Precise Manipulation

John Hua Yao. Qi Wu, Yihuai Gao, Chelsea Finn, Zipeng Fu

原文摘要

Learning from human demonstrations has enabled robots to acquire a wide range of manipulation skills, but learned policies typically execute far slower than ordinary humans. This speed gap is mainly due to lack of an interface for collecting demonstration data at high speed, and the difficulty in training policies that can robustly execute high-speed motions. In this paper, we present ALOHA Lightning (Flash ALOHA), a system for learning fast and precise robotic manipulation. Our system uses kinesthetic teaching to intuitively collect near-human-speed demonstrations on a backdrivable bimanual platform, yielding natural and fast trajectories. We also present a learning pipeline that enables smooth high-speed execution through test-time action smoothing and aligns the visual data distribution between data collection and deployment with masking. Given 50 demonstrations for each task, ALOHA Lightning autonomously completes bimanual tasks such as ball tossing and catching, folding shorts, battery insertion, and bussing tables for over 80% success rates while close to human speed. Project website: https://flash-aloha.github.io/

cs.RO · cs.AI · cs.LG · eess.SY

新论文 · v1arXiv:2611.21914

A Temporal Algebraic Framework for Data Stream Evaluation Under Delayed Labels and Distribution Shifts

Ronaldo C. Prati, Elaine Ribeiro de Faria

原文摘要

In data streams, labels often arrive late or only for some instances, class prevalence drifts, and rare classes vanish from short windows. Standard evaluation then confuses a degraded decision boundary with a change in prevalence, and is biased when labels are delayed or selected by the model. We extend an algebraic framework that obtains pooled (micro) and per-class (macro) measures from one confusion matrix to streams, weighting the counts in time by exponential fading or a sliding window; the update costs O(m^2) per step, whatever the stream length, and each pooled measure stays a weighted combination of the per-class ones. Delayed and selective feedback enter as an observation mask: until its label arrives, an instance is counted with the model's calibrated class probabilities, and the bias of this interim estimate is driven by their calibration error. Audited conventions report no value for a class without verified instances. We separate prior shift from concept drift by monitoring prior-free coordinates (recall and specificity) and attribute a change in a measure to the two sources with a first-order Jacobian decomposition. On synthetic streams and four real datasets, a detector on these coordinates with a memory per class raises no excess false alarms under label shift, unlike the Jacobian alarm and error-rate detectors. The correction for delayed labels recovers recall in some real settings and is only partial for precision, specificity and F1. Twenty-five statements are checked in Lean 4, and the code is in the aggmetrics package.

cs.LG · cs.AI · cs.DB

版本更新 · v2arXiv:2609.01852

The Memory Trust Gap: Reliance on Stale Evidence Across Model Scales in LLM Agents

Jundong Hu, Shekar Ramachandran

原文摘要

Persistent-memory agents can follow stale information despite current authoritative evidence. We study stale-evidence reliance across Qwen3 0.6/1.7/4/8B checkpoints using two controlled, action-scored suites: a Benefit suite requiring stored information and a Safety suite providing current evidence as static prompt text. We distinguish stale-value reliance from net accuracy loss relative to no memory. Benefit-suite reliance is 0.92-1.00 across checkpoints. In the Safety suite, larger checkpoints resist a plain stale note but incur substantial harm when it appears newer. A full $2\times2\times2\times2$ factorial manipulates label, recency, source authority, and position. Direct cross-size contrasts identify stronger recency effects in the larger tested checkpoints and a position-effect sign reversal within Qwen3. Metadata improves accuracy at all four checkpoints, with larger gains at 4B/8B; pre-resolution yields near-ceiling accuracy across all four. Llama-Instruct comparisons provide cross-family evidence for selected effects, while RGB and MisBench corroborate stale-value reliance. A framing control shows that the failure extends beyond memory-labeled content. These results characterize cue-dependent vulnerability across tested checkpoints without isolating capability or evaluating a complete memory lifecycle.

cs.AI · cs.CL

版本更新 · v4arXiv:2610.22155

Realtime-Bench: Can Frontier Models Solve Real-time Interactive Problems?

Jingyu Tang, Xu Yang, Zhehao Zhang, Zheng Wang, Qi Cheng, Haozhe Zhao, Manling Li, Wenxuan Li 等 10 位作者

原文摘要

Real-time interaction requires agents to perceive evolving environments and act under temporal constraints, yet existing evaluations of real-time streaming models focus on conversational quality and streaming perception, leaving effective action under strict temporal constraints untested. We introduce Realtime-Bench, a suite of carefully curated dynamic tasks where delayed actions directly affect success. Under Wall Clock, where environments advance according to actual inference latency, the best success rate reaches only 44% and the best Pass^3 is only 36%, demonstrating that real-time interaction remains challenging. To separate serving latency from models' real-time streaming capability, we compare Wall Clock with Paused Clock, which removes inference delays as an idealized reference, and introduce Token Clock, which advances simulated time according to generated token counts, testing temporal awareness independently of physical serving speed. We find that temporal costs limit success in evolving environments and that higher reasoning effort has mixed effects under time pressure, reflecting a trade-off between decision quality and timely execution. Based on these results, we discuss future directions for improving real-time multimodal models, including adaptive reasoning with temporal awareness and continuous perception and action.

cs.AI

新论文 · v1arXiv:2611.21890

Domain-Tuned Diffusion Super-Resolution for 4K Facial Asset Maps

Elly Akhoundi, Mathieu Lamarre, Thibault Goyallon, Patrick Anderson, Étienne Danvoye

原文摘要

Physically based facial assets require 4K albedo and displacement maps for cinematic close-ups, yet generative pipelines commonly produce them at only 1K. Naively applying photographic super-resolution is a poor fit: facial UV maps have domain-specific statistics, and displacement is a scalar height field whose errors directly distort geometry. We present a domain-tuned diffusion upscaler for 4x (1024 to 4096) facial asset super-resolution. Starting from a bilinearly upsampled map, we adapt a frozen SDXL backbone per modality with lightweight LoRA modules, while a Tile ControlNet preserves input structure as the diffusion prior restores pore- and wrinkle-scale detail. On held-out maps, our method produces sharper, more faithful detail than bilinear and off-the-shelf GAN and diffusion super-resolvers, reducing albedo LPIPS from 0.21 to 0.08 while avoiding misplaced or excessive displacement relief. These results show that lightweight domain adaptation can produce render-ready 4K facial asset maps without a custom super-resolution architecture.

cs.CV · cs.AI · cs.GR

版本更新 · v2arXiv:2611.11010

Lightfoot: Graph Retrieval-Augmented Generation with Exact, Model-Free Document Deletion

Uday Allu

原文摘要

Graph-based retrieval-augmented generation (GraphRAG) improves multi-hop question answering, but deployed systems pay costs that research comparisons rarely measure: model calls at query time, building the graph, and removing documents correctly. We present Lightfoot, a GraphRAG system in which every entity, link, fact and synonym edge is a deterministic function of per-passage extractions that carry chunk-level provenance. Retrieval makes no language-model call, indexing makes one per passage, and deleting a document is a database transaction with no model calls: 14 seconds for 10% of an 11,402-passage corpus, after which the graph is identical to a fresh index of the surviving documents. On a corpus of near-duplicate names, HippoRAG 2's capped synonym lists leave 565 fewer edges than a rebuild, while Lightfoot remains exact. Under identical extraction, embedding and answer models on held-out questions of MuSiQue, 2WikiMultiHopQA and HotpotQA, Lightfoot is competitive with HippoRAG 2: with Holm-corrected paired tests it is better on MuSiQue at recall@5 and recall@10 (+4.8 and +5.5 points) and on 2WikiMultiHopQA at recall@10 (+3.4), and tied on HotpotQA, while HippoRAG 2 keeps a non-significant 3.0-point edge at 2WikiMultiHopQA recall@2. Seeding the walk from entities named in the question accounts for most of the 2WikiMultiHopQA gain. Answers tie on all three datasets, and the retrieval pattern holds with three larger embedders, up to Qwen3-Embedding-8B. The gain is not the reranker's: with the same cross-encoder, dense retrieval trails Lightfoot by 9.3 and 17.8 points recall@5 on the two multi-hop sets. Lightfoot also builds the cheapest index, 26-31% below HippoRAG 2; LightRAG needed 2,187 model calls and 3.7 hours for the same deletion, and fast-graphrag cannot delete.

cs.AI · cs.CL · cs.IR

新论文 · v1arXiv:2611.21880

Cancellation-Free Constrained SquareCB for Safe Controller Selection

Kartik Pandit, Sourav Ganguly, Arnob Ghosh

原文摘要

We study constrained contextual bandits, in which a learner chooses one of $K$ actions in each round to minimize loss while keeping the expected cost of its play below a budget. Existing reductions to online regression weight the cost by a Lagrange multiplier learned from past constraint residuals, so slack in some rounds can offset violations in others. We instead run SquareCB proposed in \citep{foster2020beyond} with a rectified weight on the cost: it is zero when the estimated cost is within the budget, and otherwise just large enough to bring it back to the budget. Under realizability and Slater's condition with margin $δ$, the algorithm attains regret $O(\sqrt{KT\,\mathrm{Reg}_{\mathsf{Sq}}(T)}/δ)$ and cancellation-free violation $\sum_t(V_c^{p_t}-b)^+=O(\sqrt{KT\,\mathrm{Reg}_{\mathsf{Sq}}(T)})$, where $\mathrm{Reg}_{\mathsf{Sq}}$ is the error of the regression oracle. The same bounds hold when the budget must be met in every context, potentially adversarially chosen contexts. In controller selection for a safe-control-gym quadrotor and a BOPTEST building, the violation is 6 to 9 times lower than that of primal-dual and virtual-queue baselines, with a lower or comparable regret.

cs.LG · cs.AI · stat.ML

新论文 · v1arXiv:2611.21867

FaceRiff: Data-Efficient Unified Text-to-3D Face Asset Synthesis via Geometric Conditioning

Elly Akhoundi, Mathieu Lamarre, Thibault Goyallon, Patrick Anderson, Étienne Danvoye

原文摘要

High-resolution facial datasets with consistent UV-mapped geometry, albedo, and physically based displacement are scarce, yet such assets are essential for photorealistic digital-human rendering. Existing 3D face generation methods are often slow, single-modality, or require large specialized datasets. Meanwhile, large image diffusion models encode strong facial priors but do not natively produce lighting-independent textures, pore-scale displacement, or UV-aligned geometry. We present FaceRiff, a unified, data- and parameter-efficient framework for generating complete 3D face assets (mesh, albedo, and displacement) directly from text in a shared UV space. FaceRiff adapts a pretrained SDXL backbone with task-specific LoRA adapters and uses UV normal maps as geometric conditioning: per-instance normals guide albedo and displacement synthesis, while a dataset-mean normal provides a canonical spatial anchor for mesh generation. Using only 2,049 scans from approximately 350 identities and training on a single GPU, FaceRiff updates less than 3% of the backbone parameters per task and produces render-ready 3D face assets without per-instance optimization. Compared with FaceG2E and DreamFace, FaceRiff produces greater per-prompt diversity while preserving prompt alignment and fixed-topology UV consistency, and additionally generates lighting-independent albedo suitable for PBR relighting and pore-scale displacement absent from prior text-driven baselines. A blind study with 45 industry professionals shows that FaceRiff achieves substantially higher perceived realism than DreamFace, approaching that of real captures. Overall, FaceRiff demonstrates that pretrained diffusion priors can be adapted to unified 3D face asset synthesis under limited data.

cs.CV · cs.AI · cs.GR

新论文 · v1arXiv:2611.21865

PromptPack: Training-Free Augmentation of Diffusion Editors for Intent-Aligned Spatial and Semantic Control

Sophia Zalewski, Mustafa Munir, Enting Liu, Anjul Patney, David Dynerman

原文摘要

Many image-editing tasks require confining changes to intended spatial regions while exactly preserving user-designated source content. Instruction-driven editing pipelines often leave such user-intent underspecified, while inpainting editors condition generation on explicitly defined edit support but not on the spatial assignment of individual edit concepts. PromptPack addresses both limitations by equipping diffusion editors with two explicit 'Edit Guides': an edit-role partition that labels each pixel as 'Keep,' 'Permit,' or 'Modify' to specify where edits may occur, and a regional-prompt assignment that maps spatial regions to text prompts to specify which concepts apply where. These guides may be manually specified or, for source-object edits, automatically derived from a text-grounded SAM 3 mask. Without retraining, PromptPack injects independently embedded regional prompts into the editor's fixed-length text sequence and uses the Edit Guides to spatially mask attention, gate flow updates, and restore protected source pixels. We integrate PromptPack into FlowEdit FLUX, FlowEdit SD3, and FLUX Kontext and demonstrate improved requested-concept placement and protected-content preservation across all three integrations on PIE-Bench++. In a pairwise user study, PromptPack achieves an 87% preference rate over FlowEdit FLUX, establishing PromptPack as a practical, training-free interface for intent-aligned diffusion editing.

cs.LG · cs.AI · cs.CV

版本更新 · v3arXiv:2606.21869

The Language-Energy Divide: Measuring Energy Costs of Multilingual LLM Inference

Naihao Deng, Alissa Shen, Yiming Feng, Joan Nwatu, Jae-Won Chung, Mosharaf Chowdhury, Yulong Chen, Rada Mihalcea

原文摘要

Large language models (LLMs) are increasingly deployed in multilingual settings, yet the energy costs of serving these models across different languages remain poorly understood. We present a systematic study of inference energy consumption across languages with ML.Energy framework (Chung et al., 2026). We find striking disparities: energy consumption per output token varies by up to 8.3 times across languages, while total energy for a fixed set of requests varies by up to 179 times between the cheapest (English, 17.6 kJ) and the most expensive (Pashto, 3,147 kJ) languages. Our analysis shows that this disparity is driven by two compounding factors: (1) higher per-token energy costs for languages using complex or rare scripts, and (2) more tokens generated for low-resource languages. Moreover, we find a double cost + performance penalty: languages with the highest energy footprints also tend to achieve the lowest task accuracy. We reveal that the energy divide persists across models, hardware, and tasks, suggesting a systemic energy inequity in multilingual LLM deployment. Finally, we recommend that the community treat energy as a first-class evaluation axis, extend reporting checklists and model cards to include it, and adopt deployment-side mitigations for better energy efficiency.

cs.CL · cs.AI

新论文 · v1arXiv:2611.21852

The Hidden Cost of MoE Diffusion Decoding: Profiling and Fixing the Sampler

Harshit Samrat, Ramakrishna Paramahamsa Kompella

原文摘要

Masked diffusion language models (dLLMs) decode by iteratively denoising a full canvas of tokens, so every forward pass runs a per-token sampling and stopping pipeline over the entire canvas-by-vocabulary logits tensor. We profile single-user decoding of a 26B-parameter diffusion Mixture-of-Experts model on a bandwidth-bound edge accelerator (NVIDIA GB10) and report a result that is easy to miss: after weight quantisation, this logits pipeline, not the expert GEMMs, is the single largest consumer of decode time. Unlike autoregressive decoding, where sampling touches one position's logits per committed token, diffusion decoding processes S times as many logit entries per output token, where S is the number of denoising steps. We show this pipeline is largely redundant or fusable and remove it in two lossless tiers: a bit-exact tier whose output is byte-identical to the reference decoder, and an accuracy-parity tier using a fused single-pass entropy kernel and Gumbel-max sampling. Combined with a W4A16 expert kernel, the full stack reaches 2.4-2.7 times end-to-end speedup over bf16 at unchanged benchmark accuracy on GSM8K, HumanEval, and MATH, on a single 273 GB/s device. We confirm the finding on a second, independently built diffusion MoE (LLaDA2.0-mini: pipeline 28.8% of decode, fused W4A16 3.3 times at accuracy parity) and show it disappears on a dense diffusion model (3.6%). The cost is specific to MoE diffusion, where quantisation makes the backbone cheap. Code and quantised weights are released.

cs.LG · cs.AI · cs.CL

新论文 · v1arXiv:2611.21851

AR-WAM: efficient streaming world-action modeling for autonomous driving

Baiting Luo, Shatadal Mishra, Ahmadreza Moradipari, Nejib Ammar

原文摘要

Autonomous driving requires anticipating how traffic will evolve while responding quickly to new observations. Video foundation models offer the first ability, but repeated history encoding and iterative video generation make them costly to use for driving. We present AR-WAM, an efficient streaming world-action model that combines autoregressive temporal modeling with single-pass future conditioning. Because its video backbone is causal, AR-WAM updates its visual context incrementally, reusing past computation and conditioning future prediction on observed motion. We further hypothesize that future features computed from recent observations can improve generalization in dynamic traffic without iterative video generation. AR-WAM therefore obtains these features in one forward pass and uses them to condition trajectory generation, and reinforcement learning then improves the policy through direct optimization of driving rewards. With a single front camera, AR-WAM achieves 91.5 PDMS after supervised training and 94.0 after reinforcement learning on NAVSIM-v1, outperforming the compared strong end-to-end world-action models. In a dynamic setting of continuous driving, supervised AR-WAM keeps its visual history across decisions and outperforms strong world-action baselines.

cs.CV · cs.AI · cs.RO

新论文 · v1arXiv:2611.21847

Inventa: Decentralized Self-Improving Agents for Autonomous Scientific Discovery

Prithwish Jana, Nikos Kanakaris, Purak Jain, Anoop Deoras, Sahika Genc

原文摘要

Large language model (LLM) agents can generate research ideas and improve scientific discovery through iterative experimentation. However, increasing the number of agents or iterations does not automatically translate to producing more diverse or feasible ideas. Agents may converge on similar candidates, propagate unsupported claims through shared memory components and repeatedly apply search strategies that no longer yield any improvements. We introduce Inventa, a framework for automated scientific discovery that employs autonomous research units and verified shared knowledge. Each unit creates a lineage of candidate ideas using a mutator agent. A local orchestrator agent periodically adapts the mutator's harness. When progress stalls, the orchestrator agent replaces its model or restarts the idea evolution process. Verifiers evaluate each candidate idea and record verified improvements, useful components and constraint violations in a shared knowledge, the only channel between units. We evaluate Inventa on five benchmarks spanning the research cycle, covering retrieval (ScholarCatalyst), ideation (IdeaBench), proposal writing (IdeaScientist), and scientific law discovery (EurekaBench, SciLaws-Bench). It leads all five, achieving 73.6% on EurekaBench versus 56.2% for CORAL. On IdeaBench, it discovers 26.0 novel ideas per paper, compared with 18.6 for ResearchAgent, with 82% fewer model calls.

cs.LG · cs.AI

新论文 · v1arXiv:2611.21845

EmbedGym: Letting Coding Agents Develop and Debug Embedded Firmware Without Physical Hardware

Minghui Zhao, Yiming Li, Sophie Zhu, Maxwell Khor, Lauren Ouyang, Hai "Helen" Li, Yiran Chen, Tingjun Chen

原文摘要

Modern coding agents perform well on general software benchmarks because they can run code and unit tests in a closed loop. Embedded firmware development, however, breaks this loop: a program that compiles can still fail on real hardware due to pin polarity, bus protocol, timing, or networking bugs that an agent at a text terminal cannot observe. Existing embedded coding agents either skip hardware execution, drive GPIO-only simulators with multi-second delays, or test on physical boards in a static environment. In this paper, we present EmbedGym, an interactive virtual hardware environment that lets coding agents develop and debug embedded firmware without physical hardware. EmbedGym uses instruction-level microcontroller unit (MCU) emulators to execute the same binary that would be flashed onto a physical board, connects them to 41 peripheral models, an analog circuit simulator, and virtual WiFi/Bluetooth Low Energy (BLE) endpoints, and exposes an interface for agents to change sensor inputs, press buttons, and inspect hardware states. On top of it, we build EmbedGym-Bench, which consists of 598 tasks across 6 MCU-framework pairs graded by hidden timed test vectors, using a five-stage multi-agent workflow that synthesizes peripheral models and tasks from datasheets and driver libraries. We expose EmbedGym to a coding agent at three levels and evaluate eight models over more than 14K trajectories. Even when compilation succeeds 99.7% of the time, 19.9% of compiled firmwares fail at runtime (up to 73% on individual targets). In contrast, interactive hardware testing in EmbedGym increases pass rates across models by +4.9% to +14.7% (up to +31% on individual targets). Our results also show that an agent with EmbedGym outperforms prior embedded agent frameworks, and lightweight models with EmbedGym match frontier models without it at 30-70x lower cost.

cs.SE · cs.AI

版本更新 · v2arXiv:2610.02405

When Terminal-Agent Training Stalls: Demystifying Data Generation and Verification Challenge

Xi Qin, Isabel Kurth, Xin Cui, Elin Park, Alexander Schaefer, Yaad Oren

原文摘要

Automated terminal-agent training relies on generated environments and verifiers to turn open-ended interaction into reward. Yet a runnable Docker image and executable test suite do not guarantee that the resulting reward is faithful, informative, or even attributable to agent behavior. We present a pipeline that generates containerized terminal tasks and their executable verifiers, and use it to demystify failure modes in task generation and live-environment evaluation. Our analysis separates four requirements that are often conflated: structural validity, verifier faithfulness, policy solvability, and infrastructure reliability. In a GRPO training of Qwen3.5-9B, it overfits on a relatively easy dataset within 20 steps. Conversely, in a controlled Qwen3.5-9B comparison, adding hard tasks lowers mean rollout pass@2 from 81.3% to 20.6%, restoring informative failures without changing the training configuration. We further identify near-duplicate task generation, incomplete anti-shortcut safeguards, Dockerfile failures, asynchronous pass@k bias, and concurrency-induced environment errors as distinct threats to measurement. These findings should motivate the community for a terminal agent specific evaluation harness that pairs structural checks and rollout-derived solvability with extensive verifier audits.

cs.AI

版本更新 · v2arXiv:2606.17541

Offline Preference-Based Trajectory Evaluation

Fernando Diaz

原文摘要

Offline evaluation of agentic systems often collapses trajectories into a binary terminal success value, discarding information about partial progress. This, in turn, results in widespread ties that create statistical inefficiency by reducing effective sample size and weakening the ability to distinguish systems. We propose preference-based trajectory evaluation, an evaluation measure that compares pairs of trajectories directly by using temporal preferences between time-to-return profiles. We find that, across diverse agentic and interactive benchmarks, success-based metrics result in tied comparisons on roughly 75% of instances, while trajectory-aware preferences reduce ties to roughly 35%, improving discriminative power, ranking stability, and data efficiency. Our results suggest that benchmark saturation, often portrayed as the result of poor data collection or problem difficulty, may also be explained by the design of the evaluation measure.

cs.LG · cs.AI

新论文 · v1arXiv:2611.21828

Coordination of Heterogeneous Teams with Uncertain Task Requirements

Jongann Lee, Melkior Ornik

原文摘要

Coordinating a team of heterogeneous agents to complete tasks with unknown requirements is a challenging multi-agent planning problem. Agents differ in their ability to discover task requirements and in their ability to engage tasks, so the planner must coordinate task discovery, execution, and staging. This paper introduces a hierarchical planner that uses a graph neural network policy to assign actions to agents, then finds a path to the action's location using a shortest-path algorithm. The central part of our approach is the CAAT graph, which is a graph representation of the planning context consisting of the agents, tasks, capabilities, and actions. We demonstrate that our learned planner can jointly coordinate a heterogeneous team to simultaneously discover task requirements and complete tasks in a shorter time compared to baselines that optimize task discovery and task engagement independently. Furthermore, we show that our policy can zero-shot plan on out-of-distribution settings by training only on an outdoor map then transferring without finetuning to an indoor map with more agents, tasks, and capabilities than seen during training.

cs.MA · cs.AI · cs.LG

版本更新 · v2arXiv:2607.22697

Test-Time Coverage: Test-Conditioned Data Curation for Deployment-Aware Learning

Nadine Chang, Maying Shen, Shizhe Diao, Jialiang Wang, Jingde Chen, Thomas Breuel, Pavlo Molchanov, Rafid Mahmood 等 9 位作者

原文摘要

Deployed AI systems are often trained from broad candidate data pools, necessitating data curation towards the deployment test distribution. However, standard data curation methods score training-side criteria rather than directly optimizing deployment match. We introduce TTCov (Test-Time Coverage), a data-level test-conditioned curation method that uses test-side information before training instead of updating model weights at inference. TTCov decomposes deployment-conditioned curation into coverage and distribution. To represent coverage, it builds a task Atlas, a collection of LLM-based atomic propositions (APs) describing deployment-relevant concepts, seeded from open task knowledge and expanded with unmatched APs extracted from unlabeled deployment samples. To represent distribution, it instantiates the matched deployment APs with their frequencies, yielding a Knowledge Atlas (K-Atlas) that operationalizes the deployment distribution as a curation target. TTCov then selects a budgeted training set whose deployment APs distribution approximates this target. We apply TTCov towards autonomous driving (AD), keeping adaptation off the inference path while selecting data with greater deployment-relevant coverage, closer K-Atlas matching, and stronger downstream end-to-end driving performance than data-curation baselines, including seamless adaptability to novel domains via city-to-city expansion. Code available at https://github.com/NVlabs/TTCov.

cs.AI · cs.CV · cs.LG

新论文 · v1arXiv:2611.21827

How Much of SAR Vehicle Recognition Is the Vehicle? Background Shortcuts, Angle Gaps and Partial Libraries on MSTAR and ATRNet-STAR

Mousa Shaya

原文摘要

Deep networks score near 99% on the MSTAR vehicle benchmark, and the newer ATRNet-STAR benchmark was built to be harder. We ask how much of these scores comes from the vehicle itself. With a modern network (ConvNeXt-Small, 3 seeds) we first reproduce the known MSTAR result: with the central 96x96 pixels zeroed out, accuracy is still 73.3% (chance 10%). The same shortcut is present in ATRNet-STAR, which was collected across several scenes: with the vehicle removed, accuracy is 63.3% on the 40-class SOC split (chance 2.5%). Under the official scene-change split, the background stops helping and starts hurting. Accuracy is 22.8% on full chips and 30.8% when only the vehicle is shown. A SAR foundation model, SARATR-X, leans on the background more than ConvNeXt does (+4 to +6 points with the vehicle removed, in every seed) and does not solve the scene change. We then hold the amount of training data fixed and vary the depression angles used for training. Accuracy is set mainly by the gap to the nearest training angle: 57% at 15 degrees, 39% at 30 degrees and 26% at 45 degrees. Adding more training angles does not help. Finally, when most classes have training images at the test angle and a few do not, the missing classes do worse than when no class has them: 9.0% instead of 22.4% at a 45-degree gap. Most of their images are assigned to known classes. We recommend reporting background-only and target-only controls, and the angle gap, next to any accuracy figure. Code is public.

cs.AI

新论文 · v1arXiv:2611.21823

TUSI: A Citation-Grounded Graph-Based Benchmark

Marzieh Karimi, Mehran Safayani, Alireza Basiri

原文摘要

Evaluating retrieval-augmented generation (RAG) systems in Persian law remains challenging because benchmarks lack realism, grounding, and coverage of legal reasoning. Existing resources are either synthetically generated from statutory texts, lacking natural information needs, or derived from informal legal question--answer sources without explicit linkage to authoritative legal documents, making them unsuitable for rigorous evaluation of grounded legal reasoning. To address this gap, we introduce TUSI-Bench, a structured Persian legal benchmark designed for the evaluation of RAG-based systems. TUSI-Bench is built upon a multi-layered legal knowledge framework containing 7,261 legal articles and integrating codified statutes, legal commentaries, doctrinal writings, and Supreme Court unification rulings. We also construct a legal knowledge graph linking articles and concepts for structure-aware retrieval. Based on this foundation, TUSI-Bench provides two complementary evaluation tasks: (i) TUSI-QA, an open-ended question answering benchmark containing 1,999 questions that requires grounded responses with explicit legal citations, and (ii) TUSI-Bar, a multiple-choice benchmark containing approximately 500 four-option questions derived from Iranian Bar examination questions. Furthermore, we propose a graph-based retrieval baseline that leverages relational structure within the legal knowledge graph. Experimental results demonstrate that incorporating graph-based relations improves retrieval recall by up to 8 percentage points compared to standard baselines, highlighting the importance of structured legal representations for RAG systems. The proposed benchmark provides a standardized resource for evaluating grounded legal reasoning in Persian and enables future research on structured legal intelligence.

cs.CL · cs.AI

版本更新 · v2arXiv:2501.09646

NS-Gym: A Comprehensive and Open-Source Simulation Framework for Non-Stationary Markov Decision Processes

Nathaniel S. Keplinger, Baiting Luo, Yunuo Zhang, Kyle Hollins Wray, Aron Laszka, Abhishek Dubey, Ayan Mukhopadhyay

原文摘要

Many real-world applications require decision-making where the environmental dynamics evolve over time. These non-stationary environments pose significant challenges to traditional decision-making models, which typically assume stationary dynamics. Non-stationary Markov decision processes (NS-MDPs) offer a framework to model and solve decision problems under such changing conditions. However, there are no standardized simulation frameworks for NS-MDPs, as opposed to widely popular frameworks for stationary problems. We present NS-Gym, the first simulation toolkit designed explicitly for NS-MDPs, integrated within the popular Gymnasium framework. In NS-Gym, we segregate the evolution of the environmental parameters that characterize non-stationarity from the agent's decision-making module, allowing for modular and flexible adaptations to dynamic environments. We review prior work in this domain and present a toolkit encapsulating key problem characteristics and types in NS-MDPs. This toolkit is the first effort to develop a set of standardized interfaces and benchmark problems to enable consistent and reproducible evaluation of algorithms under non-stationary conditions. We also benchmark several algorithmic approaches from prior work on NS-MDPs using NS-Gym. We envision that NS-Gym will enable researchers to study decision-making under non-stationarity by providing standardized interfaces, simulation frameworks, and benchmark problems. Project documentation, webpage, and code can be found at https://nsgym.io and https://github.com/scope-lab-vu/ns_gym

cs.AI

版本更新 · v2arXiv:2611.04886

Skill erosion, individual memory and the reversibility of AI adoption

Renaud Lambiotte

原文摘要

Generative AI can improve immediate performance but may erode the human skills needed for future work. We consider a population of agents who perform a task either with their own skill or with the help of AI. Agents tend to imitate more successful peers, and their skills evolve in time: improving with practice and eroding with reliance on AI. We study, analytically and with simulations, the collective consequences of AI adoption. We show that the mechanism of skill erosion makes adoption bistable and that the long-term behaviour of the population depends on how skills are distributed: populations with the same adoption and mean skill can reach opposite asymptotic states, depending on how expertise is shared between AI users and non-users. A deterministic description that follows the mean skills of the two groups and the variance of skills among non-users predicts the outcome of individual runs as accurately as the simulations allow, whereas the means alone are not sufficient. We also consider the situation when the benefit of AI improves in time. In that case, adoption can jump abruptly, and a temporary increase in benefit, such as in a period of low prices, can leave a population relying on AI after the benefit has returned to its original value. Skill erosion therefore creates a history-dependent switching cost that can make widespread AI adoption difficult to reverse.

cs.AI · physics.soc-ph

新论文 · v1arXiv:2611.21811

ARC-2: Continuous Compliance Through Delta Management

Tord Johan Eide, Einar Holt

原文摘要

ARC 1 introduced a three-layer governance architecture for autonomous systems comprising model safety validation, cognitive certification (CRAB), and operational authorization (LEA). However, it did not address how compliance should be maintained after deployment. This paper introduces Delta Management, a continuous governance framework that measures deviations between a system's current operational state and its certified baseline, defined as a region rather than a fixed point. Compliance is assessed through continuous monitoring instead of static certification. Four extensions are proposed. First, CRAB evaluation incorporates the worst verified behavioral deviations, with isolated outliers quarantined pending confirmation to prevent unjustified restrictions. Second, a layered trigger taxonomy distinguishes hard safety violations, critical single-dimension failures, correlated multi-system degradation, and risk-weighted composite indicators, each triggering proportionate responses. Third, a graduated safe transition protocol separates withdrawal of operational authorization from physical shutdown, consistent with established safety standards and EU AI Act requirements. Fourth, environmental delta is introduced as a distinct baseline validity condition: environmental drift beyond the certified Operational Design Domain (ODD) triggers recertification or ODD revision rather than automatic authority contraction. Together, these extensions transform ARC from a deployment-time certification architecture into a continuous framework for monitoring, evaluating, and enforcing compliance throughout the operational lifecycle of autonomous systems. The framework addresses the continuous compliance obligations established by EU AI Act Article 72(2).

cs.RO · cs.AI · cs.CY

新论文 · v1arXiv:2611.21809

Evaluating the Evaluation of Music Generation Models with Generative Psychometrics

João Carlos Santana Silveira, Caio Von Rondow Morais, Frederico Pedrosa, Lucas N. Ferreira

原文摘要

Subjective user studies are one of the major methods for evaluating music generation models. However, subjective evaluation protocols vary widely and are rarely specified in enough detail to be reproduced, hindering comparison across studies. In this paper, we map subjective evaluation methods in ISMIR music generation papers from 2020-2025 and apply generative psychometrics to assess the stability of their items. We extract 284 items from the papers' methodological descriptions, using an LLM and manual validation to standardize them. We represent these items as contextual embeddings and apply Unique Variable Analysis to remove redundant, locally dependent items and Bootstrap Exploratory Graph Analysis to estimate their latent dimensions. Results show the network converges onto a taxonomy of six stable quality dimensions: Rhythm, Melody, Audio, Composition, Structure, and Lyrics, pointing to a unified methodology for conducting user studies to evaluate music generation models.

cs.SD · cs.AI

新论文 · v1arXiv:2611.21808

SimAI: Mechanistic User Simulation through Psychological Network Models

Cheng Li, Adhyyan Narang, Jiayi Yuan, John F. Buss, Jina Suh, Devon Sandel-Fernandez, Yulia Tsvetkov, Adam M. Kuczynski 等 9 位作者

原文摘要

Realistic user simulation enables the study of human interactions when direct experimentation is costly or potentially harmful. However, existing simulators largely overlook the psychological mechanisms that drive user behavior. We introduce SimAI, a framework for mechanistic user simulation that models how psychological factors interact to shape user responses. To overcome the challenge of supervising open-ended behavior, we propose a graph reconstruction objective that provides graph-level rewards based on how well simulated conversations reflect the underlying psychological relationships. To support this approach, we construct PSYCHE, a dataset of 18,576 profile-network pairs derived from 516 psychotherapy cases. Automatic evaluations show that SimAI achieves greater simulation and behavioral diversity than existing patient simulators while maintaining strong profile alignment. In an interactive study with 30 mental-health professionals and trainees, SimAI receives the highest average ratings for conversational authenticity, realistic help-seeking behavior, and clinical training utility. Beyond simulation quality, SimAI enables realistic clinical training and uncovers safety failures in prolonged AI interactions with vulnerable users. Code is available at https://github.com/behavioral-data/ANGEL-UserSim.

cs.CL · cs.AI · cs.HC