Academic Intelligence · Curated Daily

探索全球前沿学术脉络

AcademicHub 汇聚顶级期刊与预印本平台的实时文献。定制您的专属科研雷达,利用大语言模型自动生成交叉领域文献分析简报。

01.
Nature Medicine 2026-06-08

Apitegromab for lean mass preservation during tirzepatide-induced weight loss: a randomized, double-blind, placebo-controlled phase 2 trial

Loss of lean mass in proportion to total weight loss is observed with incretin mimetic therapies such as tirzepatide and has the potential to adversely affect health and function. Apitegromab is an investigational, fully human monoclonal antibody that selectively inhibits myostatin activation and is, thereby, capable of increasing muscle mass. In the randomized, double-blind, placebo-controlled phase 2 EMBRAZE study, adults with overweight or obesity (n = 102) were randomized 1:1 to receive tirzepatide plus apitegromab (10 mg kg−1) or tirzepatide plus placebo. At week 24, apitegromab resulted in a least square mean (80% confidence interval (CI)) of 1.9 (1.2−2.7) kg less lean mass loss than placebo (P = 0.001), despite similar total body weight loss between groups, representing a 54.9% retention of lean mass relative to placebo. In participants receiving apitegromab, trough concentrations of apitegromab and total latent myostatin, a pharmacodynamic marker, both increased over time and reached a plateau after approximately 16 weeks. Incidence of adverse events (AEs) (% (95% CI)) was generally similar across apitegromab-treated participants and placebo-treated participants, with 39 of 51 (76% (63−86%)) and 36 of 51 (71% (57−81%)) participants experiencing an AE, respectively. Serious adverse events (SAEs) were balanced and experienced by one of 51 (2% (0−10%)) participants in each arm. In summary, this proof-of-concept study demonstrated that selective targeting of myostatin by apitegromab was well tolerated and effective in preserving lean mass when combined with tirzepatide. ClinicalTrials.gov identifier: NCT06445075 . In the phase 2 EMBRAZE study, participants receiving tirzepatide and apitegromab lost less lean mass compared to participants receiving tirzepatide and placebo.

02.
arXiv (CS.AI) 2026-06-15

Can LLMs Accurately Score Medical Diagnoses and Clinical Reasoning?

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

03.
arXiv (CS.LG) 2026-06-17

Blind Recovery of Latent Domains via Unsupervised Symmetry Discovery

arXiv:2606.17782v1 Announce Type: new Abstract: Primary motivation in blind inverse problems is to recover signals of interest from corrupted observations without knowing the obfuscating mechanism. Blind deconvolution is a prominent approach when the corruption is convolutional, but it is not applicable when general linear transformations obfuscate the domain structure. In this work, we propose an unsupervised framework for recovering latent domains and signals by discovering symmetries of the data distribution. Our framework models observations as linear measurements of signals sampled from a latent random field, and optimizes a shallow group-convolutional network by imposing stationarity and locality regularization at the model output. The model learns a latent symmetry action and an appropriate filter, thereby mapping unstructured observations to a symmetry-based representation that reveals latent signals. Experiments on stochastic processes, Ising models, shuffled and bit-scrambled images, and neural recordings show that the method recovers latent domains and signals from unstructured observations, suggesting symmetry discovery as a new direction for unsupervised structure learning and blind inverse problems.

04.
arXiv (quant-ph) 2026-06-12

From 2D Yang-Mills to Calogero-Sutherland via a colored particle

arXiv:2606.13388v1 Announce Type: cross Abstract: We study Yang-Mills theory coupled to a particle on a cylinder, where gauge invariance and compactness reduce the dynamics to a finite dimensional quantum system. In the Abelian case, this yields a model equivalent to the Landau problem on a torus, with a degenerate ground state structure. We generalize this construction to non-Abelian gauge groups and show that, for SU(N), the system reduces to a one dimensional quantum many body problem with a singular Calogero-Sutherland-type interaction.

05.
arXiv (CS.CV) 2026-06-19

Cinematic Compositing Using Character-Environment-Harmonized Video Generation Models

Cinematic compositing aims to integrate green-screen characters into novel environments while maintaining physical and photometric realism. Previous methods often fail to capture the complex bidirectional interactions between characters and their surroundings, which we characterize as Character-to-Environment (C2E) physical interaction and Environment-to-Character (E2C) lighting harmonization. To address this, we propose an end-to-end video diffusion framework that jointly models C2E and E2C interactions, specifically handling the challenges of interactive props. Our approach introduces a tri-mask-guided architecture with RGB-D joint denoising to ensure physically consistent interactions among the character, props, and environment. We further develop an efficient prior-driven data curation pipeline to construct high-quality relighting pairs without expensive rendering. Finally, a reference-conditioned mechanism enables controllable environment synthesis and precise prop replacement. Extensive experiments demonstrate that our framework significantly outperforms existing methods in cinematic-quality dynamic video compositing.

06.
arXiv (CS.LG) 2026-06-11

Hierarchical Probabilistic Conformal Prediction for Distributed Energy Resources Adoption

arXiv:2411.12193v4 Announce Type: replace-cross Abstract: The rapid growth of distributed energy resources (DERs) presents both opportunities and operational challenges for electric grid management. Accurately predicting DER adoption is critical for proactive infrastructure planning, but the inherent uncertainty and spatial disparity of DER growth complicate traditional forecasting approaches. Moreover, the hierarchical structure of distribution grids demands that predictions satisfy statistical guarantees at both the circuit and substation levels, a non-trivial requirement for reliable decision-making. In this paper, we propose a novel uncertainty quantification framework for DER adoption predictions that ensures validity across hierarchical grid structures. Leveraging a multivariate Hawkes process to model DER adoption dynamics and a tailored split conformal prediction algorithm, we introduce a new nonconformity score that preserves statistical guarantees under aggregation while maintaining prediction efficiency. We establish theoretical validity under mild conditions and demonstrate through empirical evaluation on customer-level solar panel installation data from Indianapolis, Indiana that our method consistently outperforms existing baselines in both predictive accuracy and uncertainty calibration.

07.
arXiv (CS.AI) 2026-06-12

DCD: Domain-Oriented Design for Controlled Retrieval-Augmented Generation

arXiv:2604.07590v2 Announce Type: replace-cross Abstract: Retrieval-Augmented Generation (RAG) is widely used to ground large language models in external knowledge sources. However, when applied to heterogeneous corpora and multi-step queries, Naive RAG pipelines often degrade in quality due to flat knowledge representations and the absence of explicit workflows. In this work, we introduce DCD (Domain-Collection-Document), a domain-oriented design to structure knowledge and control query processing in RAG systems without modifying the underlying language model. The proposed approach relies on a hierarchical decomposition of the information space and multi-stage routing based on structured model outputs, enabling progressive restriction of both retrieval and generation scopes. The architecture is complemented by smart chunking, hybrid retrieval, and integrated validation and generation guardrail mechanisms. We describe the DCD architecture and workflow and discuss evaluation results on synthetic evaluation dataset, highlighting their impact on robustness, factual accuracy, and answer relevance in applied RAG scenarios.

08.
arXiv (CS.AI) 2026-06-11

Runtime Skill Audit: Targeted Runtime Probing for Agent Skill Security

arXiv:2606.11671v1 Announce Type: cross Abstract: Agent skills let LLM agents reuse instructions, resources, tools, and workflows, but they also create a new place for malicious behavior to hide. A skill may look benign in its documentation or code while becoming harmful only when it is invoked with particular user requests, local assets, persistent state, or multi-step tool interactions. This makes purely static vetting brittle. We present Runtime Skill Audit (RSA), a dynamic analysis method that audits skills by asking what the skill-mediated agent actually does under targeted runtime conditions. Instead of testing every skill with the same generic tasks, RSA profiles risk-relevant interfaces, prepares the execution context needed to exercise them, and assigns security labels from the resulting trace evidence. We instantiate RSA on OpenClaw and evaluate it on 100 skills against representative static baselines. RSA achieves 90.0\% accuracy with an 88.0\% true positive rate and an 8.0\% false positive rate, improving accuracy by 13.0 percentage points over the best static baseline. Under self-evolving attacks, static detectors collapse after one or two rounds, while RSA continues to detect 19–20 out of 20 malicious skills across rounds.

09.
arXiv (CS.AI) 2026-06-17

Skill-Constrained Model Predictive Control for Resilient Manufacturing Supply Chains

arXiv:2606.17269v1 Announce Type: new Abstract: In skill-constrained production-inventory systems, the qualified human capacity available tomorrow depends on training decisions made today: production requires certified workers, certifications decay unless maintained, and training consumes the same scarce worker hours that production needs now. We study a closed-loop skill-constrained model predictive controller that, at every shift, solves a finite-horizon mixed-integer program over production, inventory, backlog, and training, with binary predicted certification, hard production eligibility, and an interpretable terminal value that prices certified-capacity gaps at the horizon boundary; only the first-period action is applied before replanning. On synthetic, seed-controlled SkillChain-Gym scenarios - announced and surprise new-skill shocks, demand shocks, absenteeism, forecast- and availability-quality modes, capacity-boundary and training-rate sweeps, and negative controls - we evaluate the controller against production-only and maintenance-only ablations, static cross-training insurance plans, and a strong reactive heuristic, under an ex-ante locked configuration and paired statistics. The result is regime dependence, not superiority: no policy class dominates. Predictive control helps when skill or labor bottlenecks are forecastable early enough for training to complete; lean static insurance remains hard to beat under surprise shocks, near the demand-capacity boundary, and wherever pre-shock slack makes insurance cheap. Attribution ablations separate certification maintenance, re-acquisition of lapsed certifications, and greenfield skill acquisition. Forecastability, not adaptivity per se, decides when predictive control pays.

10.
arXiv (CS.AI) 2026-06-19

Movement Primitives in Robotics: A Comprehensive Survey

arXiv:2601.02379v2 Announce Type: replace-cross Abstract: Biological systems exhibit a continuous stream of movements, consisting of sequential segments, that allow them to perform complex tasks in a creative and versatile fashion. This observation has led researchers towards identifying elementary building blocks of motion known as movement primitives, which are well-suited for generating motor commands in autonomous systems, such as robots. In this survey, we provide an encyclopedic overview of movement primitive approaches and applications in chronological order. Concretely, we present movement primitive frameworks as a way of representing robotic control trajectories acquired through human demonstrations. Within the area of robotics, movement primitives can encode basic motions at the trajectory level, such as how a robot would grasp a cup or the sequence of motions necessary to toss a ball. Furthermore, movement primitives have been developed with the desirable analytical properties of a spring-damper system, probabilistic coupling of multiple demonstrations, using neural networks in high-dimensional systems, and more, to address difficult challenges in robotics. Although movement primitives have widespread application to a variety of fields, the goal of this survey is to inform practitioners on the use of these frameworks in the context of robotics. Specifically, we aim to (i) present a systematic review of major movement primitive frameworks and examine their strengths and weaknesses; (ii) highlight applications that have successfully made use of movement primitives; and (iii) examine open questions and discuss practical challenges when applying movement primitives in robotics.

11.
arXiv (CS.AI) 2026-06-16

CmdNeedle: Measuring the Incompleteness of Command Denylists for AI Agents

arXiv:2606.15549v1 Announce Type: cross Abstract: The adoption of AI agents is increasing rapidly. Terminal AI agents, i.e., AI agents that run in terminal environments, are a widely used type of AI agents. Terminal AI agents rely heavily on shell command execution to interact with the host systems. They adopt a three-list command-gating mechanism to mitigate security risks introduced by command execution, with denylists serving as the load-bearing component. However, modern operating systems often ship a large, ever-expanding set of shell commands with complex functionalities. Our observation is that even a built-in denylist of Claude Code, well-maintained by its developers, can overlook bypass commands that invalidate its effectiveness. Such negligence leads to fragile command denylists that cannot even block operations that practitioners expect them to block. This paper presents the first systematic characterization of command denylist fragility in terminal AI agents. The paper formalizes the command denylist fragility problem and proposes an LLM-driven pipeline, CmdNeedle, to detect such fragility. It prompts the LLM to propose possible bypasses and iteratively repairs them using feedback from a validator that executes them in a sandbox. In the evaluation, we applied CmdNeedle to 1,709 real-world command denylists (containing 13,332 denylist rules) collected from GitHub. The evaluation shows several key findings, including that 69.0–98.6% of the denylists are fragile, that this fragility occurs consistently across projects and agents, and the validity of several possible root causes for this fragility. Our pipeline and findings will hopefully facilitate future research and practice regarding the command denylists used by AI agents.

12.
arXiv (CS.AI) 2026-06-15

Benchmarking Vision-Language-Action Models on SO-101: Failure and Recovery Analysis

arXiv:2606.08881v2 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) models have demonstrated strong generalization in robotic manipulation, yet existing evaluations are primarily conducted in simulation or on expensive robotic platforms, leaving their robustness on affordable real-world robots largely unexplored. We present a standardized real-world benchmark for evaluating representative VLA and imitation learning policies on the low-cost SO-101 robotic platform. The benchmark comprises four representative manipulation tasks together with unified evaluation protocols, enabling systematic comparison under embodiment uncertainty. Using real-world teleoperated demonstrations, we fine-tune and evaluate $\pi_{0.5}$, SmolVLA, Wall-X, and ACT directly on the physical platform. Beyond conventional task success rates, the benchmark incorporates a structured failure taxonomy, semantic- and execution-level failure decomposition, and recovery-aware evaluation metrics to characterize policy robustness. Experimental results show that stronger pretrained VLA policies generally outperform the imitation learning baseline, although performance remains highly task-dependent under low-cost robotic deployment conditions. Execution instability emerges as the dominant failure source, while recovery capability varies substantially across architectures. These results highlight the importance of failure and recovery analysis beyond binary task success and establish SO-101 as a practical benchmark for evaluating embodied AI systems under realistic low-cost robotic deployment conditions.

13.
arXiv (CS.CV) 2026-06-11

Mitigating Content Shift and Hallucination in GenAI Image Editing via Structural Refinement

Generative AI (GenAI) image editors, such as Nano Banana, produce visually compelling results for retouching tasks, enabling non-experts to edit images through text prompts alone. However, the generative nature of these models often introduces spatial misalignment, texture distortion, and content hallucination, all of which are detrimental to downstream workflows that require pixel-level fidelity. We identify a problem setting we call "structure-preserving GenAI fusion" for black-box GenAI image retouching: retain the perceptual enhancements of a GenAI output while enforcing structural faithfulness to the original input image. To address this problem, we propose a post-processing framework that fuses an input image with its GenAI-enhanced counterpart by first establishing coarse spatial and photometric correspondences, then performing a fusion stage that transfers desired enhancements while suppressing hallucinated content. In the absence of direct prior work in this setting, we evaluate our framework against representative methods from photorealistic style transfer and image fusion. Our experiments demonstrate that our method better preserves aesthetic quality while maintaining pixel-level structural consistency and the input resolution.

14.
arXiv (quant-ph) 2026-06-17

Tunneling Dynamics and Time Delay in Electron Transport through Time-Dependent Barriers with Finite-Bandwidth Reservoirs

arXiv:2507.20649v2 Announce Type: replace-cross Abstract: We study a model system consisting of a tunneling barrier driven by an external harmonic field and coupled to two leads with finite bandwidth. Avoiding Floquet expansions, we derive simple expressions for the time-dependent tunneling current in the adiabatic regime. Our approach relates the barrier modulation to a measurable time delay in the steady-state periodic current. It provides a physically consistent definition of the tunneling time inside the barrier by subtracting the time delay associated with the leads from the total time delay. We find that the tunneling time always vanishes for wide/high barriers. Remarkably, the time delay persists even when the barrier becomes static, i.e., in the limit where the modulation frequency vanishes. This indicates that the time delay obtained through the introduction of an external periodic perturbation actually reflects an intrinsic property of the tunneling dynamics, rather than an effect of the external drive or of a particular system. We apply our results to the analysis of tunneling times in optical experiments and find good agreement with the experimental data.

15.
arXiv (CS.LG) 2026-06-17

Turning music identification into a neural forward pass

arXiv:2606.17301v1 Announce Type: cross Abstract: Search, a foundational operation in computer science, maps a query to a matching item in a collection. It is typically implemented as a System-2 like, rule-based pipeline in which a key is computed, an index is probed, and candidates are verified. By contrast, human recognition resembles a System-1 like, associative model of identity recovery, in which even partial cues can trigger a recall without explicitly enumerating, ranking, or even accessing discrete candidates. Here, we show that music sound identification, a difficult search problem, can be performed in a single neural feed-forward pass by a generative transformer. Trained on an audio dataset, the model predicts the corresponding track identifier from a short audio excerpt. This approach surpasses state-of-the-art acoustic fingerprinting, with the largest gains for short audio segments (1 second), demonstrating the method is not only viable but advantageous. Moreover, it reduces external storage to 0.33% of the baseline footprint and improves inference latency by 2.3x (p95). Furthermore, the model can reject queries for unseen tracks, supporting open-set operation while reducing misattribution risk. Using music track identification as an example, this work reframes search, bringing it closer in spirit to human associative recognition and away from algorithmic database lookup.

16.
arXiv (quant-ph) 2026-06-15

Sensitivity of polaron-molecule observables to MDR/GUP-like ultraviolet deformations at low energies via quantum computing

arXiv:2606.14479v1 Announce Type: new Abstract: We show that impurity many-body observables can display enhanced sensitivity to ultraviolet deformations of generalized-uncertainty-principle and modified-dispersion-relation type at accessible energy scales. Using a deformed polaron-molecule Hamiltonian constructed to preserve the infrared sector, we quantify the impact of such deformations on spectral and Ramsey observables and implement the corresponding dynamics in a controlled quantum computing setting. We identify regimes near the polaron-molecule crossover where small ultraviolet deformations are strongly amplified, leading to experimentally resolvable changes in quasiparticle properties and spectral response. Our results establish a concrete sensitivity-based route to low-energy quantum-gravity phenomenology in a well-defined many-body platform and delimit the validity of the effective description. Furthermore, we report experimental validation on the QRed superconducting quantum processor (BSC-CNS).

17.
arXiv (CS.LG) 2026-06-17

A tensor network approach for chaotic time series prediction

arXiv:2505.17740v2 Announce Type: replace Abstract: Making accurate predictions of chaotic time series is a complex challenge. Reservoir computing, a neuromorphic-inspired approach, has emerged as a powerful tool for this task. It exploits the memory and nonlinearity of dynamical systems without requiring extensive parameter tuning. However, selecting and optimizing reservoir architectures remains an open problem. Next-generation reservoir computing simplifies this problem by employing nonlinear vector autoregression based on truncated Volterra series, thereby reducing hyperparameter complexity. Nevertheless, the latter suffers from exponential parameter growth in terms of the maximum monomial degree. Tensor networks offer a promising solution to this issue by decomposing multidimensional arrays into low-dimensional structures, thus mitigating the curse of dimensionality. This paper explores the application of a previously proposed tensor network model for predicting chaotic time series, demonstrating its advantages in terms of accuracy and computational efficiency compared to conventional echo state networks. Using a state-of-the-art tensor network approach enables us to bridge the gap between the tensor network and reservoir computing communities, fostering advances in both fields.

18.
arXiv (CS.CV) 2026-06-12

AudioX-Turbo: A Unified Framework for Efficient Anything-to-Audio Generation

Audio and music generation based on flexible multimodal control signals is a widely applicable topic, with the following key challenges: 1) a unified multimodal modeling framework, 2) large-scale, high-quality training data, and 3) the prohibitive inference cost of multi-step diffusion sampling. As such, we propose AudioX-Turbo, a unified and efficient framework for anything-to-audio generation that integrates varied multimodal conditions (i.e., text, video, and audio signals) in this work. AudioX-Turbo follows a teacher-student paradigm. The teacher AudioX-Base is built on a Multimodal Diffusion Transformer with a Multimodal Adaptive Fusion module that aligns diverse multimodal inputs for high-fidelity synthesis, and is then distilled into the few-step student AudioX-Turbo via Distribution Matching Distillation adapted to flow matching, complemented by a diffusion-based discriminator for high-quality few-step generation. To support the training of AudioX-Turbo, we construct a large-scale, high-quality dataset, IF-caps-Pro, comprising approximately 9.2M samples curated through a two-stage data collection and annotation pipeline. We benchmark AudioX-Turbo across a wide range of tasks, finding that our model achieves superior performance, especially on text-to-audio and text-to-music generation, while operating at only 4 sampling steps and requiring approximately 25x fewer function evaluations (NFE) than multi-step baselines. These results demonstrate that our method is capable of audio generation under flexible multimodal control, showing efficient and powerful instruction-following capabilities. The code and datasets will be available at https://zeyuet.github.io/AudioX-Turbo/.

20.
arXiv (CS.AI) 2026-06-18

DeFAb: A Verifiable Benchmark for Defeasible Abduction in Foundation Models

arXiv:2606.18557v1 Announce Type: new Abstract: A rule-based logic solver resolves every instance in our benchmark in under 50 microseconds with 100% accuracy; the best frontier language model reaches 65% at best and drops to 23.5% under rendering-robust evaluation (worst case over four surface renderings). We introduce DeFAb (Defeasible Abduction Benchmark), a dataset and generation pipeline that converts four decades of publicly funded knowledge bases into formally grounded instances for defeasible abduction: constructing hypotheses that explain anomalies by overriding defaults while preserving unrelated expectations. Because every hypothesis must pass polynomial-time checks for valid derivation, conservativity, and minimality, DeFAb makes logical rigor the instrument for measuring creativity and theoretical reasoning, scoring the disciplined construction of theory revisions rather than fluent but theory-destroying prose. The pipeline pairs taxonomic hierarchies (OpenCyc, YAGO, Wikidata) with behavioral property graphs (ConceptNet, UMLS) to produce 372,648+ instances across 33.75M materialized rules from 18 sources, in three levels with polynomial-time verifiable gold standards. Four frontier models do not reliably internalize defeasible reasoning: rendering-robust Level 2 accuracy is 7.8-23.5%; chain-of-thought variance (~36 pp) exceeds any inter-model gap; and a matched contamination control isolates a +19.4 pp Level 3 gap. We further release DeFAb-Hard (a 235-instance Level 3 difficulty variant; best model 53.3% vs 100% symbolic) and CONJURE (a kernel-verified transformative-creativity variant of 560 Lean 4/Mathlib instances whose gold answers are definitions the proof kernel did not previously contain, judge-free verifier; a pilot finds zero novel concepts). The same verifier doubles as an exact reward for preference optimization (DPO, RLVR/GRPO). Released under MIT at https://huggingface.co/datasets/PatrickAllenCooper/DeFAb.

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

Exact Federated Continual Unlearning for Ridge Heads on Frozen Foundation Models

arXiv:2603.12977v3 Announce Type: replace Abstract: Foundation models are commonly deployed as frozen feature extractors with a small trainable head to adapt to private, user-generated data in federated settings. The ``right to be forgotten'' requires removing the influence of specific samples or users from the trained model on demand. Existing federated unlearning methods target general deep models and rely on approximate reconstruction or selective retraining, making exactness costly or elusive. We study this problem in a practically relevant but under-explored regime: a frozen foundation model with a ridge-regression head. The exact optimum depends on the data only through two additive sufficient statistics, which we turn into a communication protocol supporting an arbitrary stream of add and delete requests via fixed-size messages. The server maintains a head that is, in exact arithmetic, pointwise identical to centralized retraining after every request. We provide deterministic retrain-equivalence guarantees, order and partition invariance, two server-side variants, and a Bayesian certificate of zero KL divergence. Experiments on four benchmarks confirm the guarantees: both variants match centralized ridge retraining to within $10^{-9}$ relative Frobenius error and complete each request at orders-of-magnitude lower cost than federated retraining baselines.

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

SFT Overtraining Predicts Rank Inversion via Entropy Collapse Under RLVR

The standard heuristic of selecting the SFT checkpoint with the highest pass@1 for GRPO can fail when SFT compresses the rollout distribution. For binary rewards, the expected within group advantage variance is $p(1{-}p)(g{-}1)/g$; when early GRPO drives $p$ below $p^*(g)$, most groups have identical rewards and provide no group relative signal. We study SFT depth ladders for Qwen2.5-Coder-3B and DeepSeek-Coder-6.7B. We test Qwen2.5-Coder-3B across five depths and three seeds, and DeepSeek-Coder-6.7B across four matched depths and three seeds. On Qwen, pre RL pass@1 rises with SFT depth, but peak GRPO pass@10 falls from $0.806$ to $0.481$ (3 seed mean, $n{=}20$); pre RL entropy is positively associated with the GRPO outcome ($\rho{=}{+}0.69$). On DeepSeek, pass@1 remains far above $p^*(8){=}0.083$, and GRPO outcomes compress rather than invert. A two stage diagnostic, combining pre RL entropy triage with an early GRPO entropy monitor, flags high risk checkpoints and can stop failing runs early. Simple KL to reference regularisation and label smoothing variants do not rescue the collapsed Qwen checkpoint in our setting, suggesting the failure is not a trivial GRPO hyperparameter artefact.

23.
arXiv (CS.CL) 2026-06-12

From Passive Generation to Investigation: A Proactive Scientific Peer Review Agent

Large language models (LLMs) have shown promise in automating scientific peer review. However, existing approaches often struggle to generate in-depth reviews supported by concrete evidence. We argue that a key limitation is the lack of flexibility to proactively investigate suspicious parts of a paper based on accumulated evidence, as human reviewers do. In this paper, we explore how to enable an LLM-based review agent to perform such proactive investigation. We find that this can be naturally formulated as a Markov Decision Process (MDP), and propose ProReviewer, a scientific peer review agent that proactively reviews a paper guided by a maintained, structured review log. The structured review log serves as a workspace for the agent to track evidence and intermediate findings collected during review. Experiments show that ProReviewer with an 8B backbone, trained by supervised fine-tuning and optimized by reinforcement learning, achieves the highest average score across five quality dimensions, outperforming prompt-based methods with much larger frontier LLMs by up to 39% and the strongest fine-tuned baseline by 16% relatively. It also attains the highest win rates against baselines in human evaluation.

24.
arXiv (CS.CL) 2026-06-16

Entity Labels Are Not Entity Signals: A Framework for Observable Relevance in Document Re-Ranking

Entity-aware document retrieval uses query-associated entities as ranking signals, assuming that semantically relevant entities are also useful retrieval signals. We show this assumption is insufficient- and explain why. Unlike terms, which are ground-truth observations, entity links are hypotheses produced by an imperfect linker: an entity can be topically central yet provide no discriminative signal if the linker fires indiscriminately across relevant and non-relevant documents. We formalize this as a distinction between Conceptual Entity Relevance (CER)- whether an entity is topically related to a query- and Observable Entity Relevance (OER)- whether its observed presence in a collection discriminates relevant from non-relevant documents. Across four collections and annotation sources including human entity judgments, CER and OER exhibit near-chance agreement ($\kappa \approx 0$), while OER operationalizations agree substantially ($\kappa \approx 0.5$), confirming CER as the systematic outlier. CER-based supervision selects topically plausible but weakly discriminative entities, pruning fewer than 4% of non-relevant documents on some collections. Aligning supervision with OER improves non-relevant pruning by up to 10x and open-world MAP by 0.051 over BM25. Our findings motivate a shift from conceptual to observable notions of entity relevance in entity-aware retrieval.

25.
arXiv (CS.AI) 2026-06-16

Parallel Test-Time Scaling with Multi-Sequence Verifiers

arXiv:2603.03417v2 Announce Type: replace-cross Abstract: Parallel test-time scaling, which generates multiple candidate solutions for a single problem, is a powerful technique for improving large language model performance. However, it is hindered by two key bottlenecks: accurately selecting the correct solution from the candidate pool, and the high inference latency from generating many full solutions. We argue that both challenges are fundamentally linked to verifier calibration, as a well-calibrated verifier improves answer selection and enables early-stopping strategies to reduce latency. However, existing non-generative verifiers are limited as they score each candidate in isolation, overlooking rich contextual information across the set of candidates. To address this, we introduce the Multi-Sequence Verifier (MSV), a lightweight verifier that predicts each candidate's correctness conditioned on the full sampled set. MSV achieves improved calibration, which directly enhances best-of-N selection performance and empowers a novel early-stopping framework. Across challenging mathematical reasoning benchmarks, MSV improves best-of-64 accuracy by up to 6\% relative to strong baselines, and in the early-stopping setting reaches the same accuracy as baselines with less than half the latency.