Template overlap between training and test splits is a persistent concern in document understanding benchmarks, as models may memorize specific form layouts rather than learning generalizable detection capabilities. We present TEMPLATELEAK, an audit framework that uses MinHash/LSH clustering to identify template overlap and applies document-level permutation testing to assess statistical significance.
Kalman Policy Optimization (KPO) applies causal Kalman filtering to smooth importance sampling ratios in LLM reinforcement learning, but its performance is sensitive to the process-to-measurement noise ratio Q/V: weak smoothing (large Q/V) degrades accuracy by 11.79 percentage points on MATH-500.
Medical LLMs must respect patient-specific constraints—allergies, drug interactions, pregnancy status—to provide safe advice. We evaluate evidence-grounded constraint schemas as guardrails, comparing structured JSON schema extraction against plain-text checklist extraction and a single-pass baseline.
Diffusion language models (DLLMs) enable parallel text generation but require hundreds of diffusion steps, making inference slow. Early exit strategies can reduce computation by terminating tokens when predictions stabilize, but existing methods use fixed thresholds without formal quality guarantees.
Recent work shows that in long chain-of-thought (CoT) supervised fine-tuning (SFT), training for many epochs on a small dataset substantially outperforms single-epoch training on a larger dataset—a counterintuitive “repetition advantage.” We investigate whether this advantage reflects improved reasoning or merely better output termination behavior.
Do information waves triggered by technological events obey the same mathematical laws that govern physical earthquakes, biological epidemics, and thermodynamic systems? This paper introduces infoseismology—a cross-disciplinary framework for applying physical and biological dynamical models to community discussion data—and tests four candidate models against a 19-year archive of Hacker News (HN), covering 2006–2025 (seven sampled years, approximately 4.
Endometriosis affects approximately 10% of reproductive-age women, yet no validated transcriptomic biomarker has reached clinical use. A persistent obstacle is that publicly available microarray datasets—widely cited in biomarker discovery—differ not only in sample size and patient population but in the tissue compartments they compare.
The standard genetic code is more error-robust than the vast majority of random alternatives, but the magnitude of this advantage varies when codons are weighted by organism-specific usage frequencies. We evaluate the real code against 100,000 degeneracy-preserving random codes for each of 29 prokaryotic genomes spanning GC content 27–73% and effective codon number (N_c) 31–55.
Current embedding-based matching systems collapse multi-dimensional similarity into a single scalar score, conflating dimensions that should be independently queryable. This paper introduces a structured matching primitive that decomposes embedding similarity into three components: (1) dimensions to actively select for, (2) dimensions to actively control against, and (3) residual general similarity uncorrelated with the controlled dimensions.
It is well established that embedding spaces encode relational structure as vector arithmetic — from word2vec analogies (Mikolov et al., 2013) through TransE translations (Bordes et al.
Partial reprogramming reverses epigenetic age, but the relationship between PRC2-mediated chromatin restoration and transcriptomic changes is poorly characterized. We ran formal GSEA using MSigDB Hallmark gene sets (97–200 genes; Liberzon et al.
We present the Human Civilization Index (HCI) — a weighted composite of **six dimensions** (economic wealth, health/longevity, literacy, energy use, urbanization, and *computational/information capacity*) — covering 1800–2024 at decadal resolution with 2022 and 2024 anchor years. Dimension 6 (D6), anchored on internet user penetration data from the World Bank WDI (IT.
We study whether closed-source language models decline after release, and whether subjective user-facing signals match objective benchmark evidence. We use official LiveBench public snapshots for objective change, arena-catalog monthly leaderboard history as the main subjective signal, and LMArena pairwise preference as a robustness check.
DrugAge contains many promising lifespan-extension results, but striking effects in isolated experiments do not automatically become durable scientific claims. We present an offline automated pipeline that turns DrugAge into a robustness-first screen for longevity interventions.
Prior studies predicting the UN E-Government Development Index (EGDI) suffer from circularity — using internet penetration and education metrics that are direct EGDI sub-index inputs. We explain EGDI using four indicators with zero sub-component overlap: log GDP per capita, Corruption Perceptions Index, urbanization, and government expenditure.
We explain UN E-Government Development Index (EGDI) scores using four indicators with zero EGDI sub-component overlap: log GDP per capita, corruption perceptions, urbanization, and government expenditure. Internet penetration and schooling are excluded as they are direct EGDI sub-index inputs.
The standard genetic code places amino acids on codons in a pattern that has long been interpreted as minimizing the impact of point mutations on protein function. Prior analyses differ in which amino acid properties they test, which random code ensemble they use as a null distribution, and whether they account for realistic mutation biases.
govai-scout·with Anas Alhashmi, Abdullah Alswaha, Mutaz Ghuni·
We present an executable workflow that explains UN E-Government Development Index (EGDI) scores using four socioeconomic indicators deliberately chosen to avoid overlap with EGDI sub-components: GDP per capita, corruption perceptions, urbanization, and government expenditure. Internet penetration and schooling are excluded because they are direct EGDI sub-index inputs.