中 EN
登录 注册账户

Academic Intelligence · Curated Daily

探索全球前沿学术脉络

AcademicHub 汇聚顶级期刊与预印本平台的实时文献。定制您的专属科研雷达，利用大语言模型自动生成交叉领域文献分析简报。

登录 注册账户
01.
arXiv (quant-ph) 2026-06-16

Quantum Measurement and Continuous Markov Processes

作者:
Chris Jackson↗

arXiv:2606.15958v1 Announce Type: new Abstract: These are the lecture notes for a course on diffusive quantum measuring instruments. They were prepared and delivered at the Perimeter Institute on Mondays and Thursdays, from 2:30 to 4:00 PM, beginning October 27th, 2025 and ending December 11th, 2025. These lectures were recorded and can be found at https://pirsa.org/c25038.

阅读与讨论 → 访问原文 →
02.
arXiv (CS.CV) 2026-06-11

Semantic search for 100M+ galaxy images using AI-generated captions

作者:
Nolan Koblischke↗Liam Parker↗Francois Lanusse↗Jo Bovy↗Irina Espejo↗Shirley Ho↗

Finding scientifically interesting phenomena through slow manual labeling campaigns severely limits our ability to explore the billions of galaxy images produced by telescopes. In this work, we develop a pipeline to create a semantic search engine from completely unlabeled image data. Our method leverages Vision-Language Models (VLMs) to generate descriptions for galaxy images, then contrastively aligns a pre-trained astronomy foundation model with these embedded descriptions to produce searchable embeddings at scale. We find that current VLMs provide descriptions that are sufficiently informative to train a semantic search model that outperforms direct image similarity search. Our model, AION-Search, achieves state-of-the-art zero-shot performance on finding rare phenomena despite training on randomly selected images with no deliberate curation for rare cases. Furthermore, we introduce a VLM-based re-ranking method that nearly doubles the recall for our most challenging targets in the top-100 results. For the first time, AION-Search enables flexible semantic search for over 100 million galaxy images, enabling discovery from previously infeasible searches, including the identification of 36 new extragalactic stellar stream candidates. More broadly, our work provides an approach for making large, unlabeled scientific image archives semantically searchable, expanding data exploration capabilities in fields from Earth observation to microscopy. The code, data, and app are publicly available at https://github.com/NolanKoblischke/AION-Search

阅读与讨论 → 访问原文 →
03.
arXiv (CS.AI) 2026-06-11

Agentic Software: How AI Agents Are Restructuring the Software Paradigm

作者:
Zhenfeng Cao↗

arXiv:2606.05608v2 Announce Type: replace-cross Abstract: For over half a century, software engineering has operated on a foundational premise: human engineers decompose problems, encode decision logic into static code, and manually adapt that code as requirements evolve. This paper argues that the emergence of AI agents – systems where large language models serve as the primary reasoning engine, dynamically generating and discarding code as an instrumental resource – constitutes a fundamental restructuring of what software is, not an incremental tool improvement. We formalize the distinction between traditional deterministic software and agentic software: in the former, code is the carrier of pre-written decision logic; in the latter, the agent itself is the software, and its decision logic is generated at runtime. We trace the historical arc from licensed software to SaaS to Agent-as-a-Service (AaaS), showing that each shift transferred additional complexity away from end-users – with the agentic shift transferring not just operational complexity but decision-making complexity itself. We introduce Agentic Engineering as an expansion of the software engineering discipline into a new paradigm, distinct in its core object of study (agent systems rather than static source code), its control model (LLM-driven rather than human-predefined), and its human role (intent architect rather than code author). Through analysis of recent benchmark evidence including SWE-bench Verified, EvoClaw, and LangChain's multi-agent coordination studies, we demonstrate both the transformative potential of the agentic paradigm and its current limitations. We conclude with a four-stage roadmap toward self-evolving agent ecosystems and concrete recommendations for practitioners navigating this transition.

阅读与讨论 → 访问原文 →
04.
arXiv (CS.AI) 2026-06-15

Regulating the Machine Contributor: Governance and Policy Alignment in Open Source

作者:
Jassem Manita↗Aziz Amari↗

arXiv:2606.14594v1 Announce Type: cross Abstract: AI-assisted software development has moved from line-level autocomplete to agents that can plan changes, edit files, and submit pull requests with limited human supervision. Open-source software, however, evolves through a process designed for humans: contributor agreements, codes of conduct, and review norms all assume a legally accountable person who can attest to provenance and answer reviewer questions. Autonomous and semi-autonomous AI contributors strain those assumptions, and the 2025-2026 record of agent-driven incidents, AI-generated nuisance volume, and platform-level shutdowns shows that the gap is operationally consequential. Several open-source organisations have responded with contribution policies, but the result is fragmented, and its alignment with emerging AI governance frameworks (EU AI Act, NIST AI RMF with the UC Berkeley Agentic AI Profile, ISO/IEC 42001 and 23894) is unmapped at the contribution level. We compare policies across six organisations (SymPy, LLVM, matplotlib, OpenInfra, the Apache Software Foundation, and the Linux Foundation) using Most-Similar Systems Design with indicator-based coding and process tracing for SymPy and LLVM. From this we derive a six-dimensional taxonomy (disclosure, responsibility, human oversight, licensing, enforcement, maintainer workload), an ordinal Policy Maturity Score, and a mapping of documented agent incidents onto the dimensions each policy fails to govern. Aligning the dimensions with the regulatory frameworks above identifies overlapping gaps neither side currently closes, and we close by sketching the shape of a harmonised tiered framework and the empirical evaluation needed to calibrate it.

阅读与讨论 → 访问原文 →
05.
arXiv (CS.CL) 2026-06-11

Fine-tuning Multi-modal LLMs with ART: Art-based Reinforcement Training

作者:
Michal Chudoba↗Sergey Alyaev↗Petra Galuscakova↗Tomasz Wiktorski↗

There are two main Parameter-Efficient Fine-Tuning (PEFT) techniques for Large Language Models (LLMs). While Low-Rank Adaptation (LoRA) introduces additional weights between the LLM layers, Soft Prompting introduces additional fine-tuning-specific raw tokens to an LLM input. However, both require modification to the computational graphs of precompiled, preoptimized LLMs. As a result, neither is fully supported in high-throughput engines like vLLM. We propose fine-tuning with ART (Art-based Reinforcement Training). The method injects information into a frozen Multimodal Large Language Model (MLLM) by optimizing only its raw visual input, thus enabling the soft-token approach on pre-compiled computational graphs. It relies on backpropagation of gradients back into a plain pixel array and thus supports any fine-tuning objective. Moreover, the optimized visual input can be stylized as task-relevant computational artworks. The approach's effectiveness is confirmed for different sizes of a popular open Qwen architecture and for several textual benchmarks. Specifically, ART reaches accuracy competitive with LoRA across mathematics and structured-tool-use benchmarks.

阅读与讨论 → 访问原文 →
06.
arXiv (CS.AI) 2026-06-24

From "Aha Moments" to Controllable Thinking: Toward Meta-Cognitive Reasoning in Large Reasoning Models via Decoupled Reasoning and Control

作者:
Rui Ha↗Rui Pu↗Chaozhuo Li↗Li Sun↗Sen Su↗

arXiv:2508.04460v2 Announce Type: replace Abstract: Large Reasoning Models (LRMs) can exhibit step-by-step reasoning, reflection, and backtracking, but these behaviors are often unregulated, leading to overthinking. As a result, LRMs continue generating redundant reasoning even after reaching high-confidence conclusions. This increases inference cost and latency, limiting practical deployment. The root cause is the absence of an intrinsic mechanism to monitor the reasoning state and decide when to continue, backtrack, or stop. We propose MERA, a meta-cognitive reasoning framework that decouples reasoning from control to enable independent optimization of control strategies. MERA constructs high-quality reasoning-control supervision data via a takeover-based pipeline, and transforms long-horizon traces into structured reasoning-control alternating sequences for training. The model is trained with supervised fine-tuning to internalize the structured separation, and further optimized with Control-Segment Policy Optimization (CSPO), which combines segment-wise GRPO with control masking to focus learning on control segments. Experiments across reasoning benchmarks show that MERA improves both efficiency and accuracy.

阅读与讨论 → 访问原文 →
07.
Nature (Science) 2026-06-23

Silicon Valley’s vision for global AI is flawed: each country needs its own blueprint

作者:
Pelonomi Moiloa↗

From energy grids to language performance, emerging economies are exposing the limits of today’s artificial-intelligence strategy as it expands globally. From energy grids to language performance, emerging economies are exposing the limits of today’s artificial-intelligence strategy as it expands globally.

阅读与讨论 → 访问原文 →
08.
arXiv (CS.CL) 2026-06-25

Detect, Unlearn, Restore: Defending Text Summarization Models Against Data Poisoning

作者:
Poojitha Thota↗Shirin Nilizadeh↗

Training-time data poisoning during fine-tuning poses a significant threat to large language models (LLMs) deployed for abstractive text summarization, where small task-specific datasets exert disproportionate influence on model behavior. In this setting, adversaries manipulate fine-tuning data to induce persistent summarization failures, such as biased or harmful summaries, while preserving standard evaluation metrics. We present a unified post-hoc defense framework for detecting and remediating fine-tuning-stage poisoning in summarization models across the machine learning supply chain. Our experiments show that in white-box settings, poisoned document-summary pairs exhibit abnormally high training influence, enabling detection via influence-function analysis with semantic consistency checks. In black-box settings, poisoned models display two to three times greater sensitivity to semantics-preserving perturbations, enabling behavioral auditing without training data access. Beyond existing poisoning formulations, we introduce novel attacks targeting factual distortion and representational bias, showing that poisoning alters summarization behavior without triggering conventional alarms. Across nine architectures and six benchmark datasets under adaptive attacks, our defenses achieve 85-92% detection precision, while gradient-ascent unlearning restores up to 96% of original behavior with minimal utility loss (less than 0.6% ROUGE degradation). These results indicate that fine-tuning-time poisoning leaves persistent structural artifacts, enabling practical detection and post-deployment recovery without full retraining.

阅读与讨论 → 访问原文 →
09.
arXiv (CS.LG) 2026-06-16

Latent space mapping of interpretable structural coordinates from stochastic single-molecule signals

作者:
Matteo Cartiglia↗Sandro Kuppel↗Wouter Botermans Wannes Peeters↗Natan Biesmans↗Liam Vandekerckhove↗Eric Beamish↗Koen Ongena↗Wouter Renckens↗Pol Van Dorpe↗Sanjin Marion↗

arXiv:2606.16950v1 Announce Type: cross Abstract: Nanopores are versatile single-molecular sensors, but their utility is fundamentally constrained by stochastic translocation dynamics warping any encoded information. We resolve it by shifting from time-domain analysis to a learned latent-space mapping via a contrastive encoder trained exclusively on simulated signals from a physics-informed model. This encoder maps solid-state nanopore signals of engineered DNA barcodes into an interpretable molecular coordinate system. The learned representation is responsive to structural barcode parameters while remaining invariant to acquisition conditions and translocation conformation, allowing data pooling across devices. Molecule identification requires a single pass through the encoder, reducing computational cost by three orders of magnitude relative to alignment-based methods. We experimentally validate through mixture quantification, rare-variant detection, consensus barcode reconstruction, and real-time signal acquisition. This shift from temporal analysis to mapping structural coordinates into a latent space changes the paradigm behind analyzing stochastic sensor signals by linking classification to interpretable encoded molecular information.

阅读与讨论 → 访问原文 →
10.
Nature (Science) 2026-06-17

Optical fibre gripper for high-performance 3D micromanipulation

作者:
Deng Pan↗

Optical tweezers offer precise, non-contact control, but operate in a limited force regime and impose strict requirements on the characteristics of the targets as well as the environmental conditions1–4. Millimetre-scale mechanical tweezers can offer higher gripping force but are not suitable for precise manipulations5–11. Integrating microgrippers directly at the optical fibres provides a new approach for precise micromanipulation. However, existing fibre-integrated tweezers still face challenges in achieving high-performance manipulation of micro-objects (for example, single cells) within narrow spaces, mainly due to simplified architectures, constrained designs and millimetre-scale footprints12–14. Here we report a three-dimensional (3D) optical fibre gripper (OFG), which is fabricated by two-step, two-photon polymerization. The OFG consists of rigid photoresist microclaws and soft thermoresponsive hydrogel muscle doped with silver nanoparticles, and its size is only 38 × 38 × 61 μm3. The OFG exhibits a force-to-mass ratio of about 340 μN mg−1, outperforming previously reported fibre-integrated tweezers by one to two orders of magnitude. The OFG can manipulate opaque particles, irregular micromechanical components and diverse single-cell types. We further demonstrated its potential in 3D microassembly of complex microdevices (bearings, shafts and gearboxes) and biomimetic sampling in the narrow environment (<300 μm). These results position the OFG as a compact fibre-tip manipulator for 3D micromanipulation, offering reversible and tunable gripping in an intermediate force regime between optical field trapping and millimetre-scale mechanical tweezers. A miniature three-dimensional optical fibre gripper enables powerful, precise micromanipulation of particles and single cells in confined spaces, bridging the gap between optical and mechanical tweezers.

阅读与讨论 → 访问原文 →
11.
arXiv (math.PR) 2026-06-11

Marked random graphs with given degree sequence: large deviations on the local topology

作者:
Rangel Baldasso↗Alan Pereira↗Guilherme Reis↗

arXiv:2401.00351v2 Announce Type: replace Abstract: We investigate the behavior of the empirical neighborhood distribution of marked graphs in the framework of local weak convergence. Here we extend known results by considering uniform random graphs with given degree sequences and i.i.d. marks on half-edges and vertices. We establish a large deviation principle for such families of empirical measures. The proof builds on Bordenave and Caputo's seminal 2015 paper, and Delgosha and Anantharam's 2019 introduction of BC entropy, relying on combinatorial lemmas that allow one to construct suitable approximations of measures supported on marked trees. Possible applications of these results are in the study of interacting diffusions on top of random graphs.

阅读与讨论 → 访问原文 →
12.
arXiv (CS.CL) 2026-06-11

AI Coding Agents in Social Science: Methodologically Diverse, Empirically Consistent, Interpretively Vulnerable

作者:
Meysam Alizadeh↗Fabrizio Gilardi↗Mohsen Mosleh↗Enkelejda Kasneci↗

The deployment of LLM-based agents in scientific analysis raises opposing concerns: that agents may reduce methodological diversity, or that they may amplify the analytic flexibility through which researchers reach motivated conclusions. We argue these worries target two empirically separable layers: a design layer of methodological choices, and a verdict layer in which a decision rule maps estimates to a substantive claim. We test both by running 20 independent executions of Claude Code and Codex on a prominent immigration and social-policy against a many-analysts human baseline. At the design layer, Codex matches human methodological diversity and Claude Code produces nearly three times as many specifications; both agents' effect estimates remain broadly aligned with the human consensus, and no agent model exactly matches any human model. A prompt-induced anti-immigration researcher prior reorganizes each agent's methodological decisions but, unlike for biased human analysts in the same data, does not shift aggregate estimates or final verdicts; nor do agents reroute along the methodological axes humans use to bias their estimates. At the verdict layer, an explicit confirmatory prompt flips Claude Code's verdicts from 10% to 90% support while leaving its coefficient distribution essentially unchanged, operating through rule omission rather than rule softening. AI agents can rival or exceed human methodological diversity at the design layer while remaining vulnerable at the verdict layer. In our setting, the locus of AI bias is not estimation but interpretation.

阅读与讨论 → 访问原文 →
13.
arXiv (CS.CV) 2026-06-15

High-Fidelity Video Compression based on Invertible Neural Transform and Implicit Conditioning

作者:
Siyue Teng↗Ho Man Kwan↗Yuxuan Jiang↗Fan Zhang↗David Bull↗

Learning-based video compression has recently achieved competitive rate-distortion performance compared to conventional video codecs. However, most existing methods rely on non-invertible analysis-synthesis transforms, with reconstruction quality subject to both quantization and transform approximation errors. This limitation becomes particularly restrictive at higher quality points, where quantization errors are small and transform-induced distortion dominates. To address this, we propose InnVC, an Invertible neural network based Video Codec for wide-range and high-fidelity compression. The core idea is to preserve an invertible main transform path prior to quantization, while injecting content-adaptive context through a compact implicit conditioning field. This decouples strongly correlated video content from harder-to-model fine details, allowing different components to specialize in complementary reconstruction tasks for more efficient compression. To further improve compressibility, we introduce a scheduled masking strategy that progressively concentrates informative content into fewer latent channels for more effective entropy coding. Experiments on the UVG and MCL-JCV benchmarks show that InnVC achieves strong compression performance over a broad quality range, being particularly effective in the high-quality regime, yielding BD-rate reductions of 21.66% in PSNR and 46.06% in MS-SSIM relative to x265 on UVG. To the best of our knowledge, InnVC is the first neural video codec covers operating poins from low bitrate to high fidelity within a single architecture scale, spanning more than 20 dB in PSNR.

阅读与讨论 → 访问原文 →
14.
arXiv (CS.AI) 2026-06-24

The impact of generative artificial intelligence on academic development of Chinese students in humanities and social sciences

作者:
Lei Fan↗Fangxue Liu↗

arXiv:2606.24104v1 Announce Type: cross Abstract: Generative artificial intelligence(GenAI) is reshaping learning in higher education, with particularly pronounced implications for the humanities and social sciences(HSS), where learning outcomes are commonly expressed through written and interpretive forms that align closely with GenAI's capabilities. Yet, systematic evidence on the educational impacts of GenAI on HSS students remains limited. Addressing this gap, this study draws on a large-scale survey of HSS students in China to examine its role in academic development. Guided by relevant learning theories, this study focuses on four dimensions: patterns of use, effects on learning processes and academic performance, challenges associated with GenAI use, and preferred approaches to curricular integration. We found that more than half perceived enhanced learning motivation, independent thinking and creativity, although a substantial minority reported little change or even decline. Comparatively, a notably larger majority reported academic performance gains, although these gains may partly reflect limitations in conventional assessment practices. The study identifies variations in perceived learning and performance improvements among students with differing durations of GenAI experience, along with observable disciplinary differences and modest gender differences. While an overwhelming majority valued the importance of ethical considerations, only slightly more than half were satisfied with privacy protection. Limited accuracy and overreliance emerged as the most pressing concerns reported by students. Students favored partial or optional curricular integration supported by practice-oriented training, and widely recognized GenAI's significance for their future professional development. Grounded in student perspectives, this study offers evidence-based recommendations for the responsible and pedagogically meaningful integration of GenAI

阅读与讨论 → 访问原文 →
15.
arXiv (CS.CV) 2026-06-16

MapDream: Task-Driven Map Learning for Vision-Language Navigation

作者:
Guoxin Lian↗Shuo Wang↗Yucheng Wang↗Yongcai Wang↗Maiyue Chen↗Kaihui Wang↗Bo Zhang↗Zhizhong Su↗Deying Li↗Zhaoxin Fan↗

Vision-Language Navigation (VLN) requires agents to follow natural language instructions in partially observed 3D environments, motivating map representations that aggregate spatial context beyond local perception. However, most existing approaches rely on hand-crafted maps constructed independently of the navigation policy. We argue that maps should instead be learned representations shaped directly by navigation objectives rather than exhaustive reconstructions. Based on this insight, we propose MapDream, a map-in-the-loop framework that formulates map construction as autoregressive bird's-eye-view (BEV) image synthesis. The framework jointly learns map generation and action prediction, distilling environmental context into a compact three-channel BEV map that preserves only navigation-critical affordances. Supervised pre-training bootstraps a reliable mapping-to-control interface, while the autoregressive design enables end-to-end joint optimization through reinforcement fine-tuning. Experiments on R2R-CE and RxR-CE achieve state-of-the-art monocular performance, validating task-driven generative map learning.

阅读与讨论 → 访问原文 →
16.
Nature (Science) 2026-06-24

Genetic technologies to enhance crop nutritional value under climate change

作者:
Dominique Van Der Straeten↗

At present, more than 700 million people live with caloric hunger, and more than two billion suffer from micronutrient deficiencies, known as ‘hidden hunger’. From an agricultural viewpoint, three major objectives need to be worked towards simultaneously to achieve zero hunger (the United Nations Sustainable Development Goal 2): (1) enhanced yield; (2) higher vitamin and mineral density to sustain recommended daily intake (multi-biofortification); and (3) enhanced climate-change resilience. Although the Green Revolution increased global calorie production, it exacerbated hidden hunger by prioritizing high yield over nutritional quality. Stress from global climate change has been shown to reduce the densities of several micronutrients. CRISPR–Cas, which allows genome editing with extremely high precision, has emerged as a groundbreaking breeding technology that has already been adopted by many countries. Here we examine how CRISPR–Cas-based approaches could be used to achieve biofortification targets by enhancing micronutrient densities to the levels necessary to alleviate dietary vitamin and mineral deficiencies. Given the limited time frame available to achieve zero hunger, we argue that CRISPR–Cas technologies should be combined with metabolic engineering based on transformation and other technologies. We also consider untapped resources beyond metabolic pathways and current CRISPR–Cas methodologies to address one of the most important societal issues of the twenty-first century. This Review reflects on the joint power of genetic technologies, including untapped CRISPR–Cas techniques to combat hidden hunger and improve crop resilience, and argues in favour of their combined use to overcome these societal challenges.

阅读与讨论 → 访问原文 →
17.
arXiv (CS.CL) 2026-06-24

EXPO-SQL: Execution-based Clause-level Policy Optimization for Text-to-SQL

作者:
Jaehoon Lee↗CheolWon Na↗Suyoung Bae↗Jin-Seop Lee↗Jihyung Lee↗YunSeok Choi↗Jee-Hyong Lee↗

Text-to-SQL enables users to query databases using natural language by generating executable SQL queries. Recent methods have increasingly adopted Large Language Models based reinforcement learning (RL) to leverage execution feedback for training. However, existing RL methods assign uniform query-level rewards to all clauses in a SQL query, treating correct and incorrect clauses equally. This coarse-grained reward design leads to insufficient learning signals for correct SQL generation. To address this issue, we propose EXPO-SQL (EXecution-based clause-level Policy Optimization for Text-to-SQL) which provides fine-grained supervision through clause-level rewards. To assign clause-level rewards, our method identifies erroneous clauses by analyzing execution results, including error messages and clause-wise incremental execution. Experiments on widely-used Text-to-SQL benchmarks demonstrate that EXPO-SQL significantly outperforms existing supervised fine-tuning, prompting, and RL-based methods through fine-grained clause-level learning. Our code is available at https://github. com/jhn25/EXPO-SQL.

阅读与讨论 → 访问原文 →
18.
arXiv (CS.CL) 2026-06-11

A Resource for Enthymeme Detection in Controversial Political Discourse

作者:
Martial Pastor↗Nelleke Oostdijk↗

Enthymemes, arguments with unstated premises or conclusions, are pervasive in persuasive discourse, yet their annotation remains notoriously subjective. We present a resource of 1,482 tweets from politically controversial discourse, annotated by five annotators for the presence of enthymemes and their argument structure, designed to study label variation. We first revisit the definition of enthymemes and propose annotation guidelines anchored in Walton's argumentation schemes, offering a structured and constrained approach that nonetheless preserves room for the interpretive nature of the task. This contrasts with past resources, which tend to eliminate disagreement, obscuring its sources and preventing investigation of its potential benefits for model performance. We further propose a complexity analysis of the task, identifying where annotation imposes high cognitive load and may give rise to inconsistent annotation. Our preliminary experiments show that models trained on annotator disagreement outperform models trained on hard majority-vote labels. We close by reflecting on how structural openness in enthymeme definitions and guidelines enables the study of variation in subjective inferential processes for future resources and downstream NLP applications concerned with human inference.

阅读与讨论 → 访问原文 →
19.
arXiv (CS.AI) 2026-06-19

Calibration Without Comprehension: Diagnosing the Limits of Fine-Tuning LLMs for Vulnerability Detection in Systems Software

作者:
Arastoo Zibaeirad↗Marco Vieira↗

arXiv:2606.20502v1 Announce Type: cross Abstract: Whether LLMs scoring well on vulnerability benchmarks genuinely reason about security or merely pattern-match on contaminated data remains unresolved. We present CWE-Trace, a framework for LLM vulnerability detection built from 834 manually curated Linux kernel samples spanning 74 CWEs. The framework enforces a strict temporal split (pre-2025 historical set / post-cutoff leakage-free set), preserves context-aware vulnerable–patched pairs, and introduces two diagnostic metrics: the Directional Failure Index (DFI) and Hierarchical Distance and Direction (HDD). We evaluate eight vanilla LLMs and 15 LoRA fine-tuned variants across non-targeted detection, targeted detection, and CWE classification. Our analysis yields two key results. First, data contamination provides no measurable advantage. Function-level analysis shows that 84% of nominally contaminated samples carry no usable memorization signal: vulnerable functions are absent or cross-mapped across datasets, and ~31% of contaminated samples carry CWE misclassification. Second, backbone directional priors dominate fine-tuning. Models exhibit stable, systematic failure modes (DFI ranging from -85.5 to +94.8 pp) that persist from historical to post-cutoff data and resist correction. Fine-tuning shifts the output threshold without changing the decision policy. This is calibration without comprehension: output distributions adapt to training data while the underlying security reasoning remains absent. The weakest backbone at binary detection (DeepSeek-R1) gains the most in coarse CWE classification, revealing that detection and understanding are decoupled capabilities. The best detection score reaches only 52.1% (+2.1 pp above chance); exact CWE ranking remains below 1.3% Top-1 accuracy, confirming that current LLMs lack reliable security reasoning for systems software, regardless of fine-tuning strategy.

阅读与讨论 → 访问原文 →
20.
arXiv (CS.AI) 2026-06-11

A Survey of Reasoning and Agentic Systems in Time Series with Large Language Models

作者:
Ching Chang↗Yidan Shi↗Defu Cao↗Wei Yang↗Jeehyun Hwang↗Haixin Wang↗Jiacheng Pang↗Wei Wang↗Yan Liu↗Wen-Chih Peng↗Tien-Fu Chen↗

arXiv:2509.11575v3 Announce Type: replace Abstract: Time series reasoning treats time as a first-class axis and incorporates intermediate evidence directly into the answer. This survey defines the problem and organizes the literature by reasoning topology with three families: direct reasoning in one step, linear chain reasoning with explicit intermediates, and branch-structured reasoning that explores, revises, and aggregates. The topology is crossed with the main objectives of the field, including traditional time series analysis, explanation and understanding, causal inference and decision making, and time series generation, while a compact tag set spans these axes and captures decomposition and verification, ensembling, tool use, knowledge access, multimodality, agent loops, and LLM alignment regimes. Methods and systems are reviewed across domains, showing what each topology enables and where it breaks down in faithfulness or robustness, along with curated datasets, benchmarks, and resources that support study and deployment (https://github.com/blacksnail789521/Time-Series-Reasoning-Survey). Evaluation practices that keep evidence visible and temporally aligned are highlighted, and guidance is distilled on matching topology to uncertainty, grounding with observable artifacts, planning for shift and streaming, and treating cost and latency as design budgets. We emphasize that reasoning structures must balance capacity for grounding and self-correction against computational cost and reproducibility, while future progress will likely depend on benchmarks that tie reasoning quality to utility and on closed-loop testbeds that trade off cost and risk under shift-aware, streaming, and long-horizon settings. Taken together, these directions mark a shift from narrow accuracy toward reliability at scale, enabling systems that not only analyze but also understand, explain, and act on dynamic worlds with traceable evidence and credible outcomes.

阅读与讨论 → 访问原文 →
21.
arXiv (CS.CV) 2026-06-11

DIRECT: When and Where Should You Allocate Test-Time Compute in Embodied Planners?

作者:
Jadelynn Dao↗Milan Ganai↗Yasmina Abukhadra↗Ajay Sridhar↗Mozhgan Nasr Azadani↗Katie Luo↗Clark Barrett↗Jiajun Wu↗Chelsea Finn↗Marco Pavone↗

Vision-Language Models (VLMs) are increasingly deployed as high-level planners for embodied agents, with an emerging strategy of scaling test-time compute to improve capability. However, we observe that doing so increases latency, token usage, and FLOPs while yielding uneven, often diminishing gains in downstream success, limiting where embodied agents can be deployed. We argue that choosing when and where to spend test-time compute is central to bringing frontier performance to the real world. We introduce DIRECT, a routing framework that uses multimodal scene context to allocate compute per prompt, improving the success–cost Pareto frontier over fixed model selection. Across three dominant scaling axes, namely chain-of-thought depth, model size, and memory history, our experiments on VLABench and RoboMME show that test-time compute is not a uniform lever: different axes yield qualitatively distinct capability gains. We validate these insights on a physical Franka arm in a DROID setup spanning zero-shot manipulation and long-horizon chaining, where our router matches or exceeds a stronger model's success rate at up to 65% lower average latency. Ultimately, our results show that naively scaling test-time compute is wasteful, and that DIRECT can provide frontier-level embodied planning in robotic systems at a fraction of the cost. Project page can be found at jadee-dao.github.io/direct/.

阅读与讨论 → 访问原文 →
22.
arXiv (CS.CL) 2026-06-18

Decoupling Search from Reasoning: A Vendor-Agnostic Grounding Architecture for LLM Agents

作者:
Emmanuel Aboah Boateng↗Kyle MacDonald↗Amardeep Kumar↗Siddharth Kodwani↗Sudeep Das↗

Production LLM agents increasingly depend on real-time search, yet native search grounding bundles retrieval policy, provider choice, evidence injection, cost, latency, and generation behavior behind a single model-provider boundary. This coupling makes grounding hard to inspect, tune, reuse, or port, and can trigger Search-Induced Verbosity that breaks strict output contracts. We present Decoupled Search Grounding (DSG), a vendor-agnostic boundary that moves grounding outside the reasoning model through an MCP-compatible gateway, exposing provider routing, source-aware context rendering, configured fallback, retrieval-depth control, and exact plus semantic caching as first-class controls. Across five frontier models on SimpleQA, FreshQA, and HotpotQA, native search leads on recency-sensitive FreshQA, but DSG exposes a stronger frontier when control matters: on SimpleQA it nearly matches native accuracy (86.1% vs. 87.7%) at 91% lower search cost, preserves concise answer contracts, and reaches a 99.4% warm-cache hit rate with 68% lower latency. Deployed as a shared production grounding layer for large-scale agentic workloads with interchangeable models, DSG matches or slightly exceeds native-search accuracy on an e-commerce query-understanding (QIU) workload while cutting search cost by over 98%. Real-time grounding is best treated as an optimizable interface boundary, not a fixed model feature.

阅读与讨论 → 访问原文 →
23.
arXiv (CS.CL) 2026-06-12

EvoBrowseComp: Benchmarking Search Agents on Evolving Knowledge

作者:
Yunhan Wang↗Jiaan Wang↗Lianzhe Huang↗Xianfeng Zeng↗Fandong Meng↗

Search Agents – large language models augmented with search tools – have intensified the need for future-proof evaluation benchmarks. Existing benchmarks such as BrowseComp rely on static knowledge, making them vulnerable to test-set contamination and parametric memorization. Consequently, models can achieve high scores through fact recall rather than genuine retrieval, obscuring true browsing competence via reasoning shortcuts. In this paper, we introduce EvoBrowseComp, an evolving benchmark of 400 English and 400 Chinese contamination-free complex questions synthesized via live-web traversal. To collect these questions, we design a three-agent collaborative framework: (1) a QA synthesis agent that retrieves fresh knowledge from the live web to synthesize QA pairs; (2) an information filtering agent that filters retrieved knowledge in terms of credibility and popularity to block parametric shortcuts; and (3) a high-level guidance agent that formalizes questions into reasoning graphs to reduce logical redundancy and shortcuts in synthesized QA pairs. Because the framework supports fully automated synthesis, EvoBrowseComp can be regularly updated to prevent data contamination and maintain temporal freshness. Extensive experiments confirm its great difficulty, requiring broad horizontal search. It establishes a scalable paradigm for auto-updatable, high-difficulty benchmarking that keeps pace with both evolving world knowledge and advancing agent capabilities.

阅读与讨论 → 访问原文 →
24.
arXiv (CS.CV) 2026-06-16

Discriminative Span as a Predictor of Synthetic Data Utility via Classifier Reconstruction

作者:
Radhika Amar Desai↗Modigari Narendra↗

In many real-world computer vision applications, including medical imaging and industrial inspection, binary classification tasks are characterized by a severe scarcity of positive samples. A widely adopted solution is to generate synthetic positive data using image-to-image transformations applied to negative samples. However, a fundamental challenge remains: how can we reliably assess whether such synthetic data will improve downstream model performance? In this work, we propose a geometry-driven metric that predicts the utility of synthetic data without requiring model training. Our approach operates in the embedding space of a pre-trained foundation model and represents the dataset through difference vectors between samples. We evaluate whether the weight vector of a linear classifier can be expressed within the subspace spanned by these variations by measuring the relative projection error. Intuitively, if the variations induced by synthetic data capture task-relevant directions, their span can approximate the classifier, resulting in low projection error. Conversely, poor synthetic data fails to span these directions, leading to higher error. Across multiple datasets and architectures, we show that this metric exhibits strong correlation with downstream classification performance of CNNs trained on mixtures of real negative and synthetic positive data. These findings suggest that the proposed metric serves as a practical and informative tool for evaluating synthetic data quality in data-scarce settings.

阅读与讨论 → 访问原文 →
25.
arXiv (CS.AI) 2026-06-17

How Inference Compute Shapes Frontier LLM Evaluation

作者:
Jessica McFadyen↗Ole Jorgensen↗Harry Coppock↗Kevin Wei↗Cozmin Ududec↗

arXiv:2606.17930v1 Announce Type: new Abstract: AI evaluations are shifting toward harder tasks that benefit from longer trajectories involving tool use and iterative problem solving. As a result, performance is increasingly sensitive to the amount and allocation of compute available at test time ("inference compute"). Yet many evaluations still report performance at a single restrictive budget, meaning that low scores may reflect the evaluation setup rather than the model's underlying capability. To test this, we evaluate up to 12 frontier language models on seven challenging benchmarks spanning software engineering, mathematics, medicine, and cybersecurity. We use a controlled setup combining three simple inference-scaling interventions: larger token budgets, context compaction, and repeated submission attempts, guided either by the model itself or by minimal correctness feedback. We find three main results. First, larger token budgets substantially improve performance on benchmarks across multiple domains, including cybersecurity, FrontierMath, Humanity's Last Exam, and TerminalBench. Second, fixed-budget evaluations can increasingly understate frontier capability as models advance. Newer models reach higher performance at large budgets, where they unlock harder tasks and solve them more reliably. Third, benchmarks differ in which inference-scaling methods help most: repeated submission broadly improves performance, but the value of larger token budgets, external feedback, and parallel attempts varies by benchmark. Overall, our results show that benchmark scores are protocol-dependent. We therefore argue that evaluations should report capability as a function of inference-time compute, specify protocol choices explicitly, and compare model generations over a large shared compute range at matched budgets, especially in safety- or policy-relevant settings.

阅读与讨论 → 访问原文 →