arXiv:2607.11329v1 Announce Type: new
Abstract: Universes are central to dependent type theory, and they are notoriously difficult to handle in a way that is both correct and usable. We propose a new "fuss-free" generalised algebraic presentation for polymorphic cumulative universes that dispenses with the intricate theory of coherent universe coercions in favour of a simpler formulation, which we prove equivalent by means of a normalisation theorem for the former. Evidence for the utility of the fuss-free formulation is provided in the form of (1) an abstract specification of its bidirectional elaboration algorithm, and (2) a concrete implementation in Haskell. We also describe and implement an extension of the fuss-free universe hierarchy with a judgemental notion of datatype description from which prior notions of cumulative inductive type may be derived.
Science Journals
arXiv:2607.10298v1 Announce Type: new
Abstract: Weakly supervised video anomaly detection relies solely on video-level labels for training, making it difficult to accurately localize anomalous events in complex scenes. In real-world videos, anomalous behaviors exhibit large variations in appearance and temporal duration, while scene appearance and action dynamics are often tightly entangled. Consequently, existing models tend to rely on scene-related statistical cues rather than true behavioral deviations, resulting in unstable detection performance. To address this challenge, we propose a Structured Evidence Selection framework (SESAD) that reformulates anomaly detection as a structured reasoning process over clip-level visual evidence. Instead of directly mapping aggregated features to anomaly scores, SESAD reorganizes clip representations into semantically structured candidate evidence and performs context-conditioned selection under scene and action constraints. This mechanism adaptively emphasizes anomaly-relevant semantics while suppressing scene interference, thereby alleviating semantic entanglement under weak supervision. Furthermore, we introduce a lightweight geometric discrimination module that constructs a dual-prototype structure in the embedding space, enabling anomaly decisions through relative geometric relations. Extensive experiments on UBnormal, ShanghaiTech, and UCF-Crime show that SESAD achieves 67.92, 97.99, and 88.46 AUC, respectively, while maintaining high computational efficiency and overall consistently stable anomaly discrimination.
arXiv:2607.10575v1 Announce Type: new
Abstract: Underwater object detection is strongly affected by domain shift, where performance can vary significantly across different locations, habitats, and deployment conditions. However, detector performance is typically evaluated using aggregate metrics that hide failures in specific environments, while existing domain generalization benchmarks often rely on synthetic variations that do not reflect real-world conditions. We introduce a framework that characterizes underwater images by appearance, scene composition, and acquisition geometry to assign domain labels. Using this framework, we perform the first systematic study of how domain factors influence both human annotation quality in underwater object detection datasets and deep learning-based detector performance, revealing substantial domain-dependent discrepancies. By incorporating physically meaningful domain labels, domain shift becomes something we can characterize, measure, benchmark, and act on. We highlight how this can be used to guide data collection and annotation, design more informative benchmarks, and assess detector robustness across diverse underwater environments.
arXiv:2607.10667v1 Announce Type: new
Abstract: For a relational structure A, the Minimum Cost Constraint Satisfaction Problem is the following problem denoted by MinCostCSP(A): Given an instance of CSP(A) with rational costs on variable-value pairs, find a solution to the instance minimizing the sum of the chosen costs. For the exact minimization, a classification of MinCostCSP(A) in terms of A was established by Takhanov [STACS'10].
We focus on constant-factor approximations of MinCostCSP(A). DeHaan, Huang, and Lee recently showed that if A fails to admit a conservative near-unanimity polymorphism then MinCostCSP(A) is not constant-factor approximable [APPROX'25]. We provide a first step towards a classification, by proving a dichotomy for structures A admitting a conservative majority (also known as 3-near-unanimity) polymorphism. Our dichotomy criterion is not in terms of an algebraic condition on A but we show that this is unavoidable. We include a simple argument proving that no such condition exists.
arXiv:2607.10374v1 Announce Type: new
Abstract: Safe and socially compliant navigation remains a fundamental challenge for autonomous robots operating in human-populated environments. Beyond collision avoidance, robots must anticipate human motion and respect personal space to ensure human comfort. Model Predictive Control (MPC) offers a robust alternative to classical and data-driven methods, although its effectiveness strongly depends on accurate human motion prediction and efficient computation. This paper introduces SFM-NMPC, a Social Force Model-based Non-linear Model Predictive Control framework that embeds human motion prediction directly within the optimization loop. By incorporating the Social Force Model into the dynamic model of surrounding agents, the controller jointly predicts the trajectories of humans and robots over the prediction horizon, thereby enabling socially-aware planning. A tailored set of social cost functions guides the optimization toward human-compliant behaviors. Despite the increased model complexity, the proposed formulation runs in real time at 20 Hz. Extensive simulated testing in crowded environments demonstrates that SFM-NMPC outperforms state-of-the-art baselines in social compliance metrics while maintaining efficient and smooth navigation. Visual trajectory analysis and an ablation study further highlight the contribution of the embedded SFM dynamics and social cost terms, confirming the effectiveness of the proposed approach for real-world social navigation.
arXiv:2607.10626v1 Announce Type: new
Abstract: LLM-as-a-judge evaluation is widely used for retrieval-augmented generation (RAG), but reusing the same model family as both generator and judge makes self-leniency difficult to identify. We introduce Eval-Pair Matrix, a controlled meta evaluation protocol for source-grounded RAG. Starting from GaRAGe questions and grounding passages, we induce one hidden answer-causal contradiction per record, generate answers from perturbed passages with GPT, Grok, and Gemini models, and then use the same models as blind judges to evaluate each answer against the original passages. The experiment contains 300 core records, 897 labeled generator outputs, and 2,683 judge verdicts in a crossed 3 x 3 matrix; the primary analysis uses 275 fully validated records. Instead of comparing diagonal and off-diagonal cells across different answers, we estimate same-model effects by pairing judges on the exact same candidate answer. This changes the interpretation: diagonal and off diagonal F1 are similar, and the paired same-model recall effect is near zero (-0.5 pp; 95% cluster bootstrap CI [-2.7, +1.7]). The only robust paired gap is lower matching-judge flagging for answers that avoided the induced claim (-4.3 pp). A targeted human evaluation finds that reviewed apparent false positives are alternate source-error detections, mistakes in labeling whether the induced claim was adopted, or unclear cases; none were adjudicated as genuine false alarms. The lesson is methodological: RAG judge studies should report full matrices, answer-paired effects, behavior strata, and label-task alignment.
arXiv:2607.09712v1 Announce Type: new
Abstract: Financial control testing increasingly depends on representative enterprise resource planning (ERP) data in quality environments, yet direct production copies expose personal, supplier, banking, and commercially sensitive records. This work presents Secure ERP Quality Provisioning for Financial Control Testing (SEQ-FCT), a governed data-provisioning framework that combines deterministic masking, synthetic scenario expansion, referential tokenization, policy-based release approval, and automated validation for reconciliation, fraud-rule testing, and audit analytics. A single synthetic dataset is used for evaluation. It contains 186,000 finance-process records from six subsidiaries over 2022-2025, including accounts payable invoices, payments, general-ledger journals, accounts receivable receipts, and bank-statement lines. The dataset includes entity relationships, monetary values, approval paths, tax attributes, banking markers, exception labels, fraud-rule triggers, and control-failure outcomes. Because the dataset is synthetic, reported results demonstrate controlled internal consistency rather than production validation. Against a production-clone upper bound, static masking, rules-only synthesis, conditional tabular generative synthesis, and a hybrid baseline, SEQ-FCT achieved 0.932 reconciliation F1, 0.887 fraud-trigger recall, 0.914 control-failure F1, and an estimated leakage-risk score of 0.018. The analysis indicates that financial process behavior can be preserved more reliably when masking, synthetic data, and governance checks are evaluated as a single release pipeline instead of independent utilities.
arXiv:2607.10491v1 Announce Type: new
Abstract: Retrieval-augmented generation grounds large language models in external evidence, but most pipelines still treat retrieved passages as deterministic and mutually consistent context. In open information environments, retrieved sources may disagree because of temporal drift, source error, ambiguity, or genuine uncertainty. This paper introduces ERAG, an uncertainty-aware RAG framework that converts retrieved chunks into probabilistic evidence before generation. A lightweight evaluator extracts candidate claims and maps chunk-level support to Dirichlet evidence. A conflict-preserving Dempster-Shafer fusion rule then transfers unresolved disagreement into epistemic uncertainty rather than normalizing it away. The generator is routed to direct answering, conflict-aware answering, or abstention according to the fused uncertainty score. Experiments on CRAG, ConflictQA, and MuSiQue show that ERAG remains competitive with the strongest matched baseline on standard question answering while improving behavior under conflict. On the CRAG ambiguous subset, hallucination decreases from 45.3% for Corrective RAG to a human-calibrated estimate of 34.8%, conflict resolution increases from 35.2% to 51.2%, and expected calibration error improves to 0.122. These results suggest that evidential modeling is a practical mechanism for trustworthy information processing in foundation-model-based retrieval systems.
arXiv:2604.07643v1 Announce Type: cross
Abstract: Experienced storytellers decompose stories into local narrative strategies and how these strategies shape higher-level arcs. This decomposition helps writers recognize patterns in others' work and adapt those patterns to tell new stories. Novices, however, struggle to identify these strategies or to reuse them effectively. We present Narrix, a novel writing tool that helps novice writers recognize narrative strategies in example stories and repurpose these strategies in their own writing. Narrix analyzes strategies in example stories, highlights them with color-coded lexical cues and explanations, and situates them on an interactive story arc for exploration by emotional shifts and turning points. Writers then drag strategies onto multi-dimensional tracks and apply block-scoped edits to revise or continue their drafts through controlled generation steered by specified strategies. Through a within-subjects study (N=12), Narrix showed improved participants' retention, confidence, and creative adaptation of narrative strategies compared to a baseline chat-based writing interface.
arXiv:2512.13068v2 Announce Type: replace
Abstract: This work analyzes the convergence of sums of the form $S_{\boldsymbol{\gamma}}(m)=\sum_{v\subseteq \mathbb{N}}\gamma_v m^{|v|}$ with product and order dependent (POD) weights $\gamma_v$. We establish that for a nonnegative sequence $\{\Upsilon_j\mid j\in \mathbb{N}\}$, $$\sum_{v\subseteq \mathbb{N}} |v|! m^{|v|}\prod_{j\in v} \Upsilon_j<\infty \text{ for all } m>0 \text{ if and only if } \sum_{j=1}^\infty \Upsilon_j<\infty.$$ We further characterize the growth of $S_{\boldsymbol{\gamma}}(m)$ when $\gamma_v=(|v|!)^{\sigma}\prod_{j\in v}j^{-\rho}$ and prove that $\log S_{\boldsymbol{\gamma}}(m)$ is of asymptotic order $m^{1/(\rho-\sigma)}$ when $\rho>\sigma\geq 0$. We subsequently generalize both the convergence criterion and the asymptotic order of $\log S_{\boldsymbol{\gamma}}(m)$ to smoothness-driven product and order dependent (SPOD) weights, while noting that a full necessary-and-sufficient analogue remains open. Finally, we apply our theory to quasi-Monte Carlo (QMC) integration, showing that interlaced polynomial lattice rules achieve a dimension-independent convergence rate without a commonly imposed assumption in the QMC literature.
arXiv:2512.13840v3 Announce Type: replace
Abstract: We introduce MoLingo, a text-to-motion (T2M) model that generates realistic, lifelike human motion by denoising in a continuous latent space. Recent works perform latent space diffusion, either on the whole latent at once or auto-regressively over multiple latents. In this paper, we study how to make diffusion on continuous motion latents work best. We focus on two questions: (1) how to build a semantically aligned latent space so diffusion becomes more effective, and (2) how to best inject text conditioning so the motion follows the description closely. We propose a semantic-aligned motion encoder trained with frame-level text labels so that latents with similar text meaning stay close, which makes the latent space more diffusion-friendly. We also compare single-token conditioning with a multi-token cross-attention scheme and find that cross-attention gives better motion realism and text-motion alignment. With semantically aligned latents, auto-regressive generation, and cross-attention text conditioning, our model sets a new state of the art in human motion generation on standard metrics and in a user study. We will release our code and models for further research and downstream usage.
arXiv:2512.17776v5 Announce Type: replace
Abstract: Recent advances in large language models have enabled deep research systems that generate expert-level reports through multi-step reasoning and evidence-based synthesis. However, evaluating such reports remains challenging: report quality is multifaceted, making it difficult to determine what to assess and which criteria to use; LLM-based judges may miss errors that require domain expertise to identify; and because deep research relies on retrieved evidence, report-wide claim verification is also necessary. To address these issues, we propose DEER, a benchmark for evaluating expert-level deep research reports. DEER systematizes evaluation criteria with an expert-developed taxonomy (7 dimensions, 25 subdimensions) operationalized as 101 fine-grained rubric items. We also provide task-specific Expert Evaluation Guidance to support LLM-based judging. In addition to rubric-based assessment, we propose a claim verification architecture that verifies both cited and uncited claims and quantifies evidence quality. Experiments show that current systems produce structurally plausible, evidence-citing reports, but still struggle to fully satisfy expert-level user requests and achieve logical completeness. Beyond performance comparisons, DEER makes system strengths and limitations interpretable and provides diagnostic signals for improvement.
arXiv:2512.19311v2 Announce Type: replace
Abstract: This paper studies the training-testing discrepancy (a.k.a. exposure bias) problem for improving the diffusion models. During training, the input of a prediction network at one training timestep is the corresponding ground-truth noisy data that is an interpolation of the noise and the data, and during testing, the input is the generated noisy data. We present a novel training approach, named MixFlow, for improving the performance. Our approach is motivated by the Slow Flow phenomenon: the ground-truth interpolation that is the nearest to the generated noisy data at a given sampling timestep is observed to correspond to a higher-noise timestep (termed slowed timestep), i.e., the corresponding ground-truth timestep is slower than the sampling timestep. MixFlow leverages the interpolations at the slowed timesteps, named slowed interpolation mixture, for post-training the prediction network for each training timestep. Experiments over class-conditional image generation (including SiT, REPA, and RAE) and text-to-image generation validate the effectiveness of our approach. Our approach MixFlow over the RAE models achieve strong generation results on ImageNet: 1.43 FID (without guidance) and 1.10 (with guidance) at 256 x 256, and 1.55 FID (without guidance) and 1.10 (with guidance) at 512 x 512.
arXiv:2512.21299v2 Announce Type: replace
Abstract: Analysing the dynamics of phase-changing liquid films is essential for enhancing the performance of thermal management systems. Still, direct simulation of the full governing equations is computationally expensive. To circumvent this limitation, I derived a weighted-integral boundary-layer (WIBL) model under long-wave assumptions, weak evaporation, and strong surface tension, also accounting for variable substrate heating. In the linear regime, the WIBL reproduces growth rates and the cutoff wavenumber of unstable modes with significantly higher accuracy than commonly used Benney-type models for Re<40, as compared to the Orr-Sommerfeld equations. The linear analysis further reveals a threshold separating streamwise- and spanwise-dominated instabilities in hanging films, arising from the competition between Kapitza and Rayleigh-Taylor mechanisms; the WIBL predicts this threshold accurately for small Re and inclination angles. In the nonlinear regime, with substrate heating that varies in both space and time, the WIBL model captures the evolution of free-surface thickness and temperature within approximately 6% of the original Navier-Stokes equations. Three-dimensional simulations show that a condensing film undergoes dry-out due to Kapitza instability, whereas unsteady substrate heating promotes spanwise momentum spreading, modifies wave dynamics, and prevents dry-out. The WIBL model provides a good level of accuracy at a low computational cost, enabling extensive parametric studies, nonlinear stability analyses, and the design of optimal substrate-heating control strategies.
arXiv:2512.22140v3 Announce Type: replace
Abstract: We followed scientists who started publishing in 2000 and who continued publishing until 2020-2023 (N = 41,424). These survivors in science authored 2 million articles (N = 2,089,097) with more than 70 million cited references (N = 73,118,395) and worked in 38 OECD countries. Using a raw Scopus dataset, we examined gender disparities in publishing intensity, international collaboration, journal selection, productivity, citations, team formation, and publishing breaks in 16 STEMM and social science disciplines. Several author-level metrics were computed. Our data show a gender productivity gap for both lifetime scholarly output and annual journal prestige-normalized productivity. Surprisingly, in the context of extant literature, the data do not show a gender international collaboration gap, a gender journal selection gap, a gender citation gap, or a gender team formation gap. Men were on average 23% more productive than women cumulatively in 2000-2023 and 19% more productive in the last 5 years studied (2019-2023). Men and women published in equally prestigious journals, received the same number of citations (field-normalized), and worked in equally sized teams. In all, 80% of scientists in STEMM disciplines and 70% in the social sciences had published every year. Our data indicate interesting disciplinary differences in gender disparities.
arXiv:2512.22596v3 Announce Type: replace
Abstract: The present study investigates the attacker-defender (AD) model proposed by Brink et al. (2023), a motion model that describes the interactions between a ball carrier (attacker) and the nearest defender during ball possession. The model is based on the equations of motion for both players, incorporating resistance, goal-oriented force, and opponent-oriented force. It generates trajectories based on physically interpretable parameters. Although the AD model reproduces real dribbling trajectories well, previous studies have explored only a limited range of parameter values and relied on relatively small datasets.
This study aims to (1) enhance parameter optimization by solving the AD model for one player with the opponent's actual trajectory fixed, (2) validate the model's applicability to a large dataset from 306 J1 League matches, and (3) demonstrate distinct playing styles of attackers and defenders based on the full range of optimized parameters.
arXiv:2512.22749v2 Announce Type: replace
Abstract: We study the pricing behavior of third-party platforms facing strategic agents. Assuming the platform is a revenue maximizer, it observes market features that generally affect demand. Since only transacted quantities and prices can be observed, this presents a general demand learning problem under confounding. Mathematically, we develop an algorithm with optimal regret of $\Tilde{\mathcal{O}}(\sqrt{T}\wedge\sigma_S^{-2})$. Our results reveal that supply-side noise fundamentally affects the learnability of demand, leading to a phase transition in regret. Technically, we show that non-i.i.d. actions can serve as instrumental variables for learning demand. We also propose a novel homeomorphic construction that allows us to establish estimation bounds without assuming star-shapedness, providing the first efficiency guarantee for learning demand with deep neural networks. Finally, we use simulations and offline counterfactuals from Talabat and Lyft data to illustrate the potential revenue implications of our approach.
Decoupling Constraints from Two Directions for Evolutionary Constrained Multi-objective Optimization
arXiv:2512.23945v2 Announce Type: replace
Abstract: Real-world constrained multi-objective optimization problems (CMOPs) commonly involve multiple constraints, and understanding and exploiting their coupling relationships is crucial for efficient optimization. Recent constraint-decoupling methods handle individual constraints separately, but they generally search only in the evolutionary direction to approximate single-constraint Pareto fronts (SCPFs). In this study, we show that part or all of the constrained Pareto front (CPF) may be unrelated to any SCPF and instead be shaped by the boundaries of infeasible regions. We refer to such a portion as the independent CPF (ICPF) and introduce the reverse CPF (RCPF) to characterize its associated informative infeasible boundaries. Based on these observations, we propose a bidirectional constraint-decoupling coevolutionary algorithm named DCF2D. DCF2D dynamically identifies the constraints obstructing the main population and activates constraint-specific auxiliary populations. These populations adaptively search in the evolutionary direction for the corresponding SCPFs or in the reverse evolutionary direction for the corresponding RCPFs. Its three-stage framework integrates unconstrained global exploration, event-driven bidirectional coevolution, and final convergence refinement. Experiments on 87 benchmark instances from seven test suites and 28 real-world engineering CMOPs demonstrate that DCF2D achieves the best overall performance among nine algorithms. Code available at: https://github.com/RuiqingS/DCF2D.
arXiv:2601.01406v3 Announce Type: replace
Abstract: Face super-resolution aims to recover high-quality facial images from severely degraded low-resolution inputs, but remains challenging due to the loss of fine structural details and identity-specific features. This work introduces SwinIFS, a landmark-guided super-resolution framework that integrates structural priors with hierarchical attention mechanisms to achieve identity-preserving reconstruction at both moderate and extreme upscaling factors. The method incorporates dense Gaussian heatmaps of key facial landmarks into the input representation, enabling the network to focus on semantically important facial regions from the earliest stages of processing. A compact Swin Transformer backbone is employed to capture long-range contextual information while preserving local geometry, allowing the model to restore subtle facial textures and maintain global structural consistency. Extensive experiments on the CelebA benchmark demonstrate that SwinIFS achieves superior perceptual quality, sharper reconstructions, and improved identity retention; it consistently produces more photorealistic results and exhibits strong performance even under $8\times$ magnification, where most methods fail to recover meaningful structure. SwinIFS also provides an advantageous balance between reconstruction accuracy and computational efficiency, making it suitable for real-world applications in facial enhancement, surveillance, and digital restoration. Our code, model weights, and results are available at https://github.com/Habiba123-stack/SwinIFS.
arXiv:2601.05636v2 Announce Type: replace
Abstract: We study deletion-correcting codes in the space of length-$n$ multisets over a $q$-ary alphabet. We present an explicit cyclic Sidon-type construction for arbitrary alphabet size $q$ and deletion radius $t$, defined by a single congruence modulo $t(t+1)^{q-2}+1$. The construction has redundancy at most $\log_q(t(t+1)^{q-2}+1)$ and admits linear-time online decoding for fixed $q$ and $t$ after finite preprocessing. We prove that its syndrome classes are asymptotically balanced and compare several general upper bounds. For a single deletion, we show that the natural sum-modulo construction is asymptotically optimal for every fixed $q$. We also obtain exact results for $q=3$ and $q=4$, including uniqueness results for optimal codes in the relevant parameter ranges, and formulate conjectures for prime alphabets.
arXiv:2607.10532v1 Announce Type: new
Abstract: Multimodal Entity Alignment (MMEA) aims to identify equivalent entities across different modalities. While existing methods enhance MMEA performance through black-box context engineering strategies, their reliance on LLM parameter capacity and lack of theoretical interpretability remain unresolved. To this end, we first theoretically validate the mathematical equivalence between context engineering and model fine-tuning in MMEA tasks, demonstrating that prompt components simulate contrastive learning-based sequential fine-tuning in MMEA. Building on this foundation, we then propose PTFEA, a curriculum-learning-inspired framework that translates fine-tuning strategies into interpretable context engineering. Specifically, adaptive difficulty modulation dynamically adjusts information injection stages using confidence thresholds, establishing mathematical equivalence between curriculum learning weights and context sample selection; and three-stage progressive inference incorporates entity information from simple to complex cases, mirroring the gradient descent process in fine-tuning. Experiments on five public datasets demonstrate that PTFEA consistently outperforms strong baselines. In particular, on the ICWIKI dataset, PTFEA narrows the H@1 gap between Qwen2.5-72B and 14B to 0.6%. Moreover, compared with the representative context-engineering-based MMEA method MM-ChatAlign, PTFEA reduces the runtime of Qwen2.5-72B from 21 hours to 1 hour and lowers token consumption from 2200-3000 to 200-400, achieving over 80% reduction on the ICWIKI dataset. This work provides the first theoretical framework unifying context engineering and fine-tuning in MMEA, paving the way for future research that seeks to translate additional fine-tuning strategies into context engineering paradigms. Our code is available at https://github.com/DMiC-Lab-HFUT/PTFEA.
arXiv:2607.10555v1 Announce Type: new
Abstract: Generative Large Language Models (LLMs) have revolutionized information retrieval, yet their strictly parametric nature frequently leads to severe factual hallucinations when confronted with complex queries beyond their epistemic boundaries. While external tool-calling can mitigate this, indiscriminately invoking search tools for every document during reranking incurs prohibitive latency overheads, creating an intractable accuracy-efficiency dilemma. To address this challenge, we propose TALRanker, a novel framework that formalizes pointwise relevance scoring as an agentic Markov decision process. We optimize it via a two-stage training paradigm. An initial warm-up utilizes a language-preserving hybrid loss to prevent the catastrophic forgetting of native generative capacities. Subsequently, an asymmetric cost-aware reward equipped in reinforcement learning forces the policy to autonomously bypass tools for maximum efficiency when confident, while selectively retrieving external evidence to avert severe hallucination penalties when uncertain. Extensive evaluations demonstrate that TALRanker achieves state-of-the-art performance across standard and reasoning-intensive retrieval benchmarks, matching throughput with pointwise rerankers while outperforming parameter-heavy reasoning models.
arXiv:2601.06362v2 Announce Type: replace
Abstract: Personalized large language models are often expected to follow explicit style instructions, yet we find that such instructions can undermine the user-specific characteristics that personalization methods aim to preserve. We call this failure mode personalization collapse: explicit style control can conflict with implicit user preferences. To address this challenge, we propose PsPLUG, a lightweight plug-in that learns a user-specific residual after accounting for the requested style. PsPLUG also allows us to tune personalization strength at inference time. Our experiments show that explicit style instructions can diminish personalization in existing methods, whereas PsPLUG better preserves user preferences while providing precise control over the balance between personalization and style adherence.
arXiv:2607.10557v1 Announce Type: new
Abstract: Multimodal BrowseComp tasks require agents to combine perception, tool use, and long-horizon reasoning over dynamic web content, challenging their ability to handle compositional structure, open-world uncertainty, and multimodal integration across extended interactions. Crucially, real-world multimodal browsing involves three distinct information-flow patterns: text-only, image-to-text, and text-to-image, yet existing data construction methods cover only the text-only and image-to-text patterns, leaving text-to-image largely unaddressed and limiting agent generality and robustness. We introduce UNIBROWSE, a unified data pipeline that for the first time simultaneously generates training data covering all three patterns, augments curated knowledge graphs with live web retrieval for improved fidelity, and introduces a novel metric of exploration degree to filter low-signal instances for efficient reinforcement learning. Through this pipeline, we produce high-quality cold-start tool-use trajectories and exploration-rich QA pairs, and train a 35B-scale agent via supervised fine-tuning and exploration-aware RL.The resulting UNIBROWSE agent achieves state-of-the-art performance on multimodal BrowseComp benchmarks, attaining an average accuracy of 54.4 across five diverse benchmarks -- an improvement of 10.5 points over its base model Qwen3.5-35B-A3B -- and surpassing serveral closed-source agent workflows such as GPT-5 (42.9), Gemini-2.5 Pro (44.8), and Gemini-2.5 Flash (41.3).
arXiv:2607.10955v1 Announce Type: new
Abstract: Graduate programs in STEM disciplines are central to preparing future researchers and professors. Program requirements for students often include taking several graduate-level courses. Anecdotal evidence suggests that graduate coursework in physics in particular features outdated and ineffective pedagogical methods, with high emphasis on mathematical rigor in place of conceptual learning or connections made with authentic research. Prior physics education research indicates that students can leave graduate courses with shortcomings in conceptual understanding of covered topics. This pilot study is designed to document student perspectives on their graduate coursework in a single U.S. R1 physics Ph.D. program. A total of 14 semi-structured interviews were conducted with students enrolled in the program, and thematic analysis was conducted on five of these interviews for this paper. The resulting themes are discussed, including the prevalence of traditional passive lecture pedagogy and students placing high value on course content relevant to research.