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

+979新增论文
+781其中 AI 相关
+119已有论文修订
+95新增 AI 导读

arXiv 最近完成扫描:2026-11-20 08:48 · 全站最近入库:2026-11-20 19:58各信源进度 →

统计口径与数据新鲜度

统计 2026-11-20 01:49 — 2026-11-21 01:49(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:2026-11-20 08:48最近发现新增版本:2026-11-20 19:58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。

28,961 已收录论文24,030 AI 相关8,912 有修订112 HF 社区发现

浏览论文库

时间范围跟随排序:版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常,查看状态 →

找到 21,358 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-20 08:46(北京时间)
新论文 · v1arXiv:2611.22186

Is 4-Bit Quantization More Than Noise to a Chat Model's Safety Behaviour? Attributing Behavioural Shifts with Matched Perturbations

Ege Kaya

原文摘要

Chat models on phones and laptops are usually 4-bit copies of the model that was evaluated, and audits of such copies disagree about what quantization does to refusal. A 4-bit build is one of many weight perturbations of the same size, so a change in its behaviour could come from its specific rounding error or from any perturbation that large. In a pre-registered study of three 3-4B chat models, with paired tests, blind human labels and an 8-bit null, we find two 4-bit shifts of opposite sign: with a compliant reply prefilled, Qwen2.5-3B continues into harmful content more often, and on held-out jailbreaks Llama-3.2-3B complies 10.3 points less often. Matched perturbations separate the two. Shuffled and Gaussian versions of the error average close to zero, while scaling the real error, from its mirror image to twice its size, changes behaviour near-linearly. At half the real error, Llama's likelihood is almost unchanged, yet it already complies 5.8 points less often, and llama.cpp's Q4_K_M, at the same bits per weight, moves neither model. Refusal under 4-bit quantization depends on the specific rounding error, not on bits per weight or likelihood, so each build has to be compared with its original.

cs.CR · cs.CL

版本更新 · v2arXiv:2504.07927

Physics-Aware Zero-Shot Learning for Low-Dose CT from a Single Sinogram

Yongyi Shi, Siming Lu, Yongfeng Gao, Hao Zhang, Hengyong Yu, Ge Wang

原文摘要

Low-dose computed tomography (CT) imaging remains a persistent challenge, especially under heterogeneous acquisition conditions. Zero-shot learning, which enables models to learn directly from individual images without external supervision, represents a compelling approach to mitigate distribution mismatches arising from variations in scanner hardware, imaging protocols, and patient demographics. However, most existing zero-shot approaches employ down-sampling operations that reduce spatial resolution and cause domain shifts, which become particularly evident when applied to raw sinogram data. Here, we propose a physics-aware zero-shot CT denoising framework to exploit the conjugate symmetry property intrinsic to CT sinograms. Conjugate X-ray pencil beams traverse identical anatomical paths and ideally yield the same expected attenuation values but with independent noise realizations. By randomly mapping sinogram values between conjugate-ray pairs, we synthesize structurally consistent yet noise-diverse sinogram pairs, enabling self-supervised network training without external data. The resulting lightweight denoising network demonstrates strong generalization across low-dose simulation, clinical helical CT, and preclinical photon-counting CT datasets. Quantitative and qualitative results demonstrate competitive denoising performance, with favorable noise suppression and structural preservation across diverse acquisition settings. This work establishes a generalizable and data-efficient paradigm for CT denoising that integrates physical redundancy into learning-based image reconstruction. The code for the proposed method is openly available at: https://github.com/shiyongyi/Physics-Aware-Zero-Shot-CT-Sinogram-Denoising

eess.IV · cs.CV

新论文 · v1arXiv:2611.22183

A Compilation Model for Divergent Parallel Workloads

Alexander J Root, Christophe Gyurgyik, Devanshu Ladsaria, Rohan Sawhney, Jonathan Ragan-Kelley, Andrew Adams, Matt Pharr, Kayvon Fatahalian 等 9 位作者

原文摘要

Efficient execution on modern parallel machines requires more than abundant parallelism; it demands instruction coherence, memory coherence, and load balance to maximize SIMD utilization and memory bandwidth. Efficiently mapping \textit{divergent} parallel workloads to modern machines, where control flow, memory accesses, and unbounded iteration depend on per-thread runtime values, often requires global program restructuring that trades off increased coherence with reduced locality, parallelism, or extra work. We observe that the key optimizations used to accelerate divergent parallel programs can be expressed as compositions of well-known program transformations from the functional programming and loop optimization communities. We operationalize this insight through \textit{parallel dispatch graphs}: a compilation model that represents divergent programs by representing control flow, data flow between logical stages, and parallel loops within a single abstraction. To facilitate rapid exploration of points in the optimization space, we provide a scheduling language that includes composable transformations of data-dependent control flow and data flow, together with standard loop transformations and hardware bindings. This separation of concerns allows the core algorithm to remain portable, while the execution schedule can be specialized to the specifics of target hardware and input data values. We evaluate our system using popular applications from the domain of spatial computing, and show that the optimization strategies of hand-optimized parallel tree traversals, path tracers, and Monte Carlo PDE solvers can be reproduced as schedules in our system. Our generated code matches or outperforms the hand-optimized implementations (geomeans $1.02$--$2.95\times$ faster). Additionally, we demonstrate empirically that the best schedule varies across programs, machines, and inputs.

cs.PL

新论文 · v1arXiv:2611.22180

CueClash: Frontier Vision-Language Models Judge 3-D Relations Close to the Picture Plane

Soham Dan

原文摘要

A vision-language model (VLM) can judge 3-D relations from the picture or from the ground layout. In photographs the two usually agree. We separate them with CueClash, rendered scenes in which the 2-D cue for each of five relations is right on exactly half the statements. The shortcut gap is accuracy when the cue is right minus accuracy when it is wrong. For Claude Opus 5.5, GPT-5.5 and Gemini 3.6 Flash at low reasoning effort it is +18 to +37 percentage points. It is +7 to +29 without two relations whose cue-wrong items are confounded with borderline cases or occlusion. Given ground coordinates and each relation's definition, they answer 99 to 100% correctly. Pixel measurements with camera depths remove most of the gap for objects fitting through openings. The failure lies in the step from the picture to the layout. The maps that the models draw of the scene from above mostly keep the depth order but, like the picture, shorten depth. GPT-5.5's maps stay close to a copy of the picture. Opus 5.5's and Gemini's maps undo about 0.6 of this shortening. Gemini's then shrink all depths by a further uniform factor. Opus 5.5 follows the picture mainly on fits, and its maps correct depth about as much as human observers of real scenes do. The three models also lose accuracy on CV-Bench photographs where a reliable 2-D cue misleads. In two open models the layout information reaches the language model, but their maps lose much of it.

cs.CV · cs.AI · cs.RO

新论文 · v1arXiv:2611.22179

RAG in a Trenchcoat: When Minimal Memory Is Enough for Agentic Systems, and When It Isn't

Jingyu Liu, Zongze Li, Zach Xu, Zhanhui Zhou, Tahseen Rabbani, Dawn Song, Ce Zhang

原文摘要

Agentic memory research has accumulated extraction, supersession, reflection, and graph synthesis on top of retrieval-augmented generation. However, we found that almost none of those primitives earn their cost in our tested memory benchmarks: we show this with BRIDGE, a deliberately minimal RAG memory with chained observation notes, hybrid BM25 with cosine retrieval, and chronological rendering, which reaches parity within bootstrap CI with the prior SOTA on LongMemEval-S and BEAM, and beats the best memory baseline on EverMemBench by 5.7-6.3 points at matched reader. It runs at ~5-7K reader-prompt tokens per query, using a single LLM call for each ingestion or query. We argue that this parity is itself the result: standard benchmarks score what memory retrieves, not what memory enables when the agent acts. To test the latter, we introduce five adversarial probes (Rule Application +-, Pattern Induction Style / Procedural, Dynamic Multiturn) judged at generation time by a cross vendor verification chain, where each query can only be solved by applying a remembered rule or pattern it never mentions. Across 5 probes x 20 scenarios, no architecture wins universally: atomic fact extractors (LangMem) win rule application, while BRIDGE's block-summary memory wins pattern induction. This result holds across four domains, three graders, and three reader families. Each design fails where the other wins, for reasons a recall benchmark cannot see: atomic-fact extractors discard the form a demonstrated pattern takes, while block-summary memory retrieves the rule but the reader does not act on it, a gap that presenting the rule more prominently largely closes. What separates memory designs is not what they retrieve but what they enable an agent to do, and that is what agentic memory should be evaluated on.

cs.AI

版本更新 · v2arXiv:2608.24089

CodeHID: Learning an Addressable Hierarchical Code Index for Generative Code Retrieval

Zhen Li, Yuhong Chen, Wenhao Xu, Xiaodong Li, Hui Li

原文摘要

Code retrieval models have predominantly relied on a flat matching paradigm that treats code snippets as independent candidates, making them less capable of distinguishing similar code candidates. Generative retrieval offers a solution by constructing a learnable index over the code corpus, guiding the retriever to better understand how code candidates are semantically organized and addressed. However, naively applying generative retrieval in the code retrieval task may result in operating over an identifier space whose prefixes do not correspond to meaningful code-semantic regions. In this paper, we propose CodeHID, a generative code retrieval framework that reformulates the code retrieval task from flat candidate matching to coarse-to-fine semantic address generation. CodeHID relies on two core components. First, Pseudo-Neighbor Guided DocID Learning constructs a globally static hierarchical index by applying multi-level residual quantization and $k$-nearest-neighbor pseudo-supervision, ensuring that semantically related code snippets share prefixes while preserving target-level separability. Second, Dual-Phase DocID Generation Guidance reliably navigates this fixed index by combining training-side ranking enhancements, using hard negatives and rank distillation, with inference-side candidate constraints and prefix-aware decoding. Extensive experiments on CoSQA and ProCQA benchmarks demonstrate that CodeHID outperforms existing sparse retrieval, pre-trained code models, dense code retrieval, and generative retrieval baselines by a large margin in most cases, achieving particularly strong improvements in rank-one retrieval metrics.

cs.IR

版本更新 · v3arXiv:2608.00023

Role Steering of Language Models for Social Simulations

Isaac Song, Mohammed Rehan Parwani, Glenn Matlin, Emile Anand, Akhil Theerthala, Arjun Chatterjee, Anthony Wen-Ming Zang, Maria Kostylew 等 11 位作者

原文摘要

Social simulations built from language-model agents need role-conditioned behavior that can be checked before agents are placed into a simulated population. We introduce an activation-steering screening workflow for role-conditioned agents: define a role profile, extract a role-specific direction, sweep four steering coefficients, evaluate role-profile alignment, and pass or flag each candidate configuration. On OLMo-3-7B-Instruct, we apply the workflow to a mixed 275-role inventory with 228 role-agnostic questions, GPT-4.1-mini prompted role references, and GPT-4.1-mini judges. We also introduce CastVectors, which receive higher judged role-profile alignment than an assistant-axis directional control from prior persona-vector work, with mean overall scores of 63.2 versus 41.1 across the tested grid. They also preserve high lexical diversity, while the control drops sharply at larger coefficients. The role-level screen is the main contribution: most roles improve as steering increases, but 38 roles decline across all six measured dimensions, showing why simulation builders should choose coefficients per role rather than deploy a uniform high-strength setting. Our code is available at https://github.com/HCAI-Lab-GT/casting-call-vectors, and the role vectors and evaluation data at https://huggingface.co/datasets/HCAI-Lab-GT/casting-call-vectors.

cs.CL · cs.AI

版本更新 · v2arXiv:2609.36686

Where Root Cause Analysis Fails: A Retrieval-Reranking Decomposition

Hada Melino Muhammad, Luan Pham, Laure Barrière, Sachin Shetty, Leonardo Pulga, Flora D. Salim

原文摘要

Identifying the root cause of an anomaly among hundreds of sensors is critical for preventing safety incidents and costly downtime in complex monitored systems. Existing studies evaluate root cause analysis (RCA) methods using top@k accuracy. We show that this metric has a fundamental blind spot: it conflates two failure modes, retrieval failure, where the true cause is never considered, and reranking failure, where it is considered but ranked too low. In this work, we introduce a retrieval-reranking decomposition and audit four well-known benchmarks to expose this blind spot. Our experiments show that, on industrial CPS benchmarks, the statistical baseline scoring every sensor mis-ranks the true cause 79-100% of the time, and graph-based methods never clearly beat the best statistical baseline, whether their causal graphs are learned on short fault windows, on retrieved candidate pools guaranteed to contain the cause, or on multi-day normal-operation data. Meanwhile, on microservice benchmarks, where each injected fault drives the faulty service's own metric far above its baseline, retrieval is nearly solved (98-100%). Guided by the decomposition, we build a two-stage pipeline combining a multi-signal retriever with an LLM reranker that, as one fixed configuration, matches or exceeds the best baseline's top@1 accuracy on all six benchmark suites (by up to +12 points), with no causal graph or labeled data required. When all methods rank the same retrieved candidates with the true cause guaranteed present, adding a short system-description document lets the reranker lead the best baseline by +7 to +18 points on every benchmark. Code is available at https://github.com/cruiseresearchgroup/DecompRCA.

cs.LG

新论文 · v1arXiv:2611.22177

The Only Edge Was Memorized in Training: LLM Trading Agents Around Earnings and Bad News

Jackson Wang

原文摘要

Large language model (LLM) trading agents are tested on news, but news reaches stock prices within minutes, and a benchmark that replays headlines at their publication time finds no alpha in trading on them. Money can then remain in two places, before news whose date is known, such as earnings announcements, and after bad news, if the market's fall went too far. However, studies of LLMs and earnings read documents published with or after the announcement, and studies of reversal sort monthly portfolios, so neither tells whether an agent can trade these moments after costs. In this paper, we test LLM agents at both moments on the same look-ahead-free benchmark, before 598 earnings announcements and after 409 bad-news drops of large US stocks. Every input is cut at the decision close, staged agents are compared with single calls, simple rules and a logistic regression, and every decision is made with and without the stock's name. Across five models from four vendors, no agent beat the simple bets. Before earnings, the agents mostly bought and earned what buying before every announcement earned, and after bad news, a Gemini-based agent called almost every fall an overreaction while no agent picked better dips than buying them all. The only gains came from what the models had memorized in training. Given only the ticker and the dates, a GPT-based model called the direction of 59.2% of the announcements before its training cutoff and 53.3% after it. Agent builders should therefore compare an event agent with the simple bet and test it blind after the training cutoff.

cs.AI · cs.MA

版本更新 · v3arXiv:2511.19218

From Strategy Search to Policy Learning: MCTS-Guided Adversarial Co-Evolution for LLM Safety Alignment

Xurui Li, Kaisong Song

原文摘要

Large Language Models (LLMs) increasingly power web-facing applications, including conversational assistants and content services, where open user interaction exposes them to diverse safety threats and adaptive attacks. In particular, jailbreak attackers repeatedly reformulate harmful requests to bypass safety alignment. Although existing studies have concentrated on defending dynamic jailbreaks, they often retain discovered attacks as flat samples and discard information about how those attacks were found. In this paper, we introduce MACE, an MCTS-guided adversarial co-evolution framework that turns search into policy-learning signals for both an attacker and a defender. Its search component, GS-MCTS, explores multi-step compositions of rewriting operators and associates discovered attacks with empirical success frequency and path depth. Its optimization component, AC-TGPO, uses these statistics for difficulty-aware group-relative credit assignment and constructs a temporally asymmetric curriculum from current interactions, pairings between the latest attacker and preceding defender, and attacks that survive a defense update. Thus, MACE connects search structure to both within-round policy optimization and cross-round memory, rather than using MCTS only to generate attacks or treating each round as a flat adversarial dataset. Removing difficulty-aware weighting, asymmetric samples, or persistent hard samples increases defense ASR-LR by 3.7 %, 6.0 %, and 5.0 %, respectively. Across evaluated models, the resulting attacker achieves the highest ASR-LR in all six model--dataset settings, while the defender reduces average ASR-LR from 11.9 % for the best baseline to 8.8 %.

cs.CR · cs.AI

版本更新 · v3arXiv:2606.02879

Direct Informed Sampling on Riemannian Manifolds via Loewner Order Lower Bounds

Phone Thiha Kyaw, Jonathan Kelly

原文摘要

Informed sampling techniques accelerate sampling-based motion planners by focusing the search on promising regions of the state space, yet most existing methods rely on Euclidean heuristics that can become inadmissible under configuration-dependent Riemannian metrics. While scalar eigenvalue bounds restore admissibility by uniformly scaling the Euclidean distance, they discard the directional structure of the metric, producing overly conservative informed sets. We propose a matrix-valued admissible heuristic that exploits the Loewner order on symmetric positive definite matrices to compute the tightest constant lower bound on the metric tensor while retaining directional information discarded by scalar bounds. The Cholesky factorization of this constant bound maps the lower-bound informed set to a standard prolate hyperspheroid in Euclidean space, enabling direct, rejection-free sampling using existing algorithms. We then present an incremental algorithm for computing such a constant lower bound on the metric tensor from a sequence of pointwise evaluations. Experiments on manipulation tasks with a 6-DoF UR5, 7-DoF Franka, and 14-DoF PR2 under three distinct Riemannian metrics show that our heuristic produces consistently tighter informed sets than those produced by the Euclidean and scalar-eigenvalue heuristics, accelerating convergence across multiple state-of-the-art asymptotically optimal planners. Our implementation is available in the open-source geodex framework at https://github.com/utiasSTARS/geodex.

cs.RO

新论文 · v1arXiv:2611.22172

Towards Optimism-Pessimism Trade-off in Model-based Offline-to-Online Reinforcement Learning

Guochen Zhou, Yijun Yang, Qiqi Duan, Qing Su, Weiming Ou, Li Shen, Shihao Ji, Yuhui Shi 等 9 位作者

原文摘要

Model-based offline-to-online reinforcement learning (RL) enables sample-efficient adaptation by leveraging offline pre-training for online fine-tuning. However, the distribution shifts between offline and online stages often hinder fine-tuning performance. Many existing methods encode the optimism-pessimism trade-off through a single-objective formulation. Selecting its trade-off coefficient through online fine-tuning can require costly bi-level optimization. We identify the balance established during offline training as a key challenge: in our motivating experiments, optimistic policies adapt better to novel tasks, whereas pessimistic policies perform better on the same online tasks. To address this challenge, we propose a bi-objective formulation that captures this trade-off, yielding a Pareto policy pool during offline training. This policy pool enables adaptive selection during online fine-tuning across diverse online tasks. To generate the pool, we introduce Multi-Objective Soft Actor-critIC (MOSAIC), which solves bi-objective problems and constructs diverse Pareto policies. After offline training, a contextual bandit algorithm hierarchically selects the most suitable policy for fine-tuning at each online interaction step. Empirically, our pipeline, Hierarchical Pareto Policy Pool (HiP3), achieves the strongest aggregate performance across both same and novel online tasks and maintains its advantage under stronger distribution shifts. Comprehensive ablations further clarify the roles of the policy pool, MOSAIC-based pool construction, and the online selection mechanism.

cs.LG

版本更新 · v3arXiv:2509.06261

JigSLO: Solving the Resource Puzzle for Efficient Multi-LLM Serving

Kyungmin Bin, Seungbeom Choi, Jimyoung Son, Jieun Choi, Daseul Bae, Daehyeon Baek, Kihyo Moon, Minsung Jang 等 9 位作者

原文摘要

GPU spatial sharing co-locates instances with complementary demands, such as compute-bound prefill and memory-bound decode, promising cost-efficient multi-LLM serving. Prior systems partition compute and coordinate shared memory capacity, but do not model bandwidth contention among co-located instances of heterogeneous LLMs. This contention makes each instance's latency depend on the memory-access patterns of its co-runners, even when its own resource allocation remains unchanged. Drawing on extensive measurements, we find that the latency impact of bandwidth contention can be captured by a time-shared memory-access model with contended and uncontended periods, so the overall latency can be analytically estimated as a time-weighted average of the latency estimates for these periods. We propose JigSLO, a contention-aware multi-LLM serving system that manages memory resources and schedules each model's prefill and decode phases as separate instances, treating their co-location as a placement decision. Its contention-aware predictor estimates per-instance latency under any co-location from standalone profiles alone. Guided by these predictions, its placement algorithm jointly selects each instance's tensor parallelism, compute share, and batch configuration, pairing complementary memory-access patterns to maximize throughput within each model's latency targets. JigSLO's shared resource manager complements this on the capacity side, sharing weights, pooling KV cache, and transferring it between prefill and decode instances. Across three datasets covering chat, code generation, and long-context summarization, JigSLO outperforms state-of-the-art baselines by $2.3\times$ in SLO attainment and $1.2\times$ in throughput on average, matching their best attainment with half the GPUs on long-context workloads.

cs.DC · cs.LG

版本更新 · v3arXiv:2611.09670

Goodhart in the Planner: Sampling-Based Control Exploits Rewards Learned from Human Preferences

Soovadeep Bakshi, Apurba Bose

原文摘要

Robots increasingly plan by searching against a reward model learned from human preferences. We measure what this search does to the true objective. On HalfCheetah and Hopper with crowd-sourced labels, and on Meta-World drawer-open and plate-slide with synthetic and blind human labels, harder MPPI search cuts true return by up to 76% while held-out label accuracy stays at 0.75-0.97; with human labels the drawer opens in 0 of 20 runs, against 10 of 10 for the cloned policy the planner starts from. The effect persists with a learned dynamics model and an image scorer. We prove that $χ^2$-regularized MPPI at a temperature set by the scorer's error spread on the proposals is worst-case optimal, and that no gain can be certified once that spread exceeds the scorer's own. Calibrated from 100 ranked proposal pairs, this rule is the safest fix we test, keeping 86-94% of the cloned policy's return on drawer-open and plate-slide.

cs.RO · cs.AI · cs.LG

新论文 · v1arXiv:2611.22163

Cluster-Guided Decoding with Trust-Region Constraints for Personalized Recommendation

Gezheng Wen, Chao Shen, Raviteja Anantha, Tan Shen Cliff Chung, Layne C. Price

原文摘要

Most recommenders score each item in the catalog separately, and do not explicitly model which semantic region of the catalog a recommendation belongs to. Models expressive enough to offer that control are often too slow to run over a full catalog for online inference. In this paper, we present Cluster-Guided Generative Reconstruction (CG2R), which learns a hierarchy of semantic clusters and uses a cluster-level prior to steer item ranking for personalized recommendations. Unlike existing systems that only use clusters to pre-select candidates, CG2R adds the cluster prior directly to the item scoring rule, so that the learned cluster structure shapes the ranking itself. Because the prior is assembled from cluster centroids rather than per-item statistics, it remains defined even for cold-start users when no history is available. Across the three benchmark datasets evaluated, CG2R outperforms classical, neural, and LLM-based baselines in terms of Recall and nDCG, with its performance gains widening as the catalog grows sparser. In addition, CG2R serves recommendations in milliseconds, suggesting potential as low-latency first-stage retrieval.

cs.IR

版本更新 · v4arXiv:2606.04547

Beyond Retrieval: Learning Compact User Representations for Scalable LLM Personalization

Heng Cao, Fan Zhang, Jian Yao, Yujie Zheng, Changlin Zhao, Lu Hao, Yuxuan Wei, Wangze Ni 等 11 位作者

原文摘要

Personalizing large language models requires adapting model behavior to individual users while preserving robustness and deployment-scale efficiency. Existing approaches typically personalize LLMs either at the input level, by retrieving user histories or constructing profile prompts, or at the parameter level, by maintaining user-specific parameter-efficient modules. The former makes personalization sensitive to retrieval quality and prompt design, whereas the latter incurs storage and maintenance costs that grow with the user population. To address these limitations, we propose TAP-PER (Temporal Attentive Prefix for PERsonalization), a prefix-based framework that encodes user preferences as learnable representations, avoiding the serialization of user histories into prompts and replacing heavy per-user adapters with lightweight user-state prefix embeddings. Inspired by personalized recommendation systems, TAP-PER decomposes user modeling into user-state and query-conditioned components, and incorporates temporal signals to capture the evolving nature of user interests. Experiments on six LaMP tasks show that TAP-PER consistently outperforms prompt-based and model-based baselines across classification, rating, and generation settings. Moreover, TAP-PER uses 130x fewer per-user parameters than OPPU and roughly half the total parameter footprint of PER-PCS at the 1,000-user scale, demonstrating scalable personalization without prompt-serialized histories or heavy per-user adapters.

cs.IR · cs.CL

版本更新 · v4arXiv:2608.10450

Persistent Recursive Worlds Enable Autonomous Software Evolution

Beichen Huang, Zhenyu Liang, Bowen Zheng, Ran Cheng

原文摘要

Software systems accumulate dependencies and design commitments over time, while the agents that modify them operate within bounded sessions. How, then, can development remain cumulative across generations of transient agents? We introduce EvoX Genesis, which organizes development as a persistent recursive software world. Agents enter an accepted repository version, delegate work within local scopes, and submit changes for verification and parent review. Once accepted, those changes pass to later episodes together with the contracts and tests used to assess them. The project thus preserves both earlier work and a basis for evaluating what comes next. We examine this continuity through four studies of formation, continuation and redevelopment. Starting without compiler code, Genesis built a C compiler with 219,676 lines of Rust source through 1,019 agent episodes over 123.4 hours. Within this growing implementation, repairs increased passes on a fixed set of 100 Csmith programs from 61 to 93; a subsequent revision retained all 93 passes. Development also continued in a separate compiler lineage when DeepSeek V4 Flash inherited a GLM 5.2 snapshot. To examine continuity when the implementation itself changes, Genesis redeveloped selected MESA functionality in Rust, with numerical agreement assessed against the Fortran reference across six workloads. The WebGL trajectory, meanwhile, exposed both repairs and regressions as repeated checks recorded progress towards 2,017/2,019 conformance pages passed. A complementary EvoX Compiler case connects program-merging performance to executable graph and backend contracts. Together, these records trace how accepted work becomes a foundation for further development: later agents inherit both usable implementations and concrete ways to assess further revisions.

cs.SE · cs.AI · cs.MA · cs.NE

新论文 · v1arXiv:2611.22161

What RL on Expert-Built Tasks Teaches Nemotron 3.5 Super

Sushant Mehta, Edwin Chen

原文摘要

We post-train NVIDIA's Nemotron 3.5 Super (120B total, 12B active parameters) with reinforcement learning (RL) on five expert-built task datasets, one at a time: real-world software engineering, complex constraints, long-horizon tool use, professional chart understanding and enterprise document reasoning. Each run trains a fresh rank-32 LoRA adapter with CISPO on Tinker, with no supervised stage, and each adapter is compared with the base model on the same tasks, harness and settings. On twelve headline benchmarks, eight built by other groups and four by Surge AI, the gains range from +4.5 points on IFBench to +13.8 on GDP_pdf, which rises from 10.4% to 24.2%; SWE-bench Pro V2 rises from 86.9% to 92.1%, Terminal-Bench 2.1 from 63.9% to 70.3%, MultiChallenge from 58.3% to 66.6%, AutomationBench from 14.8% to 27.5% and Chartography from 39.4% to 50.6%. The document adapter, trained without API tools, also gains +8.0 points on AutomationBench. In attempts read side by side with the base model's on tasks each adapter newly solves, the base model's failures recur in a few modes, and each adapter does more often what the base model missed: the tool-use adapter finds the data and the rules before acting, the coding adapter writes its own test or scratch files on SWE-bench Pro V2, the constraints adapter carries earlier turns into its answers, the chart adapter measures the marks in code, and the document adapter reads AutomationBench's spreadsheets.

cs.LG · cs.AI · cs.CL · cs.SE

版本更新 · v2arXiv:2609.39340

ElectrolyteFM: Unifying Electrolyte Property Prediction through Cross-Property Knowledge Learning

Jiaxin Yu, Shuo Wang, Peng Wang, Yongcai Wang, Deying Li

原文摘要

Electrolyte formulation design requires balancing multiple physicochemical properties, yet existing models often focus on a limited subset. Learning each property in isolation can overlook transferable chemical information, whereas indiscriminate sharing can introduce cross-property interference. Our directed transfer analysis shows that jointly learning two property prediction tasks can improve or degrade prediction relative to separate training, with asymmetric transfer effects between the tasks. We propose ElectrolyteFM, a unified multi-property prediction model which can more accurately predict multiple properties of each electrolyte by effectively identifying and utilizing property-specific features and knowledge shared across properties. More specifically, ElectrolyteFM learns property-specific representations independently and captures cross-property knowledge through a separately trained expert pool. A router selects relevant shared information for each formulation and target property, and property-specific residual adapters convert this information into corrections to the corresponding representation for prediction. Experiments on Electrolyte12 show that ElectrolyteFM reduces normalized mean absolute error averaged across 12 electrolyte properties by 14.8% relative to the strongest electrolyte-specific baseline. On an independent sodium-electrolyte dataset unseen during training, it reduces conductivity mean absolute error by 6.7% relative to the best-performing baseline.

cs.LG

新论文 · v1arXiv:2611.22160

Collecting Infrastructure-Level Measures for Energy Analysis in Modern Computing Infrastructures

Isac Pasianotto, Stefano Cozzini, Niccolo Tosato

原文摘要

Modern HPC facilities are complex environments in which all components are tightly coupled; consequently, precisely modeling and accounting for energy consumption requires approaches that go beyond instrumenting individual compute nodes. Aiming for a cluster-wide vision, we designed a scalable data collection pipeline that aggregates fine-grained measurements from multiple subsystems-spanning component loads, software and hardware counters, sensors and energy-ensuring consistent sampling frequencies and synchronized data acquisition across the infrastructure. We present a dataset of comprehensive metrics collected at 1 Hz over one month from a production research data center supporting heterogeneous workloads, including HPC, Kubernetes services, and distributed storage. To illustrate what this approach makes possible in practice, we analyze two case studies-distributed I/O and large-language-model inference-showing how cluster-wide measurement reveals system-level dynamics missed by localized instrumentation. The proposed dataset and collection methodology provide a foundation for future work on predictive modeling, energy-aware resource management, and more efficient computing infrastructures.

cs.DC

版本更新 · v2arXiv:2511.18703

Delay-Aware Distributed Multi-Robot Motion Planning

Ardalan Tajbakhsh, Augustinos Saravanos, James Zhu, Evangelos A. Theodorou, Lorenz T. Biegler, Aaron M. Johnson

原文摘要

This paper examines why distributed optimization degrades when multi-robot systems coordinate under realistic communication delays. Using consensus ADMM as a representative framework for generating collision-free, dynamically feasible motion plans in both trajectory optimization and receding-horizon control settings, we show that delay degrades performance structurally: the consensus and dual updates weight stale and current information identically, so staleness enters the optimizer disguised as genuine disagreement and accumulates through the multipliers. Delay-agnostic adaptation (e.g. residual balancing) inherits this contamination and can perform worse than a fixed penalty. We introduce a Delay-Aware ADMM (DA-ADMM) variant that adapts penalty parameters to real-time delay statistics and re-derives the global consensus update as a delay-weighted average, down-weighting stale information in favor of recent updates. Simulations in 2D and 3D show that DA-ADMM improves robustness, success rate, and solution quality over fixed-parameter, residual-balancing, lower-bound, and fixed-constraint baselines. Degradation is therefore governed not by delay length or frequency, but by the optimizer's ability to reason over the age of the information it receives.

cs.RO

新论文 · v1arXiv:2611.22158

Centroid Spectral Coupling: Implicit Alignment and Representation Diversity in Self-Supervised Learning

M. Hadi Sepanj, Benyamin Ghojogh, Saed Moradi, Paul Fieguth

原文摘要

Self-Supervised Learning (SSL) aims to align augmented views while preserving representation diversity and preventing collapse. Existing non-contrastive methods often address these goals using multiple weighted loss terms. We introduce Centroid Spectral Coupling (CSC), a single-term SSL objective that implicitly couples view alignment, representation diversity, and collapse resistance. CSC constructs centroids from independently normalized augmented-view pairs and applies a ridge-based spectral objective to their uncentered second-moment matrix. Preserving centroid magnitudes encodes view agreement while encouraging balanced representation energy across spectral directions. We establish an exact decomposition into nonnegative alignment and spectral-imbalance components, characterize the geometry of global optima, and derive finite-loss guarantees for alignment and diversity. We also show that complete constant-output collapse is a strict saddle in normalized representation space. Experiments on CIFAR-10, CIFAR-100, and Imagenette show that CSC achieves the highest linear Top-1 accuracy among the evaluated non-contrastive methods on all three benchmarks, and the highest kNN Top-1 accuracy on both CIFAR datasets. A ridge-parameter sensitivity study reveals an empirical alignment--diversity trade-off: increasing $κ$ strengthens view alignment but reduces representation effective rank. These results show that a unified spectral objective yields competitive visual representations without explicit contrastive negatives or separately weighted alignment and diversity penalties.

cs.CV · cs.LG · stat.ML

版本更新 · v3arXiv:2610.14350

Natural Code Transformation Confuses Agents in Vulnerability Analysis

Jie Zhu, Ronik Bhaskar, Heather Zheng, Ben Y. Zhao, Kexin Pei

原文摘要

Having demonstrated a strong potential for automating vulnerability discovery, AI agents can also be used to attack software systems at scale. Today's software systems need mechanisms to level the playing field by slowing down potential attacks, giving them time to patch vulnerabilities and strengthen software. Along this line, our study of agent trajectories in vulnerability analysis shows a surprising finding: agents produce dramatically different results across real-world code commits that carry the same functionality and the same vulnerability. This phenomenon appears consistently in multiple code domains and vulnerability categories. It motivates us to study natural and functionality-preserving code transformations derived from real-world software development, and their potential as a defense against attacker agents. We build a corpus of natural transformations by mining recurring development patterns from real-world code commits, and apply them to evaluate impact on agentic vulnerability analysis. Our results show that even a single form of transformation can effectively "confuse" frontier agents, e.g., Claude Mythos 5.2 with Claude Code, reducing successes in vulnerability discovery. We also study different agent failures, whether transformations can be composed to provide stronger protection, and adaptive attackers who seek to reverse them. Overall, our study of natural code transformation significantly expands the research space of using code transformations to defend against malicious agents.

cs.CR · cs.AI · cs.LG

新论文 · v1arXiv:2611.22151

EgoHandWAM: Probing the Role of Hand Motion in Egocentric World Action Modeling

Kailin Wang, Haoxiang Jie, Yaoyuan Yan, Zhiyou Heng, Zhaosong Li

原文摘要

How much does hand motion tell us about what a first-person camera will see next? EgoHandWAM forecasts three-dimensional wrist and fingertip trajectories and a future visual feature from the current image and recent hand states. We train on 751,065 EgoDex windows using a frozen DINOv2 encoder. On held-out development sessions, the action-only model reduces two-second wrist endpoint error from 60.26 to 56.48~mm relative to a stationary forecast. It still underestimates long reaches. Adding future-feature supervision gives nearly identical trajectory accuracy. With the recorded hand history, the visual head reduces smooth L1 error by 1.32\% relative to copying the current feature. Shuffling histories across sessions erases this gain. In a separate four-task probe, adding recorded future displacements to observed context and future tracking-validity masks reduces visual-feature error by 3.93\%. Masks alone give little improvement. A trajectory predicted from the same observations adds little. On two simulated RoboTwin tasks, substituting aligned human examples for an auxiliary batch of robot examples gives no consistent increase in task success across three training seeds.

cs.CV · cs.AI

版本更新 · v2arXiv:2610.03587

AVL-JEPA: Preventing Causal Dynamics Information Collapse In Joint Embedding Predictive Architecture World Models

Yikang Qiao, Ling Zhang, Ziying Song, Duan Huang

原文摘要

Joint embedding predictive architectures (JEPAs) predict future latent representations without reconstructing observations, enabling world models to focus on high-level semantic dynamics. However, a JEPA can preserve high dimensional visual information while discarding information about the physical consequences of actions. We call this failure mode causal dynamics information collapse and propose action-grounded vision-invariance latent (AVL) to prevent this collapse. We first use the executed action as an auxiliary dynamics anchor that encourages the model to preserve dynamics information, and then use a vision-invariance pathway which aligns perturbed and clean latent predictions without discarding dynamics information, forcing the model to fully understand and utilize causal dynamics information. We validate AVL on four robotic control tasks (TwoRoom, PushT, OGBench Cube, and Reacher), showing that it substantially improves success rates under visual perturbations while preserving clean-environment performance. We further evaluate physical consequence alignment, clean-noisy dynamics consistency, and the causal effect of targeted transition subspace erasure. Collectively, these results indicate that dynamic information causally relevant to planning is preserved from collapse under AVL.

cs.RO

新论文 · v1arXiv:2611.22148

Long-Context Generation Is a Sampling Problem (When You Turn Up the Heat)

Allen G. Roush, Minh Nhat Nguyen, Ravid Shwartz-Ziv, Judah Goldfeder, Sanjay Basu

原文摘要

Open-weight language models are widely observed to get worse the longer they generate. Often, attempts to fix this involve more training. At high temperature, The sampler, i.e. the rule that picks the next token out of the model's probabilities, decides whether long and agentic generation survives and/or thrives. We test 29 open-weight models from 1B to 122B parameters on ten established benchmarks plus needle retrieval out to one million tokens, covering SWE-bench Verified coding agents, long chain-of-thought reasoning, long-form writing, long structured output and long-context reading. Samplers that keep their pool of candidate tokens small and steady as temperature rises don't lose accuracy. Samplers that let in low-probability tokens fall apart. At temperature 2, coding agents running top-p, top-k or Top-H resolve 0 of 100 GitHub issues, while the same agents running top-nsigma resolve 46 to 62 and win every issue the two samplers disagree on, on three Qwen3.5 models. Over all benchmarks top-nsigma keeps a median 100% of its temperature-1 score where top-p keeps 26%. Counting how many tokens each sampler lets through explains why. In one Llama-3.1-8B story, going from temperature 1 to 2 grows top-p's pool from 4 tokens to 51,845 while top-nsigma stays at 3. Across 133 combinations of sampler setting and temperature, the typical pool size predicts judged quality (Spearman correlation of -0.77), and when we pick the pool at temperature 1 and only apply temperature afterwards, Top-H, min-p and p-less stop collapsing. At the default temperature of 1 most samplers perform about the same, We release all code and data to the public.

cs.AI · cs.CL

新论文 · v1arXiv:2611.22144

When does deep reinforcement learning pay off in HVAC supervisory control? Four small Modelica case studies trained directly on FMUs

Bingnan Zhao, Harshala Gammulle, Tharindu Fernando

原文摘要

Deep reinforcement learning (DRL) is often proposed for the supervisory control of heating, ventilation and air-conditioning (HVAC) systems. The reported gains are, however, difficult to interpret, because the baselines are often weak and the attainable optimum is seldom known. This paper examines four small, reproducible case studies based on an office building in Brisbane, Australia: pre-cooling under a time-of-use tariff; chiller staging with condenser water reset; chilled water supply temperature reset with a wet cooling coil; and chilled water storage under time-of-use prices and a demand charge. The plants were modelled with the Modelica Buildings library, and in three of the cases a soft actor-critic agent was trained directly on the exported functional mock-up unit (FMU). Each agent was compared with tuned rule-based controllers and, where practicable, with a near-optimal reference from hourly exhaustive search or dynamic programming with perfect foresight. In the two cases without storage the agent reached the near-optimal reference, but the margin above a well-chosen rule was small: 1.3% for condenser water reset and 0.18% for chilled water reset. In the latter case two common reset rules consumed more energy than a fixed set point. With a chilled water storage tank the margin was considerably larger. The agent outperformed both rules (-7.8%, compared with -3.9% and -6.3%) and adopted midday charging during the solar off-peak period, yet it obtained only about 45% of the saving achieved by the perfect-foresight schedule and did little to limit the demand peak. In the pre-cooling case a hand-tuned rule remained superior (-12.0% compared with -5.6%). These results suggest that studies of DRL for HVAC control should state the margin between the best rule and a near-optimal reference. Some practical difficulties encountered in training on co-simulation FMUs are also reported.

cs.LG · cs.AI · eess.SY

版本更新 · v2arXiv:2610.18352

Let the Mind Conceive, Let the Hand Achieve: Coupling Grounded Action and Progress Evaluation for Long-Horizon Manipulation

Youpeng Wen, Gang Wang, Tianyu Huang, Weiwei Zhang, Zhijian Huang, Wei Chen, Xieyuanli Chen, Yunhui Liu

原文摘要

A robot following a multi-step instruction needs to act on the right objects in the right order, even as the scene changes. It must also maintain awareness of task progress and determine when to advance. To address these challenges, we introduce Orchestra, an agent harness that separates task planning and execution management from action generation. Orchestra decomposes a complete multimodal instruction into ordered subtasks, tracks completed steps, and updates the locations of relevant objects and destinations during execution. It maintains a shared context containing the active subtask, spatial grounding, and recent observations. An action policy uses this context to generate actions, while a separately trained progress evaluator estimates when to advance to the next subtask. To evaluate whether robots trained on atomic instructions and short trajectories can execute longer subtask sequences, we introduce a benchmark named RMBench-Compose. Orchestra achieves a mean success rate of $68.22\%$ across nine RMBench tasks, outperforming Mem-0 by $26.22$ percentage points. On RMBench-Compose, it achieves an Average Success Length (ASL) of $7.35$ subtasks, $2.46$ times that of HarnessVLA. Across four real-world multi-step tasks, it improves mean ASL by $25.0\%$ over Orchestra using the base $π_{0.5}$ policy. These results show that Orchestra helps robots carry out longer sequences of subtasks while preserving the requested subtask order and acting on the specified objects. Our code and project page are available at https://youpengwen.github.io/Orchestra

cs.RO · cs.AI

版本更新 · v2arXiv:2605.21752

PEARL: Unbiased Percentile Estimation via Contrastive Learning for Industrial-Scale Livestream Recommendation

Blake Gella, Wei Wu, Yuhao Yin, Zexi Huang, Zikai Wang, Emily Liu, Junlin Zhang, Wentao Guo 等 9 位作者

原文摘要

Recommender systems trained on user interaction data are susceptible to behavioral intensity imbalance--a systematic distortion arising from heterogeneous engagement patterns across users. This imbalance skews feedback signals such that observed interactions no longer faithfully reflect true preferences, causing models to disproportionately amplify signals from highly active users while underrepresenting others, which ultimately degrades recommendation quality and robustness at scale. To address this issue, we propose a nonparametric contrastive percentile approximation framework, PEARL, that models relative preference signals instead of absolute engagement magnitudes. Building upon relative advantage debiasing, PEARL leverages real contrastive interaction samples to approximate percentile relationships directly, without relying on auxiliary distribution estimation models. We provide theoretical justification demonstrating that such pairwise comparisons yield unbiased estimates of percentile-based preference signals. For broader applicability, we introduce a prediction-based bootstrapping mechanism for percentile smoothing to handle sparse and discrete feedback, alongside a generalized value-weighted formulation and a co-training strategy to enhance both modeling flexibility and representation learning. Extensive offline experiments demonstrate that PEARL effectively mitigates behavioral bias and consistently improves recommendation performance across multiple ranking targets. Deployed in a production livestream platform with a combined user base of billions, online A/B testing confirms substantial real-world gains: +2.10% Watch Duration, +0.80% Consumption Amount, +1.49% Interaction Rate, and -6.91% Report Rate.

cs.LG · cs.AI

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