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.
arXiv (quant-ph) 2026-06-25

Folds of one curve: the superradiant phase diagram of Dicke modes with interacting matter

arXiv:2606.26081v1 Announce Type: cross Abstract: We give a thermodynamic-limit account of Dicke models with one cavity mode coupled collectively to interacting matter. Integrating out the cavity yields an exact self-consistent functional of the magnetisation $m$, $\tilde e(m) = \lambda m^2/2 + e_mat(\lambda m)$: a classical penalty on the bare-matter energy $e_mat$ in the self-consistent field $h = \lambda m$, with $\lambda = g^2/(2\omega_c)$ the collective coupling. Supplying only that scalar field, the photon creates no phase the matter does not already possess. States holding a minimum form one connected curve, $\lambda(m) = \mu_mat^{-1}(m)/m$, so superradiant first-order transitions are folds of one equation of state not crossings of disjoint sheets, and a fold can straighten into a continuous line. The remaining rules are local, each with a spectral counterpart: onset by the leading singularity of $e_mat$ (a softening polariton), order by one bare response – the Landau quartic, or a divergent susceptibility forcing a Larkin-Pikin (LP) fold. For the Dicke-Ising model the Landau coefficients are exact, giving in closed form the second-order boundary and both zero-quartic fields, one tricritical; a $1/d$ expansion maps all four phases, with the AS-PS transition first order for $d\le d_{uc}=3=4-z$ (LP) and tricritical points in the $(d,\epsilon)$ plane above. At the degenerate quadruple point the matter is a Rydberg-blockade chain, solved by strict-blockade iDMRG: the antiferromagnetic superradiant (AS) phase persists as a finite 1D wedge, first order into the corner. Other magnets: the triangular antiferromagnet keeps a continuous superradiant-superradiant line (3D-XY, no fold forced); the compass chain a BKT-functional onset; the Heisenberg and XX chains, via a conserved operator, a spectrally silent first-order onset; and the Dicke-Heisenberg diagram an exact tricritical point at the saturation corner.

02.
arXiv (CS.AI) 2026-06-19

ScholarQuest: A Taxonomy-Guided Benchmark for Agentic Academic Paper Search in Open Literature Environments

arXiv:2606.20235v1 Announce Type: cross Abstract: Academic paper search is a core step in scientific research, and LLM-based search agents are emerging as a promising paradigm for iterative, intent-driven literature exploration. However, existing benchmarks are insufficient for systematically evaluating agentic academic search under realistic open literature environments. We propose ScholarQuest, a large-scale, taxonomy-guided benchmark for agentic academic paper search. ScholarQuest is constructed from over 1,000 computer science topics and four representative research intents, including method-oriented, setting-anchored, comparison-based, and scope-controlled queries. It further provides scalable answer construction and a shared retrieval backend ScholarBase for reproducible evaluation. Benchmarking results show that agentic methods outperform single-shot retrieval baselines, yet the best-performing agent only achieves 0.314 Recall@100 and 0.355 Recall@All, indicating substantial room for improvement. In addition, analyses of search efficiency, intent-level robustness, and failure cases further highlight the benchmark's ability to provide multi-dimensional evaluation signals for academic paper search agents.

03.
arXiv (CS.AI) 2026-06-12

The Hidden Power of Scaling Factor in LoRA Optimization

arXiv:2606.12883v1 Announce Type: new Abstract: In Low-Rank Adaptation (LoRA), the scaling factor $\alpha$ is often treated as a mere complement to the learning rate, yet its role in optimization remains poorly understood. In this paper, we reveal that the scaling factor $\alpha$ and the learning rate function differently, with $\alpha$ emerging as the dominant driver of effective optimization, delivering gains that cannot be replicated by learning rate scaling alone. Through the synergy of extensive empirical analysis and a theoretical Signal-Drift framework, we uncover three findings into LoRA's scaling mechanism: First, LoRA's spectral suppression smooths the optimization landscape, rendering standard hyperparameters overly conservative and creating an optimization gap. Second, when leveraging this smoothness to accelerate convergence, $\alpha$ outperforms the learning rate by amplifying the task signal without increasing the drift ratio. Third, the optimal scaling factor follows a sublinear relationship with the rank, well characterized by a square-root law with an unexpectedly large coefficient, revealing the insufficient scaling of existing rank-tied heuristics. Based on these insights, we propose LoRA-$\alpha$, a minimalist framework that restores $\alpha$ to its principled regime, making LoRA compatible with standard small learning rates. Extensive evaluations across diverse tasks demonstrate that LoRA-$\alpha$ consistently improves performance while streamlining hyperparameter search, unleashing the learning potential of LoRA.

04.
arXiv (CS.CV) 2026-06-18

Bridging Creative Intent and Visual Quality: Creator-Driven Recurrent Video Generation with Agentic Feedback Loops

Generative AI has made content creation increasingly accessible, but many AI-generated videos lack narrative coherence and creative direction, issues that become more substantial at longer durations. Unlike coding, where AI generation benefits from reliable feedback and techniques such as recurrent self-improvement, video generation requires subjective feedback about plot, scenes, and narrative, which naturally motivates approaches that incorporate human creative direction. We introduce CHIEF, a human-AI co-creation video generation framework that places the creator at the center of human-in-the-loop iterative video refinement, and supports them by providing automatic subjective feedback. The creator incorporates their creative direction by driving each iteration, while their revisions are incorporated by a specialized refiner agent. The feedback loop is generated by persona-conditioned multimodal LLMs that watch generated videos and produce subjective critique from the audience perspectives, providing feedback that self-evaluation alone cannot capture. To test the effectiveness of our proposed framework, we work with high school and college students with no prior filmmaking experience to create videos, from short 1-minute videos to a complete short 10-minute film with a complicated plot.

05.
arXiv (CS.CV) 2026-06-15

HiLo-Token: Input-Adaptive High-Low Frequency Token Compression for Efficient Image Editing

Creative image editing tools, such as Photoshop's Remove or Generative Fill buttons, are central to everyday customer use and account for a major share of traffic in Photoshop and Lightroom. However, current generative AI models face significant latency challenges, which become even more pronounced when transitioning from convolution-based U-Nets to Diffusion Transformers (DiTs). In our evaluation on hundreds of representative image editing samples spanning a wide range of mask ratios, the DiT module alone accounts for an average of 73% of the total model latency, even after being distilled from 50 timesteps down to 8 timesteps. To tackle this challenge, we propose $HiLo-Token$, an input-adaptive token compression framework that allocates more token budget to high-frequency, rich-context regions while assigning fewer tokens to low-frequency areas. Specifically, for the editing region specified by the user mask, we retain all tokens within a dilated mask to preserve strong locality and contextual relevance. Outside the editing region, we introduce a simple yet effective high-frequency token selection strategy based on spatial frequency to capture important local details, while using tokens from a 16x downsampled image to represent low-frequency components and preserve the blurry but global structure. Extensive experiments on production-level evaluation data validate the effectiveness of the proposed method, achieving 3.13x, 2.59x, and 1.67x DiT speedups on A100-80GB for image editing tasks across small, medium, and large mask ratio categories with average ratios of 6.38%, 15.92%, and 35.36%, respectively, without any regression in generation quality.

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

When Errors Become Narratives: A Longitudinal Taxonomy of Silent Failures in a Production LLM Agent Runtime

作者:

arXiv:2606.14589v1 Announce Type: cross Abstract: LLM agent systems increasingly run as long-lived autonomous runtimes: scheduling jobs, calling tools, maintaining memory, and pushing results to humans. We present a longitudinal study of silent failures in one such system: a personal-assistant agent runtime in continuous production since March 2026, with roughly 40 scheduled jobs, 8 LLM providers, a tool-governance proxy, and a knowledge-base memory plane, defended by 4,286 unit tests and 827 governance checks. Over eight weeks we documented 22 incidents with full root-cause postmortems, in which one meta-pattern – a failure whose error signal never reaches a human in actionable form – manifested at least 28 times. We derive a five-class, mechanism-oriented taxonomy: (A) environment and platform quirks, (B) design-assumption mismatches, (C) error swallowing and dilution, (D) chained hallucination and fabrication, (E) operational omission and forensic blind spots. Class D is unique to LLM systems and the most dangerous: the system does not merely fail to report an error – the LLM transforms it into fluent, plausible narrative delivered to the user. We term this fail-plausible: gray failure's differential observability escalated – the observer is not just blind, it is convincingly lied to by the failure itself. Three findings: about 70% of silent failures were caught by human user-view observation, not tests or audits; a retrospective audit of 15 incidents found 0% ex-ante prevention but 87% regression blocking – audits are regression engines, not prediction engines; incident latency (13 hours to 60 days) tracks failure mechanism, not code complexity – the longest-lived failures lived in the seams between components, where no test runs. We describe the resulting defense framework and distill design principles for agent systems whose failures are loud, attributable, and boring. All postmortems and artifacts are public.

07.
arXiv (math.PR) 2026-06-17

How long does it take to train an Elephant Random Walk

作者:

arXiv:2509.15049v2 Announce Type: replace Abstract: We study how conditioning on the first $k$ steps, which we think of as training, affects the long-term behavior of the Elephant Random Walk. When the elephant is conditioned to be at position $k$ at time $k$, the first return time to the origin scales as $k^{(4-4p)/(3-4p)}$ in the diffusive regime, and grows exponentially in the critical regime. We loosely interpret this as a measurement of the rate at which the elephant forgets its training.

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

A Mechanistic Understanding of Pronoun Fidelity in LLMs

Faithful and robust pronoun use is important for fair and coherent generations, yet large language models largely fail when multiple referents use different pronouns. To study the interplay of reasoning, repetition, and bias in this task, prior work relies exclusively on behavioural approaches, which may not reflect a model's internal workings. Therefore, we provide a mechanistic, model-internal perspective on pronoun fidelity, testing whether three mechanisms – group entity binding (G), recency bias (R), and stereotypical bias (S) – are causally implemented across several SOTA language models. Using Boundless Distributed Alignment Search, we find all three coexist as causal subspaces distributed across network depth. No single mechanism fully explains model behaviour, but a combination of the three consistently accounts for 91-99.5%. An attention head analysis further reveals two competing copying routes; group binding and stereotype share a localized concept-level route that retrieves a bound occupation-pronoun unit, while recency uses a distributed token-level route that repeats surface forms. In sum, pronoun fidelity arises from competition between simultaneously active causal subspaces.

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

The Reservoir Attention Network: Cross-Pass State in Pretrained Transformers via Content-Addressable Reservoir Injection

arXiv:2606.15678v1 Announce Type: cross Abstract: A feasibility and dynamics study of the Reservoir Attention Network (RAN), an architecture that injects a fixed, randomly-initialized reservoir into the mid-layer attention of a pretrained transformer to carry state across forward passes. Experiments span GPT-2 (124M, 355M) to Qwen2.5 (0.5B, 1.5B) on a single consumer GPU. The tasks are minimal probes chosen to isolate individual mechanisms; the broader always-alive agent vision is treated throughout as compute-limited future work, not a claim of this paper. The reservoir is left untrained (fixed random) by design: this isolates whether untrained recurrent dynamics alone suffice to carry usable cross-pass state, leaving trained recurrence as a complementary, more expensive direction.

10.
arXiv (CS.LG) 2026-06-18

SCOPE-FL: A Strategy-proof Chain-based Optimal pareto efficient Federated Learning System

arXiv:2606.18384v1 Announce Type: new Abstract: Hierarchical Federated Learning (HFL) enables scalable collaborative model training across distributed devices while preserving data privacy. However, existing HFL client selection mechanisms suffer from a fundamental strategic inefficiency. By prioritizing stability over Pareto efficiency (PE), they produce suboptimal resource allocations, and without strategy proofness (SP), participants are incentivized to misrepresent their true preferences, both failures degrading system overall welfare in the Pareto sense in practice. To address it, we propose SCOPE-FL (Strategy-proof Chain-based Optimal pareto efficient Federated Learning), a synchronous HFL framework that formulates client selection as a two-sided school choice problem solved through the Top Trading Cycle (TTC) algorithm that simultaneously guarantees PE and SP. For reward distribution, SCOPE-FL employs a scalable Shapley value approximation based on One-Round Reconstruction (OR), ensuring compensation proportional to each client's contribution. The entire mechanism executes via blockchain smart contracts, providing the tamper-proof environment required for the SP guarantees to hold in practice. A comprehensive evaluation on MNIST, Fashion-MNIST, and CIFAR-10 demonstrates that SCOPE-FL outperforms state-of-the-art approaches, including DA, IAS, and other methods across model accuracy, convergence rate, and reward efficiency, while achieving communication latency comparable to DA and blockchain overhead significantly lower than DA at scale.

11.
arXiv (CS.AI) 2026-06-19

Cross-Dataset, Age, and Gender Generalization: A Comprehensive Analysis of Fine-Tuning Strategies for Low-Resource Children's ASR

arXiv:2606.19791v1 Announce Type: cross Abstract: The challenge associated with recognizing dysarthric speech primarily arises from pronounced acoustic variability attributed to impaired articulatory precision. Past research has demonstrated improved recognition through the use of hybrid DNN/HMM sequence discriminative training. This paper presents a comprehensive investigation of various combinations of acoustic features tailored to different Acoustic Models, offering suitable feature selections for each. The incorporation of Pitch features notably improved recognition performance, especially for sentence recognition tasks involving dysarthric speech. Through a systematic examination of the TORGO database, we have demonstrated the potential to enhance the performance of the state-of-the-art Factorized Time Delay Neural Network (F-TDNN) model for recognizing dysarthric speech. Our methods, implemented with the F-TDNN model, resulted in a 4.65\% relative improvement in isolated word recognition and a 4.63\% relative improvement in sentence recognition for dysarthric speech, compared to previous research. This improvement effectively compensates for speech variability, attributable to our deliberate selection of the number of overlapping frames between consecutive training example chunks.

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

Reward-SQL: Boosting Text-to-SQL via Stepwise Execution-Aware Reasoning and Process-Supervised Rewards

Recent advances in large language models (LLMs) trained with reinforcement learning (RL) have improved Text-to-SQL performance. However, RL-based approaches still struggle with complex queries due to two key limitations: insufficient stepwise execution-aware reasoning grounded in database feedback, and the lack of process-level rewards for guiding reasoning optimization. To address these issues, we propose CoCTE, a divide-and-conquer and execution-aware reasoning framework that progressively composes SQL queries through intermediate view validation and structured Common Table Expressions (CTEs), improving both accuracy and interpretability. To realize a CoCTE reasoning process, we develop Reward-SQL, a unified approach with three stages: (1) model initialization, which equips LLMs with structured CoCTE reasoning capabilities; (2) process reward design, which delivers fine-grained, execution-aware supervision; and (3) process-supervised RL and inference, which integrates process rewards into training and guides the inference stage by process rewards. This paper addresses the core challenges in Reward-SQL and makes the following contributions. We introduce a process reward model (PRM) that combines execution-aware trajectory scoring with entropy-based step weighting, providing dense and interpretable supervision across reasoning steps. We integrate PRM into both RL training and inference stages, stabilizing optimization and improving trajectory exploration with process-level signals. Experiments show that Reward-SQL significantly outperforms baselines with comparable model sizes, and exhibits strong cross-domain generalization.

13.
arXiv (CS.AI) 2026-06-16

How Much Do Reviews Really Contribute? A Study on Text-Enriched Matrix Factorization for Recommendations

arXiv:2606.16973v1 Announce Type: cross Abstract: Incorporating textual reviews into a Recommender System has become a prominent strategy for enriching collaborative signals with semantic information. However, the actual contribution of review-derived representations remains an open question, particularly when strong collaborative baselines are employed. In this work, we systematically investigate the impact of textual information on Matrix Factorization by introducing and comparing three enrichment strategies over a common collaborative backbone. First, we propose a learnable gating mechanism that adaptively balances collaborative and textual signals during training. This mechanism is applied to two distinct review representations: (i) aggregated topic profiles extracted from user and item histories, and (ii) full text embedding representations derived from reviews. Additionally, we explore a cross-attention mechanism that identifies and emphasizes the most informative dimensions of the textual representation before fusion with collaborative factors. We evaluate six variants: pure, enriched with topic profiles and text via gating; enriched with topics and text via gating; and enhanced with cross-attention over textual features. Experiments across multiple review-based datasets reveal that although adaptive fusion mechanisms improve representation flexibility, the marginal contribution of textual signals remains limited compared to the collaborative backbone. These findings suggest that, under typical rating-prediction settings, collaborative information continues to dominate performance, raising important considerations for the effective integration of semantic review signals into recommendation models.

14.
arXiv (CS.AI) 2026-06-24

On the Smallness of the Large Language Models Scaling Exponents

arXiv:2606.24504v1 Announce Type: new Abstract: We discuss reasons why the scaling exponents of current Large Language Models (LLMs) applications are indicating an unsustainable regime in terms of energy resources. We further show that attributing the smallness of such exponents to a numerical bias due to the neglect of a non-zero value of the loss function in the limit of infinite data (``pedestal effect") does not remove the unsustainability issue. Finally, the effects of the smoothness (roughness) of the data on the scaling exponents is commented upon based on an analogy with phenomenological models of fluid turbulence.

15.
arXiv (CS.CL) 2026-06-24

Same Lesson, Different Story: Cross-Lingual Reconstruction of Cultural Narratives in Large Language Models

The evaluation of cultural grounding context becomes complex when multiple cultures convey the same moral lesson. This challenge is particularly relevant to large language models (LLMs), which produce narratives across a wide range of languages and cultural contexts. However, it remains uncertain whether these models preserve culturally grounded meaning when equivalent moral lessons are conveyed through distinct cultural forms. This study introduces a multilingual evaluation narrative framework that integrates a cross-linguistic collection of 414 proverbs spanning 15 languages and uses four LLMs to generate 13k narratives. By employing semantically equivalent proverbs as culturally grounded prompts, the analysis assesses whether models preserve meaning across languages, how cross-lingual conditioning influences narrative realization, and whether different model families converge on similar interpretations. Results indicate that cross-lingual prompting largely preserves proverb-level semantic meaning while systematically redistributing agency, social positioning, and narrative structure. Additionally, strong inter-model convergence is observed in both monolingual and cross-lingual settings, suggesting that multilingual LLMs rely on shared semantic abstractions despite architectural and linguistic differences. These findings shed light on the need for more comprehensive evaluations of cultural grounding. Relying exclusively on semantic similarity in multilingual narrative assessments may overestimate cultural preservation by neglecting culturally meaningful variations in narrative expression.

16.
arXiv (math.PR) 2026-06-18

Geometric obstructions to Lipschitz transport between weighted Hessian $\mathrm{CD}(\kappa,\infty)$ manifolds

arXiv:2606.11085v2 Announce Type: replace Abstract: We construct a weighted Riemannian manifold $(\mathbb R^2,g,\mu)$ satisfying $\mathrm{CD}(1/2,\infty)$, the curvature-dimension condition, with the following property: if $\gamma$ denotes a centered Gaussian measure on $\mathbb R^2$, then there is no Lipschitz map $T:(\mathbb R^2,\|\cdot\|) \to (\mathbb R^2,g)$ satisfying $T_\#\gamma=\mu$. Building on this, we prove a Weyl-type asymptotic law for the eigenvalues of the weighted Laplacian $-\Delta_{g,\mu}$ and show that they are asymptotically negligible when compared to the eigenvalues of $-\Delta_{\gamma}$. These results give strong counterexamples to two questions of E. Milman and complement the recent counterexample of Aryan.

17.
PLOS Computational Biology 2026-06-08

Assessing the inference of single-cell phylogenies and population dynamics from CRISPR lineage recordings

by Julia Pilarski, Tanja Stadler, Sophie Seidel Multicellular organisms develop from a single cell by repeated rounds of cell division, differentiation, and death, which can be represented as a single-cell phylogenetic tree. Genetic lineage tracing allows us to investigate this development by tracking the ancestry of individual cells as populations grow and change over time. However, accurate reconstruction of the cell phylogeny and quantification of the corresponding phylodynamic parameters – cell division, differentiation, and death rates – from this tracking data remains challenging and needs to be systematically evaluated. We perform simulations and assess, using the Bayesian framework, the joint inference of time-scaled cell phylogenies and phylodynamic parameters from CRISPR lineage recordings with random or sequential edits. Principally, we characterize the inference improvements as the recorder capacity increases. We observe more accurate phylogenetic reconstruction from sequential compared to random recordings, but no substantial improvement in phylodynamic inference when using the additional information contained in the order of edits. Overall, we find that CRISPR lineage recordings carry a strong signal on the rates of cell division when appropriate models are used. However, we detect biases in the inferred rates of cell division and death under phylodynamic model misspecification, i.e., when fitting classic memoryless birth-death processes to synchronous cell divisions. Moreover, for scenarios when cells differentiate into distinct types, we demonstrate that Bayesian phylodynamic analysis of sparse end-point measurements can resolve these cell differentiation trajectories by lineage and time. Under prototypical dynamics, we recover cell type-specific division and death rates, and cell type transition rates in over 80% of simulations. Overall, this simulation study explores how much information on cellular development can be extracted from state-of-the-art genetic lineage tracing data using phylogenetic and phylodynamic methodology.

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

Low-Latency Real-Time Audio Game Commentary System via LLM-Based Parallel Text Generation

We present a low-latency real-time audio game commentary system that generates spoken commentary directly from live gameplay video. In this end-to-end setting, a key bottleneck is accumulated waiting time; conventional pipelines capture frames, generate text, and synthesize speech sequentially for each utterance, and do not request the next generation until speech playback has completed. This strict sequentiality causes long and unnatural silence between utterances. To address this latency bottleneck, our system runs text generation in parallel with speech playback and buffers multiple candidate utterances ahead of time, enabling immediate synthesis at playback boundaries. Experiments on fast-paced game videos show that our parallel design reduces the mean inter-utterance silence from 9.6 seconds to 0.3 seconds compared to sequential baselines. It also improves similarity to professional speaking–silence timing patterns by over 40 %, and a user study with 120 experienced game players confirms significantly improved perceived speaking rhythm. Our demo video is available at: https://youtu.be/pmrRUlvav8M.

19.
arXiv (CS.LG) 2026-06-12

NetCause: Counterfactual Learning for Root Cause Analysis in Large-Scale Networks

arXiv:2606.13543v1 Announce Type: cross Abstract: Can a learned model capture how faults propagate through a large-scale network and use this knowledge to causally attribute customer impact to its underlying root cause? Existing root cause analysis techniques often rely on static rules, correlation heuristics, or topology-local reasoning, which struggle to generalize in dynamic environments where faults propagate across complex physical and logical dependencies. We present NetCause, a self-supervised learning-based framework that models network incidents as graph-temporal processes and uses counterfactual simulation to rank candidate root causes. This approach produces an interpretable ranking of root cause hypotheses and integrates naturally with operator-defined mitigation and remediation actions. We train the model on over 1,500 incidents collected over six months from a leading cloud provider's production network and evaluate it on 31 expert-labeled incidents. NetCause consistently improves root cause ranking quality in the regime most relevant to operational decision-making, achieving a 16.1% accuracy improvement over a rule-based heuristic baseline. While training is computationally intensive, inference is lightweight, requiring only seconds of GPU runtime per incident (well below typical telemetry collection latencies).

20.
arXiv (CS.CL) 2026-06-16

DRA-GRPO: Your GRPO Needs to Know Diverse Reasoning Paths for Mathematical Reasoning

Post-training LLMs with Reinforcement Learning, specifically Group Relative Policy Optimization (GRPO), has emerged as a paradigm for enhancing mathematical reasoning. However, standard GRPO relies on scalar correctness rewards that are often non-injective with respect to semantic content: distinct reasoning paths receive identical rewards. This leads to a Diversity-Quality Inconsistency, where the policy collapses into a narrow set of dominant modes while ignoring equally valid but structurally novel strategies. To bridge this gap, we propose Diversity-aware Reward Adjustment (DRA), a theoretically grounded framework that calibrates the reward signal using the semantic density of sampled groups. By leveraging Submodular Mutual Information (SMI), DRA implements an Inverse Propensity Scoring (IPS) mechanism that effectively de-biases the gradient estimation. This creates a repulsive force against redundancy, driving the policy to achieve better coverage of the high-reward landscape. Our method is plug-and-play and integrates seamlessly with GRPO variants. Empirical evaluations on five math benchmarks demonstrate that DRA-GRPO consistently outperforms strong baselines, achieving an average accuracy of 58.2% on DeepSeek-R1-Distill-Qwen-1.5B with only 7,000 training samples and $55 cost, highlighting the critical role of diversity calibration in data-efficient alignment. The code is available at https://github.com/xiwenc1/DRA-GRPO.

21.
arXiv (CS.CL) 2026-06-12

SafeLLM: Extraction as a Hallucination-Resistant Alternative to Rewriting in Safety-Critical Settings

Large language models (LLMs) are increasingly used to access organisational documentation, including standard operating procedures (SOPs), HR policies and institutional guidelines. However, retrieval-augmented generation (RAG) systems that rely on free-form rewriting can introduce hallucinations and unstable trade-offs between completeness and conciseness, particularly in safety- and compliance-critical settings. Objectives: To evaluate extraction as a hallucination-resistant alternative to rewriting-based RAG and compare strategies that balance precision, recall and safety across document types and model scales. Methods: We compare multiple prompting strategies, including line-number-based source selection, extraction of relevant guideline sentences with explicit safety annotations, and a multi-stage pipeline that refines draft answers using supporting evidence from source guidelines. Experiments are conducted on documents of varying length and structure, including local NHS acute care and oncology guidelines and UK-wide NICE guidelines, using both frontier-scale and locally deployable models. Performance is assessed using automatic metrics and human expert evaluation of relevance and completeness. Results: Line-number selection achieves the strongest results, outperforming direct copying and safety-focused strategies across both large and small models while maintaining high term recall (up to 95%) and close alignment with source text. Safety-oriented approaches improve precision but introduce systematic omissions, while multi-stage filtering further amplifies this trade-off. Performance varies with document structure: line-based extraction excels in protocol-like content, whereas alternative strategies perform better on more verbose documents (up to 97% term recall).

22.
bioRxiv (Bioinfo) 2026-06-24

ComplexDesign: sequence-hallucination design of protein binders bridging multiple proteins

Motivation: Designing multichain protein complexes requires coordinating the folding of component proteins with the formation of their interfaces. The existing methods, however, remain limited in their ability to satisfy these requirements simultaneously, especially for trimeric and tetrameric complexes. As an important practical scenario, designing a binder that bridges two target proteins into a ternary complex requires flexibility in the relative arrangement of the two targets, adding an additional challenge to existing design methods. Results: We present ComplexDesign, a hallucination-based approach for multichain protein design. ComplexDesign performs structure-prediction-guided sequence optimization to simultaneously fold each protein chain and form inter-chain interactions that bind them together. To provide the flexibility required to appropriately arrange these target proteins, ComplexDesign introduces a specialized masking mechanism that enables exploration of possible relative arrangements rather than being limited to the predefined ones. Across a comprehensive set of benchmarks with various chain lengths, ComplexDesign outperformed existing methods in the unconditional design of dimers, trimers, and tetramers, achieving a high design success rate exceeding 50%, supporting its capability for multichain complex design. Furthermore, in the case of multi-target binder design, ComplexDesign produced high-confidence, self-consistent ternary complexes for 8 out of 10 target pairs. These results establish ComplexDesign as an effective tool for multichain protein design, with particular utility for designing binders that bridge two target proteins. Availability and implementation: The source code of ComplexDesign will be made publicly available upon publication.

23.
arXiv (CS.CV) 2026-06-17

MoonSplat: Monocular Online Gaussian Splatting with Sim(3) Global Optimization

Online 3D reconstruction from monocular image sequences is a challenging and ongoing research topic. 3D Gaussian Splatting (3DGS), leveraging its high-quality real-time rendering capability, empowers online 3D reconstruction to represent dense scenes with enhanced expressiveness, and thus holds great promise for a wide range of applications such as robotics and AR/VR. However, existing online 3DGS methods still suffer from some key challenges: fragile camera pose estimation due to the lack of global optimization, and low optimization efficiency in large-scale or long-sequence scenarios. To address these issues, we propose a robust and efficient online voxelized 3DGS reconstruction framework integrated with global $Sim(3)$ optimization, which enables reliable camera tracking and efficient global loop closure for both camera poses and voxelized 3DGS. To accelerate the convergence of the voxelized 3DGS, we further introduce a color residual learning strategy, which not only boosts optimization speed but also enhances rendering quality. Extensive experiments on diverse indoor and outdoor datasets demonstrate that our method achieves state-of-the-art performance in both camera pose estimation accuracy and rendering quality, while retaining real-time efficiency. Additionally, we develop and deploy a real-world UAV-based active reconstruction system grounded on our proposed method, validating its robustness and generalizability for practical online 3D reconstruction tasks. Our code and data are available at https://github.com/TrickyGo/MoonSplat.

24.
arXiv (quant-ph) 2026-06-15

Conditional squeezing induced by a two-level system: arbitrary-time Magnus coefficients in the quantum Rabi model

arXiv:2508.03506v5 Announce Type: replace Abstract: We present a systematic Magnus expansion treatment of the quantum Rabi model beyond the Rotating Wave Approximation. We show that at the second order of Magnus series, the second-order evolution operator contains a term that induces conditional squeezing of the field mode depending on the state of the atom, in addition to the energy shifts. We analyze the scaling behavior of the conditional squeezing coefficient for $^{87}\mathrm{Rb}$ $5^2S_{1/2}\rightarrow5^2P_{1/2}$ transition line and show that the slow envelope of the squeezing coefficient is maximized at half-detuning cycles, and that it scales with $\frac{4g^2}{\omega_0|\Delta|}$. We also show that the quadrature squeezing angle suggests a possible route towards quantum non-demolition readouts, while further investigation is required for a full first-order suppression. We then connect our work to the well-studied AC-Stark shift and Bloch-Siegert shift using the effective Hamiltonian theory. Finally, we show how the energy shifts and the conditional squeezing arise, as a whole $\mathrm{SU}(1,1)$ algebra, and how they can be disentangled as individual unitary evolutions.