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 (math.PR) 2026-06-25

Gaussian rigidity for infinite exchangeable sequences

作者:
Yushu Zheng↗Qi Zhou↗

arXiv:2606.25976v1 Announce Type: new Abstract: We prove a Gaussian rigidity theorem for infinite exchangeable sequences of real-valued random variables: the joint Gaussianity of a single pair of entries already forces the entire sequence to be a Gaussian process. This settles a conjecture raised by Newman (2026). The main analytic ingredient in the proof is Hardy's uncertainty principle. We also obtain a finite-dimensional vector-valued extension.

阅读与讨论 → 访问原文 →
02.
arXiv (CS.CV) 2026-06-16

No One Knows the State of the Art in Geospatial Foundation Models

作者:
Isaac Corley↗Nils Lehmann↗Caleb Robinson↗Gabriel Tseng↗Anthony Fuller↗Hamed Alemohammad↗Evan Shelhamer↗Jennifer Marcus↗Hannah Kerner↗

Geospatial foundation models (GFMs) have been proposed as generalizable backbones for disaster response, land-cover mapping, food-security monitoring, and other high-stakes Earth-observation tasks. Yet the published work about these models does not give reviewers or users enough information to tell which model fits a given task. We argue that nobody knows what the current state of the art is in geospatial foundation models. The methods may be useful, but the GFM literature does not standardize evaluations, training and testing protocols, released weights, or pretraining controls well enough for anyone to compare or rank them. In a 152-paper audit, we find 46 cross-paper disagreements of at least 10 points for the same model, benchmark, and protocol; 94/126 papers with extractable pretraining data use a configuration no other paper uses; and 39% of GFM papers release no model weights. This lack of community standards can be solved. We propose six concrete expectations: named-license weight release, shared core evaluations, copied-versus-rerun baseline annotations, variance reporting, one shared evaluation harness, and data-vs-architecture-vs-algorithm controls. These gaps are a coordination failure, not a fault of any individual lab; the authors of this paper, like many others in the GFM community, have contributed to them. Rather than just critiquing the community, we aim to provide concrete steps toward a shared understanding of how to innovate GFMs.

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

An Information-Theoretic Analysis of Threshold Group Testing

作者:
Remco van der Hofstad↗Noela M\"uller↗Connor Riddlesden↗

arXiv:2606.11353v1 Announce Type: cross Abstract: We study the Threshold Group Testing (TGT) problem in the noiseless and non-adaptive setting, where the objective is to exactly recover a sparse binary vector from pooled tests, using as few tests as possible. In TGT, each test applied to a subset of items returns a positive outcome if the number of 1's (defective items) in that subset meets or exceeds a specified threshold, and has a negative outcome otherwise. We investigate how the complexity of TGT compares to that of Classical Group Testing (CGT), corresponding to the special case of the threshold equal to one, and analyse the impact of increasing the threshold on the required number of tests. Our main contribution is the derivation of a sharp information-theoretic phase transition at $c_{\mathrm{inf}}^{\mathrm{TGT}}k\log(n/k)$ (non-adaptive) tests for TGT within the constant-column test design. The threshold constant $c_{\mathrm{inf}}^{\mathrm{TGT}}$ is expressed as a function of the prevalence of defectives and the threshold value. Our upper bound is derived under an analytic assumption, and we verify that this assumption is satisfied for a threshold value of 2. The value of $c_{\mathrm{inf}}^{\mathrm{TGT}}$ reveals that TGT on the constant-column design has the same information-theoretic behaviour as CGT in the low-prevalence regime. Yet, strikingly, at higher prevalences, the threshold leads to a significant reduction in the number of tests. On the other hand, we provide evidence that when the asymptotic proportion of defective items is positive, TGT actually becomes strictly harder than CGT (excluding trivial reductions).

阅读与讨论 → 访问原文 →
04.
arXiv (CS.CL) 2026-06-24

The African Language Tax: Quantifying the Cost, Latency, and Context Penalty of Tokenizing African Languages in Frontier LLMs

作者:
Olaoye Anthony Somide↗

Commercial large language models bill, scale latency, and budget context per token. Yet tokenizers assign more subword tokens to the same meaning in some languages than in others, so speakers of languages with high token-fertility pay a structural penalty before a model is ever invoked. This penalty is documented for multilingual settings in general, but it has not been measured systematically for African languages at the level of enterprise deployment economics and cognitive context capacity. We measure it across 20 African languages spanning five language families and three scripts (Latin, Ge'ez/Ethiopic, N'Ko; 19 appear in the primary FLORES-200+ corpus, with Nigerian Pidgin measured via MAFAND-MT only), using parallel corpora so that the language effect is isolated from content. Across 11 frontier and open tokenizers on FLORES-200+, every African language carries a tokenization premium above English (median 1.88x on GPT-5 / o200k_base, up to 8.92x for N'Ko); the penalty is largest for Ethiopic and N'Ko scripts (reaching 7-9x) and is near-invariant across corpora (FLORES vs SIB-200 Pearson r = 0.9998). Translated into deployment terms, this results in up to 8.9x inference cost and an equivalent generation-latency multiplier (N'Ko vs English on GPT-5; 7.4x for Amharic), and as little as 11% of English's effective context window. The best currently available tokenizer for African languages, Gemma 4, reduces the mean premium from 3.31x (cl100k_base) to 2.38x, but no tokenizer eliminates the penalty. We release an open measurement tool (afri-fertility), a public leaderboard, a results dataset, and mitigation guidance for African builders. The penalty falls hardest on the languages whose speakers can least afford it, a digital divide encoded directly into the subword vocabulary.

阅读与讨论 → 访问原文 →
05.
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 →
06.
Science (Express) 2026-05-07

Induction of broadly neutralizing HIV antibodies by a two-step mechanism informs vaccine design | Science

作者: 未知作者

A major obstacle confronting HIV-1 vaccine and cure research is the lack of an outbred animal model for rapid and consistent induction of broadly neutralizing antibodies (bNAbs). We designed an epitope-focused simian-human immunodeficiency virus (SHIV.5MUT) that elicited broad and potent V3-glycan-targeted antibodies within a year of infection in 14 of 22 macaques compared with 0 of 14 control animals. SHIV.5MUT elicited bNAbs by a two-step mechanism, inducing an initial wave of V1-directed antibodies that selected for Envs with shortened, hypoglycosylated V1 loops, which in turn primed V3-glycan bNAb precursors. Rhesus bNAbs were immunogenetically and structurally diverse, closely resembling human V3-glycan bNAbs. Env-bNAb coevolution revealed a diverse repertoire of bNAb precursors and the Env variants that matured them, yielding a molecular blueprint for vaccine design.

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

Decision-Aware Memory Cards: Counterfactual-Inspired Context Selection and Compression for Tool-Using LLM Agents

作者:
Xinyu Guan↗Qianyang Zhao↗Yuming Deng↗

arXiv:2606.08151v2 Announce Type: replace Abstract: Modern large language model (LLM) agents do not simply need longer contexts; they need decision-relevant evidence at the moment of action. We study decision-aware context selection: ranking retrieved files, tests, traces, rules, and memories by their expected effect on an agent's next action rather than by semantic similarity alone. We present the Counterfactual-Inspired Context Layer (CICL), which builds an instance context graph, estimates decision-oriented utility for candidate units, and compresses selected evidence into typed memory cards. The same schema can be instantiated with hosted LLM judges, local surrogates, or lightweight rankers, making the selection protocol auditable across model choices. On 50 SWE-bench Verified file-retrieval instances, Qwen3.6-Plus reranking of BM25 top-50 candidates improves hit@1 from 0.58 to 0.78 and MRR@10 from 0.634 to 0.790, with all 2,500 judgments parseable. Controlled diagnostics show that CICL identifies action-critical evidence: removing the top-utility semantic unit reduces F1 from 0.245 to 0.000. In selected-then-compressed mode, memory cards save 44.93 tokens per query while preserving selected evidence. CICL provides a practical layer for measuring, ranking, and compressing decision-critical context for tool-using agents. Code is available at https://github.com/stephen-guan-researcher/CICL.

阅读与讨论 → 访问原文 →
08.
arXiv (CS.AI) 2026-06-19

Mitigating Simplicity Bias in OOD Detection through Object Co-occurrence Analysis

作者:
Boyang Dai↗Chaoqi Chen↗Yizhou Yu↗

arXiv:2605.07821v2 Announce Type: replace-cross Abstract: Out-of-distribution (OOD) detection is crucial for ensuring the reliability of deep learning models. Existing methods mostly focus on regular entangled representations to discriminate in-distribution (ID) and OOD data, neglecting the rich contextual information within images. This issue is particularly challenging for detecting near-OOD, as models with simplicity bias struggle to learn discriminative features in disentangled representations. The human visual system can use the co-occurrence of objects in the natural environment to facilitate scene understanding. Inspired by this, we propose an Object-Centric OOD detection framework that learns to capture Object CO-occurrence (OCO) patterns within images. The proposed method introduces a new OOD detection paradigm that understands object co-occurrence within an image by predicting disentangled representations for the test sample, then adaptively divides patterns into three scenarios based on object co-occurrence patterns observed in ID training data, and finally performs OOD detection in a divide-and-conquer manner. By doing so, OCO can distinguish near-OOD by considering the semantic contextual relationships present in their images, avoiding the tendency to focus solely on simple, easily learnable regions. We evaluate OCO through experiments across challenging and full-spectrum OOD settings, demonstrating competitive results and confirming its ability to address both semantic and covariate shifts. Code is released at https://github.com/Michael-McQueen/OCO.

阅读与讨论 → 访问原文 →
09.
arXiv (CS.CV) 2026-06-17

Bounding Box Label Propagation for Re-Annotation of Document Layout Analysis Datasets

作者:
Nick Jochum↗Tobias Alt-Veit↗Christian Sch\"on↗Alexander L\"uck↗Ren\'e Schuster↗Didier Stricker↗

Datasets in practical document processing scenarios typically grow over time, and their class annotations undergo continuous refinement. This creates significant re-annotation efforts, which are time-consuming and costly. A promising remedy is to re-annotate only a small subset of available documents manually and apply semi-supervised learning techniques that leverage both labelled and unlabelled data. Although there are numerous approaches to tackle this problem for classification, there exists no adaptation for the problem of re-classifying object detection instances, e.g. for document layout analysis. To this end, we propose Bounding Box Label Propagation (BBLP), a pseudo-labelling framework for object detection. An object encoder integrates visual, textual, and positional embeddings from object detection samples to come up with a joint embedding that can be used for Label Propagation on partially annotated datasets in a plug-and-play fashion. Evaluation results indicate that the proposed approach produces high-quality class annotations of bounding boxes. In the D4LA layout analysis dataset, it achieves a mAP of 54.0%, corresponding to 81.6% of fully supervised performance, while using only 10% labelled data. Our work demonstrates the potential of Label Propagation for object detection and lays the groundwork for reducing manual annotation efforts in real-world document processing applications.

阅读与讨论 → 访问原文 →
10.
arXiv (CS.AI) 2026-06-16

Posterior Twins: Distributional Behavioral Simulation for Enterprise Decisions

作者:
Ankit Das↗

arXiv:2606.16415v1 Announce Type: new Abstract: Enterprise behavioral simulation requires more than producing a plausible response. Many decisions depend on the shape of a population under a proposed action: which segments accept, defect, hesitate, or move into risk-sensitive states. This paper introduces Posterior Twins, a memory-grounded digital-twin approach that represents likely behavior as an updated distribution under a specific decision context. We evaluate a family of Twinning Labs behavioral-model operating points on a 226-example held-out behavioral-response benchmark and report both modal accuracy and Wasserstein-1 distance. The results show that modal accuracy and distributional fidelity identify different operating regimes. TL-Twin Alpha achieves the lowest observed Wasserstein-1 distance in the reported result set ($W_1 = 1.16$), while TL-Twin Delta and TL-Twin Gamma provide balanced operating points near the modal-accuracy frontier. The paper frames these results as a systems result: governed memory, behavioral model routing, scenario orchestration, distributional aggregation, and auditability are necessary for turning simulated behavior into reusable enterprise decision evidence.

阅读与讨论 → 访问原文 →
11.
arXiv (math.PR) 2026-06-25

Hoeffding-Style Concentration Bounds for Exchangeable Random Variables

作者:
Nina Maria Gottschling↗Michele Caprio↗

arXiv:2603.10190v2 Announce Type: replace-cross Abstract: We establish Hoeffding-type concentration inequalities for the lower and upper tails of finite sums of exchangeable random variable sequences. In contrast to the existing literature, our concentration bounds are expressed in terms of the largest and smallest means among the distributions in the support of the de Finetti mixing measure, rather than the population mean. Specifically, the upper-tail bound is centered at the largest such mean, while the lower-tail bound is centered at the smallest. These results bridge the gap between finite-sample and population means of exchangeable random variables, and the means of the underlying distributions in the de Finetti representation.

阅读与讨论 → 访问原文 →
12.
arXiv (CS.CL) 2026-06-16

Hidden Ghost Hand: Unveiling Backdoor Vulnerabilities in MLLM-Powered Mobile GUI Agents

作者:
Pengzhou Cheng↗Haowen Hu↗Zheng Wu↗Zongru Wu↗Tianjie Ju↗Zhuosheng Zhang↗Gongshen Liu↗

Graphical user interface (GUI) agents powered by multimodal large language models (MLLMs) have shown greater promise for human-interaction. However, due to the high fine-tuning cost, users often rely on open-source GUI agents or APIs offered by AI providers, which introduces a critical but underexplored supply chain threat: backdoor attacks. In this work, we first unveil that MLLM-powered GUI agents naturally expose multiple interaction-level triggers, such as historical steps, environment states, and task progress. Based on this observation, we introduce AgentGhost, an effective and stealthy framework for red-teaming backdoor attacks. Specifically, we first construct composite triggers by combining goal and interaction levels, allowing GUI agents to unintentionally activate backdoors while ensuring task utility. Then, we formulate backdoor injection as a Min-Max optimization problem that uses supervised contrastive learning to maximize the feature difference across sample classes at the representation space, improving flexibility of the backdoor. Meanwhile, it adopts supervised fine-tuning to minimize the discrepancy between backdoor and clean behavior generation, enhancing effectiveness and utility. Extensive evaluations of various agent models in two established mobile benchmarks show that AgentGhost is effective and generic, with attack accuracy that reaches 99.7\% on three attack objectives, and shows stealthiness with only 1\% utility degradation. Furthermore, we tailor a defense method against AgentGhost that reduces the attack accuracy to 22.1\%. Our code is available at \texttt{anonymous}.

阅读与讨论 → 访问原文 →
13.
Nature Medicine 2026-06-25

<b>HPV vaccination linked to dramatic reduction in cervical cancer deaths</b>

作者:
Karen O’Leary↗

A new study examines the impact of the national HPV vaccination program in England, initiated almost two decades ago — showing huge declines in cervical cancer mortality among young women. A new study examines the impact of the national HPV vaccination program in England, initiated almost two decades ago — showing huge declines in cervical cancer mortality among young women.

阅读与讨论 → 访问原文 →
14.
bioRxiv (Bioinfo) 2026-06-19

Evaluation of analysis modes for RNA coexpression in single-cell and bulk tissue

作者:
Pavlidis↗Chu↗Elazzabi↗Garreau↗Xu↗Xiang Yu↗Morin↗

Coexpression of transcripts presents the most common means of computational inference of transcription factor regulation, and is often combined with other data types to infer regulatory networks. With the growing popularity of single-cell approaches, there are questions about how best to extract coexpression information from the data. Recently we reported a simulation study that explored the differences among coexpression performed at different levels: across single cells (xCell, per cell type), across subjects from pseudobulked single-cell data (xSubject, per cell type), or across subjects using bulk tissue samples (xBulk). Here we test predictions made by those models using real data. We consider both preservation (consistency of coexpression findings across different levels of analysis of the same data) and replicability across independent studies, as well as biological interpretability. We find that preservation across levels is limited, indicating the choice of analysis level will affect outcomes. We show that xCell coexpression is more replicable across studies compared to xSubject. xBulk coexpression is dominated by patterns driven by variability in cellular composition and fails to capture much coexpression that is reliably detected at finer resolutions. While all modes of analysis exhibit some enrichment for known regulatory relationships, it was highest with the xCell mode. Finally, we present a case study of the effect of analysis modes on a schizophrenia-associated pattern, reinforcing the importance of analytic choices in the interpretation and replicability of coexpression analyses. Together with our modeling study, this work emphasizes the importance of understanding sources of expression covariation as they relate to the goals of the analysis, and recommend single-cell-based data with biological replicates should be the focus of attempts to infer dynamic regulatory interactions that are more likely to be replicable by others.

阅读与讨论 → 访问原文 →
15.
arXiv (quant-ph) 2026-06-25

Distillation of supersinglet states

作者:
Saeed Ahmad↗Shuang Li↗Jonathan Raghoonanan↗Kaixuan Zhou↗Valentin Ivannikov↗Tim Byrnes↗

arXiv:2509.20962v2 Announce Type: replace Abstract: We introduce an entanglement distillation (purification) protocol for supersinglet states composed of N qubits. The supersinglet state we target is a total spin zero state with zero spin variance, and has a fully entangled structure involving all qubits. In our distillation protocol, three copies of an initial spin zero state are measured in the local total spin basis such that a higher fidelity supersinglet state is generated upon postselection. The initial state can be prepared using conventional Bell state distillation methods distributed in a way to target the supersinglet symmetries. The protocol uses only local operations and classical communications, and is suitable for long-distance applications such as quantum clock synchronization and cryptography, and avoids a high dimensional Schur transform such that it can be used for tasks such as quantum metrology.

阅读与讨论 → 访问原文 →
16.
arXiv (CS.AI) 2026-06-18

Data Intelligence Agents: Interpreting, Modeling, and Querying Enterprise Data via Autonomous Coding Agents

作者:
Anoushka Vyas↗Aarushi Dhanuka↗Sina Khoshfetrat Pakazad↗Henrik Ohlsson↗

arXiv:2606.19319v1 Announce Type: cross Abstract: Production data integration is bottlenecked by repeated, lossy handoffs between data owners, engineers, and analysts who must collaboratively discover, structure, and query enterprise data. We present Data Intelligence Agents (DIA), a system of three agents (Data Interpreter, Schema Creator, and Query Generator) that compresses this workflow by treating autonomous coding agents (ACAs) as a first-class abstraction: rather than emitting text, the agents generate, execute, validate, and repair concrete artifacts, draw on a shared memory for experience reuse, and surface each for review by domain experts. DIA is deployed in production for enterprise customers. We study the Query Generator in depth and evaluate it in fully autonomous mode across seven SQL benchmarks spanning four task categories and four dialects. It matches or surpasses the best published results on all seven, demonstrating that an architecture grounded in execution, built on ACAs and a shared memory, generalizes across the data intelligence workload with adaptation confined to natural-language instructions.

阅读与讨论 → 访问原文 →
17.
arXiv (CS.CL) 2026-06-12

Helping Figures Tell their Story! Paper-Grounded Video Generation Explaining Complex Scientific Figures

作者:
Ishani Mondal↗Javad Baghirov↗Jordan Boyd-Graber↗

Scientific figures compress complex pipelines into a single canvas, yet understanding them requires paper-grounded, step-by-step narration aligned with visual highlights a capability missing from current video generation systems and benchmarks. To address this, we introduce paper-grounded figure-to-video generation: generating narrated, region-grounded walkthrough videos from a figure and its paper. We propose MINARD (Multimodal Interpretation of Narrated Architecture via Region Decomposition), a pipeline that generates paper-grounded narrations and sequentially grounds them to figure regions. We also release FigTalk, a benchmark with new sequential and component-level grounding metrics derived. On FigTalk, MINARD generates humanlike, paper-faithful narrations and outperforms narration-conditioned figure spatial grounding compared to existing approaches in both automatic and human evaluation

阅读与讨论 → 访问原文 →
18.
arXiv (CS.LG) 2026-06-24

EnerInfer: Energy-Aware On-Device LLM Inference

作者:
Bohua Zou↗Nian Liu↗Binqi Sun↗Matteo Mascherin↗Debayan Roy↗Yutao Liu↗Yu Peng↗Ning Jia↗Haibo Chen↗

arXiv:2606.23001v1 Announce Type: cross Abstract: On-device LLM inference is increasingly attractive for privacy-preserving, reliable, and cost-effective deployment, yet its energy and thermal costs remain a critical bottleneck. Existing systems primarily optimize for decoding speed, implicitly assuming that faster execution is always preferable. We show instead that on-device LLM inference often has exploitable configuration slack: modestly lowering NPU and memory frequencies preserves quality of experience (QoE) while substantially improving energy efficiency and reducing heat. Realizing this opportunity in production is challenging. The most energy-efficient NPU/DDR setting varies with the model, inference engine, platform, and runtime conditions, with no stable ranking across configurations. Commercial devices further lack component-level power sensing, and shell temperature evolves with request arrivals, response lengths, and thermal history. To address these challenges, we propose EnerInfer, the first on-device LLM inference framework that jointly manages energy efficiency, throughput, and thermal comfort for LLM workloads. EnerInfer replaces per-model profiling and sensor-heavy control with disaggregated, model-structure-aware prediction and ranking-driven online feedback. It predicts throughput and power for unseen LLMs across NPU/DDR frequency settings, selects QoE-satisfying efficient configurations under runtime interference, and uses lightweight limited-horizon thermal prediction to dynamically switch between energy-optimized and thermally constrained inference. Evaluations on real-world LLMs show that EnerInfer improves energy efficiency by up to 65%, 12%, and 24% on phones, a laptop, and a development board, respectively, without QoE violation.

阅读与讨论 → 访问原文 →
19.
arXiv (CS.AI) 2026-06-24

Visualizing "We the People": Bridging the Perception Gap through Pluralistic Data Storytelling

作者:
Lisa Schirch↗Beth Goldberg↗

arXiv:2606.24635v1 Announce Type: cross Abstract: Traditional visual data storytelling relies on binary graphics that depict two simplified groups in conflict. This can increase political polarization by oversimplifying intra-group disagreements and erasing ambiguity and shared ideas or values. This can inadvertently foster "us versus them" thinking. Intentional, pluralistic design choices for AI-enabled digital platforms can produce visualizations that emphasize nuance, opinion distribution, and intergroup commonalities. To demonstrate this potential, we examine deliberative technologies that map high-dimensional opinion spaces and highlight areas of both consensus and dissensus. The paper highlights the We the People deliberation conducted by Jigsaw and the Napolitan Institute in September 2025, which engaged over 2,400 Americans across all 435 congressional districts in an AI-supported, asynchronous dialogue regarding freedom and equality. By utilizing AI to synthesize long-form, text-based participant inputs into interactive "opinion landscapes," the initiative provided an alternative format for pluralistic data storytelling that humanized diverse viewpoints and revealed hidden areas of substantial broad consensus. The paper concludes that shifting from divisive, contrast-heavy visual frameworks to distribution-focused, interactive models represents a highly scalable, low-cost intervention capable of bridging perceptual gaps and cultivating a more resilient, collaborative democratic culture.

阅读与讨论 → 访问原文 →
20.
arXiv (CS.CV) 2026-06-24

What Do Flow-Based Inverse Solvers Approximate? A Posterior-Transport View

作者:
Jian Xu↗Delu Zeng↗John Paisley↗Qibin Zhao↗

A growing family of training-free solvers – FlowDPS, FLOWER, PnP-Flow and their diffusion ancestors (DPS, DAPS) – repurpose a pretrained flow-matching prior to solve imaging inverse problems by adding a measurement-guidance term to the deterministic probability-flow ODE. Despite strong empirical results, what these per-step corrections actually approximate – and how far the resulting samples are from the true posterior $p(x\mid y)$ – has not been characterized. We give a posterior-transport account of flow-based inverse problem solving. Our starting point is a simple but consequential fact: for a deterministic flow prior, Bayesian conditioning is realized entirely by a reweighting of the source distribution, not by a drift correction; pushing the reweighted source through the unmodified velocity field yields exact posterior samples. From this we show that trajectory-guidance solvers can be read as the minimum-kinetic-energy correction field needed to morph the unconditional source into the posterior, and that FlowDPS / FLOWER / PnP-Flow correspond to distinct zeroth-order / Gaussian / proximal approximations of this single object; we bound the resulting posterior bias in Wasserstein distance. A controlled $2$D study with a closed-form posterior confirms the theory decisively: source reweighting matches the true posterior to the Monte-Carlo floor on every metric, whereas trajectory guidance incurs $200$–$800\times$ larger error and collapses posterior modes, regardless of guidance strength. Guided by the analysis we propose a cheap, principled velocity-correction solver that is competitive across two in-domain priors (AFHQ, CelebA) and two out-of-distribution settings while, unlike point-estimate source-space optimizers, producing diverse posterior samples with uncertainty that correlates with reconstruction error.

阅读与讨论 → 访问原文 →
21.
arXiv (CS.AI) 2026-06-16

Rethinking Scaffolding in LLM Tutors: The Interactional Mismatch Between Benchmarks and Real-World Deployments

作者:
Alexandra Neagu↗Jeffrey T. H. Wong↗Marcus Messer↗Rhodri Nelson↗Peter B. Johnson↗

arXiv:2606.15766v1 Announce Type: new Abstract: A central pedagogical value evaluated in AI tutor benchmarks is scaffolding: guiding students through graduated steps toward a solution. Alignment and evaluation methods for embedding scaffolding behaviour into chatbots, however, rest on an implicit assumption: that students will take up the scaffolding and engage in the conversation. To examine whether this assumption holds, we introduce an evaluation pipeline around two metrics - Chatbot Scaffolding and Student Uptake - and apply them across nine datasets of 9,490 chats, spanning AI tutor benchmarks and real-world deployments of educational chatbots. Our analysis reveals that while benchmarks assume a high-scaffolding, high-student-uptake environment, students in real-world settings exhibit lower levels of uptake overall - frequently bypassing the chatbot's pedagogical framing to drive the interaction toward their own learning goals at little interpersonal cost. We argue that bypassing scaffolding is not necessarily detrimental; rather, it frequently highlights a mismatch between a chatbot's pedagogical framing and the student's learning goals. To meaningfully evaluate the effectiveness of a chatbot's assistance, future benchmarks must move beyond the assumption that students will simply take up the scaffolding, and instead evaluate how these chatbots navigate diverse learning contexts and student-driven interaction patterns.

阅读与讨论 → 访问原文 →
22.
arXiv (CS.CL) 2026-06-24

Selective Capability Unlearning in End-to-End Spoken Language Understanding

作者:
Akanksha Singh↗Vinod Kumar Kurmi↗

Modern spoken language understanding (SLU) systems are increasingly deployed in real-world settings, where specific functionalities may need to be removed due to policy or safety constraints. In SLU, a functionality corresponds to an intent and its associated slot-generation behavior. However, in autoregressive models, suppressing a target intent does not eliminate the conditional mapping that generates slots conditioned on that intent. When the intent prefix is externally supplied, the model can reconstruct the original intent-slot structure. We identify this structural failure as capability persistence. We propose \underline{Binding \underline{S}ubspace (BSU)}, a representation-level framework that isolates and attenuates intent-conditioned directions underlying this mapping. Across SLU benchmarks, BSU substantially reduces forced-prefix recoverability while preserving retained performance.

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

From Benchmarks to Skills: Low-Rank Factors for LLM Evaluation

作者:
Aviya Maimon↗Amir DN Cohen↗Gal Vishne↗Shauli Ravfogel↗Reut Tsarfaty↗

Current evaluations of large language models (LLMs) rely heavily on a growing collection of benchmarks and on aggregate benchmark scores, yet it remains unclear what this comparison actually captures, and what these scores reveal about models' underlying capabilities. Here, we propose a new paradigm for LLM evaluation, by asking whether benchmark performance reflects many independent abilities, or rather relies on a small number of shared dimensions. To answer this, we apply Factor Analysis (FA) to a massive performance matrix of LLMs versus benchmarks \((60\times44)\) revealing an intrinsically low-rank structure of that matrix. That is, a small number of latent factors captures most of the structure in the full task space. This low-rank geometry reveals substantial redundancy across existing tasks and explains why many benchmarks appear to be measuring overlapping abilities. We further show that these latent factors correspond to coherent, skill-like, dimensions of LLM behavior. Leveraging this latent skill-space, we deliver three practical tools for LLM evaluation and downstream users: (i)~identifying redundant tasks, (ii)~profiling new models using a small subset of tasks, and (iii)~selecting models aligned with desired skill profiles. Our method provides a solid alternative to the de-facto standard of a single aggregate score, and establishes an interpretable and practical framework for understanding and benchmarking LLM core capabilities.

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

LegalHalluLens: Typed Hallucination Auditing and Calibrated Multi-Agent Debate for Trustworthy Legal AI

作者:
Lalit Yadav↗Akshaj Gurugubelli↗

AI systems deployed in legal workflows hallucinate at rates that aggregate metrics report at ~52%, but this average conceals where errors concentrate and in which direction they run, leaving compliance officers without an actionable signal for trustworthy deployment. We present LegalHalluLens, an auditing framework with three components: typed hallucination profiles across four legally-motivated claim categories (numeric, temporal, obligation/entitlement, factual) over CUAD (Hendrycks et al., 2021); a Risk Direction Index (RDI) that reduces omission-versus-invention bias to a single deployment-comparable scalar; and a typed debate pipeline calibrated to both magnitudes and directions. Across 510 contracts and 249,252 clause-level instances we measure a within-model gap of approximately 38-40 pp between obligation/numeric and temporal claims that aggregate reporting hides, and show that two systems with matched 52% rates can carry opposite RDIs. The debate pipeline reduces fabricated detections by 45% with per-category gains tracking the diagnosis, matching commercial APIs with a substantially smaller backbone (4B active parameters). Typed profiles and RDI surface failure modes that aggregate metrics hide; we further show these diagnostics serve as calibration inputs for multi-agent debate pipelines, where Skeptic challenges and asymmetric gates targeted at measured failure modes outperform generically-tuned debate. The framework supports direction-aware procurement, accountability, and agent design for legal AI deployed in the wild.

阅读与讨论 → 访问原文 →