Skip to main content
Search

University of Science and Technology Beijing

5 related articles

Subscribe to updates
  1. arXiv

    Tail calibration flags when a larger Best-of-N budget backfires, and code generation is where it shows

    Best-of-N inference returns the candidate with the highest proxy reward, so a larger budget reads the extreme tail of the reward distribution rather than its average ranking quality. The work writes the BoN value as a Beta-weighted integral of conditional true reward over within-prompt score quantiles, showing that BoN depends on scores only through their ranks and can decline only when the calibration curve in the tail turns downward. On public candidate pools, negative tail excess concentrates in code generation, and a cross-fitted tail statistic predicts whether a budget cap pays off on held-out data about as well as rank accuracy.

  2. arXiv

    SplitMoE splits the video-diffusion expert pool into semantic and generic branches, beating a same-source MoE at 14B activated parameters and showing coarse-to-fine denoising routing

    The work identifies a "uniformity trap" in which language-style token-wise MoE routing, combined with uniform expert-usage regularization, scatters neighboring patches of the same object across different experts in video diffusion, causing routing fragmentation and structural distortion; it proposes SplitMoE, which bifurcates the expert pool into semantic and generic experts and uses prototype-guided routing with pull-push regularization so tokens cluster by semantic attributes, outperforming traditional load-balanced MoE on VBench-2.0 and T2V-CompBench under an equivalent activated-parameter budget, while revealing a coarse-to-fine denoising pattern in which semantic-expert weight peaks in the early high-noise stage around steps 10-15 and then shifts toward generic experts.

  3. arXiv

    NeuroSymb-MRG generates radiology reports via differentiable abductive reasoning and active uncertainty minimization, raising BLEU-1 to 0.602 on IU X-ray and 0.487 on MIMIC-CXR

    The work presents NeuroSymb-MRG, a framework that maps image features to probabilistic clinical concepts, composes multi-hop abductive reasoning chains through a differentiable logic layer (product t-norm AND, probabilistic sum OR, learnable gating α), decodes those chains into roughly 120 clause templates augmented with retrieved evidence and constrained LLM paraphrasing, and drives clinician-in-the-loop review with active sampling based on rule-level predictive entropy (5 MC-dropout passes) plus k-center diversity (k=16 per round); on IU X-ray and MIMIC-CXR it improves BLEU-1 through BLEU-4, ROUGE-L and METEOR over the listed representative baselines, for example IU X-ray BLEU-1 of 0.602 and MIMIC-CXR BLEU-1 of 0.487.

  4. Yang and Xia prove the generalized trace ratio problem needs both a redundant constraint and scaling to close its Lagrangian duality gap

    This paper studies the generalized trace ratio problem (GTRP), which maximizes a trace-form quadratic fractional objective over the Stiefel manifold, and, using a newly established matrix S-lemma, proves that adding the redundant constraint XXT⪯I_n together with a well-chosen scaling yields an equivalent problem (GRS) with zero Lagrangian duality gap, whereas the original (GTRP), the redundant-constraint-only version (GR), and the scaling-only version (GS) can all exhibit a positive Lagrangian duality gap.

  5. Mendeley Data

    A corrosion inhibitor dataset built from 5,597 publications

    This work assembles a corrosion inhibitor dataset from 5,597 publications using a multi-agent extraction pipeline, releasing a manually verified IE_datasets.xlsx, LLM-extracted json_extracted.zip, and IE_PH_4_materials.xlsx for model construction, with fields covering reference metadata, inhibitor name and composition, anodic/cathodic/mixed type, film mechanism, SMILES, corrosion material name, grade, composition, processing and heat treatment, corrosive medium, medium type, concentration and temperature, and test temperature, test time, inhibitor concentration, test method and inhibition efficiency percentage.

You’re all caught up