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。

01.
arXiv (CS.AI) 2026-06-25

Judging the Judges: A Systematic Evaluation of Bias Mitigation Strategies in LLM-as-a-Judge Pipelines

arXiv:2604.23178v2 Announce Type: replace Abstract: LLM-as-a-Judge has become the dominant paradigm for evaluating language model outputs, yet LLM judges exhibit systematic biases that compromise evaluation reliability. We present a comprehensive empirical study comparing nine debiasing strategies across five judge models from four provider families (Google, Anthropic, OpenAI, Meta), three benchmarks (MT-Bench n=400, LLMBar n=200, custom n=375), and four bias types. Our headline practical finding is that a mid-tier model with the right debiasing can outperform frontier judges at a fraction of the cost: Gemini 2.5 Flash with the Combined Budget strategy reaches the highest agreement of any configuration we tested (71.0%, kappa=0.549) at ~$0.001 per evaluation, about 15x cheaper than the best frontier setup (Claude Sonnet 4, 69.5%, ~$0.015). Other key findings: (1) Style bias is the dominant bias (0.10-0.76 across models, favoring markdown over plain prose), far exceeding position bias (

02.
arXiv (CS.CL) 2026-06-15

CoRe: A Continuously Reward-Finetuned LLM Query Rewriter for Multi-Stage Context-Aware Relevance in Web-Scale Video Search

LLM-based query rewriters in production face a tension: the training reward must reflect how the rewrite is consumed by the production ranker, yet the training procedure must be cheap enough to support continuous redeployment as data drifts. We present CoRe (Context Relevance), such a system, redeployed weekly for over five months in a major short-video search engine. Our reward uses the deployed multimodal relevance model as its source and a multiplicative ratio form mirroring the production fusion algebra, closing the simulation-production gap that offline reward proxies leave open. A semi-online Mixed Preference Optimization loop makes this reward affordable at multi-million-instance weekly scale: a DPO-style pairwise objective restricts the gradient pass to a small top-k/bottom-k subset of sampled trajectories, and a phase structure reduces trainer/inference-server parameter syncs from per-step to per-phase. An automated promotion gate over reward-like and stability metrics detected and recovered from a real reward-hacking incident in production. Rewriter output is consumed as parallel relevance signals at recall, rawrank, and finerank without displacing the original signals, bounding rewriter-failure blast radius. Online A/B from two sequential production launches, first deploying the rewriter at finerank, then extending consumption to recall and rawrank, delivers statistically significant reductions in change-query rate on rewrite-impacted queries, with all headline relevance and engagement metrics moving in the expected direction.

03.
arXiv (CS.LG) 2026-06-17

MGUP: A Momentum-Gradient Alignment Update Policy for Stochastic Optimization

arXiv:2606.17526v1 Announce Type: new Abstract: Efficient optimization is essential for training large language models. Although intra-layer selective updates have been explored, a general mechanism that enables fine-grained control while ensuring convergence guarantees is still lacking. To bridge this gap, we propose MGUP, a novel mechanism for selective updates. MGUP augments standard momentum-based optimizers by applying larger step-sizes to a selected fixed proportion of parameters in each iteration, while applying smaller, non-zero step-sizes to the rest. As a nearly {plug-and-play} module, MGUP seamlessly integrates with optimizers such as AdamW, Lion, and Muon. This yields powerful variants such as MGUP-AdamW, MGUP-Lion, and MGUP-Muon. Under standard assumptions, we provide theoretical convergence guarantees for MGUP-AdamW (without weight decay) in stochastic optimization. Extensive experiments across diverse tasks, including MAE pretraining, LLM pretraining, and downstream fine-tuning, demonstrate that our MGUP-enhanced optimizers achieve superior or more stable performance compared to their original base optimizers. We offer a principled, versatile, and theoretically grounded strategy for efficient intra-layer selective updates, accelerating and stabilizing the training of large-scale models. The code is publicly available at https://github.com/MaeChd/MGUP.

04.
arXiv (CS.AI) 2026-06-24

Difference-Making without Making a Difference

arXiv:2606.24832v1 Announce Type: new Abstract: Over a series of seven papers, Andreas & Günther have introduced seven definitions of actual causation and have classified them as belonging to three different, competing, types of accounts: factual difference-making, counterfactual difference-making, and regularity-based. I show that their most recent - factual difference-making - definition instantiates all three types, thereby proving that these are distinctions without a difference. I further compare their novel account to the other six accounts on several crucial examples, revealing that this undermines all seven of their accounts.

05.
arXiv (CS.CL) 2026-06-24

Are LLM Evaluators Really Narcissists? Sanity Checking Self-Preference Evaluations

Recent research has shown that large language models (LLMs) favor their own outputs when acting as judges, undermining the integrity of automated post-training and evaluation workflows. However, it is difficult to disentangle which behaviors are explained by narcissism versus experimental confounds. Specifically, LLM evaluators may deliver self-preferring verdicts when comparing responses to questions they fail on; these verdicts may not depend on the identity of the author, but on evaluator quality. We correct this by directly comparing the judge's voting distribution in cases where it evaluates itself versus another model. This evaluator quality baseline reveals that only 51% of examples in previous findings retain statistical significance against this null hypothesis, covering 89.6% of total self-preference probability mass. Finally, we compare the entropy of voting distributions, suggesting uncertainty-driven overlap, and show that our procedure enables more careful documentation against the backdrop of judge-bias research.

06.
arXiv (CS.AI) 2026-06-16

STRIDE: Strategic Trajectory Reasoning via Discriminative Estimation for Verifiable Reinforcement Learning

arXiv:2606.15866v1 Announce Type: new Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has become an effective post-training paradigm for improving the reasoning abilities of large language models. However, existing RLVR methods typically rely on final-answer correctness to assign trajectory-level rewards, providing sparse supervision and treating all tokens uniformly regardless of their actual contribution to reasoning. Although recent studies introduce intermediate signals such as process rewards, high-entropy tokens, and semantic uncertainty, these signals are often not inherently verifiable and may fail to distinguish beneficial strategic patterns from harmful ones. To address this limitation, we propose STRIDE (Strategic Trajectory Reasoning with Discriminative Estimation), a fine-grained RLVR framework that derives strategic reasoning supervision from verifiable outcomes. STRIDE contrasts successful and failed trajectories within each response group to estimate the outcome-discriminative preference of each $n$-gram strategic pattern, and further combines this signal with reasoning saliency entropy to identify decision-relevant strategic patterns. These patterns are assigned differentiated advantage values during RL optimization, enabling more precise credit assignment while preserving the verifiability of RLVR. Extensive experiments demonstrate that STRIDE consistently improves reasoning performance across diverse models, tasks, and extended settings, including VLMs and agent-based systems.

07.
arXiv (CS.LG) 2026-06-15

Towards Steering without Sacrifice: Principled Training of Steering Vectors for Prompt-only Interventions

arXiv:2605.05983v2 Announce Type: replace Abstract: Recently, steering vectors (SVs) have emerged as an effective and lightweight approach to steer behaviors of large language models (LLMs), among which fine-tuned SVs are more effective than optimization-free ones. However, current approaches to fine-tuned SVs suffer from two limitations. First, they require careful selection of steering factors on a per-SV basis to balance steering effectiveness and generation quality at inference time. Second, they operate as full-sequence SVs (FSSVs), which can sacrifice generation quality regardless of factor selection due to excessive intervention on the model generation process. To address the first limitation, we propose joint training of steering factors and directions, such that post-hoc factor selection is no longer required. Using neural network scaling theory, we find that moderately large initialization sizes and learning rates for steering factors are essential for stability and efficiency of joint training. To tackle the second limitation, we draw inspiration from representation fine-tuning and introduce Prompt-only SV (PrOSV), an SV that intervenes only on a few prompt tokens. Our empirical results show that PrOSV outperforms traditional FSSVs on AxBench when using our joint training scheme. We also find that PrOSV achieves a better tradeoff between general model utility and adversarial robustness than FSSV.

08.
medRxiv (Medicine) 2026-06-24

Multifactorial tuberculosis severity score for people living with HIV based on the Rand Appropriateness Method

Background In people living with Human Immunodeficiency Virus (PLWH), tuberculosis (TB) is the leading cause of death and is often associated with substantial morbidity. Better identifying PLWH with severe forms of TB could help target early interventions to reduce mortality and severe morbidity. Existing TB severity assessment tools may be sub-optimal for assessing disease severity in PLWH, since they incompletely integrate key determinants of disease severity. We aimed to develop a consensus-based TB severity score tailored to PLWH. Methods We developed a multifactorial TB severity score (TBSS) for PLWH using a modified Delphi process with a multidisciplinary group of international TB experts as the second part of a RAND/UCLA Appropriateness Method, following a previously published systematic review. Results Eight of 15 invited experts (53%) participated in both Delphi rounds. Of 62 candidate factors, 15 reflecting TB-related characteristics, host-related characteristics as well as characteristics related to both TB and host were rated as having high appropriateness for inclusion in the final TBSS. The total score ranges from 0 (no severity) to 61 (highest severity). Conclusion This study represents a first step towards the development of a multidimensional TB severity assessment tool for PLWH. However, its clinical usefulness, feasibility, and added value compared with existing severity scores remain to be demonstrated through validation studies before routine implementation can be considered. Key words: tuberculosis, HIV, severity.

09.
arXiv (CS.LG) 2026-06-11

PCS-UQ: Uncertainty Quantification via the Predictability-Computability-Stability Framework

arXiv:2505.08784v2 Announce Type: replace-cross Abstract: As machine learning (ML) enters high-stakes domains, trustworthy uncertainty quantification (UQ) is essential for safety. In this paper we introduce PCS-UQ, a framework based on the Predictability, Computability, and Stability (PCS) principles for veridical data science. Starting with a candidate set of models or algorithms, PCS-UQ integrates a rigorous prediction-check to screen out unsuitable models in the set and utilizes bootstrap samples, in order to capture both inter-sample variability and algorithmic instability for the prediction-checked algorithms. We then introduce a novel multiplicative calibration scheme to enhance local adaptivity, which basically corresponds to a new score in conformal prediction. Moreover, we produce a compilation of 17 real-world regression datasets with manually-constructed subgroups. On this benchmark, PCS-UQ maintains the target coverage while outperforming or matching conformal methods equipped with oracle-selected algorithms in interval width. PCS-UQ achieves consistent subgroup coverage, outperforming these oracle-selected conformal methods. Notably, PCS-UQ stands out in achieving both competitive interval widths and consistent subgroup coverage.Across 6 classification datasets, PCS-UQ reduces prediction set sizes by 20\%. To scale the framework for deep learning, we propose computationally efficient variants that bypass expensive retraining. On three computer vision benchmarks, these variants reduce prediction set sizes by 20\% over conformal baselines. Finally, we provide theoretical proof that a modified PCS-UQ algorithm preserves valid coverage under exchangeability as a form of split conformal inference.

10.
arXiv (CS.CL) 2026-06-15

Can Post-Training Turn LLMs into Good Medical Coders? An Empirical Study of Generative ICD Coding

Automated International Classification of Diseases (ICD) coding is a core medical-coding task for billing, epidemiology, and clinical decision support. Generative large language models (LLMs) are often reported as weak medical coders, but this finding mainly comes from inference-time settings such as prompting, retrieval, reranking, or tool use, leaving the role of task-specific post-training underexplored. We present a controlled empirical study of post-training for generative ICD coding, comparing discriminative baselines with LLM coders across prompting, supervised fine-tuning, and reinforcement learning under a common protocol and metric set. To our knowledge, this is the first study to evaluate RL-based post-training for generative LLM coders in ICD coding. We further introduce PHI, a diagnostic curriculum that extends GRPO to refine missed-code cases. Our results show that prompting-only evaluation substantially underestimates the potential of LLMs for ICD coding. SFT provides the main capability jump, GRPO further improves code-set prediction beyond SFT, and PHI provides targeted gains on macro-level performance. These findings suggest that the main bottleneck is not the generative formulation alone, but how the model is adapted and optimized for full-taxonomy recall. We release our code, data splits, and checkpoints at https://github.com/AlexandreWANG915/LLM4ICD.

11.
arXiv (quant-ph) 2026-06-24

The Quantum Split-Step Fourier Algorithm for Nonlinear Optical Waveguides

arXiv:2606.24643v1 Announce Type: cross Abstract: We introduce the Quantum Split-Step Fourier (QSSF) algorithm for nonlinear optical waveguides, a numerical framework that combines split-step propagation of the nonlinear Schrödinger equation with a commutator-preserving Bogoliubov evolution of Gaussian quantum fluctuations. The method propagates the classical mean field together with the Bogoliubov matrices $U$ and $V$, from which reduced second moments, covariance matrices, symplectic eigenvalues, and entropic measures are constructed for arbitrary spectral windows. Applied to soliton-driven resonant radiation, QSSF shows that the selected radiation band acquires a steadily increasing von Neumann entropy and a corresponding loss of purity, quantifying its entanglement with the rest of the spectrum in the lossless Gaussian setting. The analysis also reveals a surprisingly pronounced low-dimensional structure: although the radiation occupies many Fourier bins, its reduced Gaussian state is dominated by only a few Williamson modes. QSSF therefore provides a practical information-theoretic diagnostic for quantum correlations in nonlinear frequency conversion, supercontinuum generation, and multimode squeezed-light formation in ultrafast waveguide platforms.

12.
arXiv (CS.AI) 2026-06-17

Offline Preference-Based Trajectory Evaluation

arXiv:2606.17541v1 Announce Type: cross Abstract: Offline evaluation of agentic systems often collapses trajectories to terminal success, discarding information about partial progress and inducing widespread ties, creating substantial statistical inefficiency by reducing effective sample size and weakening the ability to distinguish systems. We propose preference-based trajectory evaluation, which compares trajectories directly through temporal preferences over progress and time-to-return profiles. We find that, across diverse agentic and interactive benchmarks, standard success-based metrics produce tied comparisons on roughly 75% of instances, whereas trajectory-aware preferences reduce ties to roughly 35%, improving discriminative power, ranking stability, and data efficiency. Our results suggest that benchmark saturation, often attributed to poor data collection or problem difficulty, may also be explained by the choice of evaluation measure.

13.
medRxiv (Medicine) 2026-06-17

Clinician knowledge and self-efficacy in snakebite management: A cross-sectional assessment in Northern Uganda

Background: Snakebite envenomation (SBE) is a major public health crisis in rural Uganda, yet it remains a neglected tropical disease. Effective management is often compromised by systemic barriers and a lack of clinician training. This study assessed clinician self-efficacy and objective knowledge regarding SBE management in Northern Uganda. Methods: A descriptive, cross-sectional study was conducted between February and July 2025 among 379 healthcare workers in Gulu, Omoro, and Pader districts. A validated questionnaire was used to collect data on socio-demographics, self-reported efficacy (scale 1-10), and objective knowledge. Knowledge scores [&ge;]70% were categorized as adequate. Multivariable logistic regression identified independent predictors of adequate knowledge, and Spearmans correlation ({rho}) assessed the relationship between knowledge and self-efficacy. Results: The participants had a mean age of 35.6 years (SD {+/-}7.3), were predominantly female (56.5%, 214/379), and most (83.6%, 317/379) practiced at Health Centre III level facilities. While 53.8% (204/379) reported prior training, 48.3% (183/379) of these had not received an update in over 10 years. Adequate knowledge was demonstrated by 51.5% (195/379) of participants. In the multivariable analysis, practicing in Omoro (adjusted odds ratio [aOR]: 0.3, 95% CI: 0.1-0.6, p < 0.001) or Pader (aOR: 0.2, 95% CI: 0.1-0.4, p < 0.001) was associated with lower odds of adequate knowledge compared to Gulu district. Prior training significantly increased the odds of adequate knowledge (aOR: 2.3, 95% CI: 1.3-4.2, p = 0.006). A moderate positive correlation was observed between self-efficacy and objective knowledge (Spearmans {rho} = 0.33, p < 0.0001). Conclusion: Approximately half of the frontline healthcare workers in Northern Uganda lack adequate knowledge on SBE management, with significant geographic differences and outdated training. The gap between clinician self-efficacy and objective knowledge poses a risk to patient safety. Regular, mandatory refresher training and targeted educational outreach to remote districts are required to reduce SBE-related morbidity and mortality.

14.
arXiv (CS.AI) 2026-06-17

The Discrete-Log Clock: How a Transformer Learns Modular Multiplication

arXiv:2606.17399v1 Announce Type: cross Abstract: When small transformers grok modular multiplication, prior work reports that the learned embedding has a "dense" Fourier spectrum requiring all frequencies. This contrasts with modular addition, where only a sparse set of key frequencies suffices. We show this density is an artifact of analyzing in the wrong basis. The natural Fourier transform for multiplication is not the standard additive DFT but the multiplicative character transform, which decomposes functions on the multiplicative group $(\mathbb{Z}/p\mathbb{Z})^*$ into its irreducible representations. Applying this transform to a grokked transformer trained on $a \cdot b \bmod 113$, we find the embedding spectrum becomes highly sparse (Gini coefficient 0.58 vs. 0.07 in the additive basis) with only 4 key frequencies carrying significant energy. Furthermore, 96.9% of MLP neurons are cleanly tuned to a single multiplicative frequency, and neuron activation heatmaps reveal 2D-periodic structure when reordered by the discrete logarithm. These results demonstrate the transformer reduces multiplication to addition in discrete-log space, implementing a "Discrete-Log Clock" algorithm analogous to Nanda et al.'s Clock algorithm for addition. The methodology generalizes: matching the analysis basis to the algebraic structure of the task reveals interpretable structure where standard tools see noise.

15.
arXiv (CS.CV) 2026-06-24

Ingredient-Level Food Image Segmentation for Nutrition Awareness

Food images often contain several visible ingredients, so assigning one dish label to an entire image hides important visual structure. This work studies ingredient-level semantic segmentation on FoodSeg103, where the model predicts an ingredient class for each pixel. Two SegFormer variants were fine-tuned and evaluated under a controlled setup: SegFormer-B0 as the smaller baseline model and SegFormer-B1 as the larger final model. Both models use ImageNet-pretrained MiT backbones with newly initialized 104-class output layers. On the held-out FoodSeg103 test split of 2,135 images, B0 achieved 0.7709 pixel accuracy and 0.2521 mean IoU, while B1 achieved 0.7929 pixel accuracy and 0.3204 mean IoU. B1 improved every saved test metric, including a +0.0683 absolute gain in mean IoU. The system also converts predicted masks into visible ingredient-area percentages, giving a simple visual composition summary of the predicted meal. This summary can serve as a first-pass nutrition-awareness cue by providing a visual alternative to detailed food tracking similar to plate-based meal guidance, but it is not a direct estimate of calories, macronutrients, food mass, volume, density, or true portion size.

16.
arXiv (CS.CV) 2026-06-17

WeaveLA: Event Driven Cross-Subtask Latent Memory Weaving for Repetitive Robot Manipulation

Vision-Language-Action (VLA) policies have achieved remarkable single-step manipulation, yet they remain brittle precisely where each stage depends on what was just completed. The core issue is structural: short-window VLAs lack an explicit channel for rouxting information across sub-task boundaries, and existing memory-augmented variants either write at every frame, retrieve from demonstration-time stages, or fire at sub-goal events without performing an explicit sub-task-to-sub-task hand-off into the action expert. We identify the sub-goal completion event as the natural temporal unit for cross-subtask memory hand-off, and present WeaveLA (Weave Latent memory for Vision-Language-Action policies), a cross-subtask memory interface that, on top of a frozen VLA backbone, compresses each completed segment into latent tokens via query-driven attention pooling and routes them directly into the action-generation path of the next sub-task. This event-triggered, action-side design preserves the base policy's short-window interface while adding a lightweight cross-subtask channel. Through stratified evaluation on RoboMME with a $\pi_{0.5}$ backbone, WeaveLA's gains land exactly where the channel is needed: on the hardest repetition slice (SwingXtimes, $N{=}3$), success rises from $0\%$ to $47.8\%$, while single-execution episodes remain unchanged. Per-episode paired analysis confirms the gains are confined to tasks whose causal structure requires cross-subtask information.

17.
arXiv (CS.AI) 2026-06-18

DeepInflation: an AI agent for research and model discovery of inflation

arXiv:2601.14288v2 Announce Type: replace-cross Abstract: We present DeepInflation, an AI agent designed for research and model discovery in inflationary cosmology. Built upon a multi-agent architecture, DeepInflation integrates Large Language Models (LLMs) with a symbolic regression (SR) engine and a retrieval-augmented generation (RAG) knowledge base. This framework enables the agent to automatically explore and verify the vast landscape of inflationary potentials while grounding its outputs in established theoretical literature. We demonstrate that DeepInflation can successfully discover simple and viable single-field slow-roll inflationary potentials consistent with the latest observations (with the ACT DR6 results taken as an example) or any given $n_s$ and $r$, and provide accurate theoretical context for obscure inflationary scenarios. DeepInflation serves as a prototype for a new generation of autonomous scientific discovery engines in cosmology, which enables researchers and non-experts alike to explore the inflationary landscape using natural language. This agent is available at https://github.com/pengzy-cosmo/DeepInflation.

18.
arXiv (CS.CL) 2026-06-16

Beyond Layer Importance in Layer-wise Sparsity: An Inter-Layer Perturbation-Absorption Perspective

The considerable layer-wise redundancy in large language models (LLMs) has established non-uniform sparsity allocation across layers as the standard pruning approach for efficient compression. Existing layer-wise allocation methods that estimate allocation strategy from local signals such as activation outliers or weight spectra mainly derive from local layer importance, whereas the final post-pruning performance is also influenced by the network's subsequent compensatory capacity. In this paper, we directly characterize this property through controlled perturbation experiments. We make the following empirical findings. First, layers exhibit highly heterogeneous responses to pruning-scale perturbations. In most cases, early layers amplify perturbations, while middle and late layers actively absorb them, with relative L2 drift decreasing monotonically across depth and direction realigning toward the unperturbed hidden-state trajectory. Second, absorption is a large-perturbation phenomenon. Under small perturbations the network exhibits amplification across all layers, and the transition to absorption occurs smoothly as perturbation magnitude grows to pruning scale. This enriches the linearized accumulation theory underlying related works. Building on these findings, we define an absorption coefficient per layer and propose absorption-aware correction, an orthogonal augmentation that improves OWL and AlphaPruning by reducing perplexity by 7.13% and boosting zero-shot accuracy by 1.02% across multiple model families at 70% sparsity.

19.
arXiv (CS.CV) 2026-06-18

Quantile Transfer for Reliable Operating Point Selection in Visual Place Recognition

Visual Place Recognition (VPR) is a key component for localisation in Global Navigation Satellite System (GNSS)-denied environments, but its performance critically depends on selecting an image matching threshold (operating point) that balances precision and recall. Thresholds are typically hand-tuned offline for a specific environment and fixed during deployment, leading to degraded performance under environmental change. We propose a method that automatically selects the operating point of a VPR system to maximise recall at 100% precision. The method uses a small calibration traversal with known correspondences and transfers thresholds to deployment via quantile normalisation of similarity score distributions. This quantile transfer ensures that thresholds remain stable across calibration sizes and query subsets. Experiments with seven state-of-the-art VPR techniques across five benchmark datasets demonstrate that our proposed approach consistently outperforms existing baselines, enabling the underlying VPR technique to operate at 100% precision in approximately twice as many deployment scenarios (median improvement), while retrieving up to 29% more correct matches at that precision. The method eliminates manual tuning by adapting to new environments and generalising across operating conditions. Our code is available at https://github.com/DhyeyR-007/Quantile-Transfer-for-Reliable-VPR.

20.
arXiv (CS.CL) 2026-06-18

ASyMOB: Algebraic Symbolic Mathematical Operations Benchmark

Large language models (LLMs) are increasingly applied to symbolic mathematics, yet existing evaluations often conflate pattern memorization with genuine reasoning. To address this gap, we present ASyMOB, a high-resolution dataset of 35,368 validated symbolic math problems spanning integration, limits, differential equations, series, and hypergeometrics. Unlike prior benchmarks, ASyMOB systematically perturbs each seed problem using symbolic, numeric, and equivalence-preserving transformations, enabling a fine-grained assessment of generalization. Our evaluation reveals three key findings: (1) most models' performance collapses under minor perturbations, while top systems exhibit an apparent regime shift in robustness; (2) integrated code tools stabilize performance, particularly for weaker models; and (3) we identify examples where Computer Algebra Systems (CAS) fail while LLMs succeed, as well as problems solved only via a hybrid LLM-CAS approach, highlighting a promising integration frontier. ASyMOB serves as a principled diagnostic tool for measuring and accelerating progress toward building verifiable, trustworthy AI for scientific discovery.

21.
arXiv (CS.AI) 2026-06-16

StyleShield: Exposing the Fragility of AIGC Detectors through Continuous Controllable Style Transfer

arXiv:2605.00924v2 Announce Type: replace-cross Abstract: AI-generated content (AIGC) detectors are increasingly deployed in high-stakes settings such as academic integrity screening, yet their reliability rests on a fundamental paradox: as language models are trained on human-written corpora, the statistical boundary between AI and human writing will inevitably dissolve as models improve. Commercial incentives have further distorted this landscape – detection services and "de-AIification" tools often operate within the same supply chain, replacing evaluation of content quality with judgment of content origin. We present StyleShield, the first flow matching framework for conditional text style transfer, operating directly in continuous token embedding space via a DiT backbone with zero-initialized cross-attention adapters conditioned on frozen Qwen-7B representations. At inference, we adapt the SDEdit paradigm from image synthesis to text embeddings, with a single parameter gamma providing smooth continuous control over the evasion-preservation trade-off. On a multi-domain Chinese benchmark, StyleShield achieves 94.6% evasion against the training detector and >=99% against three unseen detectors, maintaining 0.928 semantic similarity. We further introduce RateAudit, a document-level scheduling algorithm that demonstrates detection-rate verdicts can be set to arbitrary values, directly questioning the reliability of score-based evaluation.

22.
arXiv (CS.AI) 2026-06-12

DailyReport: An Open-ended Benchmark for Evaluating Search Agents on Daily Search Tasks

arXiv:2606.12871v1 Announce Type: new Abstract: Search Agents (SAs) typically leverage large language models (LLMs) to support complex information-seeking tasks by autonomously exploring web sources and synthesizing information into comprehensive responses. For SAs evaluation, prior benchmarks mainly focus on specialized tasks that are unlikely to arise in real-world user scenarios. Moreover, their reliance on coarse task-level rubrics often limits evaluation interpretability. To bridge this gap, we introduce DailyReport, an open-ended benchmark to evaluate SA capabilities on daily search tasks. It contains 150 open-ended tasks with 3,546 associated rubrics, capturing widely discussed and timely information demands of real-world users. Each task is decomposed into subtasks and evaluated with cascade rubrics across disentangled dimensions. Through cascade performance attribution and user-centric aggregation, we derive highly interpretable scores for each dimension, along with a user preference score. Our results on 17 agentic systems show that current systems still fall short of users' expectations. To facilitate future research, our dataset and code are made publicly available at https://github.com/AGI-Eval-Official/DailyReport.

23.
arXiv (CS.AI) 2026-06-24

Sensing Intelligence as a Trainable Metamaterial Property

arXiv:2605.23967v2 Announce Type: replace-cross Abstract: In biological systems, sensing is not performed by the brain alone: the body deforms, vibrates, and filters external stimuli before they are transduced into neural signals. In engineered systems, this processing burden is placed largely on electronics and computation, while the mechanical body is usually designed only for strength and stability. Here, we present sensing intelligence as a trainable property of the body. We show that the geometry of a metamaterial can be optimized to reshape external stimuli into internal signals that are easier for a neural network to interpret. Rather than hand-designing this physical preprocessing, we let the neural network train its own body for sensing by backpropagating the sensing loss to the body's design parameters through differentiable simulation. Across numerical and experimental sensing scenarios, the optimized body improves sensing accuracy by up to fivefold or reduces the number of required electronic sensors by nearly an order of magnitude.

24.
arXiv (CS.CL) 2026-06-12

One Token to Fool LLM-as-a-Judge

Large language models (LLMs) are increasingly trusted as automated judges, assisting evaluation and providing reward signals for training other models, particularly in reference-based settings like Reinforcement Learning with Verifiable Rewards (RLVR). However, we uncover a critical vulnerability even in this reference-based paradigm: generative reward models are systematically susceptible to reward hacking. We find that superficial inputs, which we term ''master keys'' such as non-word symbols (e.g., '':'' or ''.'') or generic reasoning openers (e.g., ''Thought process:'' or ''Let's solve this problem step by step.''), can consistently elicit false positive rewards without any substantive reasoning. Our systematic evaluation demonstrates this is a widespread failure affecting a diverse range of models, including leading proprietary systems such as GPT-o1 and Claude-4. These results challenge the assumed robustness of LLM judges and pose a significant threat to their reliability. To address this, we propose a simple yet effective data augmentation strategy using truncated model outputs as adversarial negative examples. The resulting Master Reward Models (Master-RMs) demonstrate state-of-the-art robustness against these ''master key'' attacks while maintaining high performance in standard evaluation settings. We supplement these findings with a comprehensive analysis of the vulnerability across model scales, prompt variations, and common inference-time strategies, offering insights to guide future research on robust LLM evaluation. We release our robust, general-domain reward models and the synthetic training data at https://huggingface.co/sarosavo/Master-RM and https://huggingface.co/datasets/sarosavo/Master-RM.

25.
arXiv (CS.LG) 2026-06-19

Structure-Oriented Randomized Neural Networks for Poisson-Nernst-Planck and Poisson-Nernst-Planck-Navier-Stokes Systems

arXiv:2606.19912v1 Announce Type: cross Abstract: We develop a structure-oriented randomized neural network framework, termed SO-RaNN, for the Poisson-Nernst-Planck (PNP) system and the Poisson-Nernst-Planck-Navier-Stokes (PNP-NS) system. The decoupled linearized subproblems are solved iteratively by randomized neural networks in a space-time framework. For the concentration variables, a pointwise cut-off is used to enforce positivity at the value level, and discrete mass-scaling factors are computed at selected correction instants and interpolated in time, so as to ensure exact mass matching at those instants and to promote approximate mass preservation between them. To introduce an auxiliary discrete dissipation mechanism, we further employ an SAV-type post-processing correction, which yields monotonicity of the SAV auxiliary variable under the ideal SAV update. For the PNP-NS system, a structure-preserving randomized neural network (SP-RaNN) is used for the velocity field, so that the velocity approximation satisfies the incompressibility constraint pointwise by construction. On the theoretical side, we derive residual-based estimates for the raw, uncorrected RaNN solvers of the linearized subproblems, formulate a conditional local-in-time convergence result for the raw outer Picard iteration of the PNP system, and analyze the value-level positivity correction together with the mass-correction and SAV post-processing steps. For the PNP-NS system, we establish an approximation result for the SP-RaNN space and provide a conditional error statement for the corresponding linearized Oseen-type problem. Numerical experiments demonstrate approximation accuracy in the source-driven manufactured tests and illustrate the intended value-level positivity correction, selected-time mass matching, computed free-energy curves based on the final gauge-fixed potential, and divergence-free approximation in benchmark tests.