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) 2026-06-11

Super-Heisenberg Non-Equilibrium Quantum Sensing with Waveguide-Coupled Emitters

作者:
Mohammad B. Arjmandi↗

arXiv:2606.11975v1 Announce Type: new Abstract: We explore an array of quantum emitters as non-equilibrium probes, coupled to a one-dimensional photonic waveguide, aiming to estimate its properties such as wave number which encodes the waveguide frequency and dispersive characteristics. By considering transient dynamics following initial excitation, we show that the quantum Fisher information (QFI) can be significantly enhanced through careful emitter positioning. For two-emitter probes, optimal spacing stabilizes populations and coherences in the single-excitation subspace, suppressing super radiant decay and extending both the magnitude and longevity of QFI. Randomized emitter configurations also reveal that vanishing waveguide-mediated cross decay maximizes both achievable sensitivity and the temporal duration over which information about the parameter remains accessible. Extending to multipartite probes, we demonstrate that the maximum QFI and its temporal integral scale with system size, exceeding the Heisenberg limit for all positioning strategies. Our results highlight the potential of waveguide-coupled emitter arrays as versatile quantum sensors, where collective radiative dynamics can be harnessed to achieve tunable, long-lived, and enhanced precision.

阅读与讨论 → 访问原文 →
02.
arXiv (CS.LG) 2026-06-17

Stable and Steerable Sparse Autoencoders with Weight Regularization

作者:
Piotr Jedryszek↗Oliver M. Crook↗

arXiv:2603.04198v2 Announce Type: replace-cross Abstract: Sparse autoencoders (SAEs) are widely used to extract human-interpretable features from neural network activations, but their learned features can vary substantially across random seeds and training choices. To improve stability, we studied weight regularization by adding L1 or L2 penalties on encoder and decoder weights, and evaluate how regularization interacts with common SAE training defaults. On MNIST, we observe that L2 weight regularization produces a core of highly aligned features and, when combined with tied initialization and unit-norm decoder constraints, it dramatically increases cross-seed feature consistency. For TopK SAEs trained on language model activations (Pythia-70M-deduped), adding a small L2 weight penalty increased the fraction of features shared across three random seeds and roughly doubles steering success rates, while leaving the mean of automated interpretability scores essentially unchanged. Finally, in the regularized setting, activation steering success becomes better predicted by auto-interpretability scores, suggesting that regularization can align text-based feature explanations with functional controllability.

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

Artificial Intelligence in Ship Finance: Applications, Opportunities, and a Case Study in AI-Augmented Loan Origination

作者:
Lasse Dierich↗Orestis Schinas↗

arXiv:2606.11238v1 Announce Type: cross Abstract: Ship finance is a data-intensive and document-heavy segment of asset-based lending, requiring the integration of financial, technical, contractual, and regulatory information from heterogeneous and largely unstructured sources. Increasing environmental regulation and ESG reporting requirements are adding further complexity to underwriting and loan-origination processes. Recent advances in artificial intelligence (AI), particularly large language models (LLMs), create new opportunities for processing and analysing such information. This paper reviews potential applications of AI in ship finance, with a particular focus on LLM-based systems for document comprehension, information extraction, and workflow automation. We present ShipFinance.ai, a modular agentic architecture to support loan application workflows in ship finance. The proposed system combines an LLM-based extraction module, financial analysis components, external maritime data services, and a controlled document-generation module with a chatbot interface to support the preparation of standardized financing applications. The paper discusses the key challenges for using such models in production. We argue that AI-assisted systems can support maritime finance professionals in managing increasingly complex information and reporting requirements.

阅读与讨论 → 访问原文 →
04.
arXiv (CS.AI) 2026-06-16

APEX: Adaptive Principle EXtraction A Three-Layer Self-Evolution Framework for Production AI Agents

作者:
Ya-Chuan Chen↗Tien-Jen Lai↗Hsiang-Wei Hu↗

arXiv:2606.15363v1 Announce Type: new Abstract: Self-improvement in AI agents has emerged as a key research frontier: systems that modify their own prompts, workflows, and decision rules based on accumulated operational experience. The state-of-the-art Self-Harness framework [1] achieves 14–21% improvement on Terminal-Bench-2.0 by mining failure clusters and patching the agent harness. However, Self-Harness optimises only one dimension – the prompt harness – leaving behavioural principles and workflow topology unchanged. We propose APEX (Adaptive Principle EXtraction), a three-layer co-evolution framework that simultaneously evolves: (L1) the harness via failure-mode patching, (L2) behavioural principles via success-trace distillation [2], and (L3) the agent workflow topology via structural fitness-based selection [6]. We implement APEX on Joe [13], a production-grade super AI Agent built on NVIDIA Nemotron and designed as an Edge AI Agent Factory for the NVIDIA Agent Challenge 2026, managing a 15-node compute fleet using 114 real task traces collected over 18 days. APEX achieves an APEX Health Score of 0.570 (+90% vs. baseline 0.300) in a single evolutionary run, distilling 6 novel reusable principles and selecting a research-first workflow topology scoring 0.900 (+20%). Our results demonstrate that multi-dimensional co-evolution substantially outperforms single-axis harness optimisation, at a cost of only 4 LLM calls (~270 s) on a local qwen2.5-coder:32b instance.

阅读与讨论 → 访问原文 →
05.
medRxiv (Medicine) 2026-06-17

Trends in Suicide Mortality by Method among US Individuals aged 10-24 Years from 1999 to 2024

作者:
Bischops↗A. C↗Charpignon↗M.-L↗Mandl↗K. D↗Majumder↗M. S↗

Background: Suicide is the second leading cause of death in US adolescents aged 10-24. Method use strongly influences lethality and design of prevention strategies, but recent trends remain unclear. We therefore aimed to investigate trends in suicide mortality rates by method, age group, and sex. Methods: This cross-sectional study used suicide mortality data from the National Center for Health Statistics for a quarter-century period, between 1999 and 2024. All individuals aged 10-24 years at the time of death, with suicide as the underlying cause, were included. We estimated suicide mortality rates (i.e., the number of suicide deaths per 100,000 people) and annual percent change by method (firearm, asphyxiation, poisoning, other), age group (10-14, 15-19, 20-24), and sex. Changing trend time points were determined using Joinpoint regression models Results: From 1999 to 2024, 159,241 suicide deaths occurred among individuals aged 10-24. While suicide rates declined across all age groups between 2017 and 2024, the male-to-female gap narrowed by 18.9%. Among 10-14-year-olds, declining rates among males masked a consistent increase in female suicide rates since 2011. Although asphyxiation-related suicides decreased across all groups since 2018, firearm suicide rates increased for females in the 10-14 and 20-24 age groups. Albeit not as common as firearms or asphyxiation, poisoning suicide rates increased in the 15-19 and 20-24 age groups. Since 1999, suicide rates by other less common methods (e.g., jumping) showed significant increases, for both sexes, especially among individuals aged 20-24. Suicide rates were consistently highest in the 20-24 age group across all study years. Conclusion: The decrease in suicide mortality rates among individuals aged 10-24 was largely driven by declines in males and reductions in asphyxiation-related suicides. However, increasing female suicide rates in the 10-14 age group, as well as increasing rates of death by less common means, warrant close attention. While suicide prevention efforts like structural interventions and means restriction have shown effectiveness among male adolescents, priority should now be given to adapting these approaches for female adolescents, particularly those aged 10-14.

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

LLM-Powered Multi-Agent System for Automated Crypto Portfolio Management

作者:
Yichen Luo↗Yebo Feng↗Jiahua Xu↗Paolo Tasca↗Yang Liu↗

arXiv:2501.00826v3 Announce Type: replace-cross Abstract: Cryptocurrency portfolio management requires the fusion of heterogeneous multi-modal signals, including structured price and on-chain time series, unstructured news text, and technical indicators, under high-volatility and real-time constraints. While deep learning approaches show predictive capability, their opacity limits practical adoption, and single large language model (LLM) agents struggle to process the breadth of modality-specific inputs needed for robust decision-making. We propose a multi-agent system (MAS) framework in which three modality-specialised agents, a Crypto Agent for market dynamics, a News Agent for weekly news sentiment, and a Trading Agent for signal fusion and portfolio execution, decompose the task across three communication architectures: hierarchical, collaborative, and debate. We evaluate four capability configurations: zero-shot, chain-of-thought (CoT), retrieval-augmented generation (RAG), and skill-augmented. In a 52-week backtest over calendar year 2025 across the top 15 L1 blockchain native cryptocurrencies by market capitalisation as of January 2025, the best configuration, Hierarchical (Skill), achieves a cumulative return of 133.52% and a Sharpe ratio of 1.502, outperforming single-agent variants, passive benchmarks, and deep learning baselines. An ablation study identifies the Crypto Agent as the most critical component, with its removal reducing cumulative return by 42.57 percentage points. A cross-model comparison further shows that MAS outperforms the single-agent baseline under GPT-4o, GPT-5, and Claude Sonnet 4.5, suggesting that the benefit of multi-agent coordination is model-agnostic. Unlike black-box deep learning models, every portfolio decision is traceable to explicit agent reasoning, offering an interpretable and effective approach to multi-modal cryptocurrency portfolio management.

阅读与讨论 → 访问原文 →
07.
arXiv (CS.LG) 2026-06-16

SDVDiag: Multimodal Causal Discovery for Online Diagnosis in Software-defined Vehicles

作者:
Matthias Wei{\ss}↗Athreya Hosahalli Prakash↗Falk Dettinger↗Nasser Jazdi↗Michael Weyrich↗

arXiv:2606.15559v1 Announce Type: cross Abstract: The transition toward software-defined vehicles concentrates an increasing share of vehicle functionality into distributed software services, where failures propagate through service dependencies and the surface symptom is often several causal hops away from the underlying defect. Existing approaches to causal root-cause analysis in such systems address this only partially: they typically reason over a single observability modality and operate in an offline, operator-driven mode that does not match the demands of continuous vehicle operation. This paper presents SDVDiag, a multimodal causal-discovery pipeline that fuses log-based and metric-based service representations into a shared embedding space before graph construction, coupled with an anomaly-driven trigger that converts the diagnostic platform from a manually operated batch tool into a continuously running online system. Evaluation on an Autonomous Valet Parking testbed shows that the multimodal pipeline produces sparser causal graphs than a metrics-only baseline (134 vs. 182 edges on average) and consistently outperforms it in edge-weighted reward against an expert knowledge graph at every stage of human-feedback refinement, showing a 2.4-fold improvement over the baseline after 60 feedback queries. An end-to-end fault-injection scenario further demonstrates that the integrated trigger correctly recovers a true root cause located two causal hops upstream of the observable symptom.

阅读与讨论 → 访问原文 →
08.
arXiv (CS.CL) 2026-06-12

Epistemic Constitutionalism Or: how to avoid coherence bias

作者:
Michele Loi↗

Large language models increasingly function as artificial reasoners: they evaluate arguments, assign credibility, and express confidence. Yet their belief-forming behavior is governed by implicit, uninspected epistemic policies. This paper argues for an epistemic constitution for AI: explicit, contestable meta-norms that regulate how systems form and express beliefs. Source attribution bias provides the motivating case: I show that frontier models enforce identity-stance coherence, penalizing arguments attributed to sources whose expected ideological position conflicts with the argument's content. When models detect systematic testing, these effects collapse, revealing that systems treat source-sensitivity as bias to suppress rather than as a capacity to execute well. I distinguish two constitutional approaches: the Platonic, which mandates formal correctness and default source-independence from a privileged standpoint, and the Liberal, which refuses such privilege, specifying procedural norms that protect conditions for collective inquiry while allowing principled source-attending grounded in epistemic vigilance. I argue for the Liberal approach, sketch a constitutional core of eight principles and four orientations, and propose that AI epistemic governance requires the same explicit, contestable structure we now expect for AI ethics.

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

Learning Permutation Distributions via Reflected Diffusion on Ranks

作者:
Sizhuang He↗Yangtian Zhang↗Shiyang Zhang↗David van Dijk↗

arXiv:2603.17353v2 Announce Type: replace-cross Abstract: The finite symmetric group S_n provides a natural domain for permutations, yet learning probability distributions on S_n is challenging due to its factorially growing size and discrete, non-Euclidean structure. Recent permutation diffusion methods define forward noising via shuffle-based random walks (e.g., riffle shuffles) and learn reverse transitions with Plackett-Luce (PL) variants, but the resulting trajectories can be abrupt and increasingly hard to denoise as n grows. We propose Soft-Rank Diffusion, a discrete diffusion framework that replaces shuffle-based corruption with a structured soft-rank forward process: we lift permutations to a continuous latent representation of order by relaxing discrete ranks into soft ranks, yielding smoother and more tractable trajectories. For the reverse process, we introduce contextualized generalized Plackett-Luce (cGPL) denoisers that generalize prior PL-style parameterizations and improve expressivity for sequential decision structures. Experiments on sorting and combinatorial optimization benchmarks show that Soft-Rank Diffusion consistently outperforms prior diffusion baselines, with particularly strong gains in long-sequence and intrinsically sequential settings.

阅读与讨论 → 访问原文 →
10.
arXiv (CS.CL) 2026-06-12

Beyond the Commitment Boundary: Probing Epiphenomenal Chain-of-Thought in Large Reasoning Models

作者:
Daniel Scalena↗Sara Candussio↗Luca Bortolussi↗Elisabetta Fersini↗Malvina Nissim↗Gabriele Sarti↗

Chain-of-thought (CoT) reasoning is the dominant paradigm for inference-time scaling in language models, yet the causal influence of individual steps on the final answer poorly understood. We estimate each step's causal importance via early exit and use this measure to study how answers form across the reasoning traces of several model families. Across diverse tasks, we find that reasoning typically crosses a commitment boundary – a sharp transition from transient intermediate guesses to a stable, high-confidence answer. This transition often happens in a single step, well before the model's reasoning block ends, and is followed by epiphenomenal CoT steps that leave the final answer probability unaltered. Using attention probes, we show that answer-formation stages can be linearly decoded from intermediate reasoning steps with high accuracy and generalize robustly to unseen reasoning tasks. We exploit this signal to early-exit reasoning blocks at the commitment boundary, reducing the length of CoTs up to 55\% on average with negligible impact on model performance.

阅读与讨论 → 访问原文 →
11.
medRxiv (Medicine) 2026-06-15

Neural Correlates of Human Food Memory link to Microbial, Homeostatic, and Hedonic Signals: Evidence from a Prebiotic Randomized Clinical Trial

作者:
Jensen↗D. E. A↗Thieleking↗Medawar↗Reinicke↗Rolle-Kampczyk↗von Bergen↗Stumvoll↗Villringer↗Beyer↗Witte↗

Background Homeostatic and hedonic brain circuits regulate eating behavior but also shape how food memories are encoded and retrieved. Objective We examined neural correlates during food memory encoding and retrieval during functional MRI before and after a 14-day prebiotic intervention in a preregistered, double-blind crossover trial (NCT03829189). Design 55 healthy adults with overweight (19 females, age 28{+/-}6.5, BMI 25-30 kg/m2) underwent 3 Tesla task-based functional MRI before and after dietary intervention of prebiotic (30g inulin/day) or equicaloric placebo for 14 days. Peripheral metabolic, short-chain fatty acids (SCFA), and microbial markers using 16S rRNA analysis were assessed in fasting blood and feces. Results Food memory was enhanced by assigned reward value and engaged brain activity in hedonic regions, including the nucleus accumbens, orbitofrontal cortex, caudate, cingulate, dorsomedial prefrontal cortex, and ventral tegmental area, as well as homeostatic and memory-related such as the hypothalamus and the hippocampus. Higher neural activations during food encoding were related to higher Actinobacteriota abundance, fecal SCFA acetate, and creatinine levels, and lower ghrelin levels. Activations in reward-related and homeostatic brain areas partially correlated with insulin, glucagon-like peptide-1, leptin, and thyroid-stimulating hormone levels. Neural activations related to food memory decreased after prebiotic intervention. The prebiotic supplementation induced decrease of hippocampal activity during food encoding related to changes in gut microbiota Firmicutes abundance. Conclusions This study indicates that neuronal food-related memory processes depend on homeostatic and hedonic brain signals modulated by the gut-brain axis. Our findings raise implications for the treatment of obesity and substance use disorder.

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

LSTM-Based Detection of Structural Breaks in Property Insurance Loss Reserving: A Climate-Informed Approach

作者:
Thomas Mbrice↗Shashwat Panigrahi↗

arXiv:2606.11463v1 Announce Type: cross Abstract: Accurate loss reserving is foundational to insurer solvency, yet accelerating climate driven catastrophes systematically violate the stability assumptions on which traditional actuarial methods depend. This white paper presents a research program testing whether Long Short Term Memory (LSTM) neural networks can detect and adapt to these structural breaks faster and more accurately than Chain Ladder, Bornhuetter Ferguson, and Cape Cod methods. Using 15 plus years of regulatory development triangle data from Florida and Louisiana, enriched with NOAA hurricane intensity indices and sea surface temperatures, we hypothesize a targeted improvement of 15, 20% in reserve accuracy for catastrophe exposed years, a threshold grounded both in the prior neural network reserving literature and in the formal convergence results developed here. Beyond empirical validation, we develop a theoretical framework grounding LSTM structural break detection in probabilistic terms, providing formal performance guarantees that compensate for the limited number of catastrophe events in the test period. We document the research design, methodology, expected contributions, and a candid assessment of limitations.

阅读与讨论 → 访问原文 →
14.
arXiv (CS.CL) 2026-06-16

Data Augmentations for Data-Constrained Language Model Pretraining

作者:
Michael K. Chen↗Xikun Zhang↗Zhen Wang↗

As AI labs approach a data ceiling where compute capacity outpaces the rate of new high-quality text generation, language model pretraining is shifting toward a data-constrained, compute-abundant regime that demands productive multi-epoch training on fixed corpora. Standard autoregressive (AR) pretraining overfits severely in this setting, reaching its optimum early and then continuously deteriorating. We investigate data augmentation as a regularizer to mitigate this overfitting and enable productive training for hundreds of epochs on the same data. We introduce three orthogonal categories of augmentation for AR pretraining: token-level noise (masking, random replacement), sequence permutations (right-to-left prediction, Fill-in-the-Middle), and target offset prediction ($x_{t+i}$ for $i > 1$). Through systematic ablations, we find that individual augmentations delay overfitting and lower validation loss relative to the baseline, with random token replacement achieving the best minimum loss among individual methods. Combining augmentation categories further lowers the minimum validation loss. Our experiments demonstrate that data augmentations mitigate AR pretraining's data inefficiency and offer a promising solution to the data-constrained regime. All code and data are available at https://github.com/michaelchen-lab/data-augmentations-for-pretraining

阅读与讨论 → 访问原文 →
15.
medRxiv (Medicine) 2026-06-17

Sao Tome and Principe on the verge of eliminating lymphatic filariasis as a public health problem: evidence from IDA impact assessment surveys

作者:
Bakajika↗Rosario↗Fernandes↗J. V↗Diawara↗L. S↗Mangala↗J.-P. T↗Kalenga↗Gil↗V. S↗Yumba↗…

Background Accelerated efforts to eliminate lymphatic filariasis (LF) as a public health problem have been supported by the introduction of the triple-drug regimen of ivermectin, diethylcarbamazine and albendazole (IDA) in endemic settings. In Sao Tome and Principe, nationwide mass drug administration (MDA) with diethylcarbamazine and albendazole was implemented in 2018, followed by IDA in 2019 and 2020. This study assesses progress towards elimination using post-MDA impact assessment surveys conducted after cessation of treatment. Methods Cross-sectional surveys were conducted among adults aged 20 years and older in 2022 and again between December 2024 and January 2025. Circulating filarial antigen (CFA) was detected using the filarial test strip (FTS). Individuals who tested positive were examined for microfilaremia using nocturnal calibrated thick blood smear microscopy. Additionally, programme data on MDA coverage and morbidity were obtained from national surveillance records. Results Three rounds of nationwide MDA achieved high epidemiological coverage (86.4% in 2018, 74.2% in 2019 and 80.0% in 2020). The impact assessment surveys conducted in 2022 evaluated 14 132 adults, with 21 individuals (0.15%) testing positive for CFA, while the follow-up survey conducted between December 2024 and January 2025 assessed 14 653 adults and detected seven positive cases (0.05%). No microfilariae were detected among the 28 antigen-positive individuals examined using nocturnal calibrated thick blood smears. National morbidity records documented 190 cases of lymphoedema and nine cases of hydrocoele. Conclusions Infection indicators remain well below WHO decision thresholds, suggesting that LF transmission is unlikely to be sustained. Sao Tome and Principe appears to be close to eliminating LF as a public health problem. However, strengthening morbidity management services will be essential to support the preparation of the national elimination dossier.

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

Hierarchical Modeling of ICD Codes in EHR Foundation Models

作者:
Megha Thukral↗Dong Gyun Kang↗Rudra Pratap Singh↗Shruthi Kashinath Hiremath↗Katrin H\"ansel↗Thomas Pl\"otz↗

arXiv:2606.15447v1 Announce Type: new Abstract: Electronic health record foundation models typically treat ICD diagnosis codes as flat tokens, overlooking the clinically meaningful hierarchical structure that captures disease families, subcategories, and fine-grained diagnostic detail. As a result, existing EHR representation learning methods do not explicitly exploit the hierarchical structure already present in the coding system. In this work, we study ICD-10-CM hierarchy as a general inductive bias for clinical representation learning. We investigate two complementary mechanisms for incorporating hierarchy: first, by augmenting diagnosis sequences in a BERT-style transformer with tokens corresponding to different levels of the ICD hierarchy, and second, by injecting hierarchy into graph-based code representations through hierarchy-aware edges combined with diagnosis co-occurrence structure. Across these settings, we evaluate whether explicit hierarchy improves downstream prediction, which levels of the hierarchy are most useful, whether hierarchy encoding improves transfer across datasets, and how hierarchy reshapes embedding similarity structure. We conduct experiments on two large-scale real-world clinical datasets: MIMIC-IV, used for pretraining and in-domain evaluation, and eICU, used to assess cross-dataset transfer via frozen encoder probing. Our findings show that explicitly encoding ICD hierarchy improves over flat code representations in both in-domain and cross-dataset settings, while revealing that the most useful level of hierarchy depends on both the task and the modeling approach. More broadly, we focus on hierarchy-aware EHR representation learning and show that the benefits of encoding hierarchy are generalizable across modeling settings and hierarchy levels.

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

Regression Language Models for Code

作者:
Yash Akhauri↗Xingyou Song↗Arissa Wongpanich↗Bryan Lewandowski↗Mohamed S. Abdelfattah↗

We study code-to-metric regression: predicting numeric outcomes of code executions, a challenging task due to the open-ended nature of programming languages. While prior methods have resorted to heavy and domain-specific feature engineering, we show that a single unified Regression Language Model (RLM) using a frozen LLM encoder can simultaneously predict directly from text, (i) the memory footprint of code across multiple high-level languages such as Python and C++, (ii) the latency of Triton GPU kernels, and (iii) the accuracy and speed of trained neural networks represented in ONNX. In particular, a relatively small 300M parameter RLM based on T5Gemma, obtains >0.9 Spearman-rank on competitive programming submissions from APPS, and a single unified model achieves >0.5 average Spearman-rank across 24 different programming languages from CodeNet. Furthermore, the RLM can obtain the highest average Kendall-Tau of 0.46 on five classic NAS design spaces previously dominated by graph neural networks, and simultaneously predict architecture latencies on numerous hardware platforms.

阅读与讨论 → 访问原文 →
19.
PLOS Medicine 2026-05-21

Novel symptoms associated with eclampsia could improve detection and save lives

作者:
Alice Beardmore-Gray↗

by Alice Beardmore-Gray, Andrew Shennan Eclampsia is a life-threatening complication of pre-eclampsia, yet remains difficult to predict. In this Perspective, Alice Beardmore-Gray and Andrew Shennan highlight a recent study that identifies 10 novel prodromal symptoms of eclampsia, with potential to better predict which women are at risk and therefore reduce delays in intervention.

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

Substrate Asymmetry in User-Side Memory: A Diagnostic Framework

作者:
Youwang Deng↗

User-side memory in LLMs is typically scored as a single "personalization" capability: given a user's history, is the output more user-aware? We show this aggregate metric hides opposite-direction failures. Memory factorises into at least three orthogonal axes – behavioral consistency (style, voice), factual presence (recall facts in history), and factual absence (abstain when a fact is absent) – and no single substrate wins all three. Comparing per-user gamma-LoRA (a small LoRA adapter trained on each user's history; gamma denotes per-user, not per-task) against BGE-large dense top-K retrieval on a controlled 50-user synthetic corpus and a real-data probe (LaMP-3), we find gamma-LoRA decisively wins behavioral style while RAG decisively wins factual absence – and the same query-projection cells in attention layers 21-35 causally load-bear both effects in opposite directions (zeroing those LoRA weights raises absence-probe TPR by +33 pp and drops presence-probe TPR by 20 pp). On the more heavily RLHF-tuned Llama-3.1-8B-Instruct the asymmetry strengthens, not heals: parametric memory's behavioral advantage collapses while its absence-calibration deficit against retrieval widens – an alignment tax on parametric user-memory. On real-data LaMP-3, gamma-LoRA underperforms a majority baseline; a 9-condition mitigation sweep diagnoses this as instruction-following collapse, not substrate failure (a 9x2 cross-product shows the eval-time {1..5} logit mask drives main_acc to >=0.995 on every recipe), and the best training-time fix replicates bit-identically on Llama. Finally, substrate-selection routing is question-classification, not calibration: a 110M DistilBERT on the question text alone beats every logit-based router. We contribute the diagnostic framework, the diagnosed real-data negative, the alignment-tax replication, and the routing-as-classification finding.

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

FinTradeBench: A Financial Reasoning Benchmark for LLMs

作者:
Yogesh Agrawal↗Aniruddha Dutta↗Md Mahadi Hasan↗Santu Karmaker↗Aritra Dutta↗

Real-world financial decision-making is a challenging problem that requires reasoning over heterogeneous signals, including company fundamentals derived from regulatory filings and trading signals computed from price dynamics. Recently, with advances in Large Language Models (LLMs), financial analysts have begun to use them for financial decision-making tasks. However, existing financial question-answering benchmarks for testing these models primarily focus on company balance sheet data and rarely evaluate reasoning about how company stocks trade in the market or their interactions with fundamentals. To leverage the strengths of both approaches, we introduce FinTradeBench, a benchmark for evaluating financial reasoning that integrates company fundamentals and trading signals. FinTradeBench contains 1,400 questions grounded in NASDAQ-100 companies over a ten-year historical window. The benchmark is organized into three reasoning categories: fundamentals-focused, trading-signal-focused, and hybrid questions requiring cross-signal reasoning. To ensure reliability at scale, we adopt a calibration-then-scaling framework that combines expert seed questions, multi-model response generation, intra-model self-filtering, numerical auditing, and human-LLM judge alignment. We evaluate 14 LLMs under zero-shot prompting and retrieval-augmented settings and witness a clear performance gap. Retrieval substantially improves reasoning over textual fundamentals, but provides limited benefit for trading-signal reasoning. These findings highlight fundamental challenges in the numerical and time-series reasoning for current LLMs and motivate future research in financial intelligence.

阅读与讨论 → 访问原文 →
22.
arXiv (CS.AI) 2026-06-17

Memory as a Wasting Asset: Pricing Flash Endurance for Embodied Agents, and the Limits of Doing So

作者:
Josef Liyanjun Chen↗

arXiv:2606.18144v1 Announce Type: new Abstract: A robot's flash endurance is a non-renewable stock: every persisted write spends one of a few thousand program/erase cycles and never refills, yet no fielded robot memory system prices which memories are worth an erase cycle. We treat embodied memory as depreciating capital and price that stock with a single endurance shadow price $\eta$, which makes cost-minimizing placement across a RAM / on-board NVM / cloud hierarchy a threshold in a wear-augmented per-byte index. The index is cost-optimal whatever the sign of the value-write association $\chi$; only when $\chi > 0$ does the optimum turn non-monotone, sending a robot's most valuable memories off its flash. The pivot is thus empirical, and we measure $\chi$ on real robot logs at a pre-specified gate: its sign is a property of the deployment regime – positive on recurrent long-horizon manipulation ($\hat{\chi} \approx +1.0 \times 10^{-3}$, replicated at full power), null on a shorter-horizon suite, and negative on non-recurrent teleoperation. Two boundaries scope the result. The endurance budget is dormant on premium 3,000-P/E TLC at datasheet prices and binding on the commodity QLC/eMMC ($\sim$1,000 P/E) that cheaper edge robots run. And where it binds, a learned wear-aware controller only ties price-based routing on task value, because realized value is tier-invariant across RAM, NVM, and cloud: the rent governs device lifetime and cost, not task performance. Whether wear-aware placement improves task value remains open – $\chi$ is measured against a value proxy, and the non-monotone optimum, while proven, is not yet observed in data.

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

Brick: Spatial Capability Routing for the Mixture-of-Models (MoM) Paradigm

作者:
Francesco Massa↗Marco Cristofanilli↗

arXiv:2606.13241v1 Announce Type: new Abstract: Defining query difficulty is one of the hardest problems in deployment engineering. Existing LLM routers rely on surface features such as domain labels, keywords, and token count, ignoring the within-domain variance that actually determines model success. Frontier models cost ten to one hundred times more than local open-weight models, so at production scale even small per-request savings become a direct cloud-bill lever. We present Brick, a multimodal router that scores each model on six capability dimensions, combines this with a per-query difficulty estimate, and dispatches via a cost-penalized geometric rule. A continuous preference knob lets operators slide between max-quality and max-saving profiles at deploy time. On a benchmark of 5,504 queries, Brick at max-quality reaches 76.98% accuracy, beating the best single model (75.02%) and all tested routers. At a neutral cost-quality profile, Brick achieves 74.11% accuracy at 4.71x lower cost than always using the strongest model. At min-cost, it cuts cost 22.15x with 11.85 points accuracy loss. Median latency drops from 51.2s to 22.8s.

阅读与讨论 → 访问原文 →
24.
arXiv (CS.CL) 2026-06-11

ClawEnvKit: Automatic Environment Generation for Claw-Like Agents

作者:
Xirui Li↗Ming Li↗Ion Stoica↗Cho-Jui Hsieh↗Tianyi Zhou↗

Constructing environments for training and evaluating claw-like agents remains a manual, human-intensive process that does not scale. We argue that what is needed is not just a dataset, but an automated pipeline capable of generating diverse, verified environments on demand. To this end, we introduce ClawEnvKit, an autonomous generation pipeline that instantiates this formalism from natural language descriptions. The pipeline comprises three modules: (1) a parser that extracts structured generation parameters from natural language input; (2) a generator that produces the task specification, tool interface, and scoring configuration; and (3) a validator that enforces feasibility, diversity, structural validity, and internal consistency across the generated environments. Using ClawEnvKit, we construct Auto-ClawEval, the first large-scale benchmark for claw-like agents, comprising 1,040 environments across 24 categories. Empirically, Auto-ClawEval matches or exceeds human-curated environments on coherence and clarity at 13,800x lower cost. Evaluated across 4 model families and 8 agent harness frameworks, we find that harness engineering boosts performance by up to 15.7 percentage points over a bare ReAct baseline, completion remains the primary axis of variation with no model saturating the benchmark, and automated generation enables evaluation at a scale previously infeasible. Beyond static benchmarking, ClawEnvKit enables live evaluation: users describe a desired capability in natural language and obtain a verified environment on demand, turning evaluation into a continuous, user-driven process. The same mechanism serves as an on-demand training environment generator, producing task distributions that adapt to an agent's current weaknesses rather than being bounded by existing user logs.

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

TACOMORE: Exploring a replicable prompting protocol for LLM-assisted corpus analysis

作者:
Bingru Li↗Han Wang↗Nicholas Groom↗

As corpus linguistics continues to scale, researchers are facing a growing methodological bottleneck: while computational tools can easily count billions of words, the qualitative interpretation of these data remains a slow and labor-intensive human task. Large Language Models (LLMs) offer a promising way to automate this process, yet their integration into the field is often hindered by concerns over black-box unpredictability and a lack of replicability. This study introduces TACOMORE, a structured prompting framework designed to transform ad-hoc AI interactions into a standardized linguistic protocol. Built upon four foundational principles (Task, Context, Model, and Replicability), the framework guides LLMs to move beyond generic probability prediction to anchoring their reasoning in the specific co-occurrence patterns of a target corpus. We applied this framework to three core corpus tasks, i.e., the analysis of keywords, collocates, and concordances, using an open corpus of COVID-19 research abstracts. After testing three LLMs, we found that while structured prompting improves accuracy and replicability, inherent limitations regarding hallucination persist. This research offers a critical lens into the role of LLMs in corpus linguistics, highlighting their potential as complementary tools while emphasizing the irreplaceable role of human validation.

阅读与讨论 → 访问原文 →