Dataset Viewer
Auto-converted to Parquet Duplicate
instruction
string
queries
dict
survey_queries
dict
search_results
dict
search_success
bool
query_responses
dict
ea_criteria
string
raw_ea_criteria
dict
llm_criteria
string
ea_criteria_scored_raw
list
ea_criteria_scored
list
llm_criteria_scored_raw
list
llm_criteria_scored
list
ea_full_criteria
string
merged_source_indicators
list
"Generate a comprehensive technical report analyzing the computational scaling properties of pretrai(...TRUNCATED)
{"how to communicate technical findings to a broader audience":"Discover techniques for clearly conv(...TRUNCATED)
{"survey papers on Feed-Forward Network Dynamics arxiv OR biorxiv pdf":"Exploring the behaviors of f(...TRUNCATED)
{"how to communicate technical findings to a broader audience":{"all_web_results":[{"content":"Produ(...TRUNCATED)
true
{"how to communicate technical findings to a broader audience":"### Combined Advice List:\n\n1. **Kn(...TRUNCATED)
"1.The response should include a clear and descriptive title reflecting the focus on computational s(...TRUNCATED)
{"filter_raw_response":{"1":"The response would need a clear and descriptive title that reflects the(...TRUNCATED)
"1. The response should generate a comprehensive technical report.\n2. The response should analyze t(...TRUNCATED)
["To evaluate the relevance of the aspect relating to the requirement of a clear and descriptive tit(...TRUNCATED)
[7.0,9.0,7.0,9.0,9.0,6.0,7.0,3.0,6.0,10.0,9.0,10.0,10.0,2.0,9.0,10.0,10.0,2.0,6.0,6.0,9.0,9.0,3.0,9.(...TRUNCATED)
["To evaluate the relevance of the aspect \"The response should generate a comprehensive technical r(...TRUNCATED)
[ 10, 10, 9, 10, 9, 4, 10, 10, 9, 9, 3, 6, 9, 9, 10, 9, 9, 9, 9, 6 ]
"1. The response should generate a comprehensive technical report.\n2. The response should analyze t(...TRUNCATED)
["llm","llm","llm","llm","llm","llm","ea","ea","ea","ea","ea","llm","llm","llm","llm","llm","llm","l(...TRUNCATED)
README.md exists but content is empty.
Downloads last month
29