跳到正文
研踪PaperTrail

YOUR RESEARCH, IN SIGHT

让下一篇重要论文，不再错过。

覆盖计算机全学科，以 AI 为快捷入口。关注方向、作者和关键词，把发现变成持续的研究积累。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+0新增论文
+0其中 AI 相关
+1已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-07 20:41 · 全站最近入库：2026-11-07 17:43各信源进度 →

统计口径与数据新鲜度

统计 2026-11-06 21:17 — 2026-11-07 21:17（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：2026-11-07 20:41最近发现新增版本：2026-11-07 17:43库内论文最新更新时间：2026-11-06 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。

17,100 已收录论文14,661 AI 相关4,056 有修订112 HF 社区发现

浏览论文库

研究日报 / 周报 →按信源浏览 →

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 17,100 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-07 20:35（北京时间）
新论文 · v1arXiv:2611.06508

ElecLoad: Adaptive Electricity Load Forecasting Guided by Multiple Events

Huajian Zhang, Mingyue Cheng, Xingpeng Gao, Xiaoyu Tao, Zhiding Liu, Jiqian Yang, Zhi Li

原文摘要

Accurate electricity load forecasting provides critical decision support for energy procurement, generation scheduling, and supply-demand balancing, contributing to the economic efficiency and operational reliability of power systems. However, future electricity demand can be affected by the complex interplay of multiple events described in Web notices, weather alerts, and event schedules. Their effects vary over time and can reinforce or offset one another, making electricity load difficult to forecast accurately. While existing methods can incorporate context as additional input, how to use the effects of multiple events and their interactions to guide the load forecasting process itself remains to be explored. To address this gap, we propose ElecLoad, a context-aware agent for multi-event electricity load forecasting that adaptively organizes the forecasting process. Specifically, we train ElecLoad to combine historical load with context to analyze each event's relevance to the target load, the timing of its effects, and potential interactions among events. Guided by this analysis, ElecLoad adaptively chooses to forecast the entire forecast horizon directly or to partition it into segments according to changes in event effects. When the horizon is partitioned, sub-agents independently generate initial forecasts for each segment. The main agent then reviews these forecasts, and the sub-agents perform selective refinement guided by its feedback. Experiments on multiple electricity load datasets show that ElecLoad achieves state-of-the-art performance in most evaluation scenarios. Code is available at https://github.com/zhjai/ElecLoad.

cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06506

Trace2Eval: Contextualized Evaluation of Coding Agents from Real-World Traces

Ji Shi, Lirong Jie, Lexiao Zou, Yaoming Zhu, Yifan Yang, Xuezhi Cao, Xunliang Cai, Miao Zhang

原文摘要

Coding agents are increasingly used in real-world workflows across diverse domains, motivating evaluation grounded in real user tasks. Existing benchmarks, however, largely rely on static public tasks with reference patches or hidden tests. Next-generation coding-agent benchmarks should therefore move beyond correctness on fixed tasks toward continually renewed evaluation of overall artifact quality in real user workflows. We propose Trace2Eval, a dynamically updatable pipeline that constructs benchmarks from real-world interaction trajectories. Trace2Eval employs a multi-agent reconstruction system to analyze the filtered trajectory, recover the original user query, restore the workspace files to their pre-interaction state, and rebuild a trajectory-equivalent execution environment. A candidate agent then executes the reconstructed query to produce an initial artifact. To evaluate these artifacts, the pipeline iteratively refines human rubrics through inter-annotator alignment, and calibrates judge agents against human assessments using source-code and runtime evidence. Guided by feedback from the Trace2Eval pipeline, models revise their initial artifacts, yielding a mean score gain of 10.05 percentage points across 19 model-harness configurations and a maximum of 19.28 percentage points for DeepSeek-V4-Pro with Claude Code, which supports rubric feedback as actionable guidance for artifact improvement.

cs.SE

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2507.06770

Quantum Information Transmission Over Quantum Relay Channels

Yigal Ilin, Uzi Pereg

原文摘要

Quantum relays are central to both quantum communication and distributed quantum computing, enabling long-distance transmission and modular architectures. Unlike classical repeaters, quantum repeaters preserve coherence without amplifying quantum information, relying on entanglement swapping and quantum error correction to overcome loss and decoherence. In this work, we investigate the transmission of quantum information via quantum relay channels. Our three-terminal relay model captures the trade-off between repeater-assisted and repeaterless communication strategies. Specifically, we propose a partial decode-forward strategy, in which quantum ``message system" consists of two components. The first component is decoded by the relay and then sent to the destination receiver, whereas the second component is decoded by the destination receiver without the relay's help. We analyze both entanglement-assisted and unassisted scenarios. As a special case, the full decode-forward strategy is recovered, with the relay decoding, re-encoding, and forwarding the entire message. Our framework allows for different entanglement topologies between the transmitter, the relay and the destination receiver, recovering known results on entanglement-assisted and unassisted communication. Furthermore, we discuss the interpretation of coding with quantum side information. These findings provide a foundation for designing secure, efficient, and reliable quantum networks and for realizing practical quantum repeaters and long-range quantum key distribution.

quant-ph · cs.IT

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2512.10485

From Lab to Reality: A Practical Evaluation of Deep Learning Models and LLMs for Vulnerability Detection

Chaomeng Lu, Bert Lagaisse

原文摘要

Vulnerability detection methods based on deep learning (DL) have shown strong performance on benchmark datasets, yet their real-world effectiveness remains underexplored. Recent work suggests that graph neural network-based and transformer-based models, including large language models (LLMs), yield promising results when evaluated on curated benchmark datasets. These datasets are typically characterized by similar data distributions and may contain synthetic samples, heuristic labels, or labeling noise. In this study, we systematically evaluate four representative DL models---Devign, ReVeal, LineVul, and VulBERTa---across four representative datasets: Juliet, Devign, BigVul, and ICVul. Each model is trained independently on each dataset, and the graph-based and CodeBERT representations adopted by these models are analyzed using t-SNE and centroid distance to examine vulnerability-related patterns. To assess realistic applicability, we further evaluate trained ReVeal and LineVul models, along with four open-weight LLMs, on VentiVul, our newly constructed temporally separated out-of-distribution (OOD) dataset comprising 200 recent vulnerabilities from Linux and Chromium. Our experiments reveal that current representation methods struggle to distinguish vulnerable from non-vulnerable code and that trained models generalize poorly across datasets with differing distributions and characteristics. When evaluated on VentiVul, performance drops sharply, with most models failing to detect vulnerabilities reliably or distinguish vulnerable functions from their patched counterparts. These results expose a persistent gap between academic benchmarks and real-world deployment, emphasizing the value of our deployment-oriented evaluation framework and the need for more robust code representations, higher-quality datasets, and evaluation methods that account for vulnerability-fixing changes.

cs.CR · cs.LG · cs.SE

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06504

NaijaFacts: A Framework for Disentangling Language from Knowledge in Facts about Nigeria

Zainab Afolabi, Nikita Kozodoi, Jack Butler, Babs Khalidson

原文摘要

Multilingual benchmarks for African languages mostly translate generic questions into local languages, measuring linguistic competence but not whether models know facts about Africa. We introduce \naija, a benchmark of factual and statistical questions about Nigeria, authored by native speakers and grounded in authoritative official statistics, with every question available in English, Nigerian Pidgin, \yoruba, Igbo, and Hausa. Answering similar questions for the UK and US yields a country by language matrix that disentangles a language effect (the model knows the fact but degrades in a low-resource language) from a knowledge effect (the fact is absent in every language). Across eight frontier and Nigeria-adapted models, the knowledge effect dominates: English-query accuracy is 49--50\% for the US/UK but 20\% for Nigeria. The language effect is smaller and tracks web representation, degrading from English to \yoruba{} (9 points on Nigeria, up to 24 on US/UK questions). We find scale does not help: the strongest models reach 66--79\% on the UK and the US but none exceed 31\% on Nigerian statistics. Extending to all 54 African countries with our best performing frontier model shows the same profile continent-wide, 96\% on culture and 98\% on politics against 26\% on statistics, so the gap is not specific to Nigeria.

cs.CL · cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v5arXiv:2505.07681

Verified Purely Functional Catenable Real-Time Deques

Jules Viennot, Arthur Wendling, Armaël Guéneau, François Pottier

原文摘要

We present OCaml and Rocq implementations of Kaplan and Tarjan's purely functional, real-time catenable deques. The correctness of our Rocq code is machine-checked.

cs.PL · cs.DS · cs.LO

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06976

On the Opportunities and Challenges of Prefill-to-Autoregressive Cross-Model Transfer

Zhenyi Shen, Junru Lu, Wei Liu, Jiangnan Ye, Jiazheng Li, Di Yin, Xing Sun, Tameem Adel 等 11 位作者

原文摘要

Prefill-to-Autoregressive Cross-model Transfer (PACT), a form of latent communication between language models, combines a larger model's prompt representations with a smaller model's efficient generation, offering a promising way to balance capability and inference cost. However, existing systems provide limited support for independently evaluating large-small model collaboration, and improved task performance alone does not establish whether the receiver benefits from the sender's information. We introduce PLatour, an open-source PACT framework that connects frozen models through lightweight, zero-initialized linear bridges trained directly with task rewards and supports hybrid-attention architectures. PLatour significantly outperforms open-source baselines on generation tasks and surpasses Mostik's reported results on shared mathematics and coding benchmarks using a substantially smaller sender. However, controlled experiments show that the gains on some tasks can be reproduced by uninformative sender signals, self-communication, or receiver-only training. We find that trainable bridges can act as parameter-efficient adaptation mechanisms without exploiting useful sender information. We therefore propose PACT-Check to quantify how much of a trained PACT system's gain genuinely comes from communication. Finally, we discuss the opportunities and challenges of PACT systems. Code is available at https://github.com/zhenyi4/platour.

cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06501

CRESiST: Self-Enhancing Explorative Policy Learning for Simultaneous Speech Translation

Chenyang Le, Pei Zhang, Yuxiang Chen, Bing Han, Derek F. Wong, Baosong Yang, Yanmin Qian

原文摘要

Simultaneous Speech Translation (SimulST) requires balancing translation quality and latency. While Large Language Models (LLMs) excel in offline translation, current LLM-based SimulST systems predominantly rely on prefix-incremental generation. This paradigm suffers from redundant Key-Value cache recomputation and necessitates external Voice Activity Detection to process unbounded streams. To overcome these bottlenecks, we propose an end-to-end interleaved read-write framework built upon an omni-modal LLM. By interleaving speech chunks and text tokens, our architecture achieves full KV cache reuse, natively supporting segmentation-free, infinite-length inference. Furthermore, to learn optimal streaming policies without rigid offline alignments, we introduce a novel Actor-Critic self-enhancement pipeline. An RNN-T Actor dynamically samples read-write trajectories, while an LLM-based Critic evaluates these paths via intrinsic semantic and latency rewards, distilling the optimal policy back to the Actor. Extensive experiments on sentence-level (FLEURS, CoVoST2) and document-level (MuST-C, ACL 60/60) benchmarks demonstrate that our model establishes new state-of-the-art quality-latency Pareto frontiers, substantially outperforming existing cascaded and static-alignment baselines.

cs.CL · cs.SD

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06500

Tools Need Not Speak in Text: Latent Context Compression for Agents

Yuetong Liu, Zhehao Zhou, Yitong Zhang, Haorui Chen, Jia Li

原文摘要

Large language model (LLM) agents typically append raw tool outputs to their context. Over multi-step interactions, long and noisy observations incur substantial context overhead and dilute the agent's attention to task-relevant information. Existing token pruning and summarization methods remain constrained by discrete text representations, limiting the achievable compression ratios. In this work, we challenge the prevailing assumption that agents must consume tool observations as explicit text tokens. To this end, we propose Coca-Cola, a continuous latent context compression model for LLM agents. Coca-Cola compresses tool observations into latent tokens and embeds them into the input embedding space of the agent. We further introduce Action-Preserving Training, which aligns the agent's next-action distributions under full and compressed observations. This trains the compression model to learn action-sufficient latent tokens that retain decision-critical information. Experiments with web-search agents on four QA benchmarks and two model families show that Coca-Cola reduces the agent's token consumption by 70.01%-89.24% and end-to-end runtime by 24.45%-55.97%, while consistently improving task performance. These results suggest that Coca-Cola offers a promising approach for context compression in agents. Our code is available at https://github.com/THU-Agent/Coca-cola.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06499

Energy-Accumulate-Then-Act Backscatter Communications With Retransmission Under Imperfect Feedback Detection

Rui Xu, Yinghui Ye, Guangyue Lu, Liqin Shi, Gan Zheng

原文摘要

In this paper, we propose a threshold-triggered energy-accumulate-then-act backscatter communication system with retransmission under imperfect feedback detection. Under this system, harvested energy is accumulated across transmission blocks, and the backscatter device switches among different operation modes, including energy harvesting, initial transmission, and retransmission, according to its stored energy and decoded feedback. To characterize the resulting dynamics, the joint evolution of the packet-buffer and stored-energy states is modeled, based on which an embedded Markov chain is constructed to derive the steady-state energy distribution at the beginning of each update cycle. Subsequently, the average age of information (AAoI) is derived from the statistics of the inter-update interval and its correlation with the preceding packet system time, while the packet loss rate (PLR) caused by imperfect feedback detection is also obtained. Simulation results validate the analytical results and reveal that retransmission reduces PLR but may either improve or degrade AAoI, depending on packet-generation energy, data transmission rate, and feedback power. Moreover, lower feedback power shrinks the AAoI-favorable region of retransmission at low transmission rate, whereas the opposite trend emerges at higher rate.

cs.IT

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06498

Signed but Steerable: Measuring and Closing the Text-Vector Integrity Gap in Offline Retrieval-Augmented Generation

Georgios Feretzakis, Alexandros Papaspyridis

原文摘要

Offline retrieval-augmented generation (RAG) on edge devices is updated through signed knowledge packs that carry text and precomputed embedding vectors. A signature proves who published a pack, not that its vectors encode its text. We study this text-vector integrity gap. A verifier that recomputes vectors and accepts them within a cosine tolerance concedes a predictable share of queries to an attacker who controls the vectors: the hijack rate equals the distribution of angular retrieval margins at the tolerance (the margin law). Margins are small in five corpora and two embedding models, so an idealised tolerance at the level of floating-point noise leaves up to 5.5% of queries exposed (at most 1.2% in any corpus once exact duplicate documents are excluded), and a threshold calibrated from our CPU INT8-FP32 comparison of a small embedding model (at the empirical 0.1% quantile of honest cosines) leaves 98-100% exposed; on a Raspberry Pi 5 running a local language model, such an attack lowers correct answers from 0.75 to 0.39. Integrity audits of compressed vector stores fare no better: on a FAISS product-quantisation index over 57,638 documents, tampered vectors that all pass the audit put the attacker's document first for 25% of targets. Verifying quantisation codes instead, with ambiguity decided by the verifier, a calibrated guard band and canonicalisation, accepted every honest vector in our samples on three platforms and left the publisher no choice. We also show that common floating-point and dynamic-INT8 inference pipelines are not bit-reproducible across batch sizes or architectures, and report pre-registered negative results on abstention, context pruning and semantic screening. We release code and data.

cs.CR · cs.IR · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2605.30202

A Dual-Path Architecture for Scaling Compute and Capacity in LLMs

Markus Frey, Behzad Shomali, Joachim Koehler, Mehdi Ali

原文摘要

Looped transformers apply a shared block multiple times and have emerged as a parameter-efficient route to scaling compute in language models. However, at fixed FLOPs a looped model has strictly less capacity than a baseline transformer. We argue that compute, the number of sequential operations applied to a hidden state, and capacity, the parameters available at a single step, are distinct scaling axes that standard transformer layers conflate. We propose a dual-path block that exposes both axes as parallel pathways within a single layer: a deep sublayer re-applied $K$ times with shared parameters, and a wide sublayer with an enlarged feed-forward network applied once, combined by independent per-token gates. At two FLOP budgets, the best dual-path configuration surpasses both iso-FLOP single-axis controls on aggregate language modelling, commonsense, and math metrics, while using 17-33% fewer parameters than the width-scaled baseline. The learned gates are directly interpretable and show systematic per-token allocation: function words and lexical content trend wide, while punctuation, symbols, and arithmetic tokens trend deep. Our results suggest that compute and capacity can be scaled separately, with the model balancing the two token by token.

cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06496

LibriSpeech-Real-QA: A Benchmark for Streaming Question Answering

Zewen Zhan, Hui Lu, Xueyuan Chen, Huimeng Wang, Yikai Huang, Xixin Wu, Shiyin Kang, Bin Xia 等 9 位作者

原文摘要

Most speech large language models (SLMs) still process speech in an offline manner. They only start responding after they receive the complete speech segment. However, many real-world applications require streaming speech interaction, for which the audio continuously flows in and the machine is required to answer questions based on the so far received audio chunks. For example, a user may ask the system to continuously count how many times a person's name has been mentioned while they are speaking. We observe a lack of benchmarks and evaluation protocols for real-time speech interaction. Motivated by this, we propose LibriSpeech-Real-QA, a dataset and a benchmark for real-time speech interaction. We construct different types of real-time question-answer pairs from LibriSpeech to evaluate whether speech LLMs can identify target words in streaming audio. We evaluate several popular real-time commercial models and find that they still tend to respond only after the speech turn has finished. We demonstrate that we can enable a pretrained SLM to perform this task, where we explore several modeling designs and establish a solid baseline for future research.

cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06494

A $2.81^n$ upper bound on the number of stable matchings

Ebrahim Ghorbani, Matthias Mnich

原文摘要

We study the maximum number ${\rm SM}(n)$ of stable matchings in an instance with $n$ men and $n$ women. Karlin, Oveis Gharan, and Weber (STOC 2018) proved the first single-exponential upper bound; their argument gives ${\rm SM}(n)\le 2^{17n}$ for sufficiently large $n$. Palmer and Pálvölgyi (FOCS 2021) improved this bound to $3.55^n+O(1)$. We prove that ${\rm SM}(n)<2.81^n$. This brings the exponential base closer to $2.28$, the base of the best known lower bound. More precisely, we bound the number of stable matchings in each instance in terms of each person's number of stable partners, where a stable partner is someone matched to that person in at least one stable matching. Our proof assigns each person a weight determined by this number and uses these weights to bound the number of stable matchings by induction.

cs.DS · cs.DM · math.CO

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06491

FPTR: Feasibility-Preserving Transactional Refinement for Deadline-Constrained Joint Beam and Resource Allocation

Minghao Kong, Zirong Liu, Jianhao Wu, Muyao Chen, Jianxin Wu, Rongjie Wang

原文摘要

Short-interval wireless schedulers must return a complete, feasible beam-and-resource allocation before each execution deadline, and a late or invalid output gives no utility in the current control cycle. Searching longer can increase transmitted traffic, but it may also miss the deadline or leave only an infeasible intermediate solution. We propose Feasibility-Preserving Transactional Refinement (FPTR), which always keeps a validated feasible allocation, the incumbent. Each refinement stage builds a separate candidate and leaves the incumbent unchanged. A candidate replaces the incumbent only if it is complete, passes the feasibility validator before its stage cutoff, and strictly improves the objective; otherwise it is discarded. Four refinement stages use this rule: buffer-aware resource selection, compatibility-group (CG) user sharing, beam-mask rebuilding from remaining demand, and joint reassignment of two resources. On 150 fixed synthetic instances, the full pipeline (Full) raises traffic by 3.83% on average over a lightweight beam-first reference and by 1.79% over its Base constructor. In a separate suite with matched inputs, Full exceeds six independent search baselines (ALNS, tabu search, GA, simulated annealing, iterated local search, and GRASP) by 6.68-8.35%, where outputs later than 100 ms receive zero utility. All 750 Full runs in the internal suite returned feasible allocations, and none exceeded 100 ms. Separating mandatory output construction from optional refinement lets spare computation time improve the allocation without endangering the incumbent. The open-source implementation supports independent reproduction.

cs.AI · cs.AR · cs.PF

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2602.22968

Certified Circuits: Stability Guarantees for Mechanistic Circuits

Alaa Anani, Tobias Lorenz, Bernt Schiele, Mario Fritz, Jonas Fischer

原文摘要

Understanding how neural networks arrive at their predictions is essential for debugging, auditing, and deployment. Mechanistic interpretability pursues this goal by identifying circuits--minimal subnetworks responsible for specific behaviors. However, existing circuit discovery methods are brittle: circuits depend strongly on the chosen concept dataset and often fail to transfer out-of-distribution, raising doubts whether they capture the concept or merely dataset-specific artifacts. We introduce Certified Circuits, which provide provable stability guarantees for circuit discovery. Our framework wraps any black-box discovery algorithm with randomized data subsampling to certify that inclusion decisions over circuit components--neurons or edges of the model graph, depending on the base algorithm--are invariant to bounded edit-distance perturbations of the concept dataset. Unstable components are abstained from, yielding circuits that are more compact and more accurate. We validate across three architectures (ResNet, ViT, GPT-2) on vision (ImageNet and four OOD datasets) and language (IOI, IOI-Hard, Greater-Than) tasks. Certified circuits achieve up to 56% higher accuracy and up to 80% fewer components, and remain reliable where baselines degrade. Certified Circuits puts circuit discovery on formal ground by producing mechanistic explanations that are provably stable and better aligned with the target concept. Code: https://github.com/AlaaAnani/certified-circuits.

cs.AI · cs.CV · cs.CY

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06489

Risk-Sensitive Non-Parametric and Kernel Bandits

Devdan Dey, Sujoy Bhore, Avishek Ghosh

原文摘要

We study non-parametric bandits in which arms are evaluated using Cumulative Prospect Theory (CPT) rather than expected reward. We consider continuous action spaces with arm-dependent noise and propose Non-Parametric CPT Bandits ($\mathtt{NPCB}$), an optimistic algorithm that decouples mean estimation from CPT-based decision-making. We prove CPT regret of order $\widetilde{\mathcal O}\bigl(\sqrt{T\,\dim_E(\mathcal F,T^{-2})\,\log N(\mathcal F,T^{-2},\|\cdot\|_\infty)}\bigr)$, where $\dim_E$ is the Eluder dimension of the function class $\mathcal F$, $T$ is the time horizon, and $N$ is its covering number in the sup norm. This matches the order-wise regret of classical non-parametric bandits, showing that CPT risk sensitivity incurs no additional order-wise regret cost. We extend $\mathtt{NPCB}$ to unknown noise CDFs, retaining the known-CDF mean-learning term and adding a CDF-learning cost governed by a distributional Eluder dimension. We also study Kernel-CPT bandits and obtain information-gain-based regret bounds for a broad range of kernels, matching the corresponding risk-neutral kernel rates up to logarithmic factors.

stat.ML · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06487

Measuring the collaboration, not the agent

Xinxiao Li, Eisei Nakahara

原文摘要

Benchmarks for visualization agents score the agent alone; the human only supplies reference answers. For high-dimensional data this misses what matters: the knowledge of which variables count is held by on-site experts, not by the model, so the useful question is how well an analyst and an agent do together. We study one specific form of collaboration, the agent proposes a view, the analyst accepts, rejects, or modifies it, and those judgments are kept for later sessions, and propose five measures for it: complementarity, intervention efficiency, appropriate reliance, knowledge retention, and agency. The first four are computed from interaction logs alone. We test the measures, not people: in a testbed for parallel-coordinates design on public manufacturing data (UCI SECOM), with a simulated analyst whose behavior we control, all four are computable and respond to the controlled differences. Whether they reveal anything about real analysts is the next step; we release the testbed for it.

cs.HC

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06486

Model Merging Does Not Require Uniform LoRA Ranks

Zirong Xie, Zainab Afolabi, Nikita Kozodoi, Jack Butler

原文摘要

Model merging combines separately fine-tuned experts into a single model without retraining. Standard practice trains every expert at the same LoRA rank, which couples the rank choice across experts even though each domain has its own best rank. We show this coupling is a property of the merge method, not of merging. We train 36 experts on three domains (mathematics, instruction following, multilingual) at four ranks and three model scales, then merge with six methods at uniform ranks and at each domain's own best rank. Giving experts different ranks costs averaging and utility-based selection nothing, and gains them up to 1.8 pp of multi-domain accuracy, while the conflict-resolution methods are inconsistent and can lose up to 4 pp. The split has a simple cause. Merging keeps only a fixed share of the pooled spectral components, and methods that choose them by size discard many that would have reduced loss. They also tend to discard the same expert: the instruction adapter has the smallest singular values, so it is dropped almost entirely and IFEval falls by 16 to 25 pp. Choosing components by their effect on loss instead keeps all three experts, and leaves accuracy nearly flat across an 8$\times$ rank range (0.7 pp, against 6.7 pp for TIES at 4B scale). Rank can therefore be chosen per expert whenever the merge operator does not select by magnitude.

cs.LG · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.00180

Learning to Live: A Step Toward Lifelong Embodied Intelligence

Lirong Che, Peidong Wang, Yimeng Su, Jiong Wang, Yang Chen, Wenbo Fu, Zhenyu Huang, Tiancheng Xu 等 14 位作者

原文摘要

Embodied agents are usually organized around individual tasks, yet the physical world keeps changing between tasks and during execution. We introduce LIFE (Lifelong Intelligence from Experience), a framework in which a robot observes continuously, remembers selectively, and turns experience into new capabilities. RoboBeats, a family of lightweight streaming gates, tracks conditions set by the agent and discovers new events, deciding when an observation warrants a response or a memory write. Selected observations enter Plastic Memory, a flat store of images and structured records that keeps the source evidence, so new fields and indexes can be added when later questions need them. When existing tools fall short, Explore-Distill-Mount (EDM) turns solutions found during problem solving into validated, reusable tools and skills. Under matched protocols, RoboBeats substantially improves condition tracking over its base models and, at the same writing budget, retains more useful evidence for memory question answering than uniform sampling. A case study of limited scale, replaying 24 hours of continuous video recorded by a real robot, illustrates how the integrated system builds memory from a day of experience and uses it to answer everyday questions. LIFE takes a step toward robots that become more familiar with their surroundings and more capable through experience: learning not only to complete tasks, but to live in the world.

cs.RO · cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06484

Object-independent evaluation of small-data astronomical image classifiers: architecture, representation, and acquisition shortcuts

Eraldo Pereira Marinho

原文摘要

Reported performance in small astronomical image datasets can reflect object reuse, field-of-view differences, acquisition settings, and catalogue-derived properties as well as astronomical morphology. We designed \dataset{} to measure these effects in a four-class task comprising galaxies, globular clusters, nebulae, and open clusters. The dataset contains 370 catalogue-supported physical objects, each represented by three Digitised Sky Survey FITS bands, and uses a group-constrained split of 222 training, 75 validation, and 73 test objects. We compared a custom CNN--Transformer trained from scratch with ImageNet-pretrained ResNet-50 and ViT-B/16 under an adaptive full-field representation, a common angular field, and two central-region-masked controls, using three fixed seeds. In the one-pass object-independent test, mean balanced accuracy on the adaptive view was $0.773\pm0.014$, $0.874\pm0.054$, and $0.866\pm0.023$ for the custom model, ResNet-50, and ViT-B/16, respectively. No primary architecture contrast rejected its null after Holm correction; ResNet-50 and ViT-B/16 differed by 0.007 balanced-accuracy points (95\% cluster-bootstrap interval $[-0.048,0.062]$). Controls based on acquisition and catalogue metadata reached 0.721--0.735, and central-region-masked adaptive views reached 0.731--0.800. These results demonstrate generalisation to unseen physical objects, but also show that image context, acquisition settings, and catalogue-derived object properties carry substantial class information. In this regime, high accuracy cannot be interpreted as evidence that a classifier learned target morphology, and the comparison does not isolate architecture from pretraining.

astro-ph.IM · cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06483

What Does Multi-Turn Reinforcement Learning Really Improve in LLM Agents?

Kejian Zhu, Zhuoran Jin, Shangqing Tu, Tianyi Men, Yupu Hao, Jiachun Li, Kang Liu, Jun Zhao

原文摘要

While multi-turn agentic reinforcement learning (RL) excels on long-horizon tasks, its performance gains remain highly sensitive to interaction constraints. Evaluating across sampling breadth $K$ and interaction depth $T$ via $Pass@(K,T)$, we show that trajectory-level RL advantages depend on sufficient turn budgets, diminishing significantly when budgets tighten. Besides, agentic RL does not uniformly elevate single-step decision quality; rather, its strength concentrates in error recovery and long-horizon resilience. To mitigate the deployment risks caused by this budget dependence, we provide a preliminary discussion on the benefits and limitations of process supervision signals and introduce BudgetMix RL as an alternative that varies interaction budgets during training. Experiments demonstrate that BudgetMix RL achieves robust performance across diverse execution settings. Project Page: https://rethinking-agentic-rl.github.io/

cs.LG · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06482

Computing Forest Degree Realizations with Minimum Cardinality Domination

Amotz Bar-Noy, Omri Kosary, David Peleg, Dror Rawitz

原文摘要

In the Degree Realization problem with respect to a family $\mathcal{P}$ of graphs the input is a non-increasing sequence $d = (d_1,\ldots,d_n)$ of positive integers, and the goal is to decide whether there exists a simple undirected graph $G \in \mathcal{P}$, whose degrees correspond to $d$, i.e., such that $deg(G) = d$. In this paper we consider the version of Degree Realization in which the realization is required to be a forest (i.e., $\mathcal{P}$ is the family of forests). We consider \emph{optimized} Degree Realization in which the goal is to obtain a realization that minimizes an objective function $f$. That is, the goal is to find a realization $G$ that minimizes $f(G)$ among the realizations of the given input sequence. More specifically, we focus on the following functions: the size of an optimal \emph{vertex cover} and the size of an optimal \emph{dominating set}. We also consider the \emph{total} and \emph{paired} versions of both Min Vertex Cover and Min Dominating Set. We provide characterizations and linear time realization algorithms for all the above-mentioned problems. Since Min Paired Vertex Cover is closely related to Min Edge Dominating Set, our results for Degree Realization with respect to the former extend to the latter.

cs.DS · cs.DM

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06481

ArcRAG: Agentic Compositional Retrieval over Knowledge Graphs

Jyotin Goel, Abhinav Arun, Tejas Prakash Agarwal, Bhaskarjit Sarmah, Stefano Pasquali

原文摘要

Graph-based retrieval-augmented generation (GraphRAG) answers questions by retrieving over a knowledge graph built from a corpus. Yet retrieval strategies are designed around the structure their own construction pipeline produces, and whether they survive a change of construction is rarely tested. We study this cross-construction portability: whether one retrieval procedure, with no construction-specific schema, relation names, or retuning, stays effective when the graph is built differently. We introduce ArcRAG, which induces a query-local graph interface at inference time (the node labels, relation descriptors, and property keys reachable from the entities a question names), plans multi-hop traversals against it, and executes and repairs them as Cypher. We build each of the MuSiQue, HotpotQA, and TriviaQA corpora as a schema-guided graph (typed nodes, a closed relation set, source provenance) and as an open-information-extraction graph (untyped nodes, free-text relations), with the retriever's configuration fixed across them. We compare against a construction-agnostic Direct-Cypher agent, ablate each stage across retry and pass budgets on both graph families, and compare with results reported for prior GraphRAG systems, most of which design retrieval together with their own graph construction. Across the two constructions, we find that retrieval need not be co-designed with the graph: query-time interface induction keeps a construction-agnostic retriever effective when the construction changes, although construction still shapes absolute accuracy.

cs.CL · cs.AI · cs.IR

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06479

Optimal Subcubic Byzantine Agreement with Plain PKI

Marc Dufay, Roger Wattenhofer

原文摘要

We consider Byzantine Agreement (BA), where $n$ parties run a distributed protocol to agree on a common value, despite $t$ parties being byzantine. In the synchronous setting, a choice must be made between the weaker unauthenticated setting, which only allows $t < n/3$ byzantine parties, and the stronger authenticated setting that allows up to $t < n/2$ byzantine parties. For this latter setting, BA can be solved with optimal resiliency and optimal bit complexity. However, so far, all optimal solutions rely on the use of threshold signatures. This shows a gap compared to regular signatures, where only protocols with cubic complexity are known for optimal resiliency $t < n/2$. We wonder whether this cubic complexity is required without the use of threshold signatures and show that this is not the case. More specifically, given $κ$ an upper bound on the size of a hash or signature, we describe a deterministic protocol solving BA with bit complexity $\mathcal{O}(κ^{(\sqrt{13} - 1)/2} n^{(\sqrt{13} + 1)/2}) \approx \mathcal{O}(κ^{1.303}n^{2.303})$ while having optimal resiliency $t<n/2$ and round complexity $\mathcal{O}(n)$.

cs.DC

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06478

Beyond Gomory-Hu: All-Pairs $\times$ All-Pairs Mincuts in $O(n)$ Space

Koustav Bhanja, Merav Parter, Asaf Petruschka

原文摘要

For over four decades, two classical structures have offered complementary views of mincuts. The Gomory-Hu tree encodes one $(x,y)$-mincut for every vertex pair $x,y$ in $O(n)$ space, but retains only one mincut per pair. In contrast, for a fixed pair $x,y$, the Picard-Queyranne representation admits an $O(n)$ space vertex ordering that determines whether some $(x,y)$-mincut separates any given pair $u,v$ and reports such a cut. Can these complementary guarantees hold simultaneously in $O(n)$ space? We answer affirmatively. For undirected graphs, we construct an $O(n)$ space structure encoding, for every four vertices $x,y,u,v$, an $(x,y)$-mincut separating $u,v$ whenever one exists. More precisely, given any terminal pair $x,y$, the structure recovers the Picard-Queyranne ordering in $O(n)$ time. Thereafter, separability queries take $O(1)$ time, and reporting a separating mincut $C$ takes $O(|C|)$ time. Thus, information for all $Θ(n^4)$ choices of $x,y,u,v$ is encoded in $O(n)$ space. This is essentially the strongest information of this form that fits in $O(n)$ space: if one additionally requires the mincut to place $u$ on side of $x$ and $v$ on side of $y$, then $Ω(n^2)$ space is necessary even for fixed $x,y$. As a consequence, we obtain an $O(n)$ space all-pairs mincut sensitivity oracle for a single edge failure in multigraphs, with optimal $O(n)$ time to report a new mincut. This resolves an open question of Baswana and Pandey [SODA'22] on achieving $o(n^2)$ space and $o(m)$ query time simultaneously. Our construction substantially generalizes the $O(n)$ space nearest-mincut-hierarchy [SODA'27], which reports only the nearest $(x,y)$-mincut. Recovering the entire Picard-Queyranne ordering for an arbitrary pair requires new structural properties of the Dinitz-Vainshtein [STOC'94] connectivity carcass, extending its use beyond the Steiner mincuts it represents.

cs.DS

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2507.05751

SenseShift6D: Multimodal RGB-D Benchmarking for Robust 6D Pose Estimation across Environment and Sensor Variations

Yegyu Han, Taegyoon Yoon, Dayeon Woo, Sojeong Kim, Hyung-Sin Kim

原文摘要

Recent advances on 6D object pose estimation have achieved high performance on representative benchmarks such as LM-O, YCB-V, and T-Less. These datasets capture natural-occurring variations in illumination and camera setup but do not systematically explore the effects of controllable sensor configurations, such as exposure, gain, and depth-sensor modes. To bridge this gap, we introduce SenseShift6D, the first RGB-D object-pose benchmark that physically sweeps 13 RGB exposures, 9 RGB gains, auto-exposure, 4 depth-capture modes, and 5 illumination levels. For six common household objects, we acquire 198.8k RGB and 20.0k depth images (i.e., 795.4k RGB-D scenes), providing 1,380 unique sensor-lighting permutations per object pose. Experiments with state-of-the-art pretrained, generalizable pose estimators reveal substantial performance variation across lighting and sensor settings, despite their large-scale pretraining. Strikingly, even instance-level estimators-trained and tested on identical objects and backgrounds-exhibit pronounced sensitivity to environmental and sensor shifts. These findings establish sensor- and environment-aware robustness as an underexplored yet essential dimension of 6D pose estimation. To illustrate the opportunity enabled by this benchmark, we evaluate test-time sensor selection without retraining. An idealized controller yields remarkable gains of up to +16.2 pp for generalizable models, whereas a practical consistency-based proxy provides only limited improvements, highlighting substantial headroom and the need for future research on reliable sensor-aware adaptation. SenseShift6D is available at huggingface.co/datasets/Yegyu/SenseShift6D.

cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06477

Towards Early Warning of Real-World Vulnerabilities: An Evidence-Aware Decoupled Method for CVSS Vector Prediction Based on Unstructured Disclosures

Jianhong Ma, Wenhao Li, Xixing Li, Chenyu Yan, Haowen Chen

原文摘要

Early assessment of Common Vulnerability Scoring System (CVSS) vectors from Open Source Intelligence (OSINT) can reduce the delay between initial public disclosure and official National Vulnerability Database (NVD) scoring. Existing predictors, trained mainly on standardized late-stage descriptions, degrade on noisy and incomplete pre-NVD reports. We introduce PARSE, a vulnerability-disclosure lifecycle model, and VPre, an early assessment framework built on two-level Evidence-Aware Decoupling. Decoupling Level I, Security-Evidence Screening, separates security-relevant evidence from noisy early reports at the report and sentence levels. Decoupling Level II, Evidence-Matched Routing, decouples the prediction task of the eight CVSS metrics into a scattered-evidence class and an explicit-evidence class based on validation-set measurements, matching each with a targeted predictor: large-language-model reasoning for the former, lightweight supervised classification for the latter. On the 851-report held-out test split of the 5,581-report VPre-Bench, VPre achieves a mean absolute error of 0.7239 -- 44.7% below the reproduced BERT-Small-NVD early-transfer baseline; paired timestamps across the full benchmark show an average 38-day lead before official scoring (observed range 21-60 days). Under the 38:1 PARSE-Triage noise setting, STDU's report-level screening reaches 0.98 precision, 1.00 recall, and 0.99 F1. Controlled ablations support the contributions of both decoupling levels.

cs.CR

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2603.03989

Faces in Things, Faces in Nothing: Probing Face Selectivity in Vision Models with Pareidolia

Qianpu Chen, Derya Soydaner, Rob Saunders

原文摘要

An electrical outlet can look like a face without depicting a person. Face pareidolia is thus a revealing probe of how vision models turn ambiguous visual evidence into category judgments. When a model calls such an object Human, does that answer depend on its face-like arrangement? Using the FacesInThings dataset, we test six vision models, from a vision-only ViT to CLIP and generative vision-language models, with face detectors as a reference. Each face-like region is paired with three controls from the same photograph that lack the arrangement: a background crop, the region with its tiles shuffled, and its average color. A Human answer driven by the arrangement should appear less often on these controls. We also vary how the categories are described. Three findings emerge. First, models respond to illusory faces, but not as people do: face detectors rarely return a face for these regions, yet CLIP ViT-L/14 and two generative models label roughly half of them Human when the categories are described as kinds of face. Annotators, by contrast, read these faces as non-human, such as animal or cartoon faces. For these models, a face is a human face by default. Second, similar Human rates hide opposite behavior. LLaVA-1.5 answers Human less often once the arrangement is removed; Qwen2-VL calls every average-color patch Human and describes a person in each; CLIP ViT-B/32 calls shuffled regions Human more often than real faces; SmolVLM rarely answers Human, but mostly for face-like regions. Third, rewording the categories changes whether Human answers favor the face at all, removing this preference in LLaVA-1.5 and creating it in CLIP. A Human answer is not evidence of a face: vision models can keep giving it after the face is gone.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2602.05943

Orthogonal Model Merging

Sihan Yang, Kexuan Shi, Weiyang Liu

原文摘要

Merging finetuned Large Language Models (LLMs) has become increasingly important for integrating diverse capabilities into a single unified model. However, prevailing model merging methods rely on linear arithmetic in Euclidean space, which often destroys the intrinsic geometric properties of pretrained weights, such as hyperspherical energy. To address this, we propose Orthogonal Model Merging (OrthoMerge), a method that performs merging operations on the Riemannian manifold formed by the orthogonal group to preserve the geometric structure of the model's weights. By mapping task-specific orthogonal matrices learned by Orthogonal Finetuning (OFT) to the Lie algebra, OrthoMerge enables a principled yet efficient integration that takes into account both the direction and intensity of adaptations. In addition to directly leveraging orthogonal matrices obtained by OFT, we further extend this approach to general models finetuned with non-OFT methods (i.e., low-rank finetuning, full finetuning) via an Orthogonal-Residual Decoupling strategy. This technique extracts the orthogonal components of expert models by solving the orthogonal Procrustes problem, which are then merged on the manifold of the orthogonal group, while the remaining linear residuals are processed through standard additive merging. Extensive empirical results demonstrate the effectiveness of OrthoMerge in mitigating catastrophic forgetting and maintaining model performance across diverse tasks.

cs.LG

查看论文
arXiv ↗PDF ↗