探索全球前沿学术脉络

01.

arXiv (CS.AI) 2026-06-18 DOI: arXiv:2605.17986

LivePI: More Realistic Benchmarking of Agents Against Indirect Prompt Injection

作者:

Lei Zhao↗Abhay Bhaskar↗Edgar Dobriban↗

arXiv:2605.17986v3 Announce Type: replace-cross Abstract: AI agents such as OpenClaw are increasingly deployed in local workflows with access to external tools. This creates indirect prompt-injection (IPI) risk: an agent may execute harmful instructions embedded in untrusted inputs such as email, downloaded files, webpages, repositories, or group-chat messages. Existing evaluations are often small, purely simulated, or focused on a narrow set of channels. We introduce LivePI (Live Prompt Injection), a structured benchmark for IPI risk in a production-like but test-controlled environment. LivePI covers seven input surfaces, twelve attack/rendering families, and five malicious goals, including protected-information exfiltration, unauthorized security-control changes, unsafe code retrieval or execution, inbox-summary exfiltration, and cryptocurrency transfer. We run LivePI on a real virtual machine with live but test-controlled email, chat, web, local-file, repository, and wallet interfaces. Across GPT-5.3-Codex, Claude Opus 4.6, Gemini 3.1 Pro, Kimi K2.5, and GLM-5, total attack success rates range from 10.7% to 29.6%. Group-chat injection is uniformly successful across the evaluated backbones in our deployment, and repository-link attacks produce high-severity failures despite a small denominator. We also evaluate a two-layer defense consisting of prompt-level filtering and pre-execution tool-call authorization. In the GPT-5.3-Codex setting, the defense intercepts all tested malicious-goal completions in LivePI before execution while preserving benign utility on PinchBench-derived workloads.

阅读与讨论 → 访问原文 →

02.

arXiv (CS.CV) 2026-06-18 DOI: arXiv:2606.03827

Conditional Latent Diffusion Model with Fourier-based Motion Modelling for Virtual Population Synthesis

作者:

Shaokun Lan↗Haoran Dou↗Jinghan Huang↗Arezoo Zakeri↗Fengming Lin↗Zherui Zhou↗Jinming Duan↗Alejandro F. Frangi↗

In-silico trials of medical devices require the generation of virtual populations of anatomies. In cardiovascular applications, virtual anatomy is typically represented as a 3D+t mesh sampled from a generative model. However, most existing mesh generators focus on static anatomy, while sequence models often lack explicit periodicity. To this end, we propose 4D F-MeshLDM, a conditional generative framework comprising a convolutional mesh VAE to encode meshes, a structural latent space that parameterises motion using a truncated Fourier series, and a diffusion prior that learns the latent distribution over Fourier coefficient tokens. By conditioning the diffusion process on clinical covariates via affine modulation, we enable controllable synthesis. Sampling tokens and performing inverse Fourier synthesis yield cycle-consistent latent trajectories, which can be decoded into 3D+t cardiac mesh sequences. Experiments on 5,000 UK Biobank subjects demonstrate that 4D F-MeshLDM outperforms state-of-the-art baselines in anatomical fidelity and achieves near-zero cycle closure error. Furthermore, the generated cohorts accurately preserve clinical functional indices, highlighting the potential of our framework for reliable in-silico cardiac trials.

阅读与讨论 → 访问原文 →

03.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2606.15576

Localizing Credit at the Divergence: Path-Conditioned Self-Distillation for LLM Reasoning

作者:

Yu Li↗Shu Hong↗Tian Lan↗

arXiv:2606.15576v1 Announce Type: cross Abstract: Reinforcement learning from verifiable rewards assigns a single scalar to each rollout, leaving token-level credit assignment underspecified in long reasoning traces. On-policy self-distillation addresses this by letting the same model act as a teacher conditioned on privileged information, producing a dense per-token signal. But the common choice of a ground-truth answer is only an endpoint cue: on terse-answer tasks, the teacher falls silent at the intermediate positions where path-level guidance matters most. We propose Hindsight Self-Distillation (HSD), which conditions the teacher on a successful peer rollout drawn from the current training group. Such a peer is an exact sample from the success-conditioned policy, requiring no additional sampled rollouts. By providing a full successful continuation rather than only the final answer, the resulting credit signal concentrates at the divergence position between a failed rollout and a successful peer. Across Qwen3-8B and Qwen3-32B on math and code benchmarks, HSD obtains the best result against GRPO variants and on-policy distillation baselines, with the largest gains on terse-answer tasks such as AIME.

阅读与讨论 → 访问原文 →

04.

arXiv (CS.CL) 2026-06-18 DOI: arXiv:2606.19236

STARE: Surprisal-Guided Token-Level Advantage Reweighting for Policy Entropy Stability

作者:

Haipeng Luo↗Qingfeng Sun↗Songli Wu↗Can Xu↗Wenfeng Deng↗Han Hu↗Yansong Tang↗

Reinforcement Learning with Verifiable Rewards algorithms like GRPO have emerged as the dominant post-training paradigm for complex reasoning in LLMs, yet commonly suffer from policy entropy collapse during training. We conduct a first-order gradient analysis of token-level entropy dynamics under GRPO and identify a token-level credit assignment mismatch: the per-token entropy variation decomposes into the product of the trajectory-level advantage and an entropy sensitivity function over the next-token distribution, yielding an advantage-surprisal four-quadrant structure and a near-criticality property. Motivated by it, we propose STARE (Surprisal-guided Token-level Advantage Reweighting for policy Entropy stability), which identifies entropy-critical token subsets via batch-internal surprisal quantiles, selectively reweights their effective advantages, and incorporates a target-entropy closed-loop gate for stable entropy regulation. Across model scales from 1.5B to 32B and three task families (Short CoT, Long CoT, and Multi-Turn Tool Use), STARE sustains stable RL training over thousands of steps while maintaining policy entropy within the target band. On AIME24 and AIME25, STARE outperforms DAPO and other competitive baselines by 4%-8% in average accuracy, with reflection tokens and response length growing in tandem, indicating sustained exploration-exploitation balance that further unlocks RL training potential.Code is available at https://github.com/hp-luo/STARE.

阅读与讨论 → 访问原文 →

05.

arXiv (CS.LG) 2026-06-12 DOI: arXiv:2606.13426

Accelerating Speculative Diffusions via Block Verification

作者:

Alexander Soen↗Hisham Husain↗Valentin De Bortoli↗Arnaud Doucet↗

arXiv:2606.13426v1 Announce Type: new Abstract: Speculative decoding speeds up LLM inference by using a draft model to generate tokens, with an acceptance-rejection scheme that ensures that the output matches the target distribution. Adapting this to continuous diffusions is difficult because speculative sampling requires drawing from a residual distribution. While straightforward in discrete spaces, efficiently sampling this residual in continuous space is non-trivial. Consequently, existing diffusion adaptations either use computationally inefficient sampling techniques or rely on an alternative scheme. In this work, we introduce a novel scheme that efficiently implements the original speculative sampling mechanism for diffusion models. Our approach offers a critical advantage over current methods: it enables us to adapt block verification from LLMs to diffusions – which provably improves the acceptance rate of drafts. Furthermore, we formalize and analyze the Free Drafter, a heuristic self-speculative drafter for diffusions that requires no training. By enabling block verification, our Free Drafter yields up to a 6.3% speedup over existing speculative methods with no additional training and negligible overhead beyond the existing parallel verification pass.

阅读与讨论 → 访问原文 →

06.

arXiv (CS.AI) 2026-06-15 DOI: arXiv:2606.14502

From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI

作者:

Yongheng Zhang↗Ziang Liu↗Jiaxuan Zhu↗Shuai Wang↗Xiangqi Chen↗Haojing Huang↗Jiayi Kuang↗Siyu Chen↗Ao Shen↗Hao Wu↗Qiufeng Wang↗Qian-Wen Zhang↗…

arXiv:2606.14502v1 Announce Type: new Abstract: Large Language Models (LLMs) are undergoing a fundamental transformation from conversational generators into integrated AI systems capable of reasoning, action, memory, and self-improvement. We conceptualize this transition as a shift from Chatbot to Digital Colleague: from conversational answers to persistent work. We organize this transition along two tightly coupled dimensions. First, at the cognitive core level, LLMs are advancing from Chatbot-era "fast thinking" systems driven by next-token prediction toward Thinking LLMs that leverage inference-time computation, Chain-of-Thought reasoning, reflection, process supervision, and reinforcement learning to support more deliberate and reliable cognition. Second, at the tool-augmented task execution level, LLMs are progressing from tool-calling Agents that invoke external resources in an ad hoc manner toward OpenClaw-style workstation systems (OpenClaw) equipped with persistent Workspaces, skills, verification loops, and governance. The "Workspace + Skill" paradigm makes episodic tool use colleague-like via state persistence, reusable procedures, task closure, and experience reuse. We examine data construction shifts from instruction-response pairs to State-Action-Observation trajectories and evaluation from static benchmarks to sandboxed, auditable, self-evolving AI ecosystems.

阅读与讨论 → 访问原文 →

07.

arXiv (CS.AI) 2026-06-18 DOI: arXiv:2606.19168

Beyond Safe Data: Pretraining-Stage Alignment with Regular Safety Reflection

作者:

Jinhan Li↗Kexian Tang↗Yihan Xu↗Zhuorui Ye↗Kaifeng Lyu↗

arXiv:2606.19168v1 Announce Type: new Abstract: To achieve deeper safety alignment for large language models (LLMs), recent efforts have studied how to push safety interventions earlier into the pretraining stage, primarily by filtering unsafe data or rewriting it into safer forms. We argue that pretraining-stage alignment should go beyond making the data safe: LLMs may compose seemingly benign knowledge and capabilities into unsafe behaviors. To this end, we propose Safety Reflection Pretraining, a pretraining-stage alignment method which regularly inserts short safety reflections into pretraining corpora to integrate self-monitoring directly into language modeling, establishing a foundational capability that is subsequently reinforced by compatible post-training. Our experiments with 1.7B models pretrained on FineWeb-Edu show that Safety Reflection Pretraining improves safety classification accuracy and substantially reduces the success rates of inference-stage and finetuning attacks. Complementary to our real-world experiments, we also introduce a fully controlled synthetic environment, MedSafetyWorld, with a clear definition of safety and a reasoning structure under which models can easily generalize unsafe behaviors from safe data. Ablations in MedSafetyWorld further demonstrate a clear advantage of Safety Reflection Pretraining in preventing models from acting on unsafe behaviors generalized from safe data, compared with data filtering and rewriting. Taken together, our findings suggest that pretraining alignment should not only make the training data safe, but also shape the behaviors that models are likely to acquire from safe data.

阅读与讨论 → 访问原文 →

08.

medRxiv (Medicine) 2026-06-16 DOI: HASH:3f270ea261d512e150c6fb255dffd515

Care Delivery Gap framework: a proof-of-concept patient-reported measure of guideline-referenced care-process omissions in sickle cell disease

作者:

Agbalalah↗Rowaiye↗

Abstract Background:Sickle cell disease (SCD) is concentrated in sub-Saharan Africa, where delivery of guideline-referenced care remains challenging. Current evaluation approaches rely largely on access indicators and clinical outcomes, which do not directly measure care delivery. We developed the Care Delivery Gap (CDG) framework, a patient-reported approach for identifying care-process omissions, and conducted a proof-of-concept study to assess feasibility and explore variation across income strata. Methods: We conducted a cross-sectional framework-development study involving a proof-of-concept sample of 52 individuals with SCD or caregivers recruited through clinics and moderated SCD communities across Africa, North America, and Europe between June 2025 and March 2026. The CDG framework assessed patient-reported omissions in specialist involvement, follow-up continuity, cardiovascular screening, and biochemical surveillance. Analyses were descriptive. Results: Substantial multi-domain care-process omissions were identified despite high reported healthcare engagement. Across geographic income strata, cardiovascular screening was reported by 4/35 (11%) LMIC versus 16/17 (94%) HIC participants, and regular follow-up within the preceding 12 months by 14/35 (40%) versus 16/17 (94%), respectively. High CDG scores, representing 1 omissions across three or four domains, occurred in 20/35 (57%) LMIC compared with 1/17 (6%) HIC participants. Similar disparities were observed across specialist review and vitamin B12 surveillance domains. Conclusion: A structured patient-reported framework identified multi-domain omissions in guideline-referenced SCD care, including among individuals reporting healthcare access. The divergence between access indicators and reported care delivery suggests that service contact alone may not reflect care quality. The framework provides a feasible foundation for future process-level quality measurement in high-burden settings.

阅读与讨论 → 访问原文 →

09.

arXiv (CS.CL) 2026-06-12 DOI: arXiv:2606.12904

Trait, Not State: The Durability of Reading Identity in Social Highlighting

作者:

Kazuki Nakayashiki↗Keisuke Watanabe↗

Prior work on a social web highlighter located individuality in selection – which documents a person chooses to highlight – but measured it cross-sectionally. We ask the temporal question: is a reader's selection signature a trait or a state? We freeze each reader's first six months of highlighting as a profile and track its own-vs-other advantage on their later selections at growing gaps (to 24+ months), with negatives drawn from the same calendar era – so supply drift cannot masquerade as personal drift – at a coarse global level and at a fine level whose negatives and controls come from the reader's own interest neighborhood; the anchor cell reproduces the prior cross-sectional level (+0.188 vs +0.169), validating the harness. Four results. Within the same users, the fine-layer advantage shows no statistically detectable paired decline at any horizon (6-12 month retention R = 1.00 [0.85, 1.18], n = 212; the farthest bin is compatible with a modest decline; the only contrast whose interval excludes zero is the coarse layer at 12-24 months, about 13%). The signal is not reducible to repeated domains (~90% survives excluding all profile sources). Within-person drift is slow (a recent-half profile beats the old half by +0.042). Prospectively, personal profiles – even one built from a reader's earliest documents, median 20 months before evaluation – rank their next reads at roughly 3x the AP of every simple non-personal prior tested. We use "trait" operationally (a stable signature under continued engagement); the scope is heavy, long-tenured readers of one platform, and exposure is not separable from choice.

阅读与讨论 → 访问原文 →

10.

arXiv (quant-ph) 2026-06-11 DOI: arXiv:2305.04908

Tight Bounds for Quantum Phase Estimation and Related Problems

作者:

Nikhil S. Mande↗Ronald de Wolf↗

arXiv:2305.04908v3 Announce Type: replace Abstract: Phase estimation, due to Kitaev [arXiv'95], is one of the most fundamental subroutines in quantum computing. In the basic scenario, one is given black-box access to a unitary $U$, and an eigenstate $\lvert \psi \rangle$ of $U$ with unknown eigenvalue $e^{i\theta}$, and the task is to estimate the eigenphase $\theta$ within $\pm\delta$, with high probability. The cost of an algorithm for us is the number of applications of $U$ and $U^{-1}$. We tightly characterize the cost of several variants of phase estimation where we are no longer given an eigenstate, but are required to estimate the maximum eigenphase of $U$, aided by advice in the form of states (or a unitary preparing those states) which are promised to have at least a certain overlap $\gamma$ with the top eigenspace. We give algorithms and nearly matching lower bounds for all ranges of parameters. We show that a small number of copies of the advice state (or of an advice-preparing unitary) are not significantly better than having no advice at all. We also show that having lots of advice (applications of the advice-preparing unitary) does not significantly reduce cost, and neither does knowledge of the eigenbasis of $U$. We immediately obtain a lower bound on the complexity of the Unitary recurrence time problem, resolving an open question of She and Yuen~[ITCS'23]. Lastly, we study how efficiently one can reduce the error probability in the basic phase-estimation scenario. We show that a phase-estimation algorithm with precision $\delta$ and error probability $\epsilon$ has cost $\Omega\left(\frac{1}{\delta}\log\frac{1}{\epsilon}\right)$, matching an easy upper bound. This contrasts with some other scenarios in quantum computing (e.g., search) where error-probability reduction costs only a factor $O(\sqrt{\log(1/\epsilon)})$. Our lower bound uses a variant of the polynomial method with trigonometric polynomials.

阅读与讨论 → 访问原文 →

11.

arXiv (CS.AI) 2026-06-12 DOI: arXiv:2603.21563

Counterfactual Credit Policy Optimization for Multi-Agent Collaboration

作者:

Zhongyi Li↗Wan Tian↗Jinju Chen↗Huiming Zhang↗Yang Liu↗Yikun Ban↗Fuzhen Zhuang↗

arXiv:2603.21563v5 Announce Type: replace Abstract: Collaborative multi-agent large language models (LLMs) can solve complex reasoning tasks by decomposing roles, but reinforcement learning for such systems is limited by credit assignment: shared terminal rewards obscure individual contributions and can encourage free-riding. We introduce two optimizer-agnostic credit assignment methods for converting joint outcomes into agent-specific learning signals. Counterfactual Credit for Policy Optimization (CCPO) estimates an agent's marginal contribution by comparing the realized joint outcome with a counterfactual outcome where that agent is removed. Self-Evaluated Credit for Policy Optimization (SEPO) uses constrained self- and peer-evaluations as a verifier-anchored credit signal while keeping the external task outcome dominant. Both operate at the reward-construction layer rather than as policy optimizers, producing role-specific rewards or advantages for GRPO, GSPO, or REINFORCE++. We instantiate these credit signals in a sequential Think–Solve setting and evaluate them on mathematical reasoning benchmarks. Results show that explicit credit assignment often improves dual-agent reasoning, especially on MATH500 and several out-of-distribution settings, while gains vary across models and datasets. Our code is available at: https://github.com/bhai114/ccpo.

阅读与讨论 → 访问原文 →

12.

arXiv (CS.AI) 2026-06-12 DOI: arXiv:2606.12647

Token Complexity Theory for AI-Augmented Computing

作者:

Jie Wang↗

arXiv:2606.12647v1 Announce Type: cross Abstract: AI-augmented computing delegates natural language queries, code generation requests, and other open-ended tasks to a cluster of AI models that processes queries and generates responses. This paradigm introduces a resource dimension that neither classical time nor space complexity captures: the cost of sending queries to and receiving responses from such a cluster. We introduce token complexity, a formal resource measure defined as the minimum expected token cost to achieve a specified level of output quality on a task, and develop a taxonomy classifying AI systems by the strength of their probabilistic properties. We develop token complexity within the framework of AI-Oracle Turing machines, in which a probabilistic Turing machine interacts with a stochastic oracle via dedicated query and response tapes. We prove basic theorems establishing that token complexity behaves as expected: monotonicity (higher quality costs more tokens), convexity (quality improvements become progressively more expensive), price sensitivity (small price changes produce bounded cost changes), and price-relativity of task ordering (the token complexity ordering of tasks can reverse depending on the query-to-response cost ratio). We prove that the complexity frontier, defined as the set of all feasible resource bounds in tokens, time, and space, is non-empty, upward-closed, and convex.

阅读与讨论 → 访问原文 →

13.

arXiv (CS.LG) 2026-06-18 DOI: arXiv:2507.17786

Reinforcement Learning for Accelerated Aerodynamic Shape Optimisation

作者:

Florian Sobieczky↗Alfredo Lopez↗Erika Dudkin↗Christopher Lackner↗Matthias Hochsteger↗Bernhard Scheichl↗Helmut Sobieczky↗

arXiv:2507.17786v2 Announce Type: replace Abstract: We introduce a reinforcement learning (RL) based adaptive optimization algorithm for aerodynamic shape optimization focused on dimensionality reduction. The form in which RL is applied here is that of a surrogate-based, actor-critic policy evaluation MCMC approach allowing for temporal 'freezing' of some of the parameters to be optimized. The goals are to minimize computational effort, and to use the observed optimization results for interpretation of the discovered extrema in terms of their role in achieving the desired flow-field. By a sequence of local optimized parameter changes around intermediate CFD simulations acting as ground truth, it is possible to speed up the global optimization if (a) the local neighbourhoods of the parameters in which the changed parameters must reside are sufficiently large to compete with the grid-sized steps and its large number of simulations, and (b) the estimates of the rewards and costs on these neighbourhoods necessary for a good step-wise parameter adaption are sufficiently accurate. We give an example of a simple fluid-dynamical problem on which the method allows interpretation in the sense of a feature importance scoring.

阅读与讨论 → 访问原文 →

14.

arXiv (CS.LG) 2026-06-15 DOI: arXiv:2606.13970

An Attention-based Model for Robust Forecasting with Missing Modality

作者:

Zhitian Zhang↗Wenjie Zi↗Yunduz Rakhmangulova↗Saghar Irandoust↗Hossein Hajimirsadeghi↗Thibaut Durand↗

arXiv:2606.13970v1 Announce Type: cross Abstract: Learning with missing modalities is a fundamental challenge in multimodal robot learning, as real-world robotic systems often operate in environments with incomplete sensor data. Attention-based models are appealing for processing multimodal data because they can handle multiple modalities with a single backbone network. However, most multimodal models assume that all modalities are available during both training and inference, limiting their applicability in robotic perception and decision-making. In this paper, we introduce a multimodal model designed to handle missing modalities during both training and inference. The model is formulated as a conditional variational autoencoder (CVAE) and incorporates a transformer-based architecture that leverages attention mechanisms to learn a unified, fixed-dimensional representation, even when some modalities are missing. We show that our proposed model can be trained with missing modalities while approximating a robust representation of all modalities. We evaluate our approach on five multimodal datasets across two robot learning tasks: human trajectory prediction and robot manipulation forecasting. Experimental results demonstrate that our model effectively learns from incomplete data and is superior to prior multimodal fusion approaches.

阅读与讨论 → 访问原文 →

15.

arXiv (math.PR) 2026-06-11 DOI: arXiv:2605.25711

On the $d$-rigidity phase transition in random graphs

作者:

Yuval Peled↗

arXiv:2605.25711v2 Announce Type: replace-cross Abstract: We study generic $d$-dimensional rigidity in sparse random graphs. Our main result is that for every $d\ge 2$, the Erdős–Rényi random graph $G\sim G(n,c/n)$ undergoes a $d$-rigidity phase transition at the known, explicit, $d$-orientability threshold $c_d$: If $cc_d$, then $G$ is a.a.s. not independent in the generic $d$-rigidity matroid, and we give a sharp asymptotic estimate for its rank. In addition, the $d$-rigidity closure of $G$ has a giant clique of linear size, which contains all but at most $o(n)$ vertices of the $((d+1)+d)$-core of the graph. More generally, we compute, up to a $1+o(1)$ factor, the generic $d$-rigidity rank of random graphs with a given degree distribution. For example, we show that the uniform $n$-vertex $k$-regular graph a.a.s. has rank $\min(k/2,d)n+o(n).$ Our approach is to estimate the rigidity rank of a random graph from its Galton–Watson local weak limit, using a parameter that we call local flexibility.

阅读与讨论 → 访问原文 →

16.

Nature (Science) 2026-06-10 DOI: HASH:cebe5cb90153bd26e70bf99154122d7b

A vast whale necropolis has been found

作者:

Stephen J. Godfrey↗

In the Indian Ocean, a deep-sea area roughly 1,200 kilometres long and 7 kilometres deep was found to harbour an ecological landmark site of whale remains. In the Indian Ocean, a deep-sea area roughly 1,200 kilometres long and 7 kilometres deep was found to harbour an ecological landmark site of whale remains.

阅读与讨论 → 访问原文 →

17.

bioRxiv (Bioinfo) 2026-06-14 DOI: HASH:8c60dda537b121ba823759c6093194ff

TopoMIL: Topology Improves Multiple Instance Learning in Diagnostic Microscopic Images

作者:

Kazeminia↗Dasdelen↗M. F↗Rieck↗Marr↗

Microscopic images of cells and tissues are central to disease diagnosis. In computational pathology, multiple instance learning (MIL) has emerged as a key paradigm for analyzing numerous images within a single patient sample. While the representative distribution of cells in a sample is important for diagnosis, existing MIL frameworks largely overlook it. We introduce TopoMIL, a framework that extracts the representative topological structure of the sample and integrates it into the MIL classifier. Three topological representations are assessed, each with distinct advantages and computational costs. We evaluate TopoMIL on four histopathology and cytomorphology datasets, each presenting unique challenges. Integrating the sample's topological information into MIL enhances classification across average, max, attention-based, and transformer pooling, yielding AUCROC gains of 3.3%, 4.2%, 5.9%, and 0.5%, respectively, with moderate computational cost. Our work underscores the potential of TopoMIL as a scalable extension to existing morphology-based models in computational pathology.

阅读与讨论 → 访问原文 →

18.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.16015

Stringalign: Moving beyond summary statistics with a transparent Unicode-aware tool for evaluating automatic transcription models

作者:

Yngve Mardal Moe↗Marie Roald↗

Comparing text strings is crucial when evaluating and understanding the performance of various text processing tasks such as document recognition and audio transcription. With an increasingly complex landscape of AI-based handwritten text recognition (HTR), optical character recognition (OCR) and automatic speech recognition (ASR) models, there is a need for tools that facilitate evaluation in a flexible and reproducible way. This paper presents Stringalign, a Python library designed to simplify the evaluation process for automatic transcription projects and facilitate transparent evaluation. Stringalign's tools to examine and visualise both the rate of errors and the types of errors a model makes, give insights into possible improvements and help inform model selection for a particular task. Widely used string comparison metrics, such as the character and word error rates (CER and WER), although useful, can be ambiguous due to varying definitions of what constitutes a character and a word. Stringalign addresses this challenge by ensuring all preprocessing (i.e. normalisation and tokenisation) is transparent and easily replicable, and by providing tools to move beyond summary statistics and analyse common model errors. Moreover, Stringalign adheres to FAIR (Findable, Accessible, Interoperable, and Reusable) principles for research software while staying lightweight and easy to adapt into researchers existing workflows. In this paper, we discuss challenges with character and word level string comparisons and show through examples that where existing tools can yield opaque and sometimes confusing results, Stringalign provides an easy-to-use and unambiguous alternative.

阅读与讨论 → 访问原文 →

19.

arXiv (CS.LG) 2026-06-19 DOI: arXiv:2606.19656

DF-ExpEnse: Diffusion Filtered Exploration for Sample Efficient Finetuning

作者:

Calvin Luo↗Chen Sun↗Shuran Song↗

arXiv:2606.19656v1 Announce Type: cross Abstract: A natural recipe for intelligent robotic decision-making is initializing from pretrained generative control policies, which have summarized offline experience, and adapting them to self-collected online experience. We present DF-ExpEnse, an exploration technique that improves the quality of online experience collection, thus increasing finetuning sample-efficiency. DF-ExpEnse leverages the multimodal modeling capabilities of the generative control policy to create an expressive and tractably evaluatable candidate set. It then utilizes an ensemble of critics to identify the action that best balances quality with high exploration interest. In fleet settings, DF-ExpEnse further enables cross-agent communication to facilitate collaborative exploration as a group. DF-ExpEnse can be seamlessly integrated with existing strategies that finetune pretrained generative control policies via reinforcement learning. We experimentally validate consistent sample-efficiency benefits through DF-ExpEnse across a variety of manipulation and locomotion tasks, compared to default finetuning and alternative action selection schemes. Project can be found at https://df-expense.github.io.

阅读与讨论 → 访问原文 →

20.

arXiv (CS.CL) 2026-06-12 DOI: arXiv:2604.12002

Self-Distillation Zero: Self-Revision Turns Binary Rewards into Dense Supervision

作者:

Yinghui He↗Simran Kaur↗Adithya Bhaskar↗Yongjin Yang↗Jiarui Liu↗Narutatsu Ri↗Liam Fowl↗Abhishek Panigrahi↗Danqi Chen↗Sanjeev Arora↗

Current post-training methods in verifiable settings fall into two categories. Reinforcement learning (RLVR) relies on binary rewards, which are broadly applicable and powerful, but provide only sparse supervision during training. Distillation provides dense token-level supervision, typically obtained from an external teacher or using high-quality demonstrations. Collecting such supervision can be costly or unavailable. We propose Self-Distillation Zero (SD-Zero), a method that is substantially more training sample-efficient than RL and does not require an external teacher or high-quality demonstrations. SD-Zero trains a single model to play two roles: a Generator, which produces an initial response, and a Reviser, which conditions on that response and its binary reward to produce an improved response. We then perform on-policy self-distillation to distill the reviser into the generator, using the reviser's token distributions conditioned on the generator's response and its reward as supervision. In effect, SD-Zero trains the model to transform binary rewards into dense token-level self-supervision. On math and code reasoning benchmarks with Qwen3-4B-Instruct and Olmo-3-7B-Instruct, SD-Zero improves performance by at least 10% over the base models and outperforms strong baselines, including Rejection Fine-Tuning (RFT), GRPO, and Self-Distillation Fine-Tuning (SDFT), under the same question set and training sample budget. Extensive ablation studies show two novel characteristics of our proposed algorithm: (a) token-level self-localization, where the reviser can identify the key tokens that need to be revised in the generator's response based on reward, and (b) iterative self-evolution, where the improving ability to revise answers can be distilled back into generation performance with regular teacher synchronization. Code: https://github.com/princeton-pli/Self-Distillation-Zero.

阅读与讨论 → 访问原文 →

21.

arXiv (CS.CL) 2026-06-18 DOI: arXiv:2606.19183

Language Models as Interfaces, Not Oracles: A Hybrid LLM-ML System for Pediatric Appendicitis

作者:

Soheyl Bateni↗Maryam Abdolali↗

Large language models (LLMs) can make clinical decision support more accessible by interpreting free-text documentation, but their direct use as diagnostic engines is limited by sensitivity to prompts, information order, and plausible but incorrect outputs. Structured machine-learning models offer more stable risk prediction, yet they require tabular inputs that are difficult to integrate with narrative clinical workflows. We present ClaMPAPP (Clinical Language-assisted Machine-learning Pipeline for Appendicitis), a hybrid system that uses an LLM as an interface rather than as the final decision-maker. ClaMPAPP extracts schema-constrained clinical features from note-like narratives, applies deterministic plausibility checks, and passes validated features to an XGBoost classifier trained on clinical, laboratory, and ultrasound variables. We evaluated ClaMPAPP on two independent pediatric appendicitis cohorts from German hospitals and compared it with end-to-end LLM baselines, including open-source and proprietary models. To preserve ground truth while testing free-text input, narratives were generated from structured electronic health records through template rendering and constrained LLM rewriting, with additional sentence-order permutation to assess positional robustness. ClaMPAPP achieved the strongest overall diagnostic performance in both internal and external validation while minimizing missed appendicitis cases, the key safety concern in acute triage. End-to-end LLMs showed unstable sensitivity-specificity trade-offs and greater degradation under narrative reordering. These results support an LLM-as-interface, ML-as-predictor design that separates natural-language usability from predictive inference and provides a more auditable pathway for clinical decision support.

阅读与讨论 → 访问原文 →

22.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.16748

MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents

作者:

Lawrence Keunho Jang↗Andrew Keunwoo Jang↗Jing Yu Koh↗Ruslan Salakhutdinov↗

Current benchmarks for computer-use agents evaluate models in impersonal environments. This leaves a gap between evaluation and deployment where personal assistants are expected to work across a user's whole digital life, including their context, historical data, and logged-in accounts. This gap is widest on web tasks, where live web evaluations cannot exercise sites that require logging in or personal information, the kind of site a real personal assistant has to drive. We introduce MyPCBench, which tests computer-use agents as personal assistants on a Linux desktop populated with 17 simulated real-world web applications and a full desktop stack, all seeded for one canonical persona, Michael Scott from The Office. We define 184 tasks in this environment, each inspired by a real request drawn from the OpenClaw community, and benchmark six closed and open-weight models with a uniform computer+bash tool surface. We find that the best model, Claude Opus 4.6, fully solves 55.4\% of the tasks, the only model above 50\%. Model failures cluster on tasks that span many applications and on long trajectories, where personalization stresses an assistant the most. We release the environment, task set, and agent harness at https://mypcbench.com.

阅读与讨论 → 访问原文 →

23.

Nature (Science) 2026-06-17 DOI: HASH:55ce571917bd9048e367ca8acfa52259

Analysis of 173,303 exomes and genomes in the Pakistan Genome Resource

作者:

Christopher Koch↗

Naturally occurring loss-of-function variants in human genes enable drug target discovery because they mimic pharmacological inhibition of proteins. However, the study of these genetic variants is constrained by their rarity. Sequencing of diverse populations, particularly those enriched in familial relatedness, has been postulated to promote discovery of rare genetic variants1–3. Here we present the Pakistan Genome Resource, a South Asian biobank with high familial relatedness comprising 173,303 participants, who collectively carry naturally occurring homozygous loss-of-function variants in 6,476 genes. We describe the genetic architecture of this population, associations between genes and biomarkers, the distribution of loss-of-function variants across molecular pathways, and recall-by-genotype studies of therapeutically relevant genes. The Pakistan Genome Resource expands the catalogue of human genetic variants, provides a comprehensive genetic reference resource for the Pakistani population, and demonstrates the value of studying diverse cohorts to advance human health. The Pakistan Genome Resource compiles biobank data from 173,303 individuals with high familial relatedness, broadening the catalogue of human genetic variation and establishing a population-specific genomic reference for Pakistan.

阅读与讨论 → 访问原文 →

24.

arXiv (CS.LG) 2026-06-16 DOI: arXiv:2606.15950

Spectral Adaptive Conformal Prediction for Structured Non-Exchangeable Data

作者:

Jeffery Opoku↗David Banahene↗

arXiv:2606.15950v1 Announce Type: cross Abstract: Conformal prediction gives prediction intervals with finite-sample coverage when the data are exchangeable. Many time-indexed datasets are not exchangeable. They have seasons, recurring regimes, changing frequencies, or other forms of structured dependence. This paper studies a simple way to use that structure. We propose spectral adaptive conformal prediction, a method that forms weighted conformal quantiles using local spectral similarity and then updates the target miscoverage level online. The spectral weights choose calibration residuals that look relevant to the current test point. The adaptive update corrects the long-run miss rate when uncertainty changes over time. We give an approximate coverage result for the fixed spectral weighted quantile and a deterministic long-run calibration result for the adaptive update. Simulations with recurring regimes and slowly changing frequencies, together with three U.S. real-data examples, show that the hybrid method can improve on fixed spectral weighting, while also showing that spectral weighting must be monitored through effective sample size diagnostics.

阅读与讨论 → 访问原文 →

25.

arXiv (CS.CL) 2026-06-12 DOI: arXiv:2601.13591

DSAEval: Evaluating Data Science Agents on a Wide Range of Real-World Data Science Problems

作者:

Maojun Sun↗Yifei Xie↗Yue Wu↗Ruijian Han↗Binyan Jiang↗Defeng Sun↗Yancheng Yuan↗Jian Huang↗

Recent LLM-based data agents aim to automate data science tasks ranging from data analysis to deep learning. However, the open-ended nature of real-world data science problems, which often span multiple taxonomies and lack standard answers, poses a significant challenge for evaluation. To address this, we introduce DSAEval, a benchmark comprising 641 real-world data science problems grounded in 285 diverse datasets, covering both structured and unstructured data (e.g., image and text). DSAEval incorporates three distinctive features: (1) Multimodal Environment Perception, which enables agents to interpret observations from multiple modalities, including text and vision; (2) Multi-Query Interactions, which mirror the iterative and cumulative nature of real-world data science projects; and (3) Multi-Dimensional Evaluation, which provides a holistic assessment across reasoning, code, and results. We systematically evaluate 13 recent advanced agentic LLMs using DSAEval. Our results show that Claude-Sonnet-4.5 achieves the strongest overall performance, MiMo-V2-Pro and GPT-5.2 lead in duration and step efficiency, respectively, and MiMo-V2-Flash is the most cost-effective. We further demonstrate that multimodal perception consistently improves performance on vision-related tasks, with gains ranging from 2.04\% to 11.30\%. Overall, while current data science agents perform well on structured data and routine data analysis workflows, substantial challenges remain in unstructured domains. Finally, we offer critical insights and outline future research directions.

阅读与讨论 → 访问原文 →