PaperGym: Rubric-Centered Evolution for Research-Plan Generation
Paper • 2608.31119 • Published • 31
data_source stringclasses 3
values | prompt listlengths 1 1 | ability stringclasses 3
values | reward_model dict | extra_info dict |
|---|---|---|---|---|
physics | [
{
"content": "Accurate prediction of time-dependent aerodynamic drag and shape evolution for freely moving liquid drops in subcritical Weber number regimes remains a critical challenge in multiphase fluid dynamics, particularly due to nonlinear fluid-structure interactions, multi-mode oscillations, and unresolv... | physics | {
"ground_truth": "The proposed solution employs a hybrid computational framework integrating high-fidelity axisymmetric fluid dynamics simulations with data-driven temporal modeling to address the challenge of predicting unsteady aerodynamic drag and shape evolution in liquid drops under subcritical Weber number reg... | {
"arxiv_id": "2405.00897",
"filename": "2405.00897_merged_body.json",
"index": 0,
"merged_rubrics": [
{
"description": "The solution explicitly models the coupling between drop shape evolution and aerodynamic drag (e.g., through shared network outputs, physics-constrained equations, or bidirectional ... |
cs | [
{
"content": "Despite advancements in quadrupedal robot locomotion leveraging rigid-body dynamics and high-torque actuation systems, the integration of actively controlled compliant spinal mechanisms with trajectory optimization for agile jumping performance remains a critical unresolved challenge. While passiv... | cs | {
"ground_truth": "The proposed solution introduces a hybrid rigid-elastic dynamic framework for quadrupedal robot locomotion, integrating actively controlled spinal compliance with trajectory optimization to address limitations in energy transfer and actuation constraints. A mechanically lock-enabled elastic prismat... | {
"arxiv_id": "2109.00149",
"filename": "2109.00149_merged_body.json",
"index": 1,
"merged_rubrics": [
{
"description": "The solution states that an offline nonlinear MPC framework co-optimizes spring stiffness k_s, rest length l_s,rest, and motion trajectories.",
"id": 1,
"weight": 1
... |
cs | [
{
"content": "In high-speed train (HST) wireless communication systems operating within viaduct-dominated environments, beamforming optimization for massive MIMO faces unresolved challenges in balancing directivity and beamwidth while maintaining spectral efficiency and positioning accuracy under extreme mobili... | cs | {
"ground_truth": "The reference solution proposes a location-aware beamforming framework for massive MIMO systems in high-speed train environments, prioritizing computational efficiency and robustness against positioning errors. The method leverages precomputed phase excitation mappers derived from vehicle location ... | {
"arxiv_id": "1702.02121",
"filename": "1702.02121_merged_body.json",
"index": 2,
"merged_rubrics": [
{
"description": "The solution avoids real-time channel state information (CSI) acquisition by relying on location-based precomputed phase excitation mappers or offline routing tables.",
"id": ... |
cs | [
{
"content": "The integration of model-driven statistical methods (e.g., ARIMA, GARCH) with data-driven neural networks (e.g., LSTM) for financial time series analysis remains a critical challenge in machine learning, as existing approaches fail to reconcile the interpretability and theoretical guarantees of st... | cs | {
"ground_truth": "The proposed solution addresses the challenge of integrating model-driven statistical methods with data-driven neural networks for financial time series analysis by decomposing the series into distinct components based on stability properties. This decomposition enables tailored modeling strategies... | {
"arxiv_id": "2110.00082",
"filename": "2110.00082_merged_body.json",
"index": 3,
"merged_rubrics": [
{
"description": "The solution explicitly describes decomposing the time series into at least three components based on stability properties (linear stable, nonlinear stable, unstable).",
"id":... |
physics | [
{
"content": "The optical memory effect (ME), a phenomenon enabling deeper tissue imaging by exploiting angular and translational correlations in scattered light, faces unresolved theoretical and experimental challenges in reconciling its predicted limitations with empirical observations in anisotropic biologic... | physics | {
"ground_truth": "The proposed solution investigates the memory effect (ME) in anisotropic biological media through a synergistic framework of experimental measurements and computational simulations. The method addresses the limitations of isotropic diffusion models in predicting ME correlation ranges in tissues wit... | {
"arxiv_id": "1502.00270",
"filename": "1502.00270_merged_body.json",
"index": 4,
"merged_rubrics": [
{
"description": "The solution introduces an explicit metric to map anisotropic scattering dynamics to equivalent diffusive behavior (e.g., effective thickness L_eff).",
"id": 1,
"weight"... |
cs | [
{
"content": "Autoregressive models in neural machine translation (NMT) exhibit well-documented declines in sequence generation accuracy toward later positions, yet the extent to which this degradation stems from error propagation versus other linguistic or architectural factors remains contentious. While exist... | cs | {
"ground_truth": "The reference solution proposes a methodological framework to investigate the interplay between linguistic branching directionality and autoregressive sequence generation accuracy in neural machine translation (NMT) and related tasks. The approach systematically evaluates whether accuracy degradati... | {
"arxiv_id": "1809.00120",
"filename": "1809.00120_merged_body.json",
"index": 5,
"merged_rubrics": [
{
"description": "Specifies the use of teacher forcing during inference, where ground-truth tokens replace generated tokens, to isolate error propagation effects from other factors.",
"id": 1,
... |
physics | [
{
"content": "Alzheimer’s disease (AD) is marked by amyloid-β (Aβ) plaque accumulation, lipid dysregulation, oxidative stress, and inflammation, yet the molecular mechanisms linking lipid abnormalities to amyloid aggregation and neurodegeneration remain poorly understood. Current imaging techniques such as fluo... | physics | {
"ground_truth": "The study introduces a non-invasive, chemically specific imaging framework to interrogate molecular interactions in Alzheimer’s disease (AD) brain tissues by integrating label-free spontaneous Raman micro-spectroscopy with hyperspectral image factorization (Q-US/PS-NMF). This approach overcomes lim... | {
"arxiv_id": "1803.01201",
"filename": "1803.01201_merged_body.json",
"index": 6,
"merged_rubrics": [
{
"description": "The solution uses a label-free spectroscopic imaging technique (e.g., spontaneous Raman, SRS, CARS) that does not require exogenous contrast agents.",
"id": 1,
"weight":... |
cs | [
{
"content": "In dynamic bipedal locomotion control for underactuated legged systems, adaptive step timing remains a critical challenge due to computational and theoretical barriers in unifying stability guarantees with real-time adjustability. While LIPM and DCM frameworks have enabled robust gait generation t... | cs | {
"ground_truth": "The proposed solution addresses adaptive step timing in dynamic bipedal locomotion for underactuated systems by integrating a Linear Inverted Pendulum Model (LIPM)-based viability kernel analysis with hierarchical inverse dynamics control. The method operates through two tightly coupled modules: a ... | {
"arxiv_id": "1704.01271",
"filename": "1704.01271_merged_body.json",
"index": 7,
"merged_rubrics": [
{
"description": "The solution explicitly defines a mathematical relationship between adaptive step timing and DCM stability bounds (e.g., inequality or update law linking step duration to DCM dynami... |
econ | [
{
"content": "Despite the expansion of financial aid programs, socioeconomically disadvantaged students in tertiary education continue to exhibit low graduation rates and delayed degree completion, exacerbated by persistent informational barriers regarding educational returns, program eligibility, and career pa... | econ | {
"ground_truth": "The study employs a randomized controlled trial (RCT) to evaluate the causal impact of targeted information interventions on academic and career outcomes among socioeconomically disadvantaged tertiary students in Italy. The intervention design addresses information asymmetries by isolating two dist... | {
"arxiv_id": "2507.02560",
"filename": "2507.02560_merged_body.json",
"index": 8,
"merged_rubrics": [
{
"description": "The solution employs a randomized controlled trial (RCT) to establish causal inference.",
"id": 1,
"weight": 1
},
{
"description": "The proposal includes a... |
cs | [
{
"content": "Autonomous robotic systems operating in safety-critical environments increasingly rely on deep learning (DL) models for perception-driven planning, yet the deterministic nature of these models and their limited explicit handling of epistemic uncertainty hinder reliable risk assessment, particularl... | cs | {
"ground_truth": "The reference solution addresses the challenge of integrating epistemic uncertainty estimation into perception-driven planning for autonomous robotic navigation in safety-critical environments. The method employs a two-stage pipeline to bridge the gap between semantic segmentation reliability and r... | {
"arxiv_id": "1910.00101",
"filename": "1910.00101_merged_body.json",
"index": 9,
"merged_rubrics": [
{
"description": "The solution includes a composite cost function that explicitly incorporates uncertainty estimates derived from perception outputs.",
"id": 1,
"weight": 1
},
{
... |
cs | [
{
"content": "Multi-focus image fusion (MIF) in computer vision seeks to integrate multiple images with varying focal depths into a single sharp composite, yet existing methods fail to address critical challenges in preserving boundary integrity and contextual coherence for complex 3D scenes. Traditional transf... | cs | {
"ground_truth": "The proposed solution addresses multi-focus image fusion by introducing a boundary-aware deep learning framework designed to preserve focus discontinuity boundaries (FDB) and contextual coherence in complex 3D scenes. The method employs a two-stage architecture: an initial 2-channel CNN (Res56 back... | {
"arxiv_id": "1904.00198",
"filename": "1904.00198_merged_body.json",
"index": 10,
"merged_rubrics": [
{
"description": "The solution uses dual independent networks to separately refine regions near and far from the focus discontinuity boundary (FDB).",
"id": 1,
"weight": 1
},
{
... |
physics | [
{
"content": "In the sub-field of statistical physics within fluid dynamics, turbulence theory in three-dimensional Navier-Stokes systems faces unresolved challenges in understanding how Fourier-mode decimation—specifically fractal and homogeneous reductions of spectral activity—affects Lagrangian velocity fluc... | physics | {
"ground_truth": "The study investigates the impact of Fourier-mode decimation on Lagrangian intermittency and scaling exponents in three-dimensional Navier-Stokes turbulence by employing a symmetry-preserving numerical framework. A generalized Galerkin projector $\\mathcal{P}$ is applied to enforce mode reduction i... | {
"arxiv_id": "1701.00351",
"filename": "1701.00351_merged_body.json",
"index": 11,
"merged_rubrics": [
{
"description": "The solution employs a symmetry-preserving Galerkin projector that maintains Hermitian symmetry for real-valued velocity fields.",
"id": 1,
"weight": 1
},
{
... |
econ | [
{
"content": "Despite global progress in reducing under-5 child mortality, D-8 countries—particularly in Sub-Saharan Africa and South Asia—continue to face persistently high neonatal and infant mortality rates, with income inequality emerging as a critical yet underexplored socioeconomic determinant. While exis... | econ | {
"ground_truth": "The study investigates the influence of education investment and economic stability on under-five child mortality in D-8 countries, prioritizing Bangladesh, Egypt, Indonesia, Iran, Malaysia, Nigeria, Pakistan, and Turkey. A panel data analysis framework is employed to address non-stationarity and u... | {
"arxiv_id": "2512.23110",
"filename": "2512.23110_merged_body.json",
"index": 12,
"merged_rubrics": [
{
"description": "The solution identifies education investment and economic stability (GNI, inflation) as primary drivers of mortality reduction, contrasting with conventional focus on health spendi... |
cs | [
{
"content": "Full-duplex radio technology, a promising approach to double spectral efficiency in LTE Evolution and 5G systems by enabling simultaneous transmission and reception on the same frequency, remains constrained by self-interference (SI) stemming from RF transceiver imperfections such as power amplifi... | cs | {
"ground_truth": "The proposed solution addresses self-interference caused by power amplifier nonlinearities in full-duplex LTE systems through a hybrid framework combining pre-calibration-based nonlinear mitigation and linear digital cancellation. The method leverages a reversed polynomial model to pre-distort the ... | {
"arxiv_id": "1607.01912",
"filename": "1607.01912_merged_body.json",
"index": 13,
"merged_rubrics": [
{
"description": "The solution includes a pre-calibration or training phase to characterize the distortion characteristics of the power amplifier.",
"id": 1,
"weight": 1
},
{
... |
econ | [
{
"content": "In the field of econometrics and financial economics, particularly within term structure modeling and yield curve dynamics, existing frameworks such as the Nelson-Siegel (NS) and its dynamic extension (DNS) remain limited in their ability to characterize the full conditional distribution of yields... | econ | {
"ground_truth": "The proposed time-varying parameter quantile Nelson-Siegel (TVP-QR-NS) model extends the classical Nelson-Siegel framework by integrating quantile regression into a state-space specification, enabling the estimation of yield curve factors at arbitrary quantiles τ of the conditional distribution. Th... | {
"arxiv_id": "2401.09874",
"filename": "2401.09874_merged_body.json",
"index": 14,
"merged_rubrics": [
{
"description": "The solution integrates quantile regression into a Nelson-Siegel factor model to estimate quantile-specific yield curve factors.",
"id": 1,
"weight": 1
},
{
... |
End of preview. Expand in Data Studio
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
This repository contains the dataset presented in PaperGym: Rubric-Centered Evolution for Research-Plan Generation.
Project page: https://zju-real.github.io/PaperGym
- Downloads last month
- 38