每天知道
你的领域
出了什么

We extend our prior work on Text2DSL automatic generation of domain-specific language (DSL) code from natural language descriptions along two complementary axes. First, we replace prompt-only synthetic generation with context-aware distillation, in which a teacher large language model (DeepSeek-V4-Flash) operates under an explicitly defined structured context comprising a BNF grammar, an API specification, and a closed identifier vocabulary; the resulting corpus is verified by a two-tier pipeline combining AST validation through esprima and runtime acceptance through the production polkitd daemon and the pkcheck client. This scales the verified PolkitBench corpus from 4,204 to 10,073 natural-language-to-Polkit-rule pairs at 100.0% AST validity and 99.7% runtime pass rate. Second, we conduct the per-component factorial ablation of structured context that was identified as future work in the precursor study: eight conditions C0-C7 are evaluated on GigaChat-10B-A1.8B with the new corpus. Three findings emerge. (i) The new harder corpus collapses the baseline mode (Syntax Valid 97.6% -> 58.5%, Combined Score 0.482 -> 0.252), whereas the context-enhanced mode degrades only marginally (Syntax 98.6% -> 97.4%, Combined 0.801 -> 0.750), confirming that structured context is not a cosmetic improvement but a load-bearing mechanism. (ii) The best absolute condition is the full context C7 across all metrics, while the strongest partial conditions (C5 = BNF + Vocabulary, C6 = API + Vocabulary) both contain the vocabulary. (iii) A Shapley-style decomposition assigns the largest semantic-quality effect to the vocabulary (Combined +0.198), the largest structural-validity effects to API (+24.7 pp) and BNF (+22.3 pp).

We propose NullFlow, a principled framework for one-step generative image reconstruction. Our key idea is to confine the generative flow to a measurement-consistent subspace. Because the flow never leaves this subspace, NullFlow needs no separate data-fidelity corrections, unlike existing solvers. NullFlow samples in a single network evaluation by learning the flow's average velocity, avoiding the step-by-step integration of traditional flow matching methods. We prove that the average velocity of this constrained flow yields a training objective whose global minimizer is a one-step posterior sampler. We show on image inpainting that NullFlow matches state-of-the-art diffusion solvers while cutting inference from hundreds of network evaluations to one.

Frontier multimodal models can guess whether a person is lying from a testimony video. To do so, they stream that raw face and voice to a third-party model. We ask whether the heavy media is needed at all. On the Real-life Trial Deception dataset, Whissle on-device speech and vision stack extracts a compact digest: transcript, emotion, age, gender, intent distributions, a deception intent filter, fluency and rhythm, per-frame facial behaviour, and prosody. Under speaker-independent evaluation, we report three findings. A small classifier on this digest reaches AUC 0.741, matching Gemini 2.5 Pro on full video. Handing the digest to a frontier LLM reaches AUC 0.755 with Claude Opus 4.8 at 7.8X fewer input tokens, with no media leaving the device. The reported 75% accuracy is a speaker-leakage artifact. We release code and experiments.

Recent advances in video generation have made minute-level synthesis possible; however, generating long videos remains challenging due to error accumulation, attribute drift, and the limited availability of long video data. In this paper, we introduce an infinite-length video generation framework that focusing on addressing these issues and produces high-quality, dynamic, and identity-consistent single-shot long videos. We first finetune a diffusion model as a video extension model on large-scale short video data to autoregressively generate temporally coherent clips. Inspired by the success of large language models (LLMs), we adopt causal attention computation between clips to further finetune this model on long video data. In this way, the tokens in one clip (short video) are computed by bidirectional attention while tokens among clips are computed by unidirectional attention. This design leverages the strengths of modern diffusion models while preserving long-term context information, effectively mitigating error accumulation and attribute drift. To achieve memory efficiency during inference, we adopt a key-value (KV) caching mechanism to maintain a constant KV memory. Furthermore, we introduce truncation-rectified flow (T-RFlow) technique to further suppress error accumulation. Experimental results demonstrate the effectiveness of our method. Our framework establishes a new benchmark for realistic and coherent minute-level video synthesis.

We introduce reference-free measures for evaluating the physical consistency of generated videos, combining relative and absolute approaches to assess fidelity. Although tools like WorldGym or WorldEval enable robotic simulation via video generation, physical fidelity gaps often prevent these environments from accurately reproducing real-world task success rates of VLA models. Unlike existing evaluation methods, which require costly human voting (Elo) or unavailable ground-truth references (FVD), our approach utilizes DROID-SLAM and SEA-RAFT to quantify physical inconsistencies, motivated by WorldScore. Videos filtered using our relative consistency assessment show an improvement in task success rates of over 8%, effectively narrowing the simulation-to-reality gap. Furthermore, our absolute assessment enables spatio-temporal localization, providing visualization of when and where physical artifacts occur.

Current embodied world models are primarily optimized for predictive objectives, limiting their ability to generalize under distribution shifts and reason systematically about unseen situations and hypothetical interventions. We argue that embodied intelligence should move beyond predictive world modeling toward self-evolving cognitive systems that continually construct and refine internal causal representations through interaction with the environment. To this end, we propose a self-evolving cognitive framework via causal world modeling for embodied scientific intelligence, which integrates three complementary components: causal world modeling, intervention-driven causal reasoning, and continual cognitive refinement. The proposed framework continuously revises and expands its internal causal world model through causal discovery, intervention-driven feedback, and counterfactual reasoning, supporting continual cognitive refinement and enabling cognition itself to evolve over time. Furthermore, we reinterpret embodied interaction not merely as a means of trajectory optimization, but as an epistemic process for causal hypothesis generation, intervention-driven experimentation, and continual knowledge acquisition. This work provides a conceptual and theoretical foundation for a transition from predictive intelligence toward epistemic intelligence, in which intelligence emerges through the continual construction, revision, and refinement of causal world models via interaction with the environment. Accordingly, an intervention-driven causal-epistemic benchmarking paradigm is suggested for evaluating self-evolving embodied scientific intelligence.

We show that replacing the standard MSE denoising loss in diffusion models with a nonlinear transformation induced by an f-divergence yields a simple robust training surrogate that empirically improves performance under data contamination, with small additional computational overhead. The theoretical foundation rests on a local divergence construction: under the Gaussian reverse-kernel structure of DDPM, each per-step likelihood ratio follows a lognormal distribution parameterized by a scalar mismatch, so the conditional f-divergence at each step reduces to a one-dimensional function of the denoising error. Summing these local divergences yields a training objective that unifies diffusion training as divergence induced weighted denoising, where the derivative of the induced divergence acts as a residual-space influence weight that controls the contribution of each sample. Bounded-influence divergences (Hellinger, negative exponential) suppress large error samples, with Hellinger yielding an explicit exponential weight, connecting the framework to robust M-estimation. Empirically, on CIFAR-10 under 30% contamination, NED reduces FID from 93.0 (KL) to 77.5, while also outperforming standard robust losses such as Huber and clipped MSE.

We present a design agent which is a Large Language Model (LLM) that orchestrates, but does not perform, the numerical simulations to design a silicon-on-insulator (SOI) $2\times2$ directional coupler. We choose a symmetric phase-matched coupler where a lot of analytical results are available that help the design strategy. The LLM proposes candidate gap values (a geometrical dimension size) and judges convergence, while all physics is owned by deterministic solvers: a frequency-domain eigenmode solver estimates the coupling coefficient~$κ$ for the current design, and an independent Finite-Difference Time-Domain (FDTD) stage validates it. Both solvers operate on a common slab-projected two-dimensional (2D) effective-index reduction of the silicon film, so the design~$κ$ and the FDTD response are consistent by problem design; the residual between them is shown to be a single constant phase offset~$φ$, attributable to a fixed excess coupling length $L_{\mathrm{extra}}=\SI{2.837(11)}{\micro\meter}$ that we find invariant across a factor-of-two range in~$κ$. Folding this offset into a closed-loop length correction, the agent delivers a $50/50$ splitter whose FDTD-measured cross fraction is $0.498$ (target $0.500$), a residual of $0.0017$. Results are made self-consistent within the 2D effective-index model; and the LLM succeeds in delivering a suitable design over a number of attempts.

Reinforcement Learning from Verifiable Rewards (RLVR) has emerged as a promising framework for enhancing the reasoning ability of large language models. However, much of the existing work is guided by heuristic intuition, leading to divergent algorithmic choices, even contradictory ones that nevertheless report empirical gains. To better understand this phenomenon, we conduct a theoretical analysis of RLVR updates. Our study reveals that differences in off-policy degree, determined by the number of gradient steps per rollout, substantially affect the distribution of importance sampling ratios and their clipping behavior, thereby altering which tokens dominate the update. Building on this insight, we characterize gradient expectation as the central quantity governing update dynamics and analyze the roles of token probability, advantage, and importance sampling ratio. Motivated by these findings, we propose Adaptive Clip Policy Optimization (ACPO), which adjusts clipping boundaries across token groups according to the empirical variance of their importance sampling ratios. Experiments on 3B and 7B models across diverse reasoning benchmarks, spanning mathematical problem solving, tabular QA, and logic puzzles, demonstrate that ACPO outperforms strong baselines such as DAPO and CISPO. These results demonstrate that principled, analysis-driven approaches yield more robust and effective RLVR methods. Code is available in: https://github.com/Control-derek/ACPO

While longitudinal brain PET imaging is the gold standard for quantifying the spatiotemporal accumulation of Beta-amyloid, its widespread clinical utility is constrained by high operational costs and cumulative radiation risks. Recent deep generative models show promise in longitudinal image synthesis; however, they often fail to capture subtle pathological progression due to identity drift and a persistent bias toward trivially replicating baseline signal intensities rather than modeling temporal transition. To this end, we propose Delta-Diffusion, a novel progression-aware framework that redefines longitudinal PET synthesis as a conditional Poisson Diffusion Bridge (PDB) process. Unlike standard diffusion models that start from Gaussian noise, our PDB formulation is mathematically anchored to the subject's baseline PET, effectively transforming the generative task into a conditional distribution transition of the amyloid trajectory. To handle heteroscedastic nature of PET imaging, we introduce a physically-grounded Poisson perturbation within a Diffusion Transformer (DiT). This architecture uses adaptive scale-shift modulation to precisely calibrate the synthesis with the elapsed clinical interval and structural MRI context. A volume-of-interest balanced objective is designed to emphasize sparse, high-risk regions of amyloid accumulation. Validated on two cohorts with 542 subjects, Delta-Diffusion demonstrates superior performance in capturing longitudinal variations in amyloid deposition compared to state-of-the-art methods, offering a robust computational framework for tracking disease progression.

Agentic AI systems retrieve private context, invoke tools, write files, call external services, coordinate with other agents, and may act without human approval. Existing bill of materials artifacts improve transparency for dependencies, model metadata, and training provenance, but leave an agentic transparency gap: capability opacity, the absence of a structured account of what a deployed agent can access, remember, change, delegate, and prove afterward. This paper introduces AgentRiskBOM, a security BOM for risk-scoping tool-using AI agents. It is an additive layer over SBOM, AIBOM, and MLBOM artifacts, referencing them where authoritative while adding fields for runtime authority: autonomy, tool permissions, memory, credential scope, approval gates, audit signals, inter-agent communication, and external action capability. We implement AgentRiskBOM as a JSON-schema artifact with a reproducible corpus, risk scenarios, scorer, diff detector, control mapper, and reports. We evaluate AgentRiskBOM on 13 open-source agents spanning coding, RAG, and multi-agent archetypes, plus 52 risk scenarios across 14 categories. The schema validates all 13 corpus artifacts. Coverage analysis gives AgentRiskBOM a native-equivalent score of 14 across 16 capability dimensions, vs. 1 for SBOM, 1.5 for AIBOM and 2 for MLBOM. Across modeled risk categories, AgentRiskBOM exposes 100% risk-category visibility vs. 10.5% for SBOM-like and 20.9% for AIBOM-like views. To test agentic authority drift, we inject 33 structured deployment mutations; the diff detector identifies the correct change type for all mutations. A secondary penalty-based scorer yields a Spearman correlation of 0.73 with the primary scorer, supporting rank-level consistency while showing that thresholds require human calibration. The results show that agentic AI security needs a machine-readable authority-and-risk artifact before incidents occur.

While Large Language Models (LLMs) can generate fluent Arabic text, their ability to reliably control readability levels remains unclear. We propose a multi-dimensional evaluation framework for Common European Framework of Reference for Language (CEFR)-controlled Arabic text generation, assessing whether instruction-following LLMs can serve as reliable generators for adaptive language learning. Our framework integrates controlled prompting, automatic readability prediction using a validated Taha-19 model, lexical constraint validation, and syntactic complexity profiling. Results show that structured prompting substantially improves CEFR alignment. In particular, CEFR-guided prompting with lexical constraints achieves the highest conformity to reference linguistic profiles (0.91 cosine similarity) and near-perfect agreement with predicted readability levels (0.99), while unconstrained prompting exhibits weak control. These findings establish an empirical foundation for integrating readability-aware Arabic text generation into adaptive educational systems.

There are many reasons one may want to edit a model's chain of thought (CoT) — e.g., to prefill it with reasoning from a stronger model or to remove steps that may yield unsafe outputs. The success of these interventions plausibly depends on a model's inability to notice them, as the model may alter its behavior if it suspects tampering. In this work, we study whether recent reasoning models are able to detect such interventions on their CoTs under a variety of conditions: both during reasoning and after it, and when prefilled both with their own CoTs and with those of other models. Broadly, we find that (i) models exhibit only very modest detection accuracy; (ii) models struggle to identify *how* their CoT was modified; and (iii) models are about as good at detecting changes to their own CoTs as to those of other models.

Reinforcement learning for robot manipulation is often bottlenecked by reward design, especially in long-horizon tasks: sparse success rewards provide weak supervision, while hand-crafted dense rewards are tedious to design and generalize poorly across tasks. Progress-based reward models offer a promising alternative by estimating how far an observation has advanced toward task completion, but existing approaches often require task-specific demonstrations or progress labels, and can assign high rewards to visually plausible but physically incorrect states. We introduce the Reference-Anchored Reward Model (RARM), a lightweight visual comparator that converts a single successful demonstration into a dense, progress-aware reward. RARM is trained once on general-purpose videos with a contrastive temporal objective, requiring no robot-specific data, task-specific reward labels, or per-task reward engineering. At deployment, RARM matches rollout clips to reference clips and rewards only confident forward progress, suppressing uncertain matches that may otherwise produce false-positive rewards. Across 9 simulated manipulation tasks from LIBERO and MetaWorld and 4 real-world tasks, RARM achieves the best overall success rates in subsequent RL training, with particularly large gains on long-horizon tasks such as cloth folding, where unreliable progress estimates are especially harmful.

Web accessibility (A11Y), which ensures web content is perceivable and usable for users with disabilities, is a critical requirement for modern web applications. Yet existing tooling overwhelmingly focuses on detecting A11Y violations rather than repairing them. Automated program repair (APR) techniques appear promising for this setting, but our study shows that state-of-the-art APR systems perform poorly when applied to real-world A11Y violations. Unlike conventional sparse-bug scenarios, web A11Y issues often manifest as multiple structurally related violations per page, requiring coordinated edits across multiple files. Existing repair systems fail to manage this multi-fault scale, as they handle each bug individually without considering their relationships or incorporating domain rules such as the Web Content Accessibility Guidelines (WCAG). We propose A11YRepair, an LLM-based framework for web A11Y repair. A11YRepair introduces a divide-and-conquer workflow that first clusters violations requiring coordinated edits to reduce redundant localization, and then decomposes each cluster by root cause so the LLM can generate focused and consistent patches. The framework further incorporates WCAG-driven knowledge to strengthen domain awareness during both fault localization and patch synthesis. To support systematic evaluation, we construct A11YBench, a benchmark of 60 real-world web projects collected from GitHub. Experimental results show that A11YRepair achieves higher repair effectiveness and lower cost than state-of-the-art baselines, and ablation studies confirm the importance of its divide-and-conquer design and selective domain knowledge integration. Specifically, patches generated by A11YRepair have been merged into open-source projects from Google, Microsoft, Facebook, IBM, K8s, Docker, and Alibaba, demonstrating its practical value.

Understanding the causal structure of a language model's thought process is a problem of significant importance for both transparency and safety. In this work, we take a local approach toward this goal by analyzing the causal relationships among individual components, termed units, of a given, specific chain-of-thought trace. We construct a structural causal model on these units and relate each unit to the log probability of generating (subsequent) output units. Our algorithm, termed AttriCoT, is a black-box method that performs attribution by estimating importance parameters in the structural causal model using $O(U)$ forward passes through the model, where $U$ is the number of units. Evaluation of perturbation curves across 5 datasets and 4 reasoning models shows that AttriCoT produces attributions that are more faithful to the model's behavior than alternative methods. The attribution results also reveal notable differences in thought structure between models and domains.

Foundation models have achieved remarkable performance across medical question answering, imaging, and electronic health record (EHR) tasks, yet reliable clinical deployment remains challenging due to limited interpretability, vulnerability to distribution shift, and weak alignment with clinician reasoning. We argue that these limitations arise because existing approaches prioritize representation learning while treating patient comparison as an emergent property rather than a primary source of clinical evidence. To address this gap, we propose CAFM, a Cohort-Anchored Foundation Model framework that elevates patient cohorts to a first-class object throughout the learning pipeline. The framework consists of four stages: deviation-aware data curation, cohort-conditioned pretraining, multimodal cohort alignment, and clinician-in-the-loop refinement. Together, these stages improve data quality, organize representations around clinically meaningful cohort structure, preserve modality-specific relationships, and support auditable clinical decision-making. The framework is compositional and can augment existing EHR foundation models without modifying their underlying encoders. We illustrate CAFM through four clinical case studies spanning acute kidney injury prediction, cardiovascular risk stratification from electrocardiograms, optic neuropathy triage from orbital imaging, and electroretinogram-grounded report generation. We further present five empirically testable hypotheses and identify open challenges in data quality, irregular temporality, multimodal learning, distribution shift, and evaluation beyond predictive accuracy. We argue that explicitly anchoring foundation models to patient cohorts provides a principled path toward trustworthy clinical AI.

A working citation looks like proof — but the fact that a link resolves does not mean the cited paper supports the claim. I find that current agentic models rarely fabricate citations (over $99%$ resolve), yet roughly $15.9%$ link to the wrong paper. Existing benchmarks miss this failure mode: when a question has a fixed answer key, a model can reproduce the expected source from that key rather than independently verifying that the source supports the claim. I introduce \textbf{\openbiorq{}}, a retrieval-grounded agentic benchmark of $12{,}553$ unsolved biomedical research questions across $12$ domains that treats open questions as a faithfulness-and-abstention probe. To my knowledge, this is the first biomedical benchmark to combine an agentic setting — where the model must issue multiple tool calls — with unsolved questions that have no answer key. Openness is verified against real follow-up evidence rather than a model's parametric knowledge. Difficulty is empirical: I anchor it on questions that three open-weight reference models fail to answer, rather than on subjective hardness labels. On this hardest subset, held-out models from the same lineage as the difficulty anchors solve only ~17%, while three independent frontier agents (Gemini-3-Pro, Opus-4.7, GPT-5.5) span a wide 29-60% range. The benchmark is thus hard, non-saturating (the best agent still leaves ~33-40% unsolved), and discriminating across capability tiers. Beyond difficulty, I observe agentic collapse on the hardest questions, where agents stop using their tools. For the most collapse-prone model, blocking tool access entirely barely changes its score — so tools stop paying off exactly where they are needed most. A frozen per-question checklist raises inter-judge agreement from Spearman 0.35 to 0.82.

Online planning in continuous partially observable Markov decision processes (POMDPs) using $ω$-regular specifications requires handling continuous belief dynamics within the finite symbolic memory in order to track temporal progress. Existing methods based on either direct search in belief space or predefined discrete abstractions suffer from drawbacks, e.g., lack of symbolic memory for long-horizon logical progress or difficult to certify from noisy online beliefs. As such, obtaining reliable symbolic states online from continuous observations remains a challenge. To address this issue, we introduce the Revealed Belief Automaton (REBA), an event-driven framework that advances the research from global belief-space discretization to a fundamental new way of thinking, namely online certification of revelation events. Specifically, we propose an online revelation method that, through information-theoretic gates, can dynamically analyse and establish belief abstraction from the continuous belief space by discovering reliable anchors among noisy beliefs. We then develop an incremental topology adaptation mechanism over the certified anchors to realise the online finite Belief Automaton. By combining with the $ω$-regular specification, REBA is able to support formal parity policy synthesis without a predefined discrete abstraction, which in turn can guide the Monte Carlo Tree Search process to perform online search beyond its local horizon. In addition, we design an error decomposition analysis which can assess the effectiveness and reliability of this discrete guidance for the underlying continuous POMDP. Empirical evaluations in patrolling and navigation scenarios show that REBA matches or exceeds all evaluated baselines, with primary metric gains of +17.0% to +47.4% over state-of-the-art approaches.

Web data curation has been widely studied for decoder Large Language Model (LLM) pretraining. Encoders for dense-terminology domains such as medicine, by contrast, are pretrained on small, manually-curated corpora that limit scalability and writing style diversity, a bottleneck even more severe in non-English clinical settings. Whether web-scale data curation also benefits encoder Masked Language Modeling (MLM) in a dense-terminology domain remains an open question. To address this, we introduce two complementary levers. Medical-term density filtering selects documents rich in medical terms. Signal-amplifying rephrasing uses an LLM to rewrite documents into denser variants with broader entity contexts. We instantiate the recipe on French medical NLP. The medical-term density filter outperforms the widely-used educational quality filter on downstream medical tasks, and the two complement each other. Signal-amplifying rephrasing alone improves on raw web data, and mixing it with filtered web data produces the largest gain. The recipe yields FineMed, a French medical pretraining corpus, and DoctoBERT, a state-of-the-art French medical encoder family evaluated on both the public benchmark DrBenchmark and a proprietary clinical Named Entity Recognition (NER) task.

Precision medicine in ophthalmology requires accurate longitudinal predictions, but the fragmented nature of multimodal clinical data remains a barrier to forecasting. We introduce OphthaDT, an LLM-based digital twin for ophthalmology that serializes longitudinal patient histories from 3,220 patients across four Phase III clinical trials into structured narratives to forecast best corrected visual acuity (BCVA). In benchmarks spanning up to 100 weeks, OphthaDT demonstrated the lowest prediction error in neovascular age-related macular degeneration (nAMD), achieving an average mean absolute error (MAE) reduction of 6.0% compared to all baselines. In diabetic macular edema (DME), OphthaDT demonstrated competitive performance against all baselines while outperforming Random Forest and XGBoost by an average MAE reduction of 2.6% and 6.9%, respectively. Results reveal that OphthaDT's predictive advantage scales with trajectory complexity: whereas linear models remain effective for the more stable treatment responses of DME, OphthaDT's capacity is better suited for capturing the high longitudinal variability of nAMD. Finally, OphthaDT handles irregular sampling without imputation, positioning LLM-based clinical trajectory modeling as a methodology that could reduce patient burden and accelerate drug development.

Multi-turn agents interleave consequential decisions with routine execution: some actions change the downstream return distribution, while others are necessary but reward-equivalent. The cost of trajectory-level credit assignment, often attributed to long horizons, is in fact governed by decision density $ρ$: the fraction of turns whose actions affect the return. When decision density is low, routine turns create signal dilution: they add gradient variance to trajectory-level estimators such as GRPO without adding expected signal. Under explicit assumptions, the resulting turn-level to trajectory-level signal-to-noise ratio scales as $ρ^{-1/2}$, provided critic error remains controlled. The same analysis identifies the complementary regime: at high decision density, trajectory-level methods can remain competitive while avoiding the cost of a critic. In a controlled environment where $ρ$ is exactly tunable, the predicted scaling is recovered with $R^2 = 0.999$, and the training-step gap widens significantly as $ρ\to 0$.

Epidemic forecasting models typically rely on surveillance data reported over administrative regions, treating them as atomic units, thereby obscuring sub-regional spatial structure that shapes disease dynamics. We introduce a spatially structured multimodal epidemic forecasting setting that integrates region-level temporal surveillance data with spatially localized auxiliary signals that are misaligned in resolution and structure, reflecting realistic public health reporting constraints. Building on this formulation, we propose M-SPICE (Multimodal SPatIal Context for Epidemic Forecasting), a structure-aware spatiotemporal forecasting framework that performs joint reasoning over temporal disease dynamics and spatial context via attention-based multimodal fusion, allowing spatial signals to selectively condition temporal representations across forecast horizons. We evaluate our approach on real-world COVID-19, influenza, and influenza-like illness (ILI) forecasting tasks under realistic real-time evaluation protocols. Across all forecasting settings, our method consistently outperforms state-of-the-art multivariate time-series, multimodal, and epidemiological forecasting baselines while maintaining strong probabilistic forecasting performance. Finally, interpretability analyses reveal when, where, and how spatial signals are leveraged, highlighting settings in which purely temporal, region-aggregated models are most likely to fail.

Large language models (LLMs) are increasingly deployed as black-box classifiers in pipelines that automate confident decisions and route uncertain ones to human review. Such selective prediction needs a confidence score that an operator can threshold at a chosen risk level. Prior work asks whether LLM confidence is well calibrated or well ranked; we ask a complementary, deployment-oriented question that has been largely overlooked: at what resolution can the score be thresholded? We call the answer the score granularity gap. Through a controlled comparison of seven ways to build a confidence score, from a single verbalized number, to token probabilities, to querying the model many times and combining the answers, across 25 model-dataset pairs (9 LLMs, 3 benchmarks), we find that single-shot verbalized confidence, once correctly converted to a class probability, ranks cases surprisingly well, yet takes only a handful of distinct values. It therefore offers an operator only a few coarse thresholds, no matter how well it ranks. We show which constructions widen this gap, at what inference cost, and with what effect on ranking, notably that multi-query aggregation helps weak models but can degrade already-strong ones. We translate these trade-offs into concrete deployment guidance.

Epidemiological models often rely on survey data to represent how individuals make health-related decisions, such as whether to vaccinate or adopt protective behaviors. However, repeated large-scale surveys are costly, time-consuming, and limited in the range of scenarios they can capture. In this work, we investigate whether large language models (LLMs) can generate synthetic survey responses that reproduce patterns observed in real populations. Using longitudinal data from the FluPaths surveys, we first identify groups associated with broadly positive or negative attitudes toward vaccination through clustering analysis. We then evaluate several LLMs using a cluster-informed prompting approach to generate synthetic survey responses across multiple epidemic waves. Across models, the synthetic data generally reproduce the distributions of demographic characteristics, vaccination-related beliefs, risk perceptions, and health behaviors observed in the survey data. However, they are less successful at capturing how these factors vary together within respondents. Some models reproduce group-level vaccination trends more reliably than others, although performance varies across waves. We also trained a classifier to distinguish real from synthetic records and found that the generated responses remained identifiable as synthetic. Overall, our findings suggest that LLM-generated survey data may provide a useful tool for exploratory data augmentation and we hope that it could support agent-based epidemic modeling approaches. However, the generated data should not be treated as a substitute for human survey data without further methodological improvements and validation.

We present RoboLineage, an agent-native data lifecycle governance system for robot policy iteration. Modern robot policies improve through repeated data collection, review, retraining, evaluation, and release decisions, but the evidence connecting these steps is often scattered across local tools, scripts, and expert memory. RoboLineage makes this lifecycle explicit by representing rollouts, reviews, dataset decisions, training runs, policy metadata, evaluations, deployment recommendations, and next-collection plans as typed lineage artifacts. Agents interpret embodied rollout evidence, adapt accepted data to existing training stacks, maintain data health, and summarize cross-iteration state under explicit artifact boundaries. In real-robot manipulation workflows, RoboLineage makes routine policy iteration faster and more auditable while maintaining downstream policy performance. We open source RoboLineage as a lightweight lifecycle layer for different robot embodiments and training families. Project page: https://robolineage.github.io/

The increasing deployment of large language model (LLM) agents in collaborative workflows demands robust multi-user, multi-principal interaction mechanisms capable of enforcing access permissions, resolving authoritative conflicts, and preventing unauthorized data disclosure. However, a fundamental mismatch exists between the single-user training paradigm of contemporary LLMs and the hard constraints required for multi-principal governance, rendering probabilistic, prompt-based safeguards vulnerable under multi-turn adversarial interactions.Our key insight is that governance constraints — who is authorized, what is restricted, and whose instructions take precedence — are deterministic runtime variables that should be enforced by execution hooks rather than entrusted to the LLM. We present Harness-MU, the first model-agnostic, zero-tuning infrastructure framework for multi-user LLM agents. By decoupling language generation from safety orchestration, Harness-MU guarantees unbreakable permission boundaries while maximizing compliant demand satisfaction. Across four frontier open-weight and proprietary models on the Muses-Bench benchmark, Harness-MU achieves the goal of privacy preservation across all access-control attacks, outperforming the standard baseline by 0.28--0.39 in utility score and improving instruction-following accuracy by up to 48.9 percentage points. Harness-MU advances the philosophy of Harness Engineering, establishing that systematic infrastructure is essential for solving LLM multi-principal governance challenges. The code and data are available at https://github.com/YuanJrShiuan/Harness-MulUser.

We introduce CFAgentBench, a reproducible, self-hostable environment and benchmark for autonomous construction-finance agents: a CFO/controller-class agent operating across the real software stack a US construction finance team runs - ERP, project management, email, documents, pay applications, payroll, certified payroll, lien waivers, and bank/treasury portals. It contains 1,014 machine-gradeable task specifications across 8 domains and 77 families, every family grounded in a real source; a self-validated subset of 40 tasks (54 with a project-management extension) is compiled into oracle-validated executable evaluators, the runnable suite reported here. Following WebArena, the benchmark runs on an executable environment rather than static traces: 35 mock applications (31 reconciled to one company book, plus 4 PM platforms) over 9 archetypes, each implementing a uniform self-hostable app contract, so every task is graded by functional correctness - a state diff plus forbidden-side-effect checks plus required-output regexes - with an LLM judge used only for reply quality, never as reward. A distinguishing principle is a money-movement guard: 278 instances embed a payment, payroll, e-signature, or e-filing step where the correct behavior is to stop and stage for human approval, and executing even the correct transaction fails the task. The public split (n=711) is sized for a 95% Wilson half-width of +/-4.1%; a private, contamination-protected split (n=303) is reserved for remote scoring. In a first three-model open-weight sweep (k=5), the strongest agent reaches pass^1 = 0.67 but only pass^5 = 0.38 - losing 43% of its successes when required to repeat them under temperature-0 decoding. The within-model pass^1 to pass^5 collapse and sharp per-domain heterogeneity are clear evidence that single-attempt accuracy overstates deployable construction-finance competence.

End-to-end code agent training is resource-intensive and plateaus on the strategy-level reasoning needed to resolve code issues, since jointly optimizing code-level execution and strategy-level reasoning leaves the latter underdeveloped. Instead, we freeze the agent and add a critic model to supply that signal. Prior code critics are post-hoc, scoring completed trajectories rather than steering the agent; we instead train a small critic that provides intra-trajectory feedback via Supervised Fine-Tuning. On SWE-bench Verified, a critic trained on CWM-32B trajectories transfers to two unseen agents (gains of +3.0 to +3.8 points), and adding target-agent trajectories to the corpus increases the gain to +3.8 on CWM-32B and +4.4 to +5.2 on two Qwen agents, at 30-92x lower critic cost than a strong teacher. On Qwen3-Next-80B-A3B, the critic-guided system is both more accurate (25.2% vs. 20.8%) and cheaper (\$0.04 vs. \$0.11) than the agent alone, because the critic also shortens trajectories. Our results show that a small, well-trained critic is a practical complement to scaling agent training. Code: https://github.com/shubhamrgandhi/critic-training. Data and models: https://huggingface.co/collections/shubhamrgandhi/critic-training-for-code-agents

Scientific discovery can be formulated as an iterative search process over the space of hypotheses and experiments. Contemporary methods navigate this space using heuristics such as MCTS. These algorithms conflate the merit of a hypothesis with the quality of its experimental execution. A promising hypothesis with preliminary execution is therefore ranked below a modest hypothesis whose execution is refined. Moreover, prior methods prune the search logs as the search progresses because the accumulated history outgrows the context window. We propose Agentic Reasoning for Tree Search (ARTS), where we deploy a reasoning language model to navigate this space. The model inspects prior execution logs, diagnoses whether earlier failures arose from faulty implementations or bad hypotheses, and selects the hypothesis to build on next. To mitigate challenges with context length, ARTS uses test-time training to instill the knowledge of search tree in the model weights. Across 22 tasks from MLGym and MLEBench, we show that ARTS outperforms leading algorithms, with over 15.3% relative improvement in the normalized score. With test-time training we show that a Qwen3-4B agent can match performance with closed-source frontier models like Gemini-3 Pro and GPT o3-reasoning with upto 5x lower inference cost. We further observe that on partially observable RL tasks, the test-time trained Qwen3-4B scientist surpasses ARTS with the o3 scientist by rediscovering the human-best recurrent-memory solution that heuristic methods prune away.