探索全球前沿学术脉络

01.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2606.19759

Optimal Scheduling in a Question-Answering Forum of Knowledge Workers

作者:

Rohit Negi↗Mustafa Yilmaz↗

arXiv:2606.19759v1 Announce Type: new Abstract: As individuals turn to the Internet to find answers to questions they may have, several Question Answering (QA) forums have evolved, where users knowledgeable in certain topics can contribute their expertise to answering these requests for information. While these are currently volunteer based, we consider a future version employing knowledge workers who are experts in certain topics. In such a system, the request-answer processes forming the queuing system may utilize schedulers that assign requests in different topics to the experts in the forum, who may be able to answer them according to their expertise levels in different topics. With this model, we calculate the capacity of the system for handling the requests while keeping the system stable, and design schedulers that achieve capacity. We also investigate how collaboration between experts in answering requests can potentially increase capacity.

阅读与讨论 → 访问原文 →

02.

Nature Medicine 2026-06-12 DOI: HASH:fde650b09ce972374e2d68e7e4b14d0c

Efficacy and target engagement of dopamine agonist pramipexole for anhedonic depression: a randomized placebo-controlled trial

作者:

Filip Ventorp↗

Anhedonia is a core and disabling symptom of mood disorders with limited treatment options. We evaluated the efficacy and safety of the dopamine agonist pramipexole in patients with mood disorders characterized by clinically significant anhedonia. In this single-center, randomized, double-blind, placebo-controlled trial, adults with major depressive disorder, dysthymia or bipolar depression and elevated Snaith−Hamilton Pleasure Scale (SHAPS) scores were assigned (1:1) to flexible dose, once-daily oral pramipexole as add-on treatment or placebo for 9 weeks. The primary outcome was change in SHAPS score from baseline to week 9. Analyses were conducted in the modified intention-to-treat population. Eighty-five participants were randomized, and 82 were included in the analysis. The primary outcome was met: pramipexole was associated with a greater reduction in SHAPS scores compared to placebo (mean difference: −4.04, 95% confidence interval: −6.89 to −1.18, P = 0.006, Hedges’ g = 0.62). Exploratory analyses indicated that pramipexole was associated with increased light physical activity and relative preservation of reward-related ventral striatal activation. Improvements in anhedonia were sustained during a 6-month open-label extension. Pramipexole was generally well tolerated compared to placebo. Pramipexole significantly improved anhedonia and showed a favorable safety profile, supporting its potential as an augmentation strategy in mood disorders. ClinicalTrials.gov identifiers: NCT05355337 and NCT05825235 . Pramipexole, in patients with major depressive disorder, dysthymia or bipolar depression, reduced Snaith−Hamilton Pleasure Scale scores significantly compared to placebo.

阅读与讨论 → 访问原文 →

03.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.15416

Encode Errors: Representational Retrieval of In-Context Demonstrations for Multilingual Grammatical Error Correction

作者:

Guangyue Peng↗Wei Li↗Wen Luo↗Houfeng Wang↗

Grammatical Error Correction (GEC) involves detecting and correcting the wrong usage of grammar. While large language models (LLMs) with in-context learning (ICL) capabilities have shown significant progress on various natural language processing (NLP) tasks, their few-shot performance on GEC remains suboptimal. This is mainly due to the challenge of retrieving suitable in-context demonstrations that capture error patterns instead of semantic similarity. In this paper, we demonstrate that LLMs can inherently capture information related to grammatical errors through their internal states. From these states, we extract the Grammatical Error Representation (GER), an informative and semantically neutral encoding of grammatical errors. Our novel GER-based retrieval method significantly boosts performance in ICL settings on multilingual GEC datasets, improving the precision of correction. For high-resource languages, our results on 8B-sized open-source models match those of closed-source models such as Deepseek2.5 and GPT-4o-mini. For low-resource languages, our $F_{0.5}$ scores surpass the baseline by up to a factor of 1.20. This method provides a more precise and resource-efficient solution for multilingual GEC, offering a promising direction for interpretable GEC research.

阅读与讨论 → 访问原文 →

04.

arXiv (CS.CV) 2026-06-11 DOI: arXiv:2606.11507

SceneMiner: Identity-Preserving Multi-Task Fine-Tuning for Unified BEV Scene Mining

作者:

Abdalmalek Aburaddaha↗Venkatraman Narayanan↗Keval Thaker↗Samir A. Rawashdeh↗

Mining hard, safety-critical scenes from driving logs is bottlenecked by the absence of difficulty labels, and no single proxy, collision risk, trajectory ambiguity, or semantic rarity suffices to find such scenes on its own. We present SceneMiner, a unified, camera-only bird's-eye-view pipeline that emits complementary mining signals from a frozen vision-language backbone in a single forward pass, with no LiDAR or radar: a retrieval embedding for text-prompted scenario search, a multi-label scene-tag distribution, and a continuous physics-based risk score (a motion forecast is a byproduct, not a contribution). Building such a multi-head model exposes our central finding, a failure mode we term cross-task interference: adding or upgrading one head shifts a shared activation stream and degrades weight-frozen sibling heads, so freezing parameters alone is insufficient. Our contribution, identity-preserving multi-task fine-tuning, removes this interference by zero-initializing every new sub-module and freezing every parameter that feeds the shared stream. The mining heads are thereby preserved bit-identically while training only ~102k parameters. The tagging head reaches mAP 0.4614 (micro-F1 0.5557) on 20 scene tags by pooling each scene into 32 visual tokens, and the embedding head supports text-prompted retrieval, validated qualitatively. Code is available at: https://anonymous.4open.science/r/sceneminer_anonymous-64E5

阅读与讨论 → 访问原文 →

05.

arXiv (CS.CL) 2026-06-15 DOI: arXiv:2606.02320

TVIR: Building Deep Research Agents Towards Text-Visual Interleaved Report Generation

作者:

Xinkai Ma↗Zhiqi Bai↗Dingling Zhang↗Pei Liu↗Yishuo Yuan↗He Zhu↗Jiakai Wang↗Qianqian Xie↗Yifan Zhao↗Xinlong Yang↗Hao Cong↗Zhiheng Yao↗…

Deep Research Agents have shown strong capability in multi-step information retrieval, reasoning, and long-form report generation, but existing benchmarks and systems remain predominantly text-centric, with limited evaluation of whether visual elements are factually reliable and well aligned with the surrounding analysis. To address this gap, we introduce TVIR (Text-Visual Interleaved Report Generation), which includes TVIR-Bench, a benchmark of 100 expert-curated multimodal deep research tasks that require visual elements to serve specific analytical sub-goals, and TVIR-Agent, a hierarchical multi-agent framework that serves as a strong baseline for constructing outlines, retrieving images, generating charts with traceable sources, and composing reports through context-aware sequential writing. We further develop a dual-path evaluation framework that combines Textual Assessment and Visual Assessment. Experiments across nine deep research systems show that TVIR-Agent achieves strong overall performance, underscoring the importance of explicit multimodal design and evaluation for evidence-driven report generation.

阅读与讨论 → 访问原文 →

06.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2606.20264

Confidence-Aware Automated Assessment of Student-Drawn Scientific Models

作者:

Luyang Fang↗Yingchuan Zhang↗Jongchan Park↗Zhaoji Wang↗Ping Ma↗Xiaoming Zhai↗

arXiv:2606.20264v1 Announce Type: new Abstract: Student-generated drawings are widely used in science education to assess learners' conceptual understanding in modeling-based tasks aligned with the Next Generation Science Standards (NGSS). However, scoring such drawings requires expert human judgment to interpret complex visual representations, making large-scale assessment costly to implement and sustain in classroom settings. In this work, we study automated scoring of student-generated scientific drawings using a vision-based model. We evaluate a Vision Transformer (ViT) with parameter-efficient adaptation and propose a confidence-aware scoring framework that derives response-level confidence from test-time predictive distributions. This confidence signal enables selective automation by scoring high-confidence responses automatically while deferring uncertain cases for human review. Experiments on six NGSS-aligned middle school assessment items show that the proposed approach improves scoring reliability while supporting a practical trade-off between automated coverage and scoring risk, highlighting the value of confidence-aware methods for trustworthy educational assessment.

阅读与讨论 → 访问原文 →

07.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2605.02427

The Model Knows, the Decoder Finds: Future Value Guided Particle Power Sampling

作者:

Tu Nguyen↗Matthieu Zimmer↗Rasul Tutunov↗Xiaotong Ji↗Haitham Bou Ammar↗

arXiv:2605.02427v3 Announce Type: replace Abstract: A recurring pattern in "reasoning without training" is that base LLMs already assign non-trivial probability mass to correct multi-step solutions; the bottleneck is locating these modes efficiently at inference time. Power sampling provides a principled way to bias decoding toward such modes by targeting p_theta(x)^alpha with alpha > 1, but practical approximations must account for future-dependent correction factors that determine which prefixes remain promising. We introduce Auxiliary Particle Power Sampling (APPS), a blockwise particle algorithm for approximating the sequence-level power target with a bounded population of partial solutions. APPS propagates hypotheses in parallel using proposal-corrected power reweighting and refines their survival through future-value-guided selection at resampling boundaries. This redistributes finite compute across competing prefixes rather than committing to a single unfolding path, while providing a direct scaling knob in the particle count and predictable peak memory. We instantiate the future-value signal with short-horizon rollouts and also study an amortized variant that replaces rollouts with a lightweight learned selection head. AMore broadly, APPS improves the accuracy–runtime trade-off of training-free decoding, further supporting the view that inference-time power approximation can recover gains often attributed to post-training.

阅读与讨论 → 访问原文 →

08.

medRxiv (Medicine) 2026-06-22 DOI: HASH:b92e52dc024a985fd48c01e35a1a7a0e

Sequential Deep Learning to Predict Non-Central to Central Geographic Atrophy Progression from OCT Imaging

作者:

Siraz↗Kamanda↗Nabil↗A. S↗Gholami↗Rao↗N. T↗Ong↗S. S↗Alam↗

Purpose: To develop and validate a temporal deep learning framework for predicting geographic atrophy (GA) progression across multi-year horizons using longitudinal optical coherence tomography (OCT) sequences. Design: Retrospective longitudinal cohort study. Subjects, Participants, and/or Controls: A total of 91 patients with dry age-related macular degeneration (AMD) were identified from Wake Forest University School of Medicine (2013-2023), yielding 455 OCT volumes. Two prediction cohorts were defined: 32 patients with no GA (NGA) at baseline who subsequently developed GA, and 35 patients whose earliest GA manifestation was non-central GA (NCGA). Non-progressing patients served as negative controls. Methods: OCT B-scan volumes were encoded into visit-level feature representations using three pretrained architectures (ResNet-18, ResNet-50, ViT-B/16). Chronologically ordered visit embeddings, optionally augmented with inter-visit time intervals ({Delta}t), were processed through recurrent neural networks (RNN), long short-term memory networks (LSTM), and Transformer encoders to model longitudinal disease trajectories. Models were trained and evaluated independently for prediction horizons of 2, 3, 4, 5, and 6 years using patient-level stratified splits (80/20). Performance was assessed across five random seeds. Main Outcome Measures: Area under the receiver operating characteristic curve (ROC-AUC), F1-score, and accuracy for predicting two clinically critical transitions: NGA to GA onset and NCGA to central GA (CGA) involvement. Results: For NGA to GA prediction, models achieved ROC-AUC of 0.84-0.94 at 2-4 years and 1.00 at 5-6 years. For NCGA to CGA prediction, Transformer-based models achieved peak AUC of 0.95 at 4 years and 0.96 at 5 years. Longer input sequences (8 visits vs. 4 visits) consistently improved NCGA to CGA performance at extended horizons. Temporal interval encoding improved stability in several LSTM configurations.

阅读与讨论 → 访问原文 →

09.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2605.09169

Prediction Bottlenecks Don't Discover Causal Structure (But Here's What They Actually Do)

作者:

Ankit Hemant Lade↗Sai Krishna Jasti↗Indar Kumar↗Aman Chadha↗

arXiv:2605.09169v2 Announce Type: replace-cross Abstract: A Mamba state-space model trained only for next-step prediction appears to recover Granger-causal structure through a simple readout $S = |W_{out} W_{in}|$, with early experiments suggesting the phenomenon generalized across architectures and benefited from interventional data at $p < 10^{-5}$. We package the protocol used to test that claim – standardized synthetic generators (VAR/Lorenz/CauseMe-style), three intervention semantics ($do(X=c)$, soft-noise, random-forcing), edge-provenance cards on three real datasets, and size-matched control arms – as a reusable falsification benchmark, and walk the claim through it in five stages. The method-level claim does not survive: (i) a plain linear bottleneck does as well or better; (ii) tuned Lasso beats the bottleneck on synthetic CauseMe-style benchmarks, and on Lorenz-96 (the only real benchmark with unambiguous ground truth) classical PCMCI and Granger lead a tight cluster in which the bottleneck trails; (iii) the headline intervention advantage is roughly 60% a sample-size confound, and the residual disappears under standard $do(X=c)$ interventions, surviving only under a non-standard random-forcing scheme; (iv) even that residual reproduces, with a larger effect, in classical bivariate Granger – the effect is method-agnostic. What survives is a narrow characterization result; the benchmark is the lasting artifact, and each stage above is one of its control arms.

阅读与讨论 → 访问原文 →

10.

arXiv (CS.CV) 2026-06-17 DOI: arXiv:2606.17298

Reasoning Text-to-Video Retrieval for Operating Room Clips via Action-Driven Digital Twins

作者:

Yiqing Shen↗Hao Ding↗Mathias Unberath↗

Text-to-video retrieval in operating rooms (OR) is an enabling technology for OR safety, as it allows stakeholders to retrieve and inspect recordings of specific events. However, because the most safety-critical events may not follow the common structure, to unlock its full potential text-to-video retrieval must be able to handle implicit queries that require reasoning to identify the right video (e.g., the step right before clipping). However, existing methods rely on global embeddings that cannot reason over such queries. We propose OR3, a text-to-video retrieval method that converts clips into action-driven digital twins (ActDTs), grouping concurrent subject-action-object triplets under non-overlapping temporal intervals. Moreover, rather than cross-modal matching through paired encoders, OR3 performs imagination-based retrieval where an LLM generates hypothetical ActDTs from queries. This enables intra-modal matching via a single encoder trained with ActDT-tailored hard negatives. Finally, evidence-grounded refinement revises imagined ActDTs based on discrepancies with top candidates to capture procedure-specific patterns. We construct a benchmark from MM-OR with 276 implicit queries across four reasoning categories over 386 clips from robotic knee procedures. OR3 achieves 57.6 R@1 and 77.3 R@5, outperforming the strongest baseline. These results demonstrate that OR3 enables fine-grained discrimination between visually similar OR video clips through temporal action reasoning.

阅读与讨论 → 访问原文 →

11.

arXiv (CS.CV) 2026-06-24 DOI: arXiv:2606.24335

Ill-Posed by Design: Probing Evidence Use in VLMs

作者:

Boaz Meivar↗Shaked Perek↗Shani Shvartzman↗Eli Schwartz↗Shai Avidan↗

Counterfactual analysis is widely used to study evidence use in vision-language models, but its diagnostic value is limited on well-posed tasks: when several cues independently support the same answer, removing one may not change the prediction. We propose monocular metric object-size estimation as an ill-posed diagnostic setting for evidence selection: because physical size cannot be determined from a single uncalibrated image, models must rely on imperfect cues category priors, target appearance, local context, apparent image size, and scene geometry. We assemble Metric VQA ($10{,}813$ dimension queries from Objectron and $331$ tape-measured in-the-wild scenes) and evaluate $12$ open-weight VLMs ($3$–$397$\,B parameters) with counterfactual analysis decomposing six visual and language evidence channels. Even the largest VLMs tested (Qwen3-VL-235B, Qwen3.5-397B, InternVL3.5-241B) trail a text-only frontier LLM on the in-the-wild split. The diagnostic analysis shows: target identity is the most load-bearing cue, target pixels and local context help only some models, apparent size shifts predictions without a directional readout, and global scene geometry is largely unused. We analyze LoRA fine-tuning as an actionable intervention specific to metric estimation: while the task is learnable, the models do not learn to leverage scene geometry.

阅读与讨论 → 访问原文 →

12.

arXiv (CS.AI) 2026-06-24 DOI: arXiv:2603.20750

Subjective-Graph LLM Agents for Simulating Uncertainty in Classroom Social Perception

作者:

Jinming Yang↗Xinyu Jiang↗Xinshan Jiao↗Xinping Zhang↗

arXiv:2603.20750v2 Announce Type: replace Abstract: Social actors do not observe a common social world: each individual forms judgments from a partial and potentially distorted view of the surrounding network. We study whether graph-local evidence and credibility-weighted communication can generate persistent distortions in perceived academic standing, even when agents repeatedly receive objective performance signals. We introduce a data-constrained multi-agent framework in which LLM agents operate through individualized subjective graphs that determine peer visibility, evidence access, and interaction opportunities. Agents exchange uncertainty-annotated assessments, evaluate message credibility, and maintain explicit Gaussian belief states updated through Bayesian fusion. We evaluate the framework on 12 middle-school classrooms comprising 482 students, using questionnaire-derived social information and six consecutive examinations. On the Social-Observed subset (n=419), collective ranking error increases from 0.066 \pm 0.008 to 0.124 \pm 0.009 across six epochs despite repeated exam-based anchoring. Ablations associate individualized visibility and LLM-based trust gating with more stable long-horizon behavior, while constrained retrieval primarily safeguards against global-information leakage. Compared with evaluated DeGroot configurations, the proposed framework achieves lower final ranking error; those DeGroot configurations exhibit near-zero terminal opinion diversity. These findings establish subjective-graph LLM agents as a mechanism-oriented framework for data-constrained simulated social perception. Code is available at https://anonymous.4open.science/r/Rashomonomon-0126.

阅读与讨论 → 访问原文 →

13.

arXiv (CS.AI) 2026-06-18 DOI: arXiv:2606.18820

Maturing Markov Decision Processes: Decision Making under Increasing Information and Shrinking Action Sets

作者:

Jiaxi Liu↗Aiping Yang↗Yuhang Yang↗Shuqi Zhang↗Zewei Dong↗Jiangming Yang↗Xuebin Chen↗

arXiv:2606.18820v1 Announce Type: cross Abstract: Sequential decision problems often exhibit an asymmetric evolution of information and decision flexibility: as a decision cycle unfolds, the agent receives richer information while feasible actions expire due to operational cutoffs, commitments, or resource constraints. Standard MDP formulations typically flatten this structure into stage-dependent state descriptions and action masks, thereby obscuring the nested information–action asymmetry that determines which decisions are urgent and which can be deferred. We introduce Maturing Markov Decision Processes (MMDPs), a formulation built around this information–action asymmetry. We characterize one of its key consequences through an expiring-action priority principle, which identifies the actions that must be resolved before the next stage. Motivated by this structure, we develop a structure-aware reinforcement learning framework with stage-aware policy design, expiring-action abstraction, and search-augmented learning with distillation. Experiments on a controlled multi-supplier replenishment problem, simplified cash-management environments of increasing complexity, and a production-scale simulator show that explicitly modeling this asymmetry improves learning efficiency and becomes increasingly valuable as decision problems scale.

阅读与讨论 → 访问原文 →

14.

arXiv (CS.AI) 2026-06-11 DOI: arXiv:2606.07537

From Architecture to Output: Structural Origins of Hallucination in Large Language Models and the Amplifying Role of Data

作者:

Md. Rejaul Korim Sadi↗Toufiqur Rahman Tasin↗Golam Mostofa Naeem↗

arXiv:2606.07537v1 Announce Type: cross Abstract: Large language models hallucinate–producing fluent, confident, factually wrong outputs–with a consistency that persists across generations and scales. Existing taxonomies classify hallucination by output type, distinguishing intrinsic from extrinsic failures and faithfulness from factuality divergence. These frameworks are descriptively rigorous but do not identify which internal mechanism produced a given instance. This paper analyses hallucination as a structural consequence of three architectural decisions that together form a compound failure system. Self-attention's co-occurrence learning substitutes statistical proximity for semantic meaning and produces entity confusion, fact misattribution, and semantic drift. The maximum likelihood estimation training objective optimises next-token probability without factual constraint, rewarding statistically plausible outputs regardless of their truth value. Autoregressive decoding's permanent left-to-right commitment under exposure bias ensures that a single wrong token cascades forward through the entire output sequence without revision. Dataset pathologies–long-tail deficiencies, training bias, and synthetic pollution–amplify these vulnerabilities but do not independently cause them. We make three contributions. First, we map each mechanism to a specific output category in the Alansari and Luqman taxonomy, locating intrinsic hallucination in self-attention, extrinsic hallucination in MLE, and logical inconsistency in autoregressive decoding. Second, we show that each commonly cited dataset pathology exploits one of these mechanisms rather than originating hallucination independently. Third, we identify the diagnostic limitation of output-type-only classification and contrast it with inference-layer mitigation approaches.

阅读与讨论 → 访问原文 →

15.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2512.20014

Bring My Cup! Personalizing Vision-Language-Action Models with Visual Attentive Prompting

作者:

Sangoh Lee↗Sangwoo Mo↗Wook-Shin Han↗

arXiv:2512.20014v3 Announce Type: replace-cross Abstract: While Vision-Language-Action (VLA) models generalize well to generic instructions, they struggle with personalized commands such as "bring my cup," where the robot must act on one specific instance among visually similar objects. We study this setting of manipulating personal objects, in which a VLA must identify and control a user-specific object unseen during training using only a few reference images. To address this challenge, we propose Visual Attentive Prompting (VAP), a simple-yet-effective training-free perceptual adapter that equips frozen VLAs with top-down selective attention. VAP treats the reference images as a non-parametric visual memory, grounds the personal object in the scene through open-vocabulary detection and embedding-based matching, and then injects this grounding as a visual prompt by highlighting the object and rewriting the instruction. We construct two simulation benchmarks, Personalized-SIMPLER and Personalized-VLABench, and a real-world tabletop benchmark to evaluate personalized manipulation across multiple robots and tasks. Experiments show that VAP consistently outperforms generic policies and token-learning baselines in both success rate and correct-object manipulation, helping to bridge the gap between semantic understanding and instance-level control.

阅读与讨论 → 访问原文 →

16.

arXiv (CS.LG) 2026-06-15 DOI: arXiv:2606.13799

The Program Is Still There: A Conservation Law for Program Discovery

作者:

Jorge Miguel Silva↗

arXiv:2606.13799v1 Announce Type: cross Abstract: Finding the shortest program that generates a sequence is uncomputable, and for six decades that fact has been mistaken for a wall around finding any generating program. It is not a wall but a price, and this paper measures it. For every algorithm that learns about a candidate program only through its score, a class spanning Levin search, evolutionary methods, simulated annealing, and the cross-entropy method, we define the coupling width of a search problem and prove an unconditional worst-case lower bound, exponential in that width with base one less than the domain size. From it follows a conservation law: structural knowledge injected into a search trades one for one against the search it removes, and their sum can never fall below the length of the program sought. Levin's 1973 upper bound and the lower bound proved here are the two ends of one conserved quantity, closing on each other as the instruction set grows. The only escape is to read a candidate's structure rather than its score, and its price, which we prove for generic targets, is incompleteness. A deterministic engine built on this theory recovers a generating program, certified by compressing its data and predicting an unseen continuation, for 2,383 of 3,914 sequences across four independent populations, including 244 of the 256 elementary cellular automata, with measured discovery cost rising along program length more than an order of magnitude inside the score-oracle worst case.

阅读与讨论 → 访问原文 →

17.

arXiv (CS.CL) 2026-06-19 DOI: arXiv:2606.19719

Closing the Calibration Gap in Semantic Caching

作者:

Aditeya Baral↗Radoslav Ralev↗Iliya Sotirov Zhechev↗Srijith Rajamohan↗Jen Agarwal↗

Semantic caching cuts LLM inference costs by serving a cached response to semantically similar queries. Standard practice evaluates these systems using PR-AUC, a metric that only measures how well scores rank and ignores whether they are usable at a fixed threshold. We show this mismatch leads to systematically poor deployment choices, as models with the highest PR-AUC are often the worst in operation. We introduce Precision-Cache Hit Ratio (P-CHR) AUC, a cache-aware metric that measures precision across cache utilization levels, and Calibration Retention Rate (CRR), which captures how much offline ranking quality survives at deployment. We decompose the operational gap between offline and deployed quality into a recoverable calibration component and an irreducible structural component fixed by the dataset's positive rate. Our experiments show that the calibration gap is governed by the training objective rather than data scale, and post-hoc calibration only partially closes it. Ultimately, model selection for semantic caching is a calibration problem, not a ranking one, and measuring it is the first step to closing the gap.

阅读与讨论 → 访问原文 →

18.

arXiv (CS.LG) 2026-06-24 DOI: arXiv:2111.05385

Subtyping patients with chronic disease using longitudinal BMI patterns

作者:

Md Mozaharul Mottalib↗Jessica C Jones-Smith↗Bethany Sheridan↗Rahmatollah Beheshti↗

arXiv:2111.05385v3 Announce Type: replace Abstract: Obesity is a major health problem, increasing the risk of various major chronic diseases, such as diabetes, cancer, and stroke. While the role of obesity identified by cross-sectional BMI recordings has been heavily studied, the role of BMI trajectories is much less explored. In this study, we use a machine-learning approach to subtype individuals' risk of developing 18 major chronic diseases by using their BMI trajectories extracted from a large and geographically diverse EHR dataset capturing the health status of around two million individuals for a period of six years. We define nine new interpretable and evidence-based variables based on the BMI trajectories to cluster the patients into subgroups using the k-means clustering method. We thoroughly review each cluster's characteristics in terms of demographic, socioeconomic, and physiological measurement variables to specify the distinct properties of the patients in the clusters. In our experiments, the direct relationship of obesity with diabetes, hypertension, Alzheimer's, and dementia has been re-established and distinct clusters with specific characteristics for several of the chronic diseases have been found to be conforming or complementary to the existing body of knowledge.

阅读与讨论 → 访问原文 →

19.

Science (Express) 2026-05-07 DOI: 10.1126/science.aec8514

TranscriptFormer: A generative cell atlas across 1.5 billion years of evolution | Science

作者: 未知作者

Single-cell transcriptomics is revolutionizing our understanding of cellular diversity, yet comparing transcriptional programs across the tree of life remains challenging. We developed TranscriptFormer, a family of generative foundation models trained on up to 112 million cells spanning 1.53 billion years of evolution across 12 species. We demonstrate state-of-the-art performance on cell type classification, even for species separated over 685 million years of evolution, and zero-shot disease state identification in human cells. Developmental trajectories, phylogenetic relationships and cellular hierarchies emerge naturally in TranscriptFormer’s representations without any explicit training on these annotations. This work establishes a powerful framework for quantitative single-cell analysis and comparative cellular biology, thus demonstrating that universal principles of cellular organization can be learned and predicted across the tree of life.

阅读与讨论 → 访问原文 →

20.

arXiv (CS.AI) 2026-06-15 DOI: arXiv:2605.07984

Where's the Plan? Locating Latent Planning in Language Models with Lightweight Mechanistic Interventions

作者:

Nicole Ma↗Nick Rui↗

arXiv:2605.07984v2 Announce Type: replace-cross Abstract: We study planning site formation in language models – where internal representations of structurally-constrained future tokens form during the forward pass, and whether they causally drive generation. Using rhyming-couplet completion as a clean test of forward-looking constraint, we apply two lightweight methods (linear probing and activation patching) across Qwen3, Gemma-3, and Llama-3 at more than ten scales. Probing shows that future-rhyme information is linearly decodable at the line boundary, with signal that strengthens with scale in all three families. Activation patching reveals that only Gemma-3-27B causally relies on this encoding, exhibiting a handoff in which the causal driver migrates from the rhyme word to the line boundary around layer 30. Every other model we test conditions on the rhyme word throughout generation, with near-zero causal effect at the line boundary despite strong probe signal. We localize the Gemma-3-27B handoff to five attention heads through two-stage path patching that recover ~90% of the rhyme-routing capacity at the newline.

阅读与讨论 → 访问原文 →

21.

arXiv (CS.CV) 2026-06-18 DOI: arXiv:2606.18441

Reasoning as Intersection: Consensus-Frame Alignment for Visual Focus in Video-MLLMs

作者:

Chengwen Liu↗Zhe Huang↗Jisheng Dang↗Hong Peng↗Qi Tian↗Tat-Seng Chua↗

Reinforcement learning has improved the reasoning ability of large language models, but applying outcome-only rewards to video multimodal large language models (Video-MLLMs) provides limited guidance on which visual evidence should support the answer. Inspired by multisensory integration, where consistent cues can enhance the salience and reliability of perceptual estimates, we introduce Consensus Frame GRPO (CF-GRPO), a temporal-annotation-free process-level reward framework for evidence-aware video reasoning. CF-GRPO constructs a consensus frame prior from intrinsic video cues, including temporal coverage, scene-transition cues, and query-conditioned visual relevance. It then computes a model-side frame-use score from visual and response representations and optimizes their agreement through the Consensus Frame Reward (CFR). With salience-aware sparse aggregation and distribution sharpening, CFR provides a high-contrast reward signal without requiring human temporal annotations. Experiments show that VideoCFR achieves competitive performance across complex video reasoning benchmarks and improves several metrics over representative Video-MLLM and RL baselines, while the consensus prior provides an interpretable view of the evidence frames emphasized during training. The implementation is available at https://github.com/1Pansy/VideoCFR.

阅读与讨论 → 访问原文 →

22.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2602.04396

LoRDO: Distributed Low-Rank Optimization with Infrequent Communication

作者:

Andrej Jovanovi\'c↗Alex Iacob↗Mher Safaryan↗Ionut-Vlad Modoranu↗Lorenzo Sani↗William F. Shen↗Xinchi Qiu↗Dan Alistarh↗Nicholas D. Lane↗

arXiv:2602.04396v2 Announce Type: replace-cross Abstract: Distributed training of foundation models via $\texttt{DDP}$ is limited by interconnect bandwidth. While infrequent communication strategies reduce synchronization frequency, they remain bottlenecked by the memory and communication requirements of optimizer states. Low-rank optimizers can alleviate these constraints; however, in the local-update regime, workers lack access to the full-batch gradients required to compute low-rank projections, which degrades performance. We propose $\texttt{LoRDO}$, a principled framework unifying low-rank optimization with infrequent synchronization. We first demonstrate that, while global projections based on pseudo-gradients are theoretically superior, they permanently restrict the optimization trajectory to a low-rank subspace. To restore subspace exploration, we introduce a full-rank quasi-hyperbolic update. $\texttt{LoRDO}$ achieves near-parity with low-rank $\texttt{DDP}$ in language modeling and downstream tasks at model scales of $125$M–$720$M, while reducing communication by $\approx 10 \times$. Finally, we show that $\texttt{LoRDO}$ improves performance even more in very low-memory settings with small rank/batch size.

阅读与讨论 → 访问原文 →

23.

arXiv (CS.AI) 2026-06-15 DOI: arXiv:2604.14892

Can LLMs Accurately Score Medical Diagnoses and Clinical Reasoning?

作者:

Amy Rouillard↗Sitwala Mundia↗Linda Camara↗Ziyaad Dangor↗Michael Cameron Gramanie↗Ismail Kalla↗Shabir A. Madhi↗Kajal Morar↗Marlvin T. Ncube↗Haroon Saloojee↗Bruce A. Bassett↗

arXiv:2604.14892v3 Announce Type: replace-cross Abstract: Evaluating medical AI systems using expert clinician panels is costly and slow, motivating the use of large language models (LLMs) as alternative adjudicators. Here, we evaluate an LLM Jury, composed of three frontier AI models, for scoring 3334 diagnoses on 300 real-world low- and middle-income country (LMIC) hospital cases. Both LLM- and clinician-generated diagnoses are scored against expert panel diagnoses across four dimensions: diagnosis, differential diagnosis, clinical reasoning, and negative treatment risk. The LLM Jury scores are compared with expert and independent re-scoring panel scores to assess error metrics, inter-rater agreement, severe-risk errors, and the effect of post hoc calibration using isotonic regression. In our data, we find that: (i) the uncalibrated LLM Jury scores preserve ordinal agreement with the expert clinician panel scores, but are systematically lower; (ii) the probability of severe-risk errors is lower for the LLM Jury than the human expert re-score panels; (iii) the LLM Jury combined with LLM diagnoses can be used to identify diagnoses at high risk of error, enabling targeted expert review and improved panel efficiency; (iv) the calibrated LLM Jury scores and rankings of diagnosing agents show excellent agreement with those of the primary expert panels; (v) LLM Jury models show no self-preference bias, they did not score diagnoses generated by their own underlying model or models from the same vendor more (or less) favourably than those generated by other models. Together, these results provide evidence that a calibrated LLM Jury is a trustworthy and reliable proxy for expert clinician evaluation in medical AI benchmarking. Confirming these findings in other clinical settings is an important direction for future work.

阅读与讨论 → 访问原文 →

24.

arXiv (CS.CV) 2026-06-15 DOI: arXiv:2606.14383

IndustryBench-MIPU: Benchmarking Multi-Image Attribute Value Extraction for Industrial Products

作者:

Haonan Qi↗Jin Cao↗Yongqi Zhang↗Xintong Wang↗Weidong Tang↗Bin Chen↗Chengfu Huo↗Haojun Pan↗Hengyu You↗Jing Li↗Yingde Wang↗Liang Ding↗…

Industrial products such as valves and circuit breakers are defined by dense technical specifications that govern procurement, compatibility, and safety across supply chains. These specifications are scattered across multiple heterogeneous product images, including specification tables, nameplates, and technical drawings, yet whether Multimodal Large Language Models (MLLMs) can reliably recover them remains underexplored. To fill this gap, we introduce IndustryBench-MIPU, the first large-scale benchmark for multi-image industrial product understanding, built around structured attribute extraction – recovering property-value pairs from product images. This task jointly probes text recognition on specification tables and nameplates, visual reasoning over technical drawings, domain knowledge to decode industrial terminology, and cross-image evidence integration to assemble scattered specifications. Concretely, the benchmark comprises 4,559 products across 27,652 images with 103,703 annotations spanning 18 industrial categories, constructed through multi-model consensus and three-tier quality assurance. Evaluating nine MLLMs under both single-image and product-level multi-image settings reveals a stark completeness gap: models achieve high precision (86–94%) but the best recovers only 49.9% of product-level attributes; moving from single-image to multi-image extraction costs 15–34 percentage points of recall. Multi-image completeness, not single-image accuracy, is the core bottleneck. Dataset and code are publicly available.

阅读与讨论 → 访问原文 →

25.

arXiv (quant-ph) 2026-06-15 DOI: arXiv:2605.20443

Interpreting Bohm-like quantum potentials in "Computing quantum waves exactly from classical action"

作者:

Winfried Lohmiller↗Jean-Jacques Slotine↗

arXiv:2605.20443v3 Announce Type: replace Abstract: The recent posting arXiv:2605.02621 [14], commenting on the article rspa.2025.0413 [7], argues that the proof of Lemma 3.1 in [7] is missing the spatial derivative of the density, which would lead to a Bohm-like quantum potential. This technical note shows why the propagated density is independent of space in the Feynman propagator construction of Lemma 3.1. This is done by extending the proof of Lemma 3.1 explicitly with Bohm-like quantum potential terms along the stationary action paths, and then showing that these terms are exactly zero. In [7], this property can also be verified directly on most examples (double slit, Aharonov-Bohm, potential well, harmonic oscillator, tunneling, EPR, QED), as well as in the derivations of the Pauli, Dirac, and Maxwell equations. For more general nonlinear actions, a time rescaling may be required to guarantee this space independence along stationary paths. In the hydrogen atom example, this time rescaling can be computed in closed form. In contrast to the general wave of the Madelung solution [9] Lemma 3.1 of [7] is defined first for a propagator, and a general wave is then constructed in a second step. Recall that a propagator is a specific quantum wave, which is initialized at $t=0$ with a Dirac impulse at a given initial position or momentum. In turn, a general wave is constructed in a second step by superposing a distribution of initial conditions using the propagator. This key difference is why the Bohm-like quantum potential terms disappear in the construction [7] (specifically, in the first step) while the Bohm potential in the Madelung analysis does not. This fundamental difference is also consistent with the fact that the wave construction in [7] extends naturally to relativistic contexts, while Bohmian non-locality notoriously prevents such extensions. Keywords - Response to arXiv:2605.02621, in relation to rspa.2025.0413

阅读与讨论 → 访问原文 →