Academic Intelligence · Curated Daily

Explore the Frontier of Global Academia

AcademicHub aggregates real-time literature from top journals and preprint platforms. Build your personal research radar and let large language models compile cross-disciplinary analysis briefings automatically.

01.
arXiv (CS.AI) 2026-06-19

Too long; didn't solve

arXiv:2604.07593v2 Announce Type: replace Abstract: Mathematical benchmarks consisting of a range of mathematics problems are widely used to evaluate the reasoning abilities of large language models, yet little is known about how their structural properties influence model behaviour. In this work, we investigate two structural length variables, prompt length and solution length, and analyse how they relate to model performance on a newly constructed adversarial dataset of expert-authored mathematics problems. We find that both prompt and solution lengths correlate positively with increased model failure across models. We also include a secondary, exploratory analysis of cross-model disagreement. Under a difficulty-adjusted normalised analysis, both variables retain weak negative associations with realised model separation, slightly stronger for prompt length. Overall, our main robust finding is that structural length is linked to empirical difficulty in this dataset.

02.
medRxiv (Medicine) 2026-06-16

Efficacy of Ergothioneine Supplementation on Postpartum Fatigue, Sleep Quality, and Quality of Life: A Randomized, Double-Blind, Placebo-Controlled Trial

Background: Postpartum asthenia, characterized by severe fatigue, sleep disturbances, and physiological stress, lacks effective targeted interventions. Ergothioneine (EGT) is a unique, naturally occurring antioxidant that actively accumulates in mitochondria, offering a compelling therapeutic rationale for systemic recovery. This study aimed to evaluate the efficacy of EGT in accelerating postpartum functional restoration and alleviating fatigue. Methods: This single-center, randomized, double-blind, placebo-controlled trial enrolled 40 postpartum women (SF-36 total score [≤] 70) who had ceased breastfeeding. Participants were randomized (1:1) to receive either 120 mg/day of EGT or a matched placebo for 30 days. Efficacy was assessed using the SF-36, Pittsburgh Sleep Quality Index (PSQI), Fatigue Scale-14 (FS-14), and Traditional Chinese Medicine (TCM) asthenia scale. To rigorously evaluate the treatment effects, advanced statistical modeling, including Linear Mixed-Effects Models (LMM) and Analysis of Covariance (ANCOVA) adjusted for baseline covariates, was employed. Results: All 40 participants completed the trial. The EGT group demonstrated robust and accelerated functional recovery. Notably, significant improvements in sleep quality (p = 0.0361) and systemic fatigue (p = 0.0059) were observed as early as Day 15. Importantly, EGT yielded a statistically significant between-group superiority in alleviating mental fatigue compared to placebo at Day 15 (p = 0.0313). By Day 30, the EGT cohort exhibited substantial within-group improvements across all primary metrics, including SF-36 (+35.94%, p = 0.0006) and FS-14 (-27.78%, p = 0.0011). Furthermore, as an additional physiological benefit, EGT induced a selective and significant reduction in hepatic transaminases (ALT: -30.42%; AST: -17.44%) within normal limits, a trend not observed in the placebo group. EGT was exceptionally well-tolerated with no treatment-related adverse events. Conclusions: EGT supplementation (120 mg/day) safely accelerates postpartum functional recovery, offering rapid relief from mental fatigue and sleep disturbances within 15 days, while concurrently optimizing hepatic physiological status. These preliminary clinical signals warrant confirmation in larger, adequately powered cohorts. Trial Registration: ChiCTR2500114171; Prospectively registered on 2025-12-08.

03.
arXiv (CS.CV) 2026-06-16

Parameter-Efficient Adaptation of SAM 3 for Automated ITV Generation from 4DCT Images

Authors:

Four-dimensional computed tomography (4DCT) captures the full respiratory cycle of thoracic anatomy, yet current Internal Target Volume contouring workflows process each phase in isolation, discarding temporal coherence and leaving contours vulnerable to phase-specific artifacts. We present a lightweight framework that applies parameter-efficient fine-tuning to the Segment Anything Model 3 (SAM 3) via low-rank adaptation (LoRA) to align its text-prompted segmentation with the medical domain using only seven annotated 3D CT volumes. Furthermore, the framework incorporates a hard negative mining strategy to improve boundary discrimination in low-contrast thoracic regions. At inference, phase-wise predictions are refined through phase-coherent temporal filtering and spatial connectivity analysis. Since respiratory motion is continuous and periodic, genuine anatomy appears in contiguous blocks of phases, whereas transient artifacts appear sporadically and are thus effectively suppressed. Experiments on pulmonary and cardiac structures yield median Dice scores of 0.968 and 0.910 with 95th-percentile Hausdorff distances of 0.998 mm and 2.931 mm, respectively. The proposed framework effectively eliminates the severe false-positive predictions inherent in the zero-shot inference of the unadapted SAM 3. With only seven annotated volumes, the framework retains over 95% of full-data accuracy, and the entire pipeline is trainable on a single consumer-grade GPU, demonstrating a scalable, data-efficient solution for adaptive radiotherapy.

04.
arXiv (CS.AI) 2026-06-19

Lagrange: An Open-Vocabulary, Energy-Based Sparse Framework for Generalized End-to-End Driving

arXiv:2606.20274v1 Announce Type: new Abstract: Scaling end-to-end autonomous driving to complex, open-world environments requires perceptual models that generalize to anomalous scenarios and planners that produce kinematically valid trajectories. Existing paradigms face a distinct dichotomy between representational efficiency and generalization capacity. Dense models (e.g., occupancy networks), while geometrically robust, incur critical computational bottlenecks and struggle with high-level semantic reasoning. Conversely, sparse, query-based planners are efficient but reliant on closed-set definitions, rendering them vulnerable to out-of-distribution (OOD) events. Although recent Vision-Language-Action (VLA) models offer open-vocabulary reasoning, their autoregressive, discrete token generation fundamentally conflicts with the continuous, high-frequency control requirements of vehicle dynamics. To address this, we propose Lagrange, an open-vocabulary, computationally sparse driving framework based on Masked Latent Fields (MLF). Rather than relying on dense volumetric reconstructions or closed-set query mechanisms, Lagrange exploits Vision-Language Models (VLMs) to encode class-agnostic object proposals into continuous semantic visual tokens. We introduce an intent-driven masked cross-attention module that temporally filters irrelevant entities, decoding the attended tokens into an implicit continuous energy field defined over spatial coordinates. By framing decision-making as a Lagrangian action minimization problem spanning this energy field, we enforce strict compliance with vehicle kinematics while executing collision avoidance. Extensive offline evaluations on both standard (nuScenes) and long-tail (CODA) benchmarks demonstrate that Lagrange establishes a promising framework for robust, interpretable, and kinematically feasible open-world autonomy.

05.
arXiv (CS.CL) 2026-06-17

Beyond Domains: Reusing Web Skills via Transferable Interaction Patterns

Large language model (LLM) web agents are usually deployed as tool callers: each turn, the model reads a fresh page observation and emits one structured tool action. When every action is a low-level primitive, horizons grow quickly and so do policy-facing LLM completions, dominating latency and cost on benchmarks such as Mind2Web and WebArena. Recent systems therefore wrap repeated interaction fragments as web skills: callable tools built from successful trajectories or induced programs, so one call can replace several primitives. However, prior skill libraries are still triggered mainly by instruction similarity or coarse site metadata, which yields low skill reuse on held-out sites and leaves much of the potential step and token reduction on the table. We present SkillMigrator, an agent that learns reusable web skills and transfers them across sites by matching layout structure rather than specific element references. Each induced skill is stored as a transferable interaction pattern (TIP): the skill paired with a structural sketch of the snapshot at induction time. At test time, SkillMigrator retrieves TIPs by layout similarity and grounds their references on the live page. The rest of the stack is standard: accessibility-snapshot observations with stable references, and fixed tool calling over primitives plus skill invocations. Compared with the state-of-the-art approaches, SkillMigrator reduces the average LLM-action count on successful trajectories by 8-10% across both WebArena and Mind2Web at matched success rate.

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

Optimising Temporary Accommodation Placement Across London with AI-Powered SaaS in E-Governance Systems

arXiv:2606.16652v1 Announce Type: cross Abstract: Temporary accommodation has become a major fiscal and administrative pressure for English local authorities, particularly in London, where demand and costs have risen sharply. This paper documents the creation and use of DOMUS, a cloud-based, AI-enabled decision-support system built from scratch at the University of East London and customised for the needs of London Borough of Newham to support statutory Temporary accommodation placement. DOMUS integrates household case records, policy-constrained affordability and suitability rules, and live private-rental listings within a single governance-aligned workflow. The system combines transparent, rule-based filtering with large language model-assisted search to standardise the application of bedroom need, affordability thresholds, geographic preferences, and accessibility requirements, while preserving officer discretion and audibility. Household and property attributes are encoded into policy-consistent representations prior to AI-assisted ranking and explanation. A pilot deployment in Newham's secure environment evaluated operational performance relative to manual workflows. Results indicate substantial reductions in search time, improved adherence to key placement constraints, and high staff satisfaction, while maintaining statutory compliance and role-based accountability. Beyond TA, the paper frames DOMUS as replicable digital public infrastructure: a modular, cloud-native Software-as-a-Service architecture that can be deployed across other UK boroughs and adapted to other public administration tasks characterised by scarcity, rule-bound eligibility, and high stakes. The findings demonstrate the feasibility of scalable, ethically governed AI deployment in local government and contribute to debates on AI-enabled public value creation in e-governance.

07.
arXiv (CS.CV) 2026-06-16

HAFMat: Hybrid Priors Guided Adaptive Fusion for Single-Image Human Material Estimation

Physically based rendering (PBR) material estimation is a fundamental appearance decomposition task with broad applications in virtual content creation, relighting, and digital human rendering. However, estimating PBR materials from a single human image remains highly ill-posed, since illumination, geometry, and reflectance are heavily entangled in the observed appearance. To mitigate this ambiguity, we propose HAFMat, a hybrid-prior-guided framework for single-image human material estimation. Our method introduces guidance maps that encode complementary cues, including appearance, body geometry, structure, and prior material predictions from pre-trained models. A key observation is that these guidance cues are heterogeneous: some cues mainly provide texture-level constraints, while others convey higher-level semantic information. To exploit this property, we design a Multi-layer Adaptive Feature Fusion Mechanism, which adaptively fuses guidance features with decoder features at different stages. This design enables texture-dominant and semantic-dominant cues to guide material decoding at appropriate levels, leading to more accurate and physically plausible material estimation. Extensive experiments on both synthetic and real data demonstrate that our method achieves state-of-the-art performance in material estimation and downstream relighting.

08.
arXiv (CS.CV) 2026-06-16

Self-Supervised Learning as Discrete Communication

Most self-supervised learning (SSL) methods learn continuous visual representations by aligning different views of the same input, offering limited control over how information is structured across representation dimensions. In this work, we frame visual self-supervised learning as a discrete communication process between a teacher and a student network, where semantic information is transmitted through a fixed-capacity binary channel. Rather than aligning continuous features, the student predicts multi-label binary messages produced by the teacher. Discrete agreement is enforced through an element-wise binary cross-entropy objective, while a coding-rate regularization term encourages effective utilization of the constrained channel, promoting structured representations. We further show that periodically reinitializing the projection head strengthens this effect by encouraging embeddings that remain predictive across multiple discrete encodings. Extensive experiments demonstrate consistent improvements over continuous agreement baselines on image classification, retrieval, and dense visual prediction tasks, as well as under domain shift through self-supervised adaptation. Beyond backbone representations, we analyze the learned binary codes and show that they form a compact and informative discrete language, capturing semantic factors reusable across classes.

09.
arXiv (CS.CL) 2026-06-16

Not All Skills Help: Measuring and Repairing Agent Knowledge

LLM agents can improve without weight updates by accumulating natural-language skills from experience, but current systems entrust every decision about which skills to keep and how to apply them to LLM judgment alone. We argue that this conflates two distinct roles: generating a skill from experience is a creative act that judgment handles well, while deciding whether that skill actually helps requires empirical evidence across many tasks. Measuring per-skill causal contributions via randomized masking, we find that skill libraries exhibit pervasive causal heterogeneity: individual skills routinely help on some task types while hurting on others, yet their opposing effects cancel in aggregate, making them invisible to global curation methods. We propose ASSAY, a framework that separates generation from curation: it computes a per-skill causal attribution on a small development set, restructures the library offline, and suppresses skills with negative predicted effect for each test task. Across seven base models spanning four providers and two benchmarks (AppWorld and tau-bench), ASSAY consistently improves over prior skill-curation approaches. On AppWorld's hardest split, DeepSeek-V3 achieves 69.3% task-goal completion (47.4% relative improvement), a new state of the art among all published methods including weight-tuned approaches. On tau-bench retail, GPT-4.1 improves by 8.7% relative, advancing past o4-mini, o1, and GPT-4.5 on the public leaderboard without any weight modification. Ablation traces the dominant gain to per-task masking, confirming that the bottleneck is matching skills to tasks at inference time, not removing bad skills globally. Code is available at https://github.com/aiming-lab/assay.

10.
arXiv (CS.LG) 2026-06-16

MolE-RAG: Molecular Structure-Enhanced Retrieval-Augmented Generation for Chemistry

arXiv:2606.05693v2 Announce Type: replace Abstract: Large language models (LLMs) have shown promise for molecular property prediction, but their ability to reason over chemical structures remains limited, as molecular representations such as SMILES differ substantially from the natural language on which LLMs are primarily trained. To bridge this semantic and chemical knowledge gap, we propose MolE-RAG, a training-free, molecule-centric retrieval-augmented generation framework for LLM-based molecular property prediction. MolE-RAG augments each prediction with three complementary sources of inference-time context: retrieved chemistry literature, molecule-specific information including compound synonyms, identifiers, functional group annotations, and physicochemical descriptors, and structurally similar molecules retrieved from the training set. We evaluate MolE-RAG across nine molecular property prediction tasks using proprietary, chemistry-specialized, and open-source LLMs. Across general-purpose LLMs, MolE-RAG improves ROC-AUC by up to 28 percentage points on classification tasks and reduces regression RMSE by up to 67% relative to a SMILES-only baseline. We further find that the utility of each context source varies across models and tasks, with different models benefiting most from textual retrieval, molecular context, or structural retrieval. These results suggest that molecule-centric retrieval can improve LLM-based molecular property prediction without model fine-tuning while providing a flexible framework for integrating heterogeneous chemical knowledge at inference time.

11.
medRxiv (Medicine) 2026-06-10

Global and local genetic overlap among ME/CFS, irritable bowel syndrome and psychiatric traits: a hypothesis-generating analysis

Authors:

Background. Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) and irritable bowel syndrome (IBS) frequently co-occur following infection, yet shared genetic architecture at the locus level has not been systematically characterised. Aims. To estimate global and local genetic correlations between ME/CFS (including infection-onset subgroup), IBS, major depressive disorder (MDD) and loneliness/isolation, and characterise ME/CFS cell-type heritability enrichment. Method. GWAS summary statistics: DecodeME (15,579 ME/CFS; 9,738 infection-onset), FinnGen R9 (9,296 IBS), PGC MDD Wave 2 (45,396) and UK Biobank loneliness (N=455,364). LDSC for global correlations; LAVA for local correlations across 2,495 loci; MAGMA for cell-type enrichment (Descartes Human atlas); coloc.abf for colocalisation. Results. All pairwise global correlations were significant after Bonferroni correction, including ME/CFS-all-MDD (rg=0.598, 95% CI 0.46-0.74) and ME/CFS-all-IBS (rg=0.573, 0.39-0.75). Of 4,232 local tests, 16 reached FDR

12.
medRxiv (Medicine) 2026-06-16

Diurnal variation in brain-derived tau and five other blood-based biomarkers for dementia and their association with cognitive performance

Blood-based biomarkers of dementia are a promising scalable tool for early diagnosis, tracking disease progression, and evaluating therapeutic efficacy. Utility of these biomarkers will not only be dependent on the reliability of their association with pathology but also contingent on their ability to track cognitive status. Previously, we demonstrated diurnal variation in several biomarkers (amyloid beta (A{beta}) 42 and 40, 42/40 ratio, glial fibrillary acidic protein (GFAP), neurofilament light (NfL), and phosphorylated-Tau 217 (p-Tau217)) which has implications for their reliability. Here, we extend these observations to a larger cohort, include brain-derived tau (BD-Tau), which is assumed to be produced exclusively in the brain, and report endocrine measures of circadian rhythmicity. We not only assessed whether these biomarkers vary with time of day, but also whether they associate with daytime function and whether these associations vary with cognitive domain and number of repeated assessments. Data collected in 20 PLWA (72.4{+/-}5.9 years, mean{+/-}SD) and 19 controls (68.9{+/-}9.8 years) were analysed. Participants completed 14 days of home monitoring and one laboratory assessment of sleep and daytime function: mood, daytime sleepiness, reaction time, immediate and delayed memory recall, everyday memory errors. During the 27-hour residential laboratory session, 3-hourly blood samples were collected and analysed for the six blood-based biomarkers of dementia as well as melatonin and cortisol. Rhythmicity of melatonin and cortisol did not differ between groups. P-Tau217 and GFAP (p

13.
arXiv (CS.LG) 2026-06-18

The Road to Artificial SuperIntelligence: A Comprehensive Survey of Superalignment

arXiv:2412.16468v4 Announce Type: replace Abstract: The emergence of large language models (LLMs) has sparked discussion on Artificial Superintelligence (ASI), a hypothetical AI system that surpasses human intelligence. Although ASI remains hypothetical and far beyond current AI capabilities, discussing its potential and exploring its feasibility and potential risks is critical for the development of future AI systems. The idea of superalignment originates from scalable oversight, which studies how to supervise increasingly capable AI systems when direct human supervision becomes insufficient. In this paper, we focus on the superalignment problem: "The process of supervising, controlling, and governing artificial superintelligence." We first review scalable oversight paradigms-Sandwiching, Self-Enhancement, and Weak-to-Strong Generalization – then analyze the limitations of current paradigms through the lens of possibility and impossibility, discuss key challenges, and propose pathways for the safe and continual improvement of future AI systems.

14.
arXiv (CS.CL) 2026-06-12

Recursive Agent Harnesses

Recursive language models (RLMs) showed that recursion over model calls is an effective strategy for long-context reasoning, and production coding agents have begun to write code that spawns subagents at scale, most recently in Anthropic's dynamic workflows. We name and study the pattern between these two lines of work, where the recursive unit is a full agent harness with filesystem tools, code execution, and planning rather than a model call with no tools. We call this the Recursive Agent Harness (RAH) and frame it as harness recursion, the code-first extension to the model recursion of RLMs. A parent agent generates and runs an executable script that spawns subagent harnesses in parallel for fine-grained workloads and uses structured function calls for small subtasks. We provide a controlled evaluation on long-context reasoning. With the backbone held fixed at GPT-5 to match the published Codex and RLM baselines, RAH improves the Codex coding-agent baseline from 71.75% to 81.36% on Oolong-Synthetic (199 samples, 13 context-length buckets up to 4M tokens), a gain attributable to the harness rather than the model. With a stronger backbone, Claude Sonnet 4.5, the same design reaches 89.77%.

15.
medRxiv (Medicine) 2026-06-18

Artificial Intelligence-informed mobile behavioural interventions to support adolescents mental health in schools: protocol for a randomised controlled trial using the MindCraft app

Background: Children and young people (CYP) are particularly affected by mental health problems. Mobile apps provide a scalable and accessible approach to adolescent mental health support, and schools are well-positioned to address multiple risk factors and deliver large-scale interventions. By combining active (self-reported) and passive (sensor-derived) data, mobile apps can model mental states and deliver context-aware support. Artificial Intelligence (AI) enables adaptive, context-aware recommendations tailored to each user. However, there is limited research on AI-based mental health interventions in community CYP. MindCraft is a mobile app designed to monitor adolescents mental health using active and passive data and provide AI-informed recommendations ("nudges"). This study aims to investigate the effectiveness of personalised AI nudges delivered through MindCraft on improving mental health outcomes among adolescents in schools in the United Kingdom. Methods: The study is a three-arm RCT using a prospective cohort of secondary school students aged 14-19. Following informed consent, participants complete a baseline online assessment at school and download MindCraft. The primary outcome is the Strengths and Difficulties Questionnaire global and subscale scores. Secondary outcomes include the Eating Disorders Diagnostic Scale, the Sleep Condition Indicator Questionnaire, the Self-Injurious Thoughts and Behaviours Interview, the Self-Efficacy Questionnaire for Children and the World Health Organisation-Five Well-Being Index. Participants are randomised to: (1) an AI-informed intervention group receiving personalised nudges, (2) an active control receiving non-personalised nudges, or (3) a control group with self-monitoring only. Participants use the app for four weeks, with follow-up at one month. Repeated-measures analyses will assess changes across time points. Discussion: We hypothesise that AI nudges will have a greater positive effect on mental health outcomes at one month than general nudges and self-monitoring. Our findings will provide key evidence on the effectiveness of personalised mobile AI recommendations for adolescents mental health and inform school-based mental health prevention and early intervention. This study will contribute evidence on the ethical, acceptable, and scalable integration of AI-enabled digital mental health tools within public health and educational systems, with implications for the design of future digital public health interventions and policies supporting their safe integration in schools.

16.
arXiv (CS.LG) 2026-06-15

Where Black-box Drug-Target Interaction Prediction Models Look: Cross-Method Explainability

arXiv:2606.14245v1 Announce Type: new Abstract: Drug-target interaction (DTI) and affinity (DTA) predictors increasingly achieve strong benchmark scores, yet their internal use of sequence, fingerprint, and graph features often remains opaque. We present an interpretability audit of BridgeDPI architecture on three different datasets including Gao, Human, and C.elegans. This study combines gradient-based attributions – integrated gradients, saliency, layer-wise relevance propagation, SmoothGrad, and SmoothGrad-IG – with feature-wise occlusion ablation and strict intersection consensus across methods to reduce single-explainer bias. We summarize sensitivity and signed effects at raw inputs, at the bridge similarity scaffold, and through the graph convolution, including edge-level sensitivities and targeted edge removals. The results show that explainability is most informative when treated as model criticism: it reveals modality dominance, padding and special-token artifacts, dataset-dependent cooperative versus suppressive effects across layers, and chemistry-consistent fragment and composition motifs where methods agree. These analyses do not substitute for structural or experimental ground truth, yet they can provide testable hypotheses for downstream validation in computational drug discovery pipelines. More broadly, applying modern XAI to contemporary DTI/DTA models is still an early pass over the rich structure implicit in trained weights and data – yet even this first layer of scrutiny already helps researchers relate predictions to drug- and target-side representations and to prioritize external validation.

17.
bioRxiv (Bioinfo) 2026-06-10

Promera: a unified model for biomolecular structure prediction, filtering, and design

Generative models have become staple tools for modeling and designing biomolecular structures. However, although these tools have improved in structural prediction accuracy, their ability to filter designed binders—an essential use case—remains insufficient; whereas design methods have focused more on unconstrained binder generation rather than capabilities enabled by controllable design. We introduce Promera, a unified generative model that combines all-atom structure prediction with improved filtering and controllable design. We find that Promera's confidence metrics are more accurate for filtering binders from non-binders for both miniproteins and nanobodies, while its co-folding performance surpasses popular open-source models (OpenFold3-p2, Boltz-2) on therapeutically relevant categories. As a design model, Promera generates binders by predicting masked protein sequences with optional epitope, paratope, and template constraints. Remarkably, our nanobody designs match the in silico success rates from backprop-based techniques (mBER) when evaluated under co-folding confidence filters. We further provide two in silico demonstrations of the the versatile capabilities of our design method: epitope targeting of the Andes hantavirus glycoprotein with VHHs and active state stabilization of the beta-2 andrenergic GPCR. We conclude by proposing a scaling law for co-folding models, suggesting a path for further performance improvement.

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

Beyond Parallel Sampling: Diverse Query Initialization for Agentic Search

arXiv:2606.17209v1 Announce Type: new Abstract: Test-time scaling for agentic search typically increases depth (i.e., more turns and tokens per trajectory) or breadth (i.e., more parallel rollouts). Here we focus on breadth scaling, showing that standard parallel sampling yields diminishing returns, tracing this to query redundancy at the first turn. When models issue similar first queries across rollouts, the threads retrieve overlapping evidence, and subsequent turns are conditioned on this shared retrieval. We address this limitation with DivInit, a training-free intervention at the first turn. Rather than sampling k independent first queries, DivInit draws n candidates from a single call, picks k < n diverse seeds, and runs them as parallel trajectories. Across five open-weight models and eight benchmarks, DivInit consistently improves over standard parallel sampling, with average gains of five to seven points on multi-hop QA at matched compute. Code available at https://github.com/cxcscmu/diverse-query-initialization

19.
arXiv (CS.CL) 2026-06-15

An Empirical Study of Automating Agent Evaluation

Agent evaluation requires assessing complex multi-step behaviors involving tool use and intermediate reasoning, making it costly and expertise-intensive. A natural question arises: can frontier coding assistants reliably automate this evaluation process? Our study shows that simply prompting coding assistants is insufficient for this task. Without domain-specific evaluation knowledge, frontier coding assistants achieve only a 30% execution success rate and produce over-engineered evaluations averaging 12+ metrics per agent, indicating that strong coding ability does not automatically translate to reliable agent evaluation. We introduce EvalAgent, an AI assistant that automates the end-to-end agent evaluation pipeline. EvalAgent encodes evaluation domain expertise as evaluation skills (procedural instructions, reusable code and templates, and dynamically retrieved API documentation) that compose into a trace-based pipeline producing complete evaluation artifacts including metrics, executable code, and reports. To systematically assess generated evaluations, we introduce a meta-evaluation framework alongside AgentEvalBench, a benchmark comprising 20 agents, each paired with evaluation requirements and test scenarios. We further propose the Eval@1 metric to measure whether generated evaluation code both executes and yields meaningful results on the first run. Our experiments show that EvalAgent produces focused evaluations, improving Eval@1 from 17.5% to 65%, and achieving 79.5% human expert preference over baseline approaches. Further ablation studies show that evaluation skills are critical for handling complex evaluation: removing them causes Eval@1 to drop significantly from 65% to 30%.

20.
arXiv (CS.CV) 2026-06-17

The Slop Paradox: How Synthetic Standardization Erodes Clinical Uncertainty and Cross-Modal Alignment in AI-Rewritten Radiology Reports

Authors:

AI-assisted clinical documentation tools increasingly summarize, standardize, and reformat radiology reports using large language models (LLMs). We present a controlled measurement of the resulting information degradation. Using 450 chest X-ray reports from the Indiana University dataset, we generate synthetic versions via three realistic LLM rewriting tasks: EHR summarization, standardized rewriting, and teaching case preparation. We measure entity erosion (via medical NER), hedging collapse (loss of clinical uncertainty language), and cross-modal alignment degradation (via BiomedCLIP image-text similarity). Our central finding is a dissociation between information loss and cross-modal fidelity. EHR summarization is the most destructive at the content level, eroding 51.4% of clinical entities and 43.7% of hedging language, yet it preserves image-text alignment almost entirely (a 2.5% drop). The two tasks meant to produce cleaner training data, standardized rewriting and teaching case preparation, do the reverse: they preserve more entities (26.8% and 29.3% eroded) but cause 14.9-16.5% alignment drops, six to seven times those of EHR summarization. We term this the slop paradox: rewriting that makes clinical text look cleaner for multimodal training is precisely what pulls it away from the image. Contrary to our pre-specified hypothesis, rare pathologies were not preferentially degraded: across nine rare-versus-common comparisons, no difference survived multiple-comparison correction, and nominal differences ran in the opposite direction (common > rare), so contamination is invisible to condition-specific monitoring. The dominant determinant of degradation is the type of AI rewriting task, not the clinical content. These findings bear on multimodal medical AI dataset construction and the governance of AI-assisted clinical documentation.

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

SpecAlign: Efficient Specification-Grounded Alignment of Large Language Models via Synthetic Data

arXiv:2606.16276v1 Announce Type: new Abstract: As large language models (LLMs) are increasingly deployed in real-world applications, alignment is no longer governed by a single universal notion of safety or helpfulness, but instead by provider- or application-specific model specifications. These specifications are typically long, structured, and frequently updated, yet existing alignment pipelines lack a systematic mechanism to operationalize them as training signals. In this paper, we propose specification-grounded alignment, a new alignment paradigm that treats provider-authored model specifications as the primary alignment target rather than abstract principles or static benchmarks. To instantiate this paradigm, we introduce SpecAlign, a framework that synthesizes alignment data directly from specification documents. SpecAlign combines structured rule annotation, controllable specification instantiation, and multi-agent adversarial data synthesis to generate fine-grained, boundary-aware preference pairs that capture both compliant behaviors and meaningful specification violations. Experiments across multiple model specifications and backbone models demonstrate that training with SpecAlign consistently improves rule compliance while preserving general capabilities and avoiding over-conservative behavior. These results suggest that grounding alignment in explicit model specifications enables rapid, precise, and scalable adaptation of LLM behavior to evolving policy requirements.

22.
arXiv (CS.CL) 2026-06-18

Compact Geometric Representations of Hierarchies

Computing geometric representations of data is a cornerstone of modern machine learning, typically achieved by training dual encoders which map queries and documents into a shared embedding space. Recent work of You et al. [NeurIPS '25] has extended this approach to hierarchical retrieval, where relevance is determined by the ancestor-descendant relationships in a Directed Acyclic Graph (DAG). While previous work has shown that valid embeddings exist when the number of descendants is small, these bounds degrade significantly for deep hierarchies, requiring dimensions as large as the total number of nodes. In this paper, we investigate compact reachability embeddings for more general graph classes and provide theoretical guarantees for representing hierarchies using embeddings whose dimension depends on structural graph parameters. We prove that for any directed tree, there exists a reachability embedding in constant dimension 3, independent of the tree's size or depth. We generalize this result to graphs characterized by treewidth $t$, constructing embeddings of dimension $O(t \log n)$, where $n$ is the number of nodes. Complementing these upper bounds, we provide matching or near-matching lower bounds, showing that dimension $\Omega(n)$ is necessary for general DAGs and $\Omega(t/\log(n/t))$ is required for graphs of treewidth $t$. We also obtain upper and lower bounds parameterized by the number of cross-edges in the DAG. We additionally show that our embeddings can be constructed on real world datasets, and that they give much smaller dimensions in high recall regimes compared to prior embeddings with theoretical guarantees.

23.
arXiv (CS.LG) 2026-06-15

IntSeqBERT: Learning Arithmetic Structure in OEIS via Modulo-Spectrum Embeddings

arXiv:2603.05556v2 Announce Type: replace Abstract: Integer sequences in the OEIS span values from single-digit constants to astronomical factorials and exponentials, making prediction challenging for standard tokenised models that cannot handle out-of-vocabulary values or exploit periodic arithmetic structure. We present IntSeqBERT, a dual-stream Transformer encoder for masked integer-sequence modelling on OEIS. Each sequence element is encoded along two complementary axes: a continuous log-scale magnitude embedding and sin/cos modulo embeddings for 100 residues (moduli $2$–$101$), fused via FiLM. Three prediction heads (magnitude regression, sign classification, and modulo prediction for 100 moduli) are trained jointly on 274,705 OEIS sequences. At the Large scale (91.5M parameters), IntSeqBERT achieves 95.85% magnitude accuracy and 50.38% Mean Modulo Accuracy (MMA) on the test set, outperforming a standard tokenised Transformer baseline by $+8.9$ pt and $+4.5$ pt, respectively. An ablation removing the modulo stream confirms it accounts for $+15.2$ pt of the MMA gain and contributes an additional $+6.2$ pt to magnitude accuracy. A probabilistic Chinese Remainder Theorem (CRT)-based Solver converts the model's predictions into concrete integers, yielding a 7.4-fold improvement in next-term prediction over the tokenised-Transformer baseline (Top-1: 19.09% vs. 2.59%). Modulo spectrum analysis reveals a strong negative correlation between Normalised Information Gain (NIG) and Euler's totient ratio $\varphi(m)/m$ ($r = -0.851$, $p < 10^{-28}$), providing empirical evidence that composite moduli capture OEIS arithmetic structure more efficiently via CRT aggregation.

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

Valid Inference with Synthetic Data via Task Exchangeability

arXiv:2606.13629v1 Announce Type: cross Abstract: There is a proliferation of work arguing for the use of synthetic data in scientific research. For example, social scientists are arguing for the use of LLM-generated "silicon samples" in pilot studies; AI evaluations increasingly rely on "LLM-as-a-judge" outputs; and proteomics research is accelerated by generative models that produce synthetic protein structures. These developments raise an intriguing possibility: synthetic data may help researchers ask more questions, run more studies, and accelerate discovery. But they also raise a fundamental concern: synthetic data can be biased, noisy, and misspecified. In this work, we propose statistical principles for using synthetic data in scientific research with provable validity guarantees. The key insight is a new technical condition that we call task exchangeability. Informally, this is a requirement that the researcher can identify historical tasks, for which real data is available, such that their current task of interest is exchangeable with the historical tasks in an appropriate mathematical sense. We develop methods for valid inference under task exchangeability, together with extensions that provide guarantees even beyond exchangeability. We demonstrate the framework on public opinion surveys with silicon samples and AI evaluation with autoraters.

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

Diffuse AI Control on Fuzzy Tasks

arXiv:2606.08892v2 Announce Type: replace Abstract: AI models deployed in critical domains, such as AI safety research, may subtly sabotage our efforts due to misalignment. Diffuse AI Control is a subfield of AI safety concerned with mitigating risks from AI sabotage distributed over long deployment horizons (diffuse threats). These risks are particularly pernicious on fuzzy tasks, i.e. tasks which are hard to grade or require intuition. To understand diffuse threats on fuzzy tasks, we introduce a framework that considers AI control as an adversarial game between a blue team and a red team. The blue team uses a weak trusted model to construct a weak score against which they would train a strong, potentially subversive model to remove the subversion propensity if it were present. The red team then tries to find model behaviors that are rated highly by the weak score, and thus might not be trained out, but actually correspond to poor performance. We test our framework on the task of writing experimental proposals for research questions from recent ML papers. We use a language model with access to the original paper as a proxy "ground-truth" scorer. Our red team discovers subversive behaviors using multi-objective evolutionary prompt optimization. We show that Opus~4.6 can write proposals that are worse according to the ground truth proxy than those of GPT-OSS-20B, while the weak scorer rates them as highly as the best proposals from Opus 4.6. We then propose an adversarial optimization algorithm for the blue team that discovers more robust prompts for the weak model. This algorithm produces a blue team prompt that our red team optimization fails to exploit.