MERIT: Matching Expertise via Rubric-Informed Training for Reviewer Assignment
Paper • 2605.27865 • Published • 1
Error code: DatasetGenerationCastError
Exception: DatasetGenerationCastError
Message: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 4 new columns ({'anchor', 'type', 'negative', 'positive'}) and 2 missing columns ({'author', 'paper'}).
This happened while the json dataset builder was generating data using
hf://datasets/Luli3220/MERIT/data/stage2_retriever/raw/train_pc.json (at revision d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d), [/tmp/hf-datasets-cache/medium/datasets/71543835784039-config-parquet-and-info-Luli3220-MERIT-c3155f50/hub/datasets--Luli3220--MERIT/snapshots/d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage1_assessor/raw/train.json (origin=hf://datasets/Luli3220/MERIT@d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage1_assessor/raw/train.json), /tmp/hf-datasets-cache/medium/datasets/71543835784039-config-parquet-and-info-Luli3220-MERIT-c3155f50/hub/datasets--Luli3220--MERIT/snapshots/d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage2_retriever/raw/train_pc.json (origin=hf://datasets/Luli3220/MERIT@d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage2_retriever/raw/train_pc.json), /tmp/hf-datasets-cache/medium/datasets/71543835784039-config-parquet-and-info-Luli3220-MERIT-c3155f50/hub/datasets--Luli3220--MERIT/snapshots/d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage2_retriever/raw/train_rc.json (origin=hf://datasets/Luli3220/MERIT@d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage2_retriever/raw/train_rc.json)], ['hf://datasets/Luli3220/MERIT@d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage1_assessor/raw/train.json', 'hf://datasets/Luli3220/MERIT@d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage2_retriever/raw/train_pc.json', 'hf://datasets/Luli3220/MERIT@d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage2_retriever/raw/train_rc.json']
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1800, in _prepare_split_single
writer.write_table(table)
File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 765, in write_table
self._write_table(pa_table, writer_batch_size=writer_batch_size)
File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 773, in _write_table
pa_table = table_cast(pa_table, self._schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2321, in table_cast
return cast_table_to_schema(table, schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2249, in cast_table_to_schema
raise CastError(
datasets.table.CastError: Couldn't cast
anchor: struct<paper_title: string, abstract: string>
child 0, paper_title: string
child 1, abstract: string
positive: struct<score: int64, papers: list<item: struct<title: string, abstract: string>>>
child 0, score: int64
child 1, papers: list<item: struct<title: string, abstract: string>>
child 0, item: struct<title: string, abstract: string>
child 0, title: string
child 1, abstract: string
negative: struct<score: int64, papers: list<item: struct<title: string, abstract: string>>>
child 0, score: int64
child 1, papers: list<item: struct<title: string, abstract: string>>
child 0, item: struct<title: string, abstract: string>
child 0, title: string
child 1, abstract: string
type: string
to
{'paper': {'title': Value('string'), 'abstract': Value('string'), 'introduction': Value('string')}, 'author': {'papers': List({'title': Value('string'), 'abstract': Value('string')})}}
because column names don't match
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1348, in compute_config_parquet_and_info_response
parquet_operations = convert_to_parquet(builder)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 980, in convert_to_parquet
builder.download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 882, in download_and_prepare
self._download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 943, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1646, in _prepare_split
for job_id, done, content in self._prepare_split_single(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1802, in _prepare_split_single
raise DatasetGenerationCastError.from_cast_error(
datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 4 new columns ({'anchor', 'type', 'negative', 'positive'}) and 2 missing columns ({'author', 'paper'}).
This happened while the json dataset builder was generating data using
hf://datasets/Luli3220/MERIT/data/stage2_retriever/raw/train_pc.json (at revision d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d), [/tmp/hf-datasets-cache/medium/datasets/71543835784039-config-parquet-and-info-Luli3220-MERIT-c3155f50/hub/datasets--Luli3220--MERIT/snapshots/d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage1_assessor/raw/train.json (origin=hf://datasets/Luli3220/MERIT@d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage1_assessor/raw/train.json), /tmp/hf-datasets-cache/medium/datasets/71543835784039-config-parquet-and-info-Luli3220-MERIT-c3155f50/hub/datasets--Luli3220--MERIT/snapshots/d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage2_retriever/raw/train_pc.json (origin=hf://datasets/Luli3220/MERIT@d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage2_retriever/raw/train_pc.json), /tmp/hf-datasets-cache/medium/datasets/71543835784039-config-parquet-and-info-Luli3220-MERIT-c3155f50/hub/datasets--Luli3220--MERIT/snapshots/d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage2_retriever/raw/train_rc.json (origin=hf://datasets/Luli3220/MERIT@d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage2_retriever/raw/train_rc.json)], ['hf://datasets/Luli3220/MERIT@d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage1_assessor/raw/train.json', 'hf://datasets/Luli3220/MERIT@d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage2_retriever/raw/train_pc.json', 'hf://datasets/Luli3220/MERIT@d3ba7fe7eb915e546b0a3d396bfc5e03ac75fa3d/data/stage2_retriever/raw/train_rc.json']
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
paper dict | author dict |
|---|---|
{
"title": "GEAR: A Simple GENERATE, EMBED, AVERAGE AND RANK Approach for Unsupervised Reverse Dictionary",
"abstract": "Reverse Dictionary (RD) is the task of obtaining the most relevant word or set of words given a textual description or dictionary definition. Effective RD methods have applications in accessibili... | {
"papers": [
{
"title": "Human-like conceptual representations emerge from language prediction",
"abstract": "People acquire concepts through rich physical and social experiences and use them to understand the world. In contrast, large language models (LLMs), trained exclusively through next-token pr... |
{
"title": "GEAR: A Simple GENERATE, EMBED, AVERAGE AND RANK Approach for Unsupervised Reverse Dictionary",
"abstract": "Reverse Dictionary (RD) is the task of obtaining the most relevant word or set of words given a textual description or dictionary definition. Effective RD methods have applications in accessibili... | {
"papers": [
{
"title": "Human-like conceptual representations emerge from language prediction",
"abstract": "People acquire concepts through rich physical and social experiences and use them to understand the world. In contrast, large language models (LLMs), trained exclusively through next-token pr... |
{
"title": "GEAR: A Simple GENERATE, EMBED, AVERAGE AND RANK Approach for Unsupervised Reverse Dictionary",
"abstract": "Reverse Dictionary (RD) is the task of obtaining the most relevant word or set of words given a textual description or dictionary definition. Effective RD methods have applications in accessibili... | {
"papers": [
{
"title": "Static Word Embeddings for Sentence Semantic Representation",
"abstract": "We propose new static word embeddings optimised for sentence semantic representation. We first extract word embeddings from a pretrained Sentence Transformer, and improve them with sentence-level princ... |
{
"title": "GEAR: A Simple GENERATE, EMBED, AVERAGE AND RANK Approach for Unsupervised Reverse Dictionary",
"abstract": "Reverse Dictionary (RD) is the task of obtaining the most relevant word or set of words given a textual description or dictionary definition. Effective RD methods have applications in accessibili... | {
"papers": [
{
"title": "Unveiling Key Aspects of Fine-Tuning in Sentence Embeddings: A Representation Rank Analysis",
"abstract": "The latest advancements in unsupervised learning of sentence embeddings predominantly involve employing contrastive learning-based (CL-based) fine-tuning over pre-traine... |
{
"title": "GEAR: A Simple GENERATE, EMBED, AVERAGE AND RANK Approach for Unsupervised Reverse Dictionary",
"abstract": "Reverse Dictionary (RD) is the task of obtaining the most relevant word or set of words given a textual description or dictionary definition. Effective RD methods have applications in accessibili... | {
"papers": [
{
"title": "Enhancing Modern Supervised Word Sense Disambiguation Models by Semantic Lexical Resources",
"abstract": "Supervised models for Word Sense Disambiguation (WSD) currently yield to state-of-the-art results in the most popular benchmarks. Despite the recent introduction of Word ... |
{
"title": "VidPanos: Generative Panoramic Videos from Casual Panning Videos",
"abstract": "Panoramic image stitching provides a unified, wide-angle view of a scene that extends beyond the camera’s field of view. Stitching frames of a panning video into a panoramic photograph is a well-understood problem for statio... | {
"papers": [
{
"title": "PanoWan: Lifting Diffusion Video Generation Models to 360 • with Latitude/Longitude-aware Mechanisms",
"abstract": "Panoramic video generation enables immersive 360°content creation, valuable in applications that demand scene-consistent world exploration. However, existing pa... |
{
"title": "VidPanos: Generative Panoramic Videos from Casual Panning Videos",
"abstract": "Panoramic image stitching provides a unified, wide-angle view of a scene that extends beyond the camera’s field of view. Stitching frames of a panning video into a panoramic photograph is a well-understood problem for statio... | {
"papers": [
{
"title": "FB-4D: Spatial-Temporal Coherent Dynamic 3D Content Generation with Feature Banks",
"abstract": "With the rapid advancements in diffusion models and 3D generation techniques, dynamic 3D content generation has become a crucial research area. However, achieving highfidelity 4D ... |
{
"title": "VidPanos: Generative Panoramic Videos from Casual Panning Videos",
"abstract": "Panoramic image stitching provides a unified, wide-angle view of a scene that extends beyond the camera’s field of view. Stitching frames of a panning video into a panoramic photograph is a well-understood problem for statio... | {
"papers": [
{
"title": "Imagine360: Immersive 360 Video Generation from Perspective Anchor",
"abstract": "https://ys-imtech.github.io/projects/Imagine360 Figure 1. Overview of Imagine360. Imagine360 lifts standard perspective video into 360 • video, enabling dynamic scene experience from full 360 de... |
{
"title": "VidPanos: Generative Panoramic Videos from Casual Panning Videos",
"abstract": "Panoramic image stitching provides a unified, wide-angle view of a scene that extends beyond the camera’s field of view. Stitching frames of a panning video into a panoramic photograph is a well-understood problem for statio... | {
"papers": [
{
"title": "StereoCrafter: Diffusion-based Generation of Long and High-fidelity Stereoscopic 3D from Monocular Videos",
"abstract": "Stereo Crafter Stereo Crafter Input Video (left view) Generated Video (right view) Display on Different Devices Figure 1. We propose a framework to convert... |
{
"title": "VidPanos: Generative Panoramic Videos from Casual Panning Videos",
"abstract": "Panoramic image stitching provides a unified, wide-angle view of a scene that extends beyond the camera’s field of view. Stitching frames of a panning video into a panoramic photograph is a well-understood problem for statio... | {
"papers": [
{
"title": "EvoWorld: Evolving Panoramic World Generation with Explicit 3D Memory",
"abstract": "Humans possess a remarkable ability to mentally explore and replay 3D environments they have previously experienced. Inspired by this mental process, we present EvoWorld: a world model that b... |
{
"title": "On the Shelf Life of Fine-Tuned LLM Judges: Future Proofing, Backward Compatibility, and Question Generalization",
"abstract": "The LLM-as-a-judge paradigm is widely used in both evaluating free-text model responses and reward modeling for model alignment and finetuning. Recently, finetuning judges with... | {
"papers": [
{
"title": "A TRANSFER LEARNING FRAMEWORK FOR WEAK TO STRONG GENERALIZATION",
"abstract": "Modern large language model (LLM) alignment techniques rely on human feedback, but it is unclear whether these techniques fundamentally limit the capabilities of aligned LLMs. In particular, it is ... |
{
"title": "On the Shelf Life of Fine-Tuned LLM Judges: Future Proofing, Backward Compatibility, and Question Generalization",
"abstract": "The LLM-as-a-judge paradigm is widely used in both evaluating free-text model responses and reward modeling for model alignment and finetuning. Recently, finetuning judges with... | {
"papers": [
{
"title": "Toward Generalizable Evaluation in the LLM Era: A Survey Beyond Benchmarks",
"abstract": "Large Language Models (LLMs) are advancing at an amazing speed and have become indispensable across academia, industry, and daily applications. To keep pace with the status quo, this sur... |
{
"title": "On the Shelf Life of Fine-Tuned LLM Judges: Future Proofing, Backward Compatibility, and Question Generalization",
"abstract": "The LLM-as-a-judge paradigm is widely used in both evaluating free-text model responses and reward modeling for model alignment and finetuning. Recently, finetuning judges with... | {
"papers": [
{
"title": "Beyond Single-Point Judgment: Distribution Alignment for LLM-as-a-Judge",
"abstract": "LLMs have emerged as powerful evaluators in the LLM-as-a-Judge paradigm, offering significant efficiency and flexibility compared to human judgments. However, previous methods primarily rel... |
{
"title": "On the Shelf Life of Fine-Tuned LLM Judges: Future Proofing, Backward Compatibility, and Question Generalization",
"abstract": "The LLM-as-a-judge paradigm is widely used in both evaluating free-text model responses and reward modeling for model alignment and finetuning. Recently, finetuning judges with... | {
"papers": [
{
"title": "Beyond Single-Point Judgment: Distribution Alignment for LLM-as-a-Judge",
"abstract": "LLMs have emerged as powerful evaluators in the LLM-as-a-Judge paradigm, offering significant efficiency and flexibility compared to human judgments. However, previous methods primarily rel... |
{
"title": "On the Shelf Life of Fine-Tuned LLM Judges: Future Proofing, Backward Compatibility, and Question Generalization",
"abstract": "The LLM-as-a-judge paradigm is widely used in both evaluating free-text model responses and reward modeling for model alignment and finetuning. Recently, finetuning judges with... | {
"papers": [
{
"title": "Training on the Test Task Confounds Evaluation and Emergence",
"abstract": "We study a fundamental problem in the evaluation of large language models that we call training on the test task. Unlike wrongful practices like training on the test data, leakage, or data contaminati... |
{
"title": "Unified Molecule Pre-training with Flexible 2D and 3D Modalities: Single and Paired Modality Integration",
"abstract": "Molecular representation learning plays a crucial role in advancing applications such as drug discovery and material design. Existing work leverages 2D and 3D modalities of molecular i... | {
"papers": [
{
"title": "3D-MOLT5: LEVERAGING DISCRETE STRUCTURAL IN-FORMATION FOR MOLECULE-TEXT MODELING",
"abstract": "The integration of molecular and natural language representations has emerged as a focal point in molecular science, with recent advancements in Language Models (LMs) demonstrating... |
{
"title": "Unified Molecule Pre-training with Flexible 2D and 3D Modalities: Single and Paired Modality Integration",
"abstract": "Molecular representation learning plays a crucial role in advancing applications such as drug discovery and material design. Existing work leverages 2D and 3D modalities of molecular i... | {
"papers": [
{
"title": "Beyond Atoms: Enhancing Molecular Pretrained Representations with 3D Space Modeling",
"abstract": "Molecular pretrained representations (MPR) has emerged as a powerful approach for addressing the challenge of limited supervised data in applications such as drug discovery and ... |
{
"title": "Unified Molecule Pre-training with Flexible 2D and 3D Modalities: Single and Paired Modality Integration",
"abstract": "Molecular representation learning plays a crucial role in advancing applications such as drug discovery and material design. Existing work leverages 2D and 3D modalities of molecular i... | {
"papers": [
{
"title": "UNIGEM: A UNIFIED APPROACH TO GENERATION AND PROPERTY PREDICTION FOR MOLECULES",
"abstract": "Molecular generation and molecular property prediction are both crucial for drug discovery, but they are often developed independently. Inspired by recent studies, which demonstrate ... |
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
This dataset is released as part of our paper: MERIT: Matching Expertise via Rubric-Informed Training for Reviewer Assignment.
For details on data construction, training, and evaluation, please refer to the paper.