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 16:11 — 2026-11-21 16:11(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:2611.22138

RoboTeam: Benchmarking Vision-Based Multi-Agent Control of Many Robot Arms in Real Time

Changxuan Fan, Zhongwei Xie, Boyi Zhang, Shaoli Hu, Chang Xu, Xiyu Ren, Yufei Li, Qiao Xiao 等 21 位作者

原文摘要

Multimodal agents can now control robot arms by calling tools and writing code. Existing benchmarks fall short in three ways: most test one or two arms on short tasks, many pause the world while the agent thinks, and many give it ground-truth object positions. We present RoboTeam, a simulated benchmark where agents control 4 to 16 robot arms in real time from camera images alone. Its 36 tasks are long jobs that the arms must do together, such as passing timbers or building a picture from hundreds of cubes. We start from the most flexible agent system, the adaptive system, where one agent decides when to fork delegates. With it, gpt-6.1-sol and opus-5.5 complete 28 and 27 tasks, and four other models at most 3. But at a high price: a typical task takes gpt-6.1-sol 3.5 hours and 218 million input tokens, and a third of its steps fail at least once and must be retried. We then compare it with three fixed systems: single-agent, hierarchical, and decentralized. All four complete a similar number of tasks but differ in time and tokens. The hierarchical and decentralized systems are faster but use up to 1.9 times the tokens, and the single agent uses fewer tokens but takes 1.6 times as long, so the adaptive system lies between them. Even so, no model or system is yet fast, cheap, and reliable at once. RoboTeam measures these limits where earlier benchmarks cannot: long multi-arm jobs, camera images only, and a world that never pauses.

cs.RO · cs.AI

新论文 · v1arXiv:2611.22136

Big Brains, Tiny Board: Pupavision, an Agentic Physical AI Instrument for Monitoring Drosophila Pupariation and Lifecycle Transitions

Joel Sadler, Daniel Graham, Melanie Worley

原文摘要

Many biology experiments are limited by the availability and ease of use of instruments that monitor and record them. For example, the timing of pupariation in the fruit fly Drosophila melanogaster, a widely used model organism, can provide biologically relevant information about growth regulation and is often recorded manually. Here, we present Pupavision, a small physical AI instrument built from a single-board Linux computer and a day/night camera for roughly 45 to 75 US dollars in parts. The key idea is that a frontier coding agent runs its agentic harness directly on the tiny board. The model thinks in the cloud, but the agent works on the device. The agent assists image capture, writes code to repair errors, and checks the experiment every day to keep raw data capture reliable. If recording stops, the researcher gets a ping on their mobile phone. A simple deterministic capture service records one image every ten minutes and never depends on the agent. We report a 26-day pilot in a fly incubator (about 3,580 frames), including eleven days in which the agent was logged out and capture continued, and a 51-hour outage whose cause led to the layered design. A researcher-in-the-loop workflow then produces per-vial pupariation curves by tracing selected pupae backward in time from a late reference frame. Future work will validate the assay for lifecycle tracking accuracy against expert-labeled data and expand real-time computer vision analysis. Pupavision offers a blueprint for do-it-yourself agentic instruments that put frontier intelligence on a tiny board, with deterministic code underneath. We present this as an early case study from a pilot lab deployment and contribute lessons learned toward a future where scientific domain experts are empowered by DIY agentic hardware.

cs.AI · cs.HC

新论文 · v1arXiv:2611.22131

LoCaR: Localized Canvas Retrieval With Frozen Vision-Language Model

Rahim Hossain, Dipanwita Saha, Ashirbad Mishra, Xinxin Shu, Svetlana Strunjas, Binbin Li

原文摘要

Whole image embeddings can under represent small objects and local attributes that distinguish relevant images for localized queries. We present LoCaR, a training free framework for localized text to image retrieval with frozen vision language models. LoCaR constructs and utilizes independent proposals (SAM2) focusing on the object so as to get a more localized representation that independently encodes each region using the frozen visual encoder. We further study paired-support adaptation as an optional residual that uses matched text region exemplars without updating the underlying encoders. We also introduce LOCFG, a Visual Genome derived benchmark designed to evaluate localized retrieval for smaller and more challenging referents. Across several vision language backbones, our results show that improving the region representation itself is the main source of gain.

cs.AI · cs.CV · cs.LG

版本更新 · v3arXiv:2611.00597

Multi-Estimate Elicitation Does Not Uniformly Debias Language Model Judgments

An-Sheng Jhang

原文摘要

Multi-Candidate Verbalization (MCV) is a task- and model-dependent output protocol: repairing format does not guarantee better posterior accuracy, and requesting weights differs from reweighting an already generated candidate set. Across 31,672 planned/admitted evaluation units and 32,184 planned/admitted calls (excluding retries), we connect a frozen six-configuration cohort to exploratory scale/organization experiments and two independently preregistered post-cohort formal suites. Stage 1 retains 11 directional confirmations among 45 frozen tests; only increased post-outcome confirmation error on gpt-oss-120b exceeds its SESOI. P1 completes 1,920 terminal units: smaller response scales sharply lower operational string-match rates but increase Bayesian error. P2 completes 2,560 calls and 2,048 units: eliciting weights raises three errors by +2.75 / +2.05 / +3.88 pp, while within-generation weighting changes them by only -0.052 / -0.049 / -0.067 pp, a 42x-58x magnitude difference. Unconstrained joint generation with post-hoc normalization offers a local normalized-error improvement for three exhaustive options, without confirmed accuracy improvement. These results distinguish format validity, numerical accuracy and psychological susceptibility; a descriptive three-class behavioral spectrum organizes heterogeneity without identifying internal causes. Materials: https://github.com/asj252/llm-judgment-studies

cs.CL · cs.AI

版本更新 · v3arXiv:2507.14760

QUTCC: Quantile Uncertainty Training and Conformal Calibration for Imaging Inverse Problems

Cassandra Tong Ye, Shamus Li, Tyler King, Kristina Monakhova

原文摘要

While deep learning offers tremendous promise for scientific and medical imaging, any failures and hallucinations (predictions that do not coincide with reality) are hard to pinpoint and can have serious downstream consequences. Uncertainty estimation techniques, such as conformal prediction, can help by predicting statistically valid error bars for a model's prediction. However, popular conformal prediction methods were not designed for high-dimensional image-valued problems and do not take into account spatial correlations within an image during conformal calibration, resulting in larger-than-necessary uncertainty intervals. We propose a practical simultaneous quantile regression method that enables non-linear, spatially-adaptive scaling during conformal calibration. Our method, QUTCC uses a U-Net architecture with a quantile embedding to learn a full conditional quantile distribution during training, and then leverages this non-linear, learned function for spatially-adaptive conformal calibration. At test time, our method can efficiently estimate uncertainty intervals with pixel-marginal coverage guarantees. In addition, QUTCC can also predict pixel-wise conditional probability density estimates without any built-in distributional assumptions. We evaluate our method on several denoising problems, accelerated magnetic resonance imaging, and quantitative phase microscopy. Our method consistently produces tighter uncertainty intervals than prior conformal methods at the same coverage level, can predict plausible conditional distributions for different tasks, and in some cases, high-uncertainty regions can help us locate hallucinations in a model's prediction.

eess.IV · cs.AI · cs.CV · cs.LG

新论文 · v1arXiv:2611.22115

TeleChat to Xing4: Advancing Frontier Agentic Models on Domestic Chips and Frameworks

Xingchen AGI Lab, :, Baotian Hu, Bingkai Yang, Boyang Wang, Changzai Pan, Chao Lyu, Chao Wang 等 140 位作者

原文摘要

We present Xing 4, a series of large language models that extends the TeleChat lineage through coordinated advances in model architecture and training infrastructure on Ascend. This report describes the design principles and systems engineering behind the series, spanning architecture selection, large-scale pretraining, and reasoning and agentic post-training. We study how sparse activation, attention design, model depth and width, and speculative decoding balance model capacity with computation, memory access, and execution overhead. Our pretraining infrastructure builds on MindSpore HyperParallel, combining topology-aware hybrid sharding, efficient distributed optimizer execution, and coordinated management of long-sequence workloads and activation memory. CPU affinity and core isolation further reduce host scheduling interference. Together, these mechanisms support the pretraining of a 285B-parameter model on 8,192 Ascend 910C cards, providing a concrete setting for the architectural and systems studies. For post-training, we introduce TeleGym, a unified agent execution platform that connects execution, evaluation, and trajectory construction to policy optimization. Disaggregated asynchronous execution, bounded oversampling, and partial rollout accommodate heterogeneous task durations while preserving trajectory continuity across policy updates. Together, these advances establish an Ascend-oriented training foundation for the continued development of Xing 4 across model scales and increasingly demanding reasoning and agent workloads.

cs.AI

新论文 · v1arXiv:2611.22113

RegimeCouncil: Paired-Evidence Routing For Memory Conflict Governance In LLM Agents

Qingcan Kang, Mingyang Liu, Shixiong Kai, Tao Zhong, Mingxuan Yuan

原文摘要

Language agents rely on long-term memory to personalize responses across extended interactions, yet stored facts quietly go stale as the world changes, and an agent often learns that using memory hurt only through delayed feedback after acting on an outdated record. Before the memory store is updated, how far should that negative feedback propagate: should it suppress memory use only on future queries for that same fact (a local policy), or also across other stored facts (a global policy)? Existing memory systems implicitly fix one of these two sharing scopes at design time. We formulate this online scope choice under hidden validity and delayed feedback as memory conflict governance, and prove that neither fixed policy is sufficient on its own. We then propose RegimeCouncil, a lightweight paired-evidence router that retains both scopes as advisors and arbitrates between them online without retraining the LLM. Using revealed comparisons between answering with memory and answering without memory (the no-memory anchor), a Local Advisor tracks the latest outcome on the queried fact alone, while a Global Advisor pools recent outcomes across all stored facts. A Meta Governor follows the Local Advisor by default to avoid cross-fact interference and switches to the Global Advisor only when global pooling accumulates a strictly positive recent utility margin, guaranteeing finite-time switching under delayed feedback. Across LongMemEval update streams and real-world historical drift on TempLAMA and TAQA with Qwen3-30B-A3B and GPT-4o, RegimeCouncil outperforms the better fixed-scope policy by 5.2-7.3 points on primary disagreement episodes and across all 30 held-out settings, while ablations show that removing any of its three modules reduces accuracy by 2-7 points.

cs.AI

版本更新 · v5arXiv:2610.21337

Does On-Policy Distillation Need to Complete Every Rollout?

Zhengxin Zhang, Maosen Tang, Ning Wang, Changhao Li, Jingren Zhou, Bo Dai, Sainyam Galhotra, Alex Townsend 等 9 位作者

原文摘要

On-policy distillation (OPD) transfers teacher knowledge to a student through feedback on student-generated rollouts, but generating complete rollouts and computing per-token teacher feedback are computationally expensive. Existing truncation methods save computation by stopping rollouts early. However, they do not compensate for truncated token gradients, which biases the gradient estimates and changes Vanilla OPD's original training objective. Furthermore, their truncation decisions rely on assumptions about supervision quality or criteria that may not generalize across settings. We introduce Stochastic Truncated On-Policy Distillation (STOPD), which accelerates OPD by randomly truncating student rollouts and correcting the resulting sampling bias with blockwise inverse-probability weighting (IPW). At each training step, we sample a stopping position for each rollout before generation and let the student generate until that position or natural termination. We obtain teacher feedback on every generated token and use blockwise IPW to compensate for the lower probability of retaining later tokens while controlling the weights. We prove that this weighting preserves Vanilla OPD's gradient sum in expectation and derive one-step update-error bounds for SGD, Adam, and Muon. We evaluate STOPD across multiple teacher-student model pairs, task domains, and optimizers. STOPD reduces generated tokens by 25-48% while maintaining accuracy comparable to Vanilla OPD, achieving 1.3-2.2x training speedups.

cs.LG · cs.AI

版本更新 · v2arXiv:2605.14434

Efficient Generative Retrieval for E-commerce Search with Semantic Cluster IDs and Expert-Guided RL

Jianbo Zhu, Xing Fang, Jing Wang, Mingmin Jin, Bokang Wang, Guangxin Song, Zhenyu Xie, Junjie Bai

原文摘要

Generative retrieval offers a promising alternative by unifying the fragmented multi-stage retrieval process into a single end-to-end model. However, its practical adoption in industrial e-commerce search remains challenging, given the massive and dynamic product catalogs, strict latency requirements, and the need to align retrieval with downstream ranking goals. In this work, we propose a retrieval framework tailored for real-world recall scenarios, positioning generative retrieval as a recall-stage supplement rather than an end-to-end replacement. Our method, CQ-SID (Category-and-Query constrained Semantic ID), employs category-aware and query-item contrastive learning along with Residual Quantized VAEs to encode items into hierarchical semantic cluster identifiers, significantly reducing beam search complexity. Additionally, we develop EG-GRPO (Expert-Guided Group Relative Policy Optimization), a reinforcement learning approach that aligns generative recall with downstream ranking under sparse rewards by injecting ground-truth samples to stabilize training. Offline experiments on TmallAPP search logs show that CQ-SID achieves up to 26.76% and 11.11% relative gains in semantic and personalized click hitrate over RQ-VAE baselines, while halving beam search size. EG-GRPO further improves multi-objective performance. Online A/B tests confirm gains in GMV (+1.15%) and UCTCVR (+0.40%). The generative recall channel now contributes substantially in production, accounting for over 50.25% of exposures, 58.96% of clicks, and 72.63% of purchases, demonstrating a viable path for deploying generative retrieval in real-world e-commerce systems.

cs.IR · cs.AI

版本更新 · v2arXiv:2611.14912

Rethinking "RL Generalizes, SFT Memorizes": The Role of SFT Data

Yunlong Hou, Fengzhuo Zhang, Yuan Cheng, Jiachun Pan, Xingyao Li, Zhuoran Yang

原文摘要

Large language models trained with Reinforcement Learning (RL) with verifiable rewards exhibit strong reasoning ability and broad generalization, whereas models trained with Supervised Fine-Tuning (SFT) are often viewed as more prone to memorization and limited transfer. This paper rethinks this distinction through the lens of SFT training data on widely used Qwen2.5 and Qwen3 series models. First, we study the role of data source and show that it is critical: a carefully mixed SFT dataset substantially outperforms data generated solely by a larger model. Second, we study the role of data scale and show that matching the number of correct rollouts between SFT and RL greatly improves SFT generalization, while matching the total rollout budget enables SFT to generalize as well as RL. Combining these two factors further enables SFT to generalize even better than RL. Third, by using LLM annotations to characterize the solution methods in training rollouts, we show that larger datasets cover more tail methods and that these tail methods provide generalizable reasoning signals. Finally, we support these empirical findings theoretically by analyzing the training dynamics of shallow transformers under both RL and SFT.

cs.LG · cs.AI

新论文 · v1arXiv:2611.22103

From Sorry to QED: Refining LLM Math Proofs via Angelic Execution

Yanju Chen, Hanzhi Liu, Ruizhe Qian, Jiaming Shan, Xu Yang, Yufei Ding, Osbert Bastani, Yu Feng

原文摘要

Large language models (LLMs) can now generate Lean proof scripts that capture the high-level structure of many mathematical arguments. However, these scripts frequently fail on complex theorems: a single missing side condition or misapplied lemma can cause the prover to abort early and discard an otherwise viable proof outline. This paper presents MathEye, a framework that converts imperfect LLM-generated proof scripts into fully verified Lean proofs by combining symbolic reasoning with minimal human interaction. At the core of MathEye is angelic proof checking, an optimistic execution model that replaces failing proof fragments with placeholders while preserving the overall structure of the argument. This optimistic check produces an angelic proof that isolates the exact proof obligations that the LLM fails to discharge. Because these unresolved obligations are often logically interdependent, asking a mathematician to address every placeholder directly is inefficient. To minimize human effort, MathEye performs residual analysis over a proof dependence graph that captures symbol-level data flow in the angelic proof. By formulating obligation selection as a MaxSAT optimization problem, MathEye selects a minimal-cost set of essential obligations whose resolution suffices to complete the entire proof. MathEye presents these essential obligations to the mathematician in ordinary mathematical language and uses the mathematician's answers to automatically reconstruct the complete Lean proof. Experiments on 30 real-world benchmarks across analysis and algebra show that MathEye solves 93.3% of the problems while reducing the number of required user interactions by 55%.

cs.AI

版本更新 · v2arXiv:2610.02975

Reliable Self-Evolution with Imperfect Proxy Rewards

Kangjun Noh, Soyu Kim, Kyungwoo Song

原文摘要

Large language model (LLM)-based self-evolving search is a promising approach to scientific discovery. However, high-fidelity evaluation of every candidate is prohibitively expensive in some domains. Self-evolving systems in such settings therefore rely on low-cost but imperfect proxy rewards, which may assign high scores to infeasible candidates. These false positives may contaminate both the final output and the feedback used to guide subsequent generations. This motivates statistically calibrated reward intervals for more reliable self-evolving search. We propose Conformal Interval-Driven Self-Evolution (CISE), which constructs candidate-specific reward intervals using conditional conformal inference and iteration-wise online density-ratio estimation. CISE uses conservative interval-based rewards for evolutionary feedback and returns candidates only when all required property intervals lie entirely within their respective feasible regions. We derive fixed-iteration coverage results under explicit assumptions of independence and covariate shift. We evaluate CISE on three self-evolving search tasks in materials science. In our experiments, all candidates returned by CISE are true positives under high-fidelity evaluation, whereas the baselines return more candidates but include false positives. These results highlight the value of a smaller, more precise shortlist when downstream validation budgets are limited. Our repository is available at https://github.com/MLAI-Yonsei/CISE.git.

cs.AI

新论文 · v1arXiv:2611.22102

EvoDisco: Strategy Discovery for Evolving Agent Memory

Mingyang Liu, Qingcan Kang, Shixiong Kai, Kaichao Liang, Xiaojin Fu, Tao Zhong, Linqi Song

原文摘要

Large language model agents accumulate experience in memory, and how that memory evolves during interaction is as important as what it stores. Recent methods update memory online, yet both the operators that transform memory and the rules that select them remain fixed by design. We propose EvoDisco, which formulates agent memory evolution as evolutionary strategy discovery: the agent must discover which operator to apply, when to apply it, and how to instantiate it, rather than rewriting memory under a static operator schedule. A learnable policy conditions on the current memory state and recent task history, then selects and instantiates one of four semantic operators---Abstract, Refine, Bridge, and Reframe---which respectively compress related memories, decompose overloaded entries, connect weakly linked experiences, and reorganize retrieved reasoning chains. The policy is updated with a bi-level fitness. An inner signal scores each transformation using a retrieval-quality improvement regularized by memory complexity. An outer signal scores strategic efficacy over a window of past decisions, combining historical operator-utility regret, utility-prediction error, and selection diversity. Under a one-pass online plus frozen-transfer protocol on ALFWorld, WebShop, and LifelongAgentBench (OS and DB), EvoDisco improves over strong memory baselines including MemP and MemRL. The largest gains are on ALFWorld ($+8.8$ online / $+12.9$ frozen-transfer points over MemP; $+16.4$ / $+17.9$ over MemRL). Ablations attribute these gains to both fitness levels and the operator library.

cs.AI

新论文 · v1arXiv:2611.22100

Semantic Communication via Analog RF Computing: When the Radio Computes the Meaning

Christo Kurisummoottil Thomas

原文摘要

Semantic communication (SC) systems transmit continuous latent representations of meaning, but computing the semantic encoder at resource-constrained edge devices incurs substantial energy and memory costs. To address this challenge, we propose an end-to-end SC framework that executes the semantic encoder in the client's RF receiver front end using analog RF computing. A base station (BS) broadcasts encoder weights as RF waveforms, while each edge device reuses its receiver's passive mixer to compute semantic latents at ultra-low energy. These latents are transmitted over the uplink for network-side task inference. We show that analog computing noise across encoder layers and channel noise affecting transmitted latents admit a unified mathematical model. We define a task-posterior semantic distortion metric and derive its second-order decomposition into per-client, per-stage semantic sensitivities. Using this decomposition, we obtain a closed-form semantic-aware water-filling allocation of the total client energy budget across clients, encoder layers, and uplinks, assigning tighter accuracy targets to semantically sensitive stages. This allocation induces a broadcast beamforming problem that accounts for client-specific, per-layer accuracy requirements. We further refine the allocation through stochastic optimization and jointly design it with multi-task noise-aware training of the shared semantic model. MNIST simulations show that, at matched total client energy, the proposed allocation reduces network semantic distortion by 9.2\%-19.2\% compared with uniform precision and 5.1\%-13.1\% compared with client-blind allocation. Noise-aware co-design yields a further 46\% reduction.

eess.SP · cs.AI · cs.IT

新论文 · v1arXiv:2611.22094

When AUC Meets Generative Reconstruction: A Unified Principle for Balanced Visual Representation Learning

Boyu Han, Qianqian Xu, Shilong Bao, Zhiyong Yang, Qingming Huang

原文摘要

Multimodal large language models (MLLMs) require visual encoders that extract fine-grained visual features while constructing semantically disentangled and discriminative representations. Conventional methodologies encounter a fundamental trade-off: contrastive learning enhances inter-class discrimination by suppressing intra-class variability, whereas generative reconstruction emphasizes perceptual fidelity but lacks explicit comparative supervision. The naive aggregation of these objectives via composite loss functions often induces optimization conflicts and instability due to gradient interference within the shared latent space. To resolve this, we introduce a unified representation-learning paradigm that reformulates generative reconstruction as a discriminative ranking problem within the generative output manifold. Our approach maintains perceptual reconstruction as a supervisory signal and integrates a ranking-based discriminative constraint by systematically ordering matched and mismatched reconstruction outputs. This formulation embeds discriminative information seamlessly into the generative process, alleviating the risk of loss competition. Specifically, we propose AUC Generative Reconstruction (AUCGR), which optimizes the area under the ROC curve (AUC) by ensuring higher scores for matched pairs than for mismatched pairs, and Partial AUC Generative Reconstruction (pAUCGR), which addresses supervision bias by focusing on mismatched comparisons within reliable feature subspaces. The proposed objectives are designed to be plug-and-play, requiring no architectural modifications to existing generative reconstruction frameworks. Theoretical analysis establishes rigorous connections to AUC/pAUC optimization and provides generalization guarantees. Extensive experiments validate that our method advances both discriminative accuracy and fine-grained perceptual representation.

cs.CV · cs.AI · cs.LG

新论文 · v1arXiv:2611.22086

HyCoInfer: Cloud-Edge Collaborative Inference for Large Language Models on Heterogeneous GPU-FPGA Architectures

Zewen Huang, Xiangwen Zhuge, Yingkun Hu, Zheng Yang, Qiang Ma

原文摘要

Large language model (LLM) serving relies predominantly on cloud GPUs, yet autoregressive decoding is memory-bound and often leaves their arithmetic units underutilized. Edge FPGAs offer attractive decoding efficiency, but are ill suited to sustaining a high-concurrency serving workload alone. This paper presents HyCoInfer, a heterogeneous cloud-edge serving system that assigns prefill to cloud GPUs and distributes decoding across cloud GPUs and edge FPGAs. The design addresses three barriers to practical offloading: capacity balancing across heterogeneous accelerators, the large KV cache crossing a constrained wide-area link, and the latency added by heterogeneous execution. HyCoInfer incorporates FPGA decode capacity into GPU prefill/decode role partitioning and places requests according to queue load, prompt length, and network state. It further combines layer-wise pipelined transfer, decoder-side precomputation, and bandwidth-adaptive KV cache compression based on residual vector quantization. We implement HyCoInfer using two NVIDIA V100 GPUs and an AMD Xilinx Alveo U280 and evaluate it over 100-1000 Mbps links. At 500 Mbps, HyCoInfer delivers up to 4.11 times the request goodput of cloud-only vLLM and reduces mean end-to-end latency by up to 61.1%. Across different network conditions and workloads, the system improves overall serving capacity, with the largest gains on communication-intensive tasks, and reduces per-request energy.

cs.OS · cs.AI

新论文 · v1arXiv:2611.22085

ATLAS-Finance: Benchmarking AI Agents in Realistic Banking Environments

Elaine Lau, Antoine Pangas, Alexandre Finot, Collin Schweiker, Amir Bouzidi, Hui Wen Goh, Andrew Peter Martin, Jonas Mueller

原文摘要

Professional work involves under-specified assignments, where relevant information is spread across workplace applications and may conflict or change over time. To evaluate how well frontier agents can perform economically valuable work in such settings, we introduce ATLAS-Finance, a benchmark of 100 end-to-end investment-banking tasks across 13 simulated firms. As in real banking workflows, agents must search email, chat, calendars, notes, and files to identify requirements, gather information, reconcile conflicts/updates, and then build complex financial models. Tasks were developed by experienced bankers based on their real deal work and take them 15-30 hours to complete. Agent deliverables are evaluated with expert-authored rubrics (median: 124.5 criteria) covering financial rationale and results, spreadsheet construction, completeness, and formatting. Despite completing substantial portions of the requested work, current agents rarely produce correct deliverables. Across today's frontier models, the best ones (Claude Opus and Fable) pass fewer than 11% of tasks.

cs.AI

新论文 · v1arXiv:2611.22084

When Lower Harm Misleads: Auditing Concept Corrections

Tianyou Lai, Wentao Yue, Hongji Li, Chaoyuan Hao, Qilei Li

原文摘要

An expert who corrects a concept prediction, such as a BI-RADS descriptor in breast ultrasound, expects the diagnosis to improve. Objectives that reduce the harm of such corrections are judged by harm and accuracy, yet harm also falls when a model responds less to every correction or starts from a worse uncorrected prediction, the reference for harm. We propose a correction audit, a reproducible evaluation protocol that separates these explanations: it splits each correction into harm and useful gain on identical events, separates the starting prediction from the correction response, and compares objectives with amplitude baselines matched in source harm. On three breast-ultrasound sources, worst-case correction training, evaluated from the mean-case model's start, lowers harm but gives up more useful gain in all 21 source comparisons pooled over six seeds, raising the corrected negative log-likelihood (NLL); from the same start, selecting between the objectives by harm picks the less helpful model on all three sources. On CUB, a monotonic harm penalty at its reported weight lowers path-level harm in all 80 concept sets while raising the NLL in all 80, almost entirely before any correction. At a fixed start, rescaling both responses to the same source harm reproduces the penalty's mean harm, useful gain and NLL on single-concept corrections within a prespecified margin of 0.0005 nat, although the two responses point in different directions. Lower harm therefore signals better corrections only when it is read together with useful gain, absolute risk and the start it is measured from.

cs.AI · cs.CV

新论文 · v1arXiv:2611.22076

Segmentation-Guided Direct Capacity Estimation of Rooftop Photovoltaics from Aerial Imagery

Guibin Chen, Emma Fan, Yuzhang Lin

原文摘要

Distribution planners need installation-level inventories of rooftop photovoltaic (PV) capacity, but interconnection records may be incomplete or spatially aggregated. Conventional image-based estimators segment PV arrays, convert the mask to physical area, and multiply by an estimated power density, coupling capacity prediction to mask-derived area. We propose Direct Head, a segmentation-guided architecture that retains U-Net segmentation and soft-mask pooling while directly regressing nonnegative capacity from PV-region features and area descriptors. We evaluate three independent runs on 7,590 French rooftop installations under the original split and a department-based holdout. Relative to a parameter-matched area-density neural model, Direct Head reduces mean absolute error by 8.7% and 12.5%, and symmetric mean absolute percentage error by 14.0% and 20.7%, respectively. Comparisons with conventional estimators, controlled architecture variants, auxiliary-loss ablations, and representative cases show that segmentation provides useful spatial evidence without requiring capacity to be determined by an area-density product.

cs.CV · cs.AI

新论文 · v1arXiv:2611.22075

Can You Undo It? Reversibility Classes for AI Agent Actions

Deepak Waghmare, Shaad Sufi

原文摘要

Tool-using AI agents change files, databases, access rights and public information, yet permission to call a tool does not establish whether a mistake can be undone. We present Reversibility Classes, a declared five-level vocabulary with argument-sensitive rules, conservative defaults and an explicit prior-state capture contract. The pinned implementation and evidence-linked registry cover 536 tools across 20 MCP servers. All 738 rule-case checks conform to declared expectations, while native annotation mapping differs from draft registry labels for 118 of 481 comparable tools. One AI-assisted reviewer approves 148 ratings, 75 matching the registry; this is not independent agreement. On 2,158 recorded runs, the implemented session guard gates the associated action in all 29 attributable attack-positive runs, compared with 18 for a read/write proxy, while adding nine approval points across 1,209 benign runs. Controlled file, Git, SQLite and three-browser experiments show that recovery depends on captured state, later changes, editor storage and recovery route, and expose a Git merge-state exception. The evidence supports making recovery assumptions explicit and testable. It does not establish registry-wide accuracy, reduced production harm or general recovery reliability.

cs.AI

新论文 · v1arXiv:2611.22069

MedGuardBench: A Patient-Oriented Benchmark for Personalized Guardrail Development

Sara Khosravi, Xiaofei Wen, Majid Sarrafzadeh, Muhao Chen

原文摘要

Safety is personal: the same action can be safe for one person but unsafe for another. Yet current guardrails judge content without considering who is affected. Using medication safety, we introduce \textit{patient-conditioned medication guarding}, which predicts a safety verdict and 17 risk categories from a patient profile and prescription. We release \textbf{MedGuardBench}, a drug-disjoint benchmark of 5{,}001 scenarios across 87 medications. We find that standard guardrails rely on surface cues, such as wording and profile presence, rather than patient facts. To address this, we develop \textit{personalized reasoning guardrails} with three components. \textbf{Personalized Reasoning Distillation} links risks to relevant patient fields. \textbf{Context Unrolling (CU)} makes models check each field themselves, reducing false alarms on most backbones; controls show that providing the check does not yield the same benefit. A \textbf{rule-based reward (guided GRPO)} improves unsafe recall and complex-risk detection, including allergies and adverse reactions, on some backbones, but increases false alarms on others. Our models outperform base models and matched no-profile controls in accuracy, while reducing false alarms relative to no-profile controls. Profile removal and swapping confirm that gains come from patient information, not fine-tuning alone.

cs.AI · cs.LG

新论文 · v1arXiv:2611.22061

FM-Mamba: An Efficient Native 2D Selective State Space Model with Stability-Aware Discretization

Shengyuan Hu, Yiyong Sun, Weihang Shan, Yongzhe Chang

原文摘要

When applied to two-dimensional features such as images, Mamba-based methods either flatten the feature map into 1D sequences and fuse multiple scans, which breaks spatial adjacency and multiplies computation, or recurse directly on the 2D grid but rescale the recurrence with a fixed input-independent factor and aggregate only the predecessors on a single anti-diagonal, leaving the discrete step unconstrained by 2D stability theory. We propose \textbf{FM-Mamba}, a native 2D selective state space model built on the general Fornasini--Marchesini (FM) model. Its coupled recurrence evolves along both spatial axes; cross-axis coupling induces a discrete diagonal-predecessor term $\tilde A_0$, and two independent input-dependent step sizes $Δ_1(i,j)$, $Δ_2(i,j)$ govern propagation along the two axes, so that the coefficients adapt per position and per direction. A normalized ratio $r_k=a_{0,k}/(a_{1,k}a_{2,k})$ yields a closed-form piecewise bound on the two step sizes; when a configuration falls outside the stable interval the steps are projected onto this bound, and the criterion is re-verified per position after training, which suppresses systematic local amplification without global damping. The recurrence is executed by a dual-diagonal wavefront scan: each update gathers information from two anti-diagonals, and the input enters the current state through a single projection, so that neither training nor inference flattens the feature map. Code is available at https://github.com/BitSsm/FM-Mamba-Runtime

eess.SY · cs.AI · cs.CV

新论文 · v1arXiv:2611.22055

Weight-Space Programming Doesn't Need Transformers

Hyunjong Ok, Jaeho Lee

原文摘要

Standard neural architectures such as the Transformer were chosen because they train well with gradient descent and run fast on GPUs. In weight-space programming, a large language model (LLM) writes target weights by hand without gradient-based training, so those reasons no longer apply. Should we keep the Transformer? Given minimal guidance, ten labeled examples, and a sandbox, mainly on multiple-choice questions, we task three frontier LLMs with either generating the weights of a fixed architecture or designing the architecture themselves. When free to design, all three compile plain-text facts into a sparse table queried by a fixed procedure, reaching up to 54.4% test accuracy. An LLM-designed model achieves higher test accuracy than the most accurate released Transformer that the same LLM wrote while using about 1,800x fewer parameters. Without gradients, they choose a table, in which the LLM can state what each weight means and can find and rewrite a wrong one. Each wrong answer points to one row, and removing a row changes about one other answer. From this, we propose an associative-memory-based model up to 17x faster on GPUs than a same-sized Transformer. We also test pretrained weights, other benchmarks, and tasks where models generate answers. Pretrained LLMs remain more accurate; the claim concerns which structure to build when writing weights.

cs.LG · cs.AI · cs.CL

新论文 · v1arXiv:2611.22039

RLDE: A Reinforcement Learning Framework for Interpretable Iterative Document Editing

Nikhil Parkar, M. Marimuthu, Riya Joshi, Karthikeyan N

原文摘要

Improving the quality, clarity, and flow of long documents remains a challenging problem in NLP, as human editing typically occurs through iterative refinement rather than a single-pass rewrite. Most existing automated systems act as black-box generators or rely on supervised learning, limiting interpretability and failing to capture the sequential nature of revisions. In this paper, we propose RLDE, an automated end-to-end metric-driven RL framework that formulates document revision as a reference-free MDP problem (unsupervised with respect to paired document-revision targets). Working with a limited action space (NOOP, INSERT, DELETE, REPLACE), RLDE leverages Pointer Networks for dynamic span selection at the token level and sends generative actions to the frozen BART-base model to obtain context-aware suggestions. PPO agents learn to optimize a multi-objective reward function based on readability, perplexity, and grammar metrics with the help of a novel SBERT-based Semantic Preservation Penalty constraint to avoid deletion degeneracy. Evaluations on the BEA-2019 dataset showed statistically significant results (p < 0.01) across all metrics, where Flesch Reading Ease increased from 49.8 to 59.4, perplexity decreased from 63.7 to 39.8, while preserving 93% of semantic information. When compared to Chain-of-Thought (CoT) prompt-engineered models like GPT-4 and LLaMA-2, RLDE shows high performance at a very low computational inference cost. Moreover, the quality of edits generated by RLDE has been verified via human evaluations, scoring 4.9/5 for meaning preservation and 4.4/5 for edit quality.

cs.CL · cs.AI · cs.LG

新论文 · v1arXiv:2611.22033

Thinking Harder Doesn't Lead to Better Choices: Reasoning Effort Increases the Confidence but Not the Correctness of LLM Shopping Agents

Adi Krishnamoorthy, X. Y. Han

原文摘要

Large language models (LLMs) are increasingly deployed as autonomous shopping agents that process multimodal product listings (consisting of visual and text data) and recommend purchases on behalf of consumers. Naturally, one might assume that models with more thinking power would make higher-quality purchasing decisions. To test this belief, we ask two questions about such agents: does asking a model to think harder make it choose better products, and does the confidence it reports tell us the likelihood it may be wrong? We benchmark ten frontier models on three product-choice benchmarks built from WebShop, Amazon's SQID and Wayfair's WANDS platforms to see if LLMs with more thinking power actually make better-quality product purchase decisions. Across all three benchmarks, we found that increasing the thinking effort does not increase decision quality in a statistically meaningful way. The same change in thinking effort raises stated confidence at a statistically significant level, so extra thinking buys confidence rather than correctness. We also measure the calibration of these models (how well their reported confidence percentage matches their true correctness percentage) in these tasks and show that most models are not well-calibrated at every effort level. We also find that errors are strongly shared: when one model errs in a task, several others tend to err on that same task. Furthermore, on tasks where several models all err, a large number of them pick the same wrong product.

cs.AI · cs.CL · cs.CY · cs.HC · econ.GN

版本更新 · v2arXiv:2605.09079

Scaling Causal Reasoning with Increasingly Complex Causal Simulators

Nicolás Astorga, Anita Kriz, Mihaela van der Schaar

原文摘要

Despite surpassing human performance across mathematics, coding, and other knowledge-intensive tasks, large language models (LLMs) continue to struggle with causal reasoning. A core obstacle is the target data itself: causal systems are complex and often expressed in non-executable forms, while ground-truth answers to causal queries are inherently scarce. We introduce CauSim, a framework that turns causal reasoning from a scarce-label problem into a scalable supervised one. CauSim constructs increasingly complex causal simulators: executable structural causal models (SCMs), incrementally built by LLMs, that scale to globally complex systems while maintaining verifiable answers to causal queries. CauSim operates across representations by formalizing non-executable causal knowledge into code, enabling data augmentation, and translating executable SCMs into natural language, enabling supervision in previously difficult-to-supervise representations. We structure our research into two parts: (1) how to construct increasingly complex causal simulators, and (2) a systematic study of what CauSim enables, demonstrating generalization across representations, consistent gains from curriculum scaling and data volume, LLM self-improvement through self-generated simulators, and data augmentation via formalization of existing domain knowledge.

cs.AI

版本更新 · v2arXiv:2606.05464

Optimization-like Reasoning in LLMs over Expanding Search Spaces

Nicolás Astorga, Nabeel Seedat, Mihaela van der Schaar

原文摘要

Verifiable reward training has improved mathematical and coding reasoning, but these domains capture only part of step-by-step decision making. Many real-world tasks require finding a high-value feasible plan among many valid alternatives. We introduce OPT*, a scalable family of optimization-style tasks for training and evaluating LLM step-by-step optimization-like reasoning along a complexity axis: each task provides a feasibility checker and evaluator, while a complexity parameter expands the search space without requiring new human labels. This motivates studying these tasks in two regimes: (i) solver-guided online policy optimization, which uses a solver as a value oracle for partial states and applies rank-based reward shaping to reinforce better next steps, and (ii) search-based offline RL when such solvers are unavailable. Theoretically, we relate success in large search spaces to the information a reasoner extracts per unit of search budget. Empirically, we ablate the ingredients that make search efficient on OPT* and show that training on OPT* improves step-by-step optimization-like reasoning.

cs.AI

新论文 · v1arXiv:2611.22030

What Does an Agent Need? Bundle-Aware Agent Resource Composition

Yunxiao Shi, Tingwei Chen, Dannong Xu, Ahad N. Zehmakan, Jialie Shen, Zhendong Chu, Qingsong Wen

原文摘要

An agent resource's value depends on its backbone and companion resources and is known only after execution. Token-based generators remain tied to their training catalog, while independent resource scoring misses complementarities and reused bundles target other tasks. We formulate Task-Conditioned Agent Resource Composition (AgentComposition): jointly selecting a backbone model and a tool-skill bundle to maximize end-to-end execution utility. Our Retrieval-Grounded Bundle-Aware Agent Generation retrieves candidate resources and past configurations of similar tasks. A fine-tuned generator reads resource descriptions and recombines candidates through constrained beam search, allowing new resources without retraining. A bundle-aware critic ranks complete configurations using a large language model (LLM) teacher and historically successful configurations as execution proxies, and adapts to execution feedback when available. We evaluate reference recovery, offline execution with judged and native rewards, and online recommendation with real execution, building AgentSelect-Exec with 5,060 executed configurations. Our method recovers the complete reference configuration within ten candidates for 67.73% of AgentSelect queries, nearly twice the strongest baseline, and leads all baselines on SkillsBench. The proxy-trained critic raises the top-ranked agent's execution score from 7.58 to 8.37 out of 10; execution feedback increases it to 8.78, above the strongest baselines. Four target-domain queries suffice to adapt the generator to SkillsBench. Agents recommended online outperform the strongest baselines under both an LLM judge and native verifiers. These results support learning and transferring joint resource composition from task experience.

cs.AI

新论文 · v1arXiv:2611.22027

GauntletScore: Auditable Fact-Checking That Matches the Best Published Checkers on LLM-AggreFact

Ernie Hobbs, Jim Hinson

原文摘要

Checking whether a generated sentence is supported by its source document is a standard safeguard for AI-written text, and the strongest published checkers are single models. We describe GauntletScore, an evidence engine whose few rules and thresholds were set on the benchmark's development split. It puts typed questions to independent readers from different providers and records each answer as a measurement in an evidence ledger. It combines the measurements with a fixed Bayesian rule that counts a contradiction only when two readers agree. On the full LLM-AggreFact test split (29,320 claims across 11 datasets), following the leaderboard's approach, GauntletScore achieves 79.51 balanced accuracy, above every average published on the leaderboard as of 15 November 2026. That places it level with the best of them, nimble-v4-max (79.45 from its published dataset scores; the engine's 95% interval, 78.47 to 80.53, contains it). On the same claims it beats two of its three readers used alone (Readers A and B) by 1.66 and 2.17 points (95% intervals 0.80 to 2.53 and 1.29 to 3.06). Without AggreFact-CNN, described below, the leads are 0.90 and 1.32 and both intervals stay above zero. The engine also records the evidence behind every verdict, which a single model does not. Its largest margin is on full-length news articles checked against multi-sentence summaries (AggreFact-CNN), a format that resembles much regulated writing. There it scores 76.45, the best on the leaderboard, against 70.1 for the next best and 63.9 to 67.8 for the six systems with the highest averages. The test split was run once, on an engine frozen before the first test claim, under a stopping plan committed in advance. We release per-claim predictions and the scoring script.

cs.CL · cs.AI

版本更新 · v2arXiv:2510.15200

The Economics of AI Foundation Models: Openness, Competition, and Governance

Fasheng Xu, Xiaoyu Wang, Wei Chen, Karen Xie

原文摘要

Foundation model developers choose how open to make their models. Greater openness lowers downstream adaptation costs but can also make competing models easier to use. In a two-period model, an incumbent chooses its license fee and openness, a deployer adapts the model for users, and an entrant offers a competing model. Usage builds model-specific experience that lowers the deployer's future adaptation cost, creating a data flywheel that favors continued use of the incumbent. We identify conditions under which optimal openness falls and then rises as learning strengthens. The incumbent moves from full openness and a high fee to restricting openness for retention, then to a low fee that builds enough usage and experience to retain the deployer despite greater openness. A full-openness mandate can induce the incumbent to abandon this low-fee strategy. The higher fee can reduce first-period adaptation despite greater openness. An openness trap arises when this welfare loss exceeds the second-period gain from easier adaptation of the entrant's model. The trap can disappear when experience transfers across models. Full integration eliminates the fee between the incumbent and deployer but commits deployment to the incumbent, potentially forgoing a rival that is cheaper to adapt. Integration raises their joint profit, consumer surplus, and welfare when learning is sufficiently strong, but can lower them when learning is weak. A small usage subsidy can benefit both developers, the deployer, and consumers. It can instead induce a fee increase larger than the subsidy and reduced openness, raising the deployer's net payment per use and adaptation costs. The resulting reduction in adaptation lowers welfare even before subsidy expenditure is deducted. These findings connect AI policy to developers' pricing and openness responses and the portability of deployment experience.

econ.TH · cs.AI