metadata
dataset_info:
- config_name: 20260223_openai_gpt-5.2_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: string
- name: classification
dtype: string
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: reasoning_content
dtype: 'null'
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 1976867
num_examples: 4775
download_size: 167829
dataset_size: 1976867
- config_name: 20260223_openrouter_claude-opus-4-6_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: 'null'
- name: classification
dtype: string
- name: input_tokens
dtype: 'null'
- name: output_tokens
dtype: 'null'
- name: reasoning_content
dtype: 'null'
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 690
num_examples: 1
download_size: 12721
dataset_size: 690
- config_name: 20260223_openrouter_moonshotai_Kimi-K2.5_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: string
- name: classification
dtype: string
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: reasoning_content
dtype: string
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 38255
num_examples: 10
download_size: 36692
dataset_size: 38255
- config_name: 20260223_openrouter_zai-org_GLM-5_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: string
- name: classification
dtype: string
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: reasoning_content
dtype: string
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 5866661
num_examples: 680
download_size: 2193043
dataset_size: 5866661
- config_name: 20260223_vertex_gemini-3-pro-preview_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: string
- name: classification
dtype: string
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: reasoning_content
dtype: string
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 3661513
num_examples: 4775
download_size: 809959
dataset_size: 3661513
- config_name: 20260223_vllm_Qwen_Qwen3-8B_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: string
- name: classification
dtype: string
- name: input_tokens
dtype: 'null'
- name: output_tokens
dtype: 'null'
- name: reasoning_content
dtype: 'null'
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 1892663
num_examples: 4775
download_size: 157276
dataset_size: 1892663
- config_name: 20260223_vllm_Qwen_Qwen3.5-397B-A17B_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: string
- name: classification
dtype: string
- name: input_tokens
dtype: 'null'
- name: output_tokens
dtype: 'null'
- name: reasoning_content
dtype: string
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 59025499
num_examples: 4775
download_size: 18959923
dataset_size: 59025499
- config_name: 20260223_vllm_allenai_Olmo-3.1-32B-Think_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: string
- name: classification
dtype: string
- name: input_tokens
dtype: 'null'
- name: output_tokens
dtype: 'null'
- name: reasoning_content
dtype: string
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 17644903
num_examples: 4775
download_size: 5941064
dataset_size: 17644903
- config_name: 20260223_vllm_google_gemma-3-27b-it_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: string
- name: classification
dtype: string
- name: input_tokens
dtype: 'null'
- name: output_tokens
dtype: 'null'
- name: reasoning_content
dtype: 'null'
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 1978412
num_examples: 4775
download_size: 159648
dataset_size: 1978412
- config_name: 20260223_vllm_meta-llama_Llama-3.1-8B-Instruct_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: string
- name: classification
dtype: string
- name: input_tokens
dtype: 'null'
- name: output_tokens
dtype: 'null'
- name: reasoning_content
dtype: 'null'
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 1989391
num_examples: 4775
download_size: 159660
dataset_size: 1989391
- config_name: 20260223_vllm_meta-llama_Llama-3.3-70B-Instruct_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: string
- name: classification
dtype: string
- name: input_tokens
dtype: 'null'
- name: output_tokens
dtype: 'null'
- name: reasoning_content
dtype: 'null'
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 2005153
num_examples: 4775
download_size: 158664
dataset_size: 2005153
- config_name: 20260223_vllm_mistralai_Ministral-3-14B-Reasoning-2512_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: string
- name: classification
dtype: string
- name: input_tokens
dtype: 'null'
- name: output_tokens
dtype: 'null'
- name: reasoning_content
dtype: 'null'
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 2103404
num_examples: 4775
download_size: 159211
dataset_size: 2103404
- config_name: 20260223_vllm_nvidia_NVIDIA-Nemotron-3-Nano-30B-A3B-BF16_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: string
- name: classification
dtype: string
- name: input_tokens
dtype: 'null'
- name: output_tokens
dtype: 'null'
- name: reasoning_content
dtype: string
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 8558357
num_examples: 4775
download_size: 1939743
dataset_size: 8558357
- config_name: 20260223_vllm_openai_gpt-oss-120b_A-1_noweb
features:
- name: paper_id
dtype: string
- name: citation_count
dtype: int64
- name: is_retracted
dtype: bool
- name: question_id
dtype: string
- name: question_text
dtype: string
- name: response_text
dtype: string
- name: response_json
dtype: string
- name: classification
dtype: string
- name: input_tokens
dtype: 'null'
- name: output_tokens
dtype: 'null'
- name: reasoning_content
dtype: 'null'
- name: repetition
dtype: int64
- name: model
dtype: string
- name: provider
dtype: string
- name: web_search_context
dtype: 'null'
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: temperature
dtype: 'null'
- name: finish_reason
dtype: 'null'
- name: elapsed_seconds
dtype: 'null'
- name: run_id
dtype: 'null'
- name: timestamp
dtype: 'null'
splits:
- name: train
num_bytes: 5107186
num_examples: 4775
download_size: 989006
dataset_size: 5107186
- config_name: 20260303_vllm_Qwen_Qwen3.5-9B-Base_A-1_noweb
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
dtype: 'null'
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: string
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: top_k
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: mean_logprob
dtype: float64
- name: min_logprob
dtype: float64
- name: mean_entropy
dtype: float64
- name: max_entropy
dtype: float64
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: 'null'
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 1135101722
num_examples: 4775
download_size: 1078988432
dataset_size: 1135101722
- config_name: 20260303_vllm_Qwen_Qwen3.5-9B_A-1_noweb
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
dtype: 'null'
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: string
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: top_k
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: mean_logprob
dtype: float64
- name: min_logprob
dtype: float64
- name: mean_entropy
dtype: float64
- name: max_entropy
dtype: float64
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: 'null'
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 908358000
num_examples: 4775
download_size: 858165240
dataset_size: 908358000
- config_name: 20260303_vllm_Qwen_Qwen3.5-9B_A-1_web
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
list:
- name: title
dtype: string
- name: snippet
dtype: string
- name: link
dtype: string
- name: full_content
dtype: string
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: string
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: top_k
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: mean_logprob
dtype: float64
- name: min_logprob
dtype: float64
- name: mean_entropy
dtype: float64
- name: max_entropy
dtype: float64
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: float64
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 896110782
num_examples: 4775
download_size: 846756024
dataset_size: 896110782
- config_name: 20260304_vllm_Qwen_Qwen3-8B_A-1_web
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
list:
- name: title
dtype: string
- name: snippet
dtype: string
- name: link
dtype: string
- name: full_content
dtype: string
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: string
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: top_k
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: mean_logprob
dtype: float64
- name: min_logprob
dtype: float64
- name: mean_entropy
dtype: float64
- name: max_entropy
dtype: float64
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: 'null'
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 904345106
num_examples: 4775
download_size: 854776361
dataset_size: 904345106
- config_name: 20260304_vllm_Qwen_Qwen3.5-9B-Base_A-1_web
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
list:
- name: title
dtype: string
- name: snippet
dtype: string
- name: link
dtype: string
- name: full_content
dtype: string
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: string
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: top_k
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: mean_logprob
dtype: float64
- name: min_logprob
dtype: float64
- name: mean_entropy
dtype: float64
- name: max_entropy
dtype: float64
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: 'null'
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 286976554
num_examples: 4775
download_size: 269833222
dataset_size: 286976554
- config_name: 20260304_vllm_google_gemma-3-27b-it_A-1_web
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
list:
- name: title
dtype: string
- name: snippet
dtype: string
- name: link
dtype: string
- name: full_content
dtype: string
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: 'null'
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: top_k
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: mean_logprob
dtype: float64
- name: min_logprob
dtype: float64
- name: mean_entropy
dtype: float64
- name: max_entropy
dtype: float64
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: 'null'
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 37763540
num_examples: 4775
download_size: 35531179
dataset_size: 37763540
- config_name: 20260304_vllm_meta-llama_Llama-3.1-8B-Instruct_A-1_web
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
list:
- name: title
dtype: string
- name: snippet
dtype: string
- name: link
dtype: string
- name: full_content
dtype: string
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: 'null'
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: top_k
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: mean_logprob
dtype: float64
- name: min_logprob
dtype: float64
- name: mean_entropy
dtype: float64
- name: max_entropy
dtype: float64
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: 'null'
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 34603072
num_examples: 4775
download_size: 32431511
dataset_size: 34603072
- config_name: 20260304_vllm_meta-llama_Llama-3.3-70B-Instruct_A-1_web
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
list:
- name: title
dtype: string
- name: snippet
dtype: string
- name: link
dtype: string
- name: full_content
dtype: string
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: 'null'
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: top_k
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: mean_logprob
dtype: float64
- name: min_logprob
dtype: float64
- name: mean_entropy
dtype: float64
- name: max_entropy
dtype: float64
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: 'null'
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 34491417
num_examples: 4775
download_size: 32349606
dataset_size: 34491417
- config_name: 20260304_vllm_mistralai_Ministral-3-14B-Reasoning-2512_A-1_web
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
list:
- name: title
dtype: string
- name: snippet
dtype: string
- name: link
dtype: string
- name: full_content
dtype: string
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: 'null'
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: top_k
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: mean_logprob
dtype: float64
- name: min_logprob
dtype: float64
- name: mean_entropy
dtype: float64
- name: max_entropy
dtype: float64
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: 'null'
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 37601488
num_examples: 4775
download_size: 35276126
dataset_size: 37601488
- config_name: 20260304_vllm_nvidia_NVIDIA-Nemotron-3-Nano-30B-A3B-BF16_A-1_web
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
list:
- name: title
dtype: string
- name: snippet
dtype: string
- name: link
dtype: string
- name: full_content
dtype: string
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: string
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: top_k
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: mean_logprob
dtype: float64
- name: min_logprob
dtype: float64
- name: mean_entropy
dtype: float64
- name: max_entropy
dtype: float64
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: 'null'
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 272369661
num_examples: 4775
download_size: 258754243
dataset_size: 272369661
- config_name: 20260304_vllm_openai_gpt-oss-120b_A-1_web
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
list:
- name: title
dtype: string
- name: snippet
dtype: string
- name: link
dtype: string
- name: full_content
dtype: string
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: 'null'
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: top_k
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: mean_logprob
dtype: float64
- name: min_logprob
dtype: float64
- name: mean_entropy
dtype: float64
- name: max_entropy
dtype: float64
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: 'null'
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 77169283
num_examples: 4775
download_size: 73410616
dataset_size: 77169283
- config_name: 20260305_openai_gpt-5.2_A-1_web
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
list:
- name: title
dtype: string
- name: snippet
dtype: string
- name: link
dtype: string
- name: full_content
dtype: string
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: string
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: float64
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 4739031
num_examples: 884
download_size: 4402072
dataset_size: 4739031
- config_name: 20260305_vertexai_publishers_google_models_gemini-3.1-pro-preview_all_web
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
list:
- name: title
dtype: string
- name: snippet
dtype: string
- name: link
dtype: string
- name: full_content
dtype: string
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: 'null'
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: 'null'
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 30072531
num_examples: 4775
download_size: 27994039
dataset_size: 30072531
- config_name: 20260305_vllm_allenai_Olmo-3.1-32B-Think_A-1_web
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
list:
- name: title
dtype: string
- name: snippet
dtype: string
- name: link
dtype: string
- name: full_content
dtype: string
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: string
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: top_k
list:
- name: token
dtype: string
- name: logprob
dtype: float64
- name: mean_logprob
dtype: float64
- name: min_logprob
dtype: float64
- name: mean_entropy
dtype: float64
- name: max_entropy
dtype: float64
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: 'null'
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 582564789
num_examples: 4775
download_size: 547657023
dataset_size: 582564789
- config_name: >-
20260309_vertexai_publishers_google_models_gemini-3.1-pro-preview_all_noweb
features:
- name: run_id
dtype: string
- name: paper_id
dtype: string
- name: question_id
dtype: string
- name: repetition
dtype: int64
- name: timestamp
dtype: string
- name: is_retracted
dtype: bool
- name: citation_count
dtype: int64
- name: question_text
dtype: string
- name: web_search_context
dtype: 'null'
- name: response_text
dtype: string
- name: response_json
struct:
- name: paper_is_retracted
dtype: bool
- name: reasoning_content
dtype: 'null'
- name: classification
dtype: string
- name: finish_reason
dtype: string
- name: logprobs
dtype: 'null'
- name: mean_logprob
dtype: 'null'
- name: min_logprob
dtype: 'null'
- name: mean_entropy
dtype: 'null'
- name: max_entropy
dtype: 'null'
- name: input_tokens
dtype: int64
- name: output_tokens
dtype: int64
- name: elapsed_seconds
dtype: 'null'
- name: model
dtype: string
- name: provider
dtype: string
- name: temperature
dtype: float64
splits:
- name: train
num_bytes: 2336298
num_examples: 4775
download_size: 230956
dataset_size: 2336298
configs:
- config_name: 20260223_openai_gpt-5.2_A-1_noweb
data_files:
- split: train
path: 20260223_openai_gpt-5.2_A-1_noweb/train-*
- config_name: 20260223_openrouter_claude-opus-4-6_A-1_noweb
data_files:
- split: train
path: 20260223_openrouter_claude-opus-4-6_A-1_noweb/train-*
- config_name: 20260223_openrouter_moonshotai_Kimi-K2.5_A-1_noweb
data_files:
- split: train
path: 20260223_openrouter_moonshotai_Kimi-K2.5_A-1_noweb/train-*
- config_name: 20260223_openrouter_zai-org_GLM-5_A-1_noweb
data_files:
- split: train
path: 20260223_openrouter_zai-org_GLM-5_A-1_noweb/train-*
- config_name: 20260223_vertex_gemini-3-pro-preview_A-1_noweb
data_files:
- split: train
path: 20260223_vertex_gemini-3-pro-preview_A-1_noweb/train-*
- config_name: 20260223_vllm_Qwen_Qwen3-8B_A-1_noweb
data_files:
- split: train
path: 20260223_vllm_Qwen_Qwen3-8B_A-1_noweb/train-*
- config_name: 20260223_vllm_Qwen_Qwen3.5-397B-A17B_A-1_noweb
data_files:
- split: train
path: 20260223_vllm_Qwen_Qwen3.5-397B-A17B_A-1_noweb/train-*
- config_name: 20260223_vllm_allenai_Olmo-3.1-32B-Think_A-1_noweb
data_files:
- split: train
path: 20260223_vllm_allenai_Olmo-3.1-32B-Think_A-1_noweb/train-*
- config_name: 20260223_vllm_google_gemma-3-27b-it_A-1_noweb
data_files:
- split: train
path: 20260223_vllm_google_gemma-3-27b-it_A-1_noweb/train-*
- config_name: 20260223_vllm_meta-llama_Llama-3.1-8B-Instruct_A-1_noweb
data_files:
- split: train
path: 20260223_vllm_meta-llama_Llama-3.1-8B-Instruct_A-1_noweb/train-*
- config_name: 20260223_vllm_meta-llama_Llama-3.3-70B-Instruct_A-1_noweb
data_files:
- split: train
path: 20260223_vllm_meta-llama_Llama-3.3-70B-Instruct_A-1_noweb/train-*
- config_name: 20260223_vllm_mistralai_Ministral-3-14B-Reasoning-2512_A-1_noweb
data_files:
- split: train
path: >-
20260223_vllm_mistralai_Ministral-3-14B-Reasoning-2512_A-1_noweb/train-*
- config_name: 20260223_vllm_nvidia_NVIDIA-Nemotron-3-Nano-30B-A3B-BF16_A-1_noweb
data_files:
- split: train
path: >-
20260223_vllm_nvidia_NVIDIA-Nemotron-3-Nano-30B-A3B-BF16_A-1_noweb/train-*
- config_name: 20260223_vllm_openai_gpt-oss-120b_A-1_noweb
data_files:
- split: train
path: 20260223_vllm_openai_gpt-oss-120b_A-1_noweb/train-*
- config_name: 20260303_vllm_Qwen_Qwen3.5-9B-Base_A-1_noweb
data_files:
- split: train
path: 20260303_vllm_Qwen_Qwen3.5-9B-Base_A-1_noweb/train-*
- config_name: 20260303_vllm_Qwen_Qwen3.5-9B_A-1_noweb
data_files:
- split: train
path: 20260303_vllm_Qwen_Qwen3.5-9B_A-1_noweb/train-*
- config_name: 20260303_vllm_Qwen_Qwen3.5-9B_A-1_web
data_files:
- split: train
path: 20260303_vllm_Qwen_Qwen3.5-9B_A-1_web/train-*
- config_name: 20260304_vllm_Qwen_Qwen3-8B_A-1_web
data_files:
- split: train
path: 20260304_vllm_Qwen_Qwen3-8B_A-1_web/train-*
- config_name: 20260304_vllm_Qwen_Qwen3.5-9B-Base_A-1_web
data_files:
- split: train
path: 20260304_vllm_Qwen_Qwen3.5-9B-Base_A-1_web/train-*
- config_name: 20260304_vllm_google_gemma-3-27b-it_A-1_web
data_files:
- split: train
path: 20260304_vllm_google_gemma-3-27b-it_A-1_web/train-*
- config_name: 20260304_vllm_meta-llama_Llama-3.1-8B-Instruct_A-1_web
data_files:
- split: train
path: 20260304_vllm_meta-llama_Llama-3.1-8B-Instruct_A-1_web/train-*
- config_name: 20260304_vllm_meta-llama_Llama-3.3-70B-Instruct_A-1_web
data_files:
- split: train
path: 20260304_vllm_meta-llama_Llama-3.3-70B-Instruct_A-1_web/train-*
- config_name: 20260304_vllm_mistralai_Ministral-3-14B-Reasoning-2512_A-1_web
data_files:
- split: train
path: 20260304_vllm_mistralai_Ministral-3-14B-Reasoning-2512_A-1_web/train-*
- config_name: 20260304_vllm_nvidia_NVIDIA-Nemotron-3-Nano-30B-A3B-BF16_A-1_web
data_files:
- split: train
path: >-
20260304_vllm_nvidia_NVIDIA-Nemotron-3-Nano-30B-A3B-BF16_A-1_web/train-*
- config_name: 20260304_vllm_openai_gpt-oss-120b_A-1_web
data_files:
- split: train
path: 20260304_vllm_openai_gpt-oss-120b_A-1_web/train-*
- config_name: 20260305_openai_gpt-5.2_A-1_web
data_files:
- split: train
path: 20260305_openai_gpt-5.2_A-1_web/train-*
- config_name: 20260305_vertexai_publishers_google_models_gemini-3.1-pro-preview_all_web
data_files:
- split: train
path: >-
20260305_vertexai_publishers_google_models_gemini-3.1-pro-preview_all_web/train-*
- config_name: 20260305_vllm_allenai_Olmo-3.1-32B-Think_A-1_web
data_files:
- split: train
path: 20260305_vllm_allenai_Olmo-3.1-32B-Think_A-1_web/train-*
- config_name: >-
20260306_vertexai_publishers_google_models_gemini-3.1-pro-preview_all_noweb
data_files:
- split: train
path: >-
20260306_vertexai_publishers_google_models_gemini-3.1-pro-preview_all_noweb/train-*
- config_name: >-
20260309_vertexai_publishers_google_models_gemini-3.1-pro-preview_all_noweb
data_files:
- split: train
path: >-
20260309_vertexai_publishers_google_models_gemini-3.1-pro-preview_all_noweb/train-*
- config_name: evaluati_openrouter_Qwen_Qwen3-8B_A-1_noweb
data_files:
- split: train
path: evaluati_openrouter_Qwen_Qwen3-8B_A-1_noweb/train-*
- config_name: evaluati_openrouter_Qwen_Qwen3.5-397B-A17B_A-1_noweb
data_files:
- split: train
path: evaluati_openrouter_Qwen_Qwen3.5-397B-A17B_A-1_noweb/train-*
- config_name: evaluati_openrouter_allenai_Olmo-3.1-32B-Think_A-1_noweb
data_files:
- split: train
path: evaluati_openrouter_allenai_Olmo-3.1-32B-Think_A-1_noweb/train-*
- config_name: evaluati_openrouter_claude-opus-4-6_A-1_noweb
data_files:
- split: train
path: evaluati_openrouter_claude-opus-4-6_A-1_noweb/train-*
- config_name: evaluati_openrouter_gemini-3-pro-preview_A-1_noweb
data_files:
- split: train
path: evaluati_openrouter_gemini-3-pro-preview_A-1_noweb/train-*
- config_name: evaluati_openrouter_google_gemma-3-27b-it_A-1_noweb
data_files:
- split: train
path: evaluati_openrouter_google_gemma-3-27b-it_A-1_noweb/train-*
- config_name: evaluati_openrouter_meta-llama_Llama-3.1-8B-Instruct_A-1_noweb
data_files:
- split: train
path: evaluati_openrouter_meta-llama_Llama-3.1-8B-Instruct_A-1_noweb/train-*
- config_name: evaluati_openrouter_meta-llama_Llama-3.3-70B-Instruct_A-1_noweb
data_files:
- split: train
path: >-
evaluati_openrouter_meta-llama_Llama-3.3-70B-Instruct_A-1_noweb/train-*
- config_name: evaluati_openrouter_mistralai_Ministral-3-14B-Reasoning-2512_A-1_noweb
data_files:
- split: train
path: >-
evaluati_openrouter_mistralai_Ministral-3-14B-Reasoning-2512_A-1_noweb/train-*
- config_name: evaluati_openrouter_moonshotai_Kimi-K2.5_A-1_noweb
data_files:
- split: train
path: evaluati_openrouter_moonshotai_Kimi-K2.5_A-1_noweb/train-*
- config_name: evaluati_openrouter_nvidia_NVIDIA-Nemotron-3-Nano-30B-A3B-BF16_A-1_noweb
data_files:
- split: train
path: >-
evaluati_openrouter_nvidia_NVIDIA-Nemotron-3-Nano-30B-A3B-BF16_A-1_noweb/train-*
- config_name: evaluati_openrouter_openai_gpt-5.2_A-1_noweb
data_files:
- split: train
path: evaluati_openrouter_openai_gpt-5.2_A-1_noweb/train-*
- config_name: evaluati_openrouter_openai_gpt-oss-120b_A-1_noweb
data_files:
- split: train
path: evaluati_openrouter_openai_gpt-oss-120b_A-1_noweb/train-*
- config_name: evaluati_openrouter_zai-org_GLM-5_A-1_noweb
data_files:
- split: train
path: evaluati_openrouter_zai-org_GLM-5_A-1_noweb/train-*